Economic Integration in East Asia: Perspectives from Spatial and Neoclassical Economics

Economic Integration in East Asia Perspectives from Spatial and Neoclassical Economics Edited by Masahisa Fujita Pres...

Author: Masahisa Fujita | Satoru Kumagai | Kōji Nishikimi

9 downloads 1285 Views 8MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Economic Integration in East Asia Perspectives from Spatial and Neoclassical Economics

Edited by

Masahisa Fujita President, Research Institute of Economy, Trade and Industry, Professor, Konan University, Professor, Kyoto University, Japan

Satoru Kumagai Research Fellow, Institute of Developing Economies, Japan External Trade Organization

Koji Nishikimi Senior Research Fellow, Institute of Developing Economies, Japan External Trade Organization INSTITUTE OF DEVELOPING ECONOMIES (IDE), JETRO

Edward Elgar Cheltenham, UK • Northampton, MA, USA

© Institute of Developing Economies (IDE), JETRO 2008 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: 2008932910

ISBN 978 1 84720 912 2 Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall

Contents vii ix xi

List of contributors Preface List of abbreviations 1

Introduction Masahisa Fujita, Satoru Kumagai, and Koji Nishikimi

1

PART I BASIC VIEWPOINTS ON EAST ASIAN ECONOMIC INTEGRATION 2

3

Regional integration in East Asia: perspectives of spatial and neoclassical economics Masahisa Fujita and Nobuaki Hamaguchi

13

Specialization and agglomeration forces of economic integration Koji Nishikimi 43

PART II GENERAL VIEWS ON EAST ASIAN ECONOMIC INTEGRATION 4

Industrial clustering and MNE management in East Asia: recent progress and prospects for the Asian Triangle Akifumi Kuchiki

77

5

Evolution of institutions and policies for economic integration in East Asia: history and prospects Yoshihiro Otsuji and Kunihiko Shinoda 106

6

Economic integration in Asia: European perspectives Anthony J. Venables, L. Alan Winters and Linda Yueh

PART III

7

143

CASE STUDIES OF EAST ASIAN ECONOMIC INTEGRATION

Structure and determinants of intra-regional trade in East Asia Satoru Kumagai 173 v

vi

8

9 10 11

Contents

Location choices of Korean MNEs in East Asia: escaping the nutcracker Ho Yeon Kim

203

Location choices of Japanese MNEs in East Asia Toshitaka Gokan

249

Economic opening and industrial agglomeration in China Zhao Chen, Yu Jin and Ming Lu

276

The home market effect in ASEAN countries Ikumo Isono

316

Index

343

Contributors Zhao Chen, Professor, China Center for Economic Studies, Fudan University. Masahisa Fujita, President, Research Institute of Economy, Trade and Industry; Professor, Konan University; Professor, Kyoto University. Toshitaka Gokan, Research Fellow, Institute of Developing Economies, JETRO. Nobuaki Hamaguchi, Professor, Research Institute for Economics and Business Administration, Kobe University. Ikumo Isono, Research Fellow, Institute of Developing Economies, JETRO. Yu Jin, Professor, Department of Economics, Iowa State University. Ho Yeon Kim, Associate Professor, Sungkyunkwan University. Akifumi Kuchiki, Professor, Nihon University College of Bioresource Sciences. Satoru Kumagai, Research Fellow, Institute of Developing Economies, JETRO. Ming Lu, Professor, Employment & Social Security Research Center; China Center for Economic Studies, Fudan University. Koji Nishikimi, Senior Research Fellow, Institute of Developing Economies, JETRO. Yoshihiro Otsuji, Director-General, Chubu Bureau of Economy, Trade and Industry, Ministry of Economy, Trade and Industry. Kunihiko Shinoda, Director (APEC & FTA), Trade Policy Bureau, Ministry of Economy, Trade and Industry. Anthony J. Venables, Professor, University of Oxford; Centre for Economic Policy Research.

vii

viii

Contributors

L. Alan Winters, Professor, University of Sussex; Research Fellow, CEPR and IZA. Linda Yueh, Fellow in Economics, St Edmund Hall, University of Oxford.

Preface This book originated from the international joint research project on economic integration in East Asia conducted by the Institute of Developing Economies, Japan External Trade Organization, (IDE-JETRO), Chiba, Japan for the fiscal years 2005–06. Based on the perspectives of spatial and neoclassical economics, we examined the main economic forces working in the process of economic integration and tried to gain a clear view of the current progress and future prospects of East Asian integration. In East Asia, various infrastructures for trade and transportation have been rapidly developed in recent years. In the mainland part of Southeast Asia, for example, the Greater Mekong Subregion Economic Cooperation Program has been carried out in a framework of international cooperation on the initiative of the Asian Development Bank. In this program, 11 projects have been launched mainly for the purpose of development of road transportation, such as the East–West, North–South, and Southern economic corridors. The East–West economic corridor connects Danang in Vietnam, Savannakhet in Laos and Mawlamyine in Myanmar. The North–South corridor links Kunming in China to Hanoi in Vietnam and Chiang Rai in Thailand, and the Southern corridor connects Ho Chi Minh City in Vietnam, Phnom Penh in Cambodia and Bangkok in Thailand. These corridors will form a large highway network connecting many industrial zones in the Greater Mekong Subregion and are expected to promote economic integration by increasing international trade and investment in that area. Development of the logistic system in East Asia will bring about a substantial decrease in the costs of cross-border transportation. In addition, both tariff and non-tariff trade barriers will be reduced by concluding increasing numbers of free trade agreements, FTAs, and economic partnership agreements, EPAs. These changes in transport and trade conditions will enable firms in East Asia to supply their products to customers at more remote places in the world. On the one hand, this stimulates international division of labor and thereby leads to more efficient allocation of resources in East Asia. It will thus provide East Asian countries with substantial opportunities for economic growth. The transport and trade facilitation noted above, on the other hand, will allow firms to decide on their location, with overriding priority to production rather than transport conditions. ix

x

Preface

This may result in firms becoming more localized at a small number of industrial sites which exhibit distinct production advantages. In other words, integration may intensify regional inequalities in East Asia. Thus the topic attracts much attention, with mixed feelings of high expectations of economic growth and fear of growing regional inequalities. With these perceptions in mind, we attempt in this book to give a clear picture of East Asian integration, focusing on various aspects, such as structure of intra-regional trade; industrial location pattern, especially of multinational enterprises; formation of industrial agglomeration; and development of political and institutional frameworks for integration. Of course, the process of economic integration in East Asia has only just started, and few results are observable so far. Thus the present book is an interim report of our trial-and-error efforts toward our ultimate goal, and we would like to go further. We would appreciate readers’ comments. As part of this IDE-JETRO research project, we held an international workshop in Chiba, Japan in 2006. Also, an IDE-JETRO symposium entitled ‘Globalization and Regional Integration: From the Viewpoint of Spatial Economics’ was held in Tokyo in 2004. The discussions in both conferences helped deepen and enrich our arguments. We are grateful to Young Han Kim, Hisaki Kono, Paul R. Krugman, Tomohiro Machikita, Kiyoshi Matsubara, Toyojiro Maruya, Bhanupong Nidhiprabha, Keola Souknilan, Yu Yongding, and all the participants in the conferences for their valuable input. Masahisa Fujita, Satoru Kumagai and Koji Nishikimi December 2007

Abbreviations ACFTA ADB AEC AFTA AICO AJCEP APEC ASEAN CAP CEPR CEPT c.i.f. CLMV E&E EAFTA EAS ECSC EDC EEC EPA EPC ESCAP EU Euratom FDI f.o.b. FTA FTZ GATT GDP GMS GSP HME

ASEAN–China Free Trade Area Asian Development Bank ASEAN Economic Community ASEAN Free Trade Area ASEAN Industrial Cooperation ASEAN–Japan Comprehensive Economic Partnership ASEAN–Pacific Economic Cooperation Association of South East Asian Nations Common Agricultural Policy (EU) Centre for Economic Policy Research (Oxford) Common Eﬀective Preferential Tariff cost, insurance, freight Cambodia, Laos, Myanmar and Vietnam electric and electronic East Asian Free Trade Area East Asia Summit European Coal and Steel Community European Defence Community European Economic Community economic partnership agreement European Political Community Economic and Social Commission for Asia and the Pacific European Union European Atomic Energy Community foreign direct investment free on board free trade agreement free trade zone General Agreement on Tariffs and Trade gross domestic product Greater Mekong Subregion Generalized System of Preferences home market effect

xi

xii

IDE-JETRO IPR IZA LCD M&A MNE NAFTA NEFTA NIE ODA PPF PPP R&D RFID ROW RTA SEZ SME SMI SOE WTO

List of Abbreviations

Institute of Developing Economies, Japan External Trade Organization intellectual property rights Institute for the Study of Labor (Bonn) liquid crystal display mergers and acquisitions multinational enterprise North American Free Trade Agreement Northeast Asian FTA newly industrializing economy overseas development assistance production possibility frontier purchasing power parity research and development radio frequency identification rest of the world regional trade agreement special economic zone small to medium-sized enterprise Single Market Initiative (EU) state-owned enterprise World Trade Organization

1.

Introduction Masahisa Fujita, Satoru Kumagai and Koji Nishikimi

The past few decades have seen a signiﬁcant acceleration in the integration of the world economy. This has been mainly the result of reductions in transport and communication costs, the lowering of trade barriers, and an increasing mobility of capital and labor. From the 1990s, these trends in the global market have been reinforced by institutional development based on free trade and economic partnership agreements (FTAs and EPAs) that have been rapidly negotiated among many countries. By the end of 2002, GATT/WTO had received notiﬁcation of some 250 regional trade agreements (RTAs), including various FTAs and EPAs, and an additional 70 RTAs are reportedly now under negotiation.1 Due to this progress of economic integration, an increasing number of consumer goods are being imported from all over the world. Also, many producers are operating aﬃliated factories in various countries and expanding their global networks of input procurement and product distribution. Accordingly, the world export–GDP ratio has risen, on average, from 15.0 percent in 1990 to 26.2 percent in 2005. Figure 1.1 shows the major RTAs signed before December 2007.2 Two giants, the EU (European Union) and NAFTA (North American Free Trade Agreement), produce respectively US$15.0 and 15.9 trillion. In Asia, ten ASEAN (Association of South East Asian Nations) countries participate in AFTA (ASEAN Free Trade Area), but their total GDP is about US$2.3 trillion. This is equivalent to 15 percent of that of the EU and NAFTA. However, China, Japan and Korea, the three large economies in East Asia, are now eagerly establishing FTAs with ASEAN countries and also with each other. If these three countries are included, East Asia accounts for US$11.3 trillion of GDP. The three major integrated economies that include the EU, NAFTA and East Asia produce about 80 percent of the world’s GDP. This is remarkable given that these countries comprise only 46 percent of the world’s population. Economic integration usually proceeds in several steps: (1) trade liberalization; (2) capital/ﬁnancial liberalization; (3) labor mobilization; and 1

NAFTA

EU

Brunei-Japan Cambodia-Japan Indonesia-Japan Japan-Malaysia Korea-USA ASEAN-Korea Japan-Philippines ASEAN-China Japan-Singapore Japan-Thailand

25 countries Pop: 497 mil. GDP: US$ 14,953 bil.

GCC

2

6 countries Pop: 37mil. GDP: US$517bil.

AFTA SAFTA

7 countries Pop: 1,595mil. GDP: US$517bil.

SACU

ANZCER

5 countries Pop: 55mil. GDP: US$709bil.

2 countries Pop: 24mil. GDP: US$829bil.

Note: Circles indicate the member countries of each RTA. Source: Drawn by authors, based on METI (2006).

Figure 1.1

3 countries Pop: 445 mil. GDP: US$ 15,857 bil.

Major RTAs in the world

10 countries Pop: 582 mil. GDP: US$2,330 bil.

DR-CAFTA 7 countries Pop: 349mil GDP: US$14,085bil.

MERCOSUR 4 countries Pop: 233mil. GDP: US$1,296bil.

3

Introduction

Table 1.1

Trade and production in the integrated economiesa Year E. Asiab EU-15 NAFTA

(1) Share of intra-regional trade in total export of each region (%) (2) East Asia’s share in total import to each region (%) (3) Share in world’s exports (%) (4) GDP share (%) (5) GDP per capita at 1990 price (US$) (Growth rate; % annum)

1990 2005

39.9 51.1

55.1 56.3

1990 2005 1990 2005 1990 2004 1990 2004

39.9 51.1 20.4 27.0 18.9 25.9 2 426 3 152 (1.9)

9.2 11.8 37.5 31.8 30.3 29.0 18 869 23 812 (1.7)

32.8 34.4

ROW

World

25.9 34.3

– –

29.6 16.3 29.8 22.0 16.3 25.7 14.1 27.1 29.0 21.8 34.0 11.1 17 968 1 850 23 174 1 750 (1.8) (0.4)

20.4 27.0 100.0 100.0 100.0 100.0 4 345 4 982 (1.0)

Notes: a All shares in rows (1), (2) and (3) are calculated from import data. b East Asia is composed of Brunei, Cambodia, China (including Taiwan), Indonesia, Japan, Korea, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand and Vietnam. Sources: United Nations (2006), IMF (2007), Republic of China (2005).

(4) currency integration.3 The actual degree of progress in economic integration varies from one group to another. Generally speaking, the EU takes the lead, NAFTA follows, and East Asia comes last. In East Asia, liberalization of trade and investment has been signiﬁcantly progressing, but labor mobilization and currency integration are still far behind. The path of integration in East Asia is somewhat unusual and has not relied on rigid institutions, as in the EU and NAFTA. It was only after 2002 that FTAs were signed between ASEAN and other East Asian countries,4 and FTAs among China, Japan and Korea are still under negotiation. However, this does not mean that integration has been particularly slow. In fact, as shown in the ﬁrst row of Table 1.1, the share of intra-regional trade in East Asia rapidly increased from 39.9 percent in 1990 to 51.1 percent in 2005, which marked the highest growth rate among the three integrated regions, East Asia, EU and NAFTA. During this period, many multinational enterprises (MNEs) have rapidly expanded their production networks in East Asia, and this has substantially promoted the economic integration of East Asian countries. Thus East Asia’s economic integration in this period is often called de facto integration. East Asian countries have signiﬁcantly gained competitiveness in exports on the basis of de facto integration. As shown in row (2) of Table 1.1, during

4

Economic integration in East Asia

1990–2005, the share of East Asian products in total imports to each region increased in each of the EU, NAFTA, East Asia and the rest of the world (ROW).5 This led to an expansion of East Asia’s export share in the world market from 20.4 percent to 27.0 percent, while the EU’s and NAFTA’s share decreased, as shown in row (3). As further seen in row (4), growth in exports stimulated regional production in East Asia and enlarged its GDP share from 18.9 percent in 1990 to 25.9 percent in 2004. As a result, the average GDP per capita in this region increased 1.3 times from US$2426 to US$3152 (both at 1990 prices). The average growth rate in the 14 years was 1.9 percent per annum. This is almost twice that of the world average and higher than any other region in the world. There have been comprehensive studies of the economic eﬀects of regional integration in the ﬁeld of international economics. The neoclassical theory of comparative advantage has shown how trade liberalization can bring about economic growth.6 If a country has an advantage in the production of a speciﬁc commodity, it can gain advantage by providing other countries with its specialty in exchange for their specialties. Hence international trade enables eﬃcient use of products across borders and thereby leads to better allocation of resources. Traditionally, this kind of analysis has focused on trade of consumer goods, but similar processes may apply to trade of intermediate inputs and components. Nowadays, various ﬁrms build up their global production networks by splitting production processes into several elements and relocating these in diﬀerent countries. As a result of this business movement (often called fragmentation), many kinds of parts, components and even R&D outputs are often traded across countries on an intra-ﬁrm network. For both types of trade (consumption and production goods), any trade barriers (including tariﬀs and transport costs) present major obstacles to eﬃcient factor allocation and may deprive the countries of growth opportunities. Economic integration as a result of lowering trade barriers should promote eﬃciency in international resource allocation and provide a signiﬁcant source of economic growth.7 The traditional theory of comparative advantage also presents an optimistic view of the inﬂuences of integration on regional inequalities. Since international trade tends to narrow international gaps in prices of products and production factors, promotion of economic integration should reduce the income diﬀerences among member countries. However, there is widespread concern that integration may intensify inequalities among integrated countries as well as among domestic regions in each country.8 As is generally known, East Asian countries are quite diversiﬁed in economic size, levels of income and industrial technology, as well as in other socio-economic aspects. For example, the coeﬃcient of variation for GDP per capita in East Asia in 2004 was 1.140, three times the coeﬃcient for the

Introduction

5

EU-15 in the same year, 0.354.9 These large disparities among East Asian countries lead to a fear that countries with unfavorable conditions may be given only a limited opportunity for growth and left behind by the other growing countries in the market under keener competition created by integration. In addition, substantial concerns have been raised about increasing inequalities among domestic regions within each country. As examined in Chapter 2, the diﬀerences in regional income in China are signiﬁcant and expanding, particularly between coastal and inland areas. There is also a large diﬀerence in production level. The GDP share of the 11 coastal provinces increased from 53.3 percent in 1990 to 61.3 percent in 2003, while these provinces occupy only 12.4 percent of land in China.10 These facts have led to serious anxiety relative to whether or not regional disparities in East Asia will become intensiﬁed by further integration, with more FTAs and EPAs coming into eﬀect in the near future. When there are scale economies in production, ﬁrms tend to agglomerate, as extensively discussed in spatial economics.11 This may cause regional disparities as discussed above. The basic scenario for the integration– disparity nexus is as follows: suppose industrial technologies are characterized by scale economies. Then, productivity and competitiveness of a ﬁrm are crucially dependent on the size of production. Therefore ﬁrms will tend to locate in regions with large markets in order to fully exploit the scale beneﬁts. This eﬀect is often called the ‘home market eﬀect’ (HME). Workers and consumers, on the other hand, are attracted to such regions because there are many jobs and a wide variety of consumer goods is easily available. As a result of the circular eﬀects of ﬁrm-attraction and consumer-attraction, large industrial agglomerations may appear in which both ﬁrms and consumers enjoy location advantages.12 In such situations, economies will most likely exhibit ‘lumpy’ landscapes with signiﬁcant regional disparities.13 The spatial economic approach allows analysis of the agglomeration forces of integration, which cannot be captured by the neoclassical theory of comparative advantage. Due to the HME noted above, the market expansion caused by economic integration attracts a disproportionately large number of ﬁrms to the integrated area and encourages economic growth there. At the same time, ﬁrms attracted to that area may agglomerate in some particular countries and regions that have large domestic markets. For example, the growing market in China appears very attractive to many multinational enterprises (MNEs) that plan to locate in East Asia. This powerful attraction sometimes leads to serious fears among other Asian countries, such as members of ASEAN. However, as discussed above, China also encounters a serious internal problem of regional disparity, speciﬁcally the rapidly expanding inequality between coastal and

6

Economic integration in East Asia

inland areas. If economic disparities among member countries and/or among domestic regions become too wide, East Asian countries may not be able to sustain further progress in economic integration. It is necessary to study how to create a harmonious integration of signiﬁcantly diversiﬁed countries in East Asia, where economic integration is expected to progress further on both de facto and de jure bases. This book focuses on economic integration in East Asia from a combined viewpoint of neoclassical and spatial economics. The remainder of the book is composed of three parts: Part I (Chapters 2 and 3) includes a discussion of the book’s primary viewpoint relative to Asian economic integration along with a brief review of its theoretical basis. Based on this framework, Part II (Chapters 4–6) includes an overview of the development process and major obstacles to integration in East Asia. These are also compared with the case of EU integration. Part III (Chapters 7–11) is devoted to an examination of speciﬁc aspects of Asian integration such as development of intra-regional trade, location behavior of MNEs, and formation of industrial agglomerations. Chapter 2 includes a comparison of the spatial economic approach with the neoclassical view of comparative advantage. With the presence of scale economies, industrial agglomeration appears at various levels of geography from small business districts to the core–periphery structure at the global scale. We shall see that the industrial structure and trade pattern in East Asia switched in the early 1990s from the traditional ﬂying-geese pattern with a single core (Japan) to the multi-cored intra-regional division of labor. Agglomeration economies appeared more signiﬁcantly in the latter phase, reﬂecting substantial reduction of trade costs. Chapter 3 provides theoretical bases of the two views discussed in Chapter 2. In particular, two major eﬀects of economic integration, ‘specialization’ and ‘agglomeration’ of industries, are examined. Specialization forces represent the inﬂuences of comparative advantages and tend to cause the dispersion of industries among integrated countries. By contrast, agglomeration forces are caused by scale economies in production (via the HME) and create regional disparities among integrated economies and among domestic regions within each country. It should be noted that the views of spatial economics and comparative advantage theory are not mutually exclusive. In fact, both specialization and agglomeration forces may be at work simultaneously, and the total eﬀect on regional disparity is determined by their relative size. Maintaining the above viewpoints, the recent development of industrial clusters in East Asia is discussed in Chapter 4. An overview of 12 major clusters in the area called the ‘Asian Triangle’ is presented, with special attention to the expansion of global production networks of MNEs.

Introduction

7

Industrial zones, capacity-building, and anchor ﬁrms appear to be primary development keys for building a successful industrial cluster. For further promotion of economic integration in East Asia, it seems essential to develop cluster-to-cluster linkages. An overview of the development of institutional and political frameworks for East Asian integration is presented in Chapter 5. In addition, major issues that China, Japan and members of ASEAN currently encounter in promotion of economic integration in this area are examined. It appears that ASEAN countries must develop ‘soft’ infrastructures for system harmonization (such as establishment of intellectual property rights) and coordination of infrastructures (such as international road networks). These may then enable a facilitation of trade and investment in the area by signiﬁcantly reducing trade costs and lead times. China also needs to set up soft infrastructures, including rules to secure transparency. In addition, international coordination is necessary for energy and environmental issues which are becoming more and more serious as a result of the rapid growth of the Chinese economy. Japan should contribute to such joint eﬀorts, utilizing its long experience in energy saving and environmental preservation. Japanese ODA (oﬃcial development assistance) has been biased toward hard infrastructures, but higher priority must be given to soft infrastructures in the areas noted above. Chapter 6 includes a comparison of the above states and trends of East Asian integration with the European experience. Economic integration of the EU is reviewed, and the key similarities and diﬀerences between European and Asian models are discussed. Europe has been able to achieve ‘deep integration’ because of the vision of a united Europe, the political balance within Europe, and the development of the institutions of the EU. None of these features is present in Asia. Complementarities among Asian economies create gains from trade liberalization, but it seems unlikely that Asia will follow the European path of ‘deep integration’ or attain the consequent economic beneﬁts. In this sense, East Asia needs to ﬁnd its own path to a harmonious integration. To look more closely at individual issues in East Asian integration, we ﬁrst examine in Chapter 7 the structure and determinants of the intraregional trade during 1985–2004, the period when de facto economic integration rapidly progressed in East Asia. The result of analysis shows that the key determinants of intra-regional trade diﬀer by industry, and this yields a complicated pattern of trade in East Asia. Roughly speaking, the HME plays a signiﬁcant role in determining the trade patterns of the chemical products sector and the transport equipment sector, but a tricklingdown dispersion pattern (like the ﬂying-geese pattern) is observed in the machinery sector.

8

Economic integration in East Asia

Chapters 8 and 9 examine the location choice of, respectively, Korean and Japanese MNEs in East Asia. As discussed in Chapter 4, expansion of production networks of MNEs has a large inﬂuence on intra-regional trade in East Asia and also on the formation of industrial clusters in certain sectors that have signiﬁcant HMEs. Therefore investigation of location choices of MNEs should provide essential information to anticipate the consequence of further progress in economic integration and also develop a prescription for harmonious integration. Generally speaking, MNEs of the two countries are attracted to large markets and low-cost sites, but Korean ﬁrms exhibit a particular tendency to locate in the three northeast provinces of China. Chapters 10 and 11 include examination of key factors in the formation of industrial agglomeration in China and ASEAN respectively. An overview of the evolution of provincial trade and production in China since 1978 is provided in Chapter 10, and causes of industrial agglomeration, using provincial panel data gathered in the period 1987–2001, are explored. Trade liberalization policies and the size of regional markets appear to play signiﬁcant roles in the development of industrial agglomeration. The HME in ASEAN countries is investigated in Chapter 11. As previously discussed, presence of the HME provides a key for creation of an agglomeration and is also a source of regional disparity. The size of the HME in each industry is measured by assuming equilibrium in a spatial economic model. Results suggest that the HME is relatively large in the sectors of: (1) textile and leather products; (2) pulp, paper and printing; and (3) transport equipment. This is consistent with the ﬁndings in Chapter 7.

NOTES 1. 2. 3. 4.

5.

More than 170 notiﬁed RTAs are currently in force. For details, see http://www.wto.org/ english/tratop_e/region_e.htm. In addition to the RTAs shown in Figure 1.1, ASEAN–Japan EPA was concluded in November 2007. It will become eﬀective in autumn 2008. Trade liberalization involves both commodity and service trades. Hence liberalization policies include institution-building such as the establishment of intellectual property rights as well as the lowering of trade barriers. With the exception of those within China (such as Hong Kong, China and Macau, China), the eﬀective FTAs in East Asia as of August 2007 are: (1) AFTA (came into eﬀect in January 1992; ASEAN 10 members); (2) Japan–Singapore (November 2002); (3) ASEAN–China (July 2005); (4) Japan–Malaysia (July 2006); (5) ASEAN–Korea (July 2006); (6) Korea–Singapore (March 2006); (7) Japan– Philippines (September 2006); (8) Japan–Thailand (April 2007); (9) Brunei–Japan (June 2007); (10) Cambodia–Japan (June 2007); and (11) Indonesia–Japan (August 2007). In addition, ASEAN–Japan is scheduled to come into force in autumn 2008. In Table 1.1, values for the EU are given for 15 rather than 25 countries because the EU expanded from 15 to 25 members in 2005. The number of member countries further increased to 27 with the entry of Bulgaria and Romania in January 2007.

Introduction 6. 7. 8.

9.

10. 11. 12.

13.

9

Chapter 3 includes more details on neoclassical and spatial economics. If there are signiﬁcant trade costs, equilibrium factor prices may diﬀer among countries. In such cases, international transfer of production factors generally improves allocation eﬃciency. For details, see Chapter 3. There is no clear evidence of expansion of income diﬀerence among the EU-15 in 1990–2004. In this period, the ratio of the highest to the lowest GDP per capita in the EU-15 expanded from 4.06 to 4.32, and the coeﬃcient of variation rose slightly, from 0.350 to 0.354. However, the Gini coeﬃcient decreased from 0.112 to 0.087. The average GDP per capita of Cambodia, Laos and Myanmar was US$283 in 2004. For Japan, it was US$36 501. The max–min ratio of GDP per capita in 2004 was 166.7 in East Asia and 4.32 in the EU-15. The Gini coeﬃcient was 0.668 in East Asia and 0.087 in the EU-15. Both indices clearly show that income disparity was much larger in East Asia than in the EU-15. The 11 provinces include Beijing, Tianjin, Hebei, Shanxi, Liaoning, Shanghai, Jiangsu, Zhejing, Shandong, Fujian and Guangdong. In this book, the terms ‘spatial economics’ and ‘new economic geography’ are used interchangeably. For details of studies in this ﬁeld, see Fujita et al. (1999), Fujita and Thisse (2002) and Baldwin et al. (2003). Workers and consumers may not have easy mobility across the country. However, as discussed in Chapter 3, regional disparity can be created solely by the mobility of ﬁrms with scale economies. Further, as seen in Chapter 2, a similar circular mechanism can be generated between producers of ﬁnal and intermediate goods, both of which are internationally mobile. In addition to agglomeration, some authors argue that there are other forces, such as those related to political economic factors, that contribute to the globalization–inequality nexus. See Nissanke and Thorbecke (2006), for example.

REFERENCES Baldwin, R., R. Forslid, P. Martin, G. Ottaviano and F. Robert-Nicoud (2003), Economic Geography and Public Policy, Princeton, NJ: Princeton University Press. Fujita, M. and J.-F. Thisse (2002), Economics of Agglomeration: Cities, industrial location, and regional growth, Cambridge: Cambridge University Press. Fujita, M., P. Krugman, and A. Venables (1999), The Spatial Economy: Cities, regions and international trade, Cambridge, MA: MIT Press. IMF (2007), Direction of Trade Statistics (CD-ROM), Washington, DC: IMF. Ministry of Economy, Trade and Industry, Japan (METI) (2006), ‘Keizai Renkei (EPA) no Torikumi ni tuite’ [‘Progress in the Economic Partnership Agreement (EPA)’], www.meti.go.jp/policy/trade policy/epa/data/ 060307suishin.pdf. Nissanke, M. and E. Thorbecke (2006), ‘Channels and policy debete in the globalization–inequality–poverty nexus’, World Development, 34, 1338–60. Republic of China (2005), Statistical Yearbook of the Republic of China, Taipei: Republic of China. United Nations (2006), Statistical Yearbook, New York: United Nations.

PART I

Basic Viewpoints on East Asian Economic Integration

2.

Regional integration in East Asia: perspectives of spatial and neoclassical economics Masahisa Fujita and Nobuaki Hamaguchi

2.1

INTRODUCTION

Globalization has highlighted the importance of geography. Firms can choose locations more freely and internationally so as to produce and deliver more eﬃciently, taking into consideration locational advantages in labor cost, infrastructure quality and accessibility to major markets. The reduction of trade costs in the broad sense (Anderson and Wincoop 2004) has allowed ﬁrms to supply a much wider area of markets from a smaller number of production locations. This, in turn, has provoked a concern over widening regional income disparities because some regions seem to attract a disproportionately large share of productive employment while others are left with only a small share. In the early 1990s, concern about the future of European integration after the establishment of the European Union (EU) in 1993 encouraged some theorists to develop models of the location of economic activities; this has become known as ‘spatial economics’. A natural question of interest was, ‘Given the regional integration of the EU as a single market, what will be the spatial distribution of economic activities?’ Since then, spatial economics has reached a certain theoretical consolidation, as we can see in Fujita et al. (1999), Baldwin et al. (2003) and Henderson and Thisse (2004). This book examines the future of the economic geography of East Asia, which is also increasing its degree of integration. Casual observation identiﬁes a complex combination of both a dispersion of industries from higher-wage regions to lower-wage ones, and a concentration in larger markets and advanced technology centers. As we noted in Chapter 1, two major forces are involved in this process. The neoclassical theory of comparative advantage explains that the reduction of international trade barriers beneﬁts both higher- and lower-income developing countries by allowing the latter to specialize in labor-intensive goods and the 13

14

Basic viewpoints on East Asian economic integration

former in capital-intensive ones. On the other hand, if we explicitly consider the movement of productive factors and scale economies, spatial economic analysis, which emphasizes agglomeration economies, describes the geographic concentration of production in larger markets. It turns out that deepening regional integration increases the relevance of spatial economic analysis. This chapter introduces the perspective of spatial economics, which is useful in interpreting the ongoing process of regional economic integration in East Asia. The next section sets out the basic theoretical concept of spatial economics. Section 2.3 will look at the characteristics of the evolution of economic integration in East Asia. This will be followed by an interpretation of the formation of industrial agglomerations in Section 2.4, and some policy implications are set forth in the ﬁnal section.

2.2 2.2.1

THE SPATIAL ECONOMICS PERSPECTIVE Neoclassical Framework

Regional integration promotes production specialization across nations and sub-regions. Neoclassical economic theory argues that a framework of comparative advantage leads to the globalized distribution of productive activities. Due to natural diﬀerences in immobile productive resources among diﬀerent locations, the theory argues that trade liberalization induces each location to concentrate its productive resource in sectors where it is more eﬃciently used. In this context, natural diﬀerences in resource endowments and income levels promote international trade and the more eﬃcient use of productive resources in the globally integrated economy as a whole. The manifestation of this argument in the East Asian context has come to be known as the ﬂying-geese pattern of economic development, which will be taken up in Section 2.3. The wide disparity in income levels across the nations of the region has served as the basis for the greater division of labor which has paved the way for the entry of late industrializing economies in the catching-up process based on the availability of cheap labor. 2.2.2

The Basic Framework of Spatial Economics

In contrast to the neoclassical economic theory of comparative advantage, spatial economics focuses on backward and forward linkages which attract mobile productive resources to particular locations. The role of spatial

Regional integration in East Asia

15

economics in terms of regional integration is to explain the agglomeration or clustering, in a large variety of forms, of economic activities as the result of the cumulative eﬀects of interactions between the linkages and concentration of productive resources owing to economies of scale. Agglomeration occurs at diﬀerent geographical levels. At one extreme is the core–periphery structure at the global level. In 2005, NAFTA was responsible for 32 percent of the world’s GDP, the EU (15 countries) 30 percent, and East Asia (ASEAN-10 plus China, Japan, South Korea, Hong Kong and Taiwan) 20 percent. These three regions account for 82 percent of the world GDP, while their share of world population and land area corresponds to 45 percent and 25 percent respectively. Below the global level, strong regional disparities can be observed within each country. For example, Fujita et al. (2004) point out that regional economies in Japan developed typical core – periphery structures, which remain essentially the same today. While the ‘hollowing out’ of manufacturing has been in progress since the mid-1960s toward less developed regions, the economic cores in Japan have maintained high ratios of GDP and higher productivity. Large metropolises constituting the core regions are highly diversiﬁed in that they nest many kinds of economic activities that are not related through direct linkages, but interaction of their diversiﬁed and highly qualiﬁed human resources tends to enhance productivity and chances of innovation.1 Finally, at the detailed extreme of the spectrum, agglomeration within cities arises in the form of bustling commercial and industrial districts made up of small enterprises, providing a location for the one-stop purchasing of a wide variety of products. To some extent, spatial concentration of economic activities can be explained by such natural features as topography, climate, mineral deposits and historical events; these are known as attributes of the ‘ﬁrst nature’. However, the impact of these ﬁrst-nature attributes on the spatial distribution of economic activities is not diﬃcult to explain within traditional economic theory. It certainly helps us to understand why a particular agglomeration was formed in a certain location, but does not give a full account of why the agglomeration is maintained there. For example, although the automobile industry in Toyota City originated from the founding of the ﬁrst automobile assembly division of the Toyota Automatic Loom Company in the mid-1930s in a small town then called Koromo that specialized in textiles, the ﬂourishing of the location as a motor city should be understood as the result of mutual stimuli between suppliers of intermediate parts and assemblers of ﬁnal products. Similarly, the current status of Singapore as the economic center of Southeast Asia cannot be fully explained by its hub port role in transportation, which overlooks the country’s growing importance as a ﬁnancial and business center.

16

Basic viewpoints on East Asian economic integration

Spatial economic models focus on such endogenous mechanisms leading to agglomeration; these are attributes of the ‘second nature’. In order to understand why agglomeration exists without relying on a priori assumptions about uneven distribution of resources, we need to specify the endogenous agglomeration forces and the counterbalancing dispersion forces. In this regard, spatial economic models explicitly consider plant-level indivisibilities (economies of scale), transport costs, and mobility of workers and consumers. The assumption of indivisibility is necessary in order to avoid the case where the economy will degenerate into ‘backyard capitalism’ in which each household or small group produces most items by itself. In fact, without scale economies at the ﬁrm level, there is no need to concentrate the production of each good at the same location. Economies of scale, in turn, lead to a market structure characterized by imperfect competition. Here, we can introduce the concept of ‘heterogeneity’ (product diﬀerentiation under monopolistic competition) in order to express the agglomeration of a large number of ﬁrms rather than a single large-scale plant. Due to the existence of transportation costs, a ﬁrm is penalized by the loss of demand the further its distance from the market, while consumers (or users) are eager to enjoy the wider variety of products that comes from locating closer to producers. If goods are suﬃciently diﬀerentiated from each other, their suppliers can locate in proximity without suﬀering severe loss of demand in other local markets. Clearly, if it were not for transport costs, location would not matter. Finally, the migration of workers ( consumers) is a prerequisite for the mutual agglomeration of workers and ﬁrms which constitutes the core market. To be more concrete, suppose that a large variety of consumer goods is produced in a city. Because of transport costs, they can be purchased at lower cost in the city. Given a nominal wage, real income is higher there. This, in turn, induces more workers to migrate to the city. The resulting increase in the number of consumers ( workers) creates greater demand for consumer goods, which induces even more ﬁrms to locate in the city, magnifying the variety of consumer goods there. A circular causation of the agglomeration of ﬁrms and workers in the city is created through forward linkage (the eﬀect of a greater variety of consumption on workers’ real income) and backward linkage (the eﬀect of market size in attracting more ﬁrms). Likewise, suppose that a large variety of intermediate goods is available in a city. In this case, the forward linkage is the eﬀect of the variety of inputs that lowers user-ﬁrm production costs, which in turn induces more user ﬁrms to locate in the city, creating the backward linkage (the eﬀect of greater user-ﬁrm demand which attracts more intermediate-good ﬁrms amplifying the amount of variety). This leads to a similar kind of circular

Regional integration in East Asia

17

causation of agglomeration through backward and forward linkages between intermediate goods and user (ﬁnal-goods) ﬁrms (and their workers). Taking producer services as an example of intermediate goods, this can partly explain the cluster of high-technology ﬁrms in Silicon Valley, or the concentration of headquarters of big enterprises in New York, Tokyo and Singapore. Finally, focusing on the diversity of people with diﬀerentiated talents, the greater agglomeration of heterogeneous people in a city leads to higher productivity in innovation activity through the complementarities of these people (forward linkage), which, in turn, induces greater localization of innovation activities creating larger demand for diﬀerentiated talent (backward linkage). The forward linkage implies knowledge spillover, mainly through face-to-face communications of tacit knowledge. Although we have considered the three types of agglomeration forces separately, they actually work together as the driving power for the growth of cities. On the other hand, forces of dispersion result from the rise of production costs and living in large agglomerations, which induce high land prices and wage rates as well as congestion eﬀects (pollution and traﬃc jams). Firms may also be attracted to a local market which has far less intense competition than the core market. From the complicated balance of agglomeration and dispersion forces, a variety of agglomeration patterns emerges. Obviously, this tug-of-war can be analyzed only within a general equilibrium framework where both forces are simultaneously determined. Thus the spatial economic perspective can be diﬀerentiated from traditional international economics in several ways. Based on the a priori assumption of an uneven ﬁrst nature (factor endowment or technological diﬀerence), the latter provides a useful framework to explain the dispersion of economic activity based on factor-price diﬀerentials. However, it cannot explain the location of production in an increasingly borderless world/ regional economy. For spatial economic models, geographical distribution of factors and the resulting trade structure and factor-price diﬀerentials are simultaneously determined, providing a more general framework for analyzing economies in this era of globalization. But this beneﬁt comes at the expense of highly complex models that involve the possibility of multiple equilibria. 2.2.3

Impact of Decreasing Transport Costs (Economic Integration)

As mentioned in Chapter 1, international transport costs (broadly deﬁned) have declined substantially with the spread of free trade agreements and advances in transportation and communication technology. This has

18

Basic viewpoints on East Asian economic integration

resulted in at least two new phenomena in the pattern of the international location of industries. One is fragmentation. Under high transport costs, a series of production processes consisting of diﬀerent levels of labor intensity should be concentrated in a single location. However, if transports costs are suﬃciently lowered, the cost saving from separating each production process and locating them in adequate locations taking advantage of wage diﬀerence across countries can exceed the cost of organizing fragmented production in diﬀerent locations. In other words, fragmentation is a ﬁner division of labor based on the comparative advantage principle mentioned in Section 2.2.1. As will be pointed out in Section 2.3, the rapid increase of intra-regional trade in East Asia has been boosted by the exchange of intermediate goods, and the driving force behind this has clearly been the growing production networks of multinational ﬁrms. Another issue of interest is the formation of industrial agglomeration, although the relationship between economic integration and agglomeration is not necessarily obvious. As noted in Section 2.1, spatial economics has been inspired by increasing economic integration during the current phase of globalization. The creation of the EU was an early manifestation of this phenomenon. In North America, the NAFTA came into being in 1994; and in East Asia, although there is no region-wide free trade agreement, de facto regional integration has been progressing. By deﬁning this process as the steady reduction of trade costs (broadly deﬁned transport costs), spatial economic models can provide us with some insights to facilitate our understanding of the likely impacts of regional integration on the spatial distribution of production and income. For ease of understanding, ﬁrst let us consider an extreme case in which transport costs are prohibitively high. Since inter-regional trade is essentially impossible, the potential market of each region is exactly the same as its own population size, leading to the dispersion of economic activities by the same proportion. Now suppose that trade costs decrease gradually and inter-regional trade starts growing. The dispersed spatial structure will become unstable because economies of scale at the ﬁrm level make it attractive to consolidate production in fewer locations. A region exhibits high market potential if it is endowed with its own large population or because of its advantageous geographic position which endows it with good accessibility to a large number of consumers in other regions, or both. Such regions tend to attract more ﬁrms due to the scale economy (home market eﬀect: HME), and the larger demand for labor will raise the nominal wage. This, in turn, tends to boost real income because more goods are now available within the same region, relieving consumers of transport costs. Such a situation naturally leads to labor (consumer) inﬂow, which further increases the market potential and induces more ﬁrms to locate there. This

Regional integration in East Asia

19

positive feedback mechanism leads to the self-organization of an agglomeration from any dispersed spatial structure. Like the agglomeration forces based on the variety of intermediate goods and of talented people, the reduction of transport costs on intermediate goods and people will have a similar eﬀect. Once such circular causation through backward and forward linkages sets in, it cannot be easily reversed (lock-in eﬀect). Hence a temporary shock may change the spatial structure quite drastically and have a permanent eﬀect.2 It ought to be clear that a large agglomeration can emerge only when trade costs are suﬃciently low. A question naturally arises as to what may happen if, in the most extreme case, trade costs become close to zero.3 In such a case, geography no longer matters because the market potential will be the same no matter where the producers and consumers are. Hence there is no need to maintain an agglomeration. Therefore, in general terms, the gradual decrease of trade costs ﬁrst leads to agglomeration, and then, at a certain point, to re-dispersion. In reality, however, despite the progress of information and communication technology, as well as the more agile and ﬂexible modes of transport, it is still too early to declare the ‘death of distance’, and we are still in a time when technological progress tends to make major metropolises and advanced regions more dominant in the world economy.

2.3

REGIONAL INTEGRATION IN EAST ASIA

Although spatial economics has provided a plausible scenario for the future of globalization, it has developed along the lines of theoretical modeling, leaving us plenty of room for reﬂection on real-world issues. Admittedly, spatial economic empirical studies are faced with a lack of qualiﬁed data on transport costs (in the broad sense), but we should be able to test the validity of their theoretical implications.4 In the context of East Asia, the transformation of economic geography should start with the traditional ﬂying-geese pattern of economic development. In this pattern, Japan has played the role of core economy. The international division of labor has developed as Japan has become increasingly specialized in technologically advanced industries while successively shedding industries in which it no longer holds a comparative advantage; these industries, in turn, moved to nearby less developed countries (the Asian NIEs – newly industrializing economies). Over time the ‘following geese’, in turn, upgrade their own industrial structures following the ‘lead-goose’ trajectory while shedding outdated industries to the neighboring less developed countries (ASEAN and China).

20

Basic viewpoints on East Asian economic integration

The trickle-down phase of the ﬂying-geese analogy can be understood within the framework of neoclassical economics because shedding activities seek the advantage of a low-cost location, and scale economies do not play a decisive role. However, it is not clear why Japan remained as the cuttingedge ‘lead goose’. According to Fujita (2005), in 1990 Japan accounted for 72 percent of the total GDP in East Asia, while within Japan, economic core regions represented 40 percent of the national total. This implies that Japan’s core regions, with a mere 0.18 percent of the total area and 2.5 percent of the total population of East Asia, represented 29 percent of the total regional GDP, showing remarkable geographical concentration. It is reasonable to assume that the three kinds of agglomeration forces noted in Section 2.2.2 (i.e. market magniﬁcation, production linkage, and diversity of people) partly explains the dominant position of Japan during the early phase of the ﬂying-geese pattern of economic development in East Asia. On the whole, the ﬂying-geese analogy describes well the catching-up industrialization in East Asia. Economic liberalization has provided lowincome economies with opportunities to become integrated into the regional production network, enabling the sequential take-oﬀ of industrialization in these economies. Some factors changed in the 1990s. First, the Asian NIEs caught up with Japan in certain technological areas, such as semiconductors and information and telecommunication equipment (i.e. notebook computers and mobile telephones). Leading Asian NIE and Japanese ﬁrms involved in these frontier technologies compete intensely with each other in the global market. In these countries, agglomeration economies based on input linkages and human resource diversity have already taken eﬀect. Second, China emerged not only as a location for cheap labor but also as a huge consumer market. Due to China’s scale economy, unlike ASEAN, where each country is a small economy, backward and forward linkages generated agglomeration economies in two ways. First, factor accumulation, initially triggered by foreign capital inﬂow when China opened up, induced the expansion of demand potential, which promoted further industrialization. Second, input–output linkages meant that while ﬁnalgood production expanded, owing to sizable input demand, parts and components suppliers were established locally. Thus the East Asian economy has been transformed from a traditional ﬂying-geese pattern with a single-core structure to one with a multi-core. The region has multiple technological centers and co-agglomerations undertaking ﬁnal-good and intermediate-good production, with intermediate goods moving bi-directionally between countries. Let us examine some characteristics of economic relationships in East Asia. Table 2.1 shows trends in the relative changes in economic power of

21

Regional integration in East Asia

Table 2.1

Share of world GDP and trade (%) Japan

GDP* Exports** Imports**

1990 2005 1990 2005 1990 2005

14.6 10.3 8.3 5.8 6.6 4.8

NIE-3 2.4 3.0 6.2 7.4 5.8 6.9

ASEAN

China

1.6 2.0 4.0 6.1 4.5 5.3

1.9 5.0 1.8 7.3 1.5 6.1

East Asia 20.5 20.2 20.4 26.6 18.4 23.1

Note: NIE-3 includes Hong Kong, South Korea and Taiwan. Singapore is included in ASEAN. Source: * IMF, World Economic Outlook Database (http://www.imf.org); ** IMF, International Financial Statistics (CD-ROM) 70..D and 71..D. Data on Taiwan are from the Statistical Yearbook of the Republic of China, 2005.

East Asian countries in terms of GDP and world trade shares, 1990 to 2005. The ﬁrst two rows indicate that the world GDP share of East Asia as a whole has not changed during this period. Within the region, however, we can see the rise of China and the decline of Japan, while the shares of the NIE-3 and ASEAN have remained the same. As a consequence, Japan’s ratio of GDP in the world declined from 14.6 percent in 1990 to 10.3 percent in 2005, whereas that for China increased from 1.9 percent to 5.0 percent. This implies that although Japan remains in the dominant position, China has been catching up very rapidly over the last two decades. In contrast, East Asia’s share of trade presents a signiﬁcantly diﬀerent picture. As shown in Table 2.1, East Asia’s share of world exports grew from 20.4 percent in 1990 to 26.6 percent in 2005. Its composition also changed over the same period. The world shares for both exports and imports of the NIE-3, ASEAN and China grew much faster than their GDP. As a consequence, they have accounted for an even greater share of world trade than Japan. Moreover, all four countries/regions of East Asia have signiﬁcantly smaller shares of imports than exports. This means that East Asia as a whole has built up an export platform that ships to the rest of the world. The growth of East Asia’s share of world trade has coincided with the increase in the share of intra-regional trade,5 which represents the share of international trade within the region over total trade (exports and imports) with all countries in the world. It can be seen from Figure 2.1 that the share of intra-regional trade is the highest in the EU-15, hovering around 60 percent of total exports and imports. This is not surprising, since many juxtaposed rich countries with small land areas have been cooperating in

22

Basic viewpoints on East Asian economic integration

unison over the past half-century, making regional integration considerably deeper and wider. In East Asia and North America intra-regional trade has shown steady growth over the past couple of decades. East Asia’s share of intra-regional trade, in particular (except during the period of the Asian ﬁnancial crisis in 1997–98), has been increasing progressively, approaching that of the EU-15, while NAFTA has seen its share of intra-regional trade decline gradually in recent years. It should be noted that this growth of intra-regional trade in East Asia has not been driven by any region-wide free trade agreement as in the case of the EU and subsequently NAFTA. Rather, it should be seen as a market-driven de facto integration promoted by the development of the region’s transportation infrastructure and evolving inter-regional divisions of labor within the production network (‘value chain’) in which multinational ﬁrms play a major role. Figure 2.1 also shows the change in the share of intra-regional trade of the ASEAN-10 and of Northeast Asia (made up of China, Japan, South Korea and Taiwan). We can see from the ﬁgure that although neither ASEAN nor Northeast Asia alone represents a suﬃciently integrated region, the two sub-regions together constitute an integrated region with strong interdependence. The growth in the share of intra-regional trade for the whole of East Asia has been accompanied by a gradual increase in trade within each sub-region. In fact, trade among the countries of Northeast Asia has been growing more rapidly recently. This reﬂects the geographic situation in Northeast Asia, where China is endowed with a huge pool of cheap labor, and the capital-intensive economies of Japan, South Korea and Taiwan are located within short distances, thus enabling vertical production integration. The trade pattern in East Asia is characterized by an intra-regional division of labor. We can see from Figure 2.2 that China is by far the biggest net exporter of consumer goods, while importing parts, processed materials and primary goods. On the other hand, Japan is the major supplier of capital goods and parts in a complementary relationship with China. South Korea and Taiwan have characteristics similar to Japan, although they are still net exporters of consumer goods. Singapore is specialized in capital goods related to information technology and processed materials derived from petrochemicals. Other ASEAN countries are net exporters of consumer goods, while also showing a comparative advantage in capital goods (the Philippines and Malaysia in computers, Thailand in automobiles), and in processed and unprocessed primary commodities (Indonesia). Figure 2.3 presents a schematic of the trade pattern in the Asia-Paciﬁc region with the focus on East Asia and the USA. Japan, the Asian NIEs, China and ASEAN together have developed a highly integrated production system. Intermediate goods are moving back and forth while the consumer

% 70 60

64.4 EU-15 EU-15, 58.7

56.4

East Asia, 52.4

East Asia

50

NAFTA, 44.6 40 30

34.9 NAFTA

33.2

23

China, Japan, Korea, Taiwan, 25.8

ASEAN-10 20 10

15.9

China, Japan, Korea

ASEAN-10, 22.2

13.9

0 1980

1985

1990

1995

2000

2003

Note: East Asia corresponds to Japan, South Korea, China, Taiwan plus Hong Kong and ASEAN-10 combined. Sources: IMF, Direction of Trade Statistics, CD-ROM. For Taiwan, ROC Statistical Yearbook.

Figure 2.1 The share of intra-regional trade as a percentage of the total trade (exports imports) of each region, 1980–2005

24

Basic viewpoints on East Asian economic integration

China

Consumer goods Capital goods Parts Processed materials Primary goods Consumer goods Japan

Capital goods Parts Processed materials Primary goods 0

50000

100000

150000

200000

Consumer goods Korea

Capital goods Parts Processed materials Primary goods

Taiwan

Consumer goods Capital goods Parts Processed materials Primary goods

Singapore

Consumer goods Capital goods Parts Processed materials Primary goods 0

20000

40000

60000

Source: Author’s own calculations based on METI (2006), Figure 2.3.1, and IMF Direction of Trade Statistics (CD-ROM).

Figure 2.2 Composition of Asian country exports and imports by types of goods, 2003 (US$ million) goods assembled in China and ASEAN (as well as in Japan and the NIEs) are exported to the USA (as well as to the rest of the world (ROW)).6 China, in particular, has become the largest export platform to the ROW. Here it must be pointed out that a large share of this trade activity is conducted by aﬃliates of multinational enterprises (MNEs). For example,

25

Regional integration in East Asia

Indonesia

Consumer goods Imports (2003)

Capital goods

Exports

Parts Processed materials Primary goods

Philippines

Consumer goods Capital goods Parts Processed materials Primary goods

Malaysia

Consumer goods Capital goods Parts Processed materials Primary goods

Thailand

Consumer goods Capital goods Parts Processed materials Primary goods 0

10000

20000

30000

40000

50000

according to oﬃcial Chinese data,7 57 percent of importing and exporting was conducted by MNEs in 2004. Over the last two decades, there has been a rapid growth in the inﬂow of foreign direct investment (FDI) into East Asia. According to the UNCTAD World Investment Report 2006, East Asia, excluding Japan, received US$155.3 billion of FDI, which represents 17 percent of the world total. This is large when compared to the region’s 10 percent share of world GDP. Of particular note is that China alone

26

Basic viewpoints on East Asian economic integration Japan and NIEs

Final goods Intermediate goods

Final goods (parts) Intermediate goods Capital goods (consumption goods)

China ASEAN

USA (and the ROW) Final goods (parts)

Source: Fujita (2005).

Figure 2.3

The triangular trade pattern in the Asia-Pacific region

received US$72.4 billion in FDI in 2005, corresponding to almost half of the region’s total. While China and ASEAN can be placed in the same position within the regional trade linkage depicted in Figure 2.3, the Chinese economy is much more vigorous. Table 2.2 shows that the share of FDI ﬂowing to East Asian countries has increased for China and decreased for ASEAN, and that the share of ﬁxed capital formation has followed the same trend. It should be pointed out that China’s share of ﬁxed capital formation was smaller than its share of FDI in the 1990s, but the former has been well above the latter in recent years. ASEAN’s share of ﬁxed capital formation in recent years has fallen below that of South Korea, although the ﬂow of FDI into the latter is substantially smaller. These observations imply that local capital investment in the Chinese economy, involving the emergence of a large variety of local intermediate-goods suppliers, is very vigorous, which implies a mutual stimulation between foreign and domestic investment, leading to agglomeration. On the other hand, ASEAN remains dependent on foreign capital, and its local capital is not playing the leading role in capital accumulation. Table 2.2 also shows that Japan’s dominant position in ﬁxed capital formation in the 1990s is being challenged by China’s rapid gains in the same way as seen with GDP. To summarize, the growth of intra-regional trade in East Asia has developed in the triangular trade pattern depicted in Figure 2.3. Historically, this specialization structure has developed through the ﬂying-geese pattern of

27

Regional integration in East Asia

Table 2.2

Share of FDI and fixed capital formation in East Asia

FDI inflow (% for East Asia, excluding Japan)

Fixed capital formation (% for East Asia, excluding Japan)

(% for East Asia, including Japan)

1990–2000 (annual average)

2003

2004

2005

China South Korea ASEAN

42 4

58 4

46 6

47 5

31

22

20

24

China South Korea ASEAN

38 19

58 17

61 16

59 17

23

17

15

15

China South Korea ASEAN Japan

14 7

31 9

32 9

35 10

9 62

9 46

8 47

9 41

Source: UNCTAD, World Investment Report 2006.

catching-up industrialization. A stronger concentration of activities is now seen in China, where FDI is ﬂowing in, and local investment is also being stimulated. This process can be interpreted as the rise of market potential catalyzed by FDI followed by domestic investment that further increased market potential. This interaction between market potential and factor inﬂow is characteristic of spatial economic-type industrial agglomeration. Middle-income Southeast Asian economies are concerned about the mobility of capital, which is tending to concentrate in China. Moreover, there is increased uncertainty about whether economic liberalization and strengthening regional integration will oﬀer the low-income ASEAN countries the same opportunities for economic growth as the ﬂying-geese pattern once did.

2.4

INDUSTRIAL AGGLOMERATION IN EAST ASIA

Having looked at some of the features of the growing regional integration in East Asia and its expanding intra-regional trade and FDI, which are stimulating diﬀering intensities of factor accumulation and production

28

Basic viewpoints on East Asian economic integration

growth, our next task is to relate these phenomena to the spatial economic perspective that emphasizes the concentration of economic activities. 2.4.1

Automobile Industry

The automobile industry is concentrated globally in three markets: North America, Europe and East Asia. As Figure 2.4 shows, the ratio of concentration in these three markets stood at 90 percent in 2005, a slight decline from 93 percent in 1990. This can be explained by the increasing returns to scale in production and the eﬀects of transport costs, called the HME, by which larger-scale markets tend to become net exporters of these goods.8 Among the three regions, production is dispersed because the relatively high transport costs make it reasonable to produce closer to consumers. Within East Asia, the Japanese automobile industry held a dominant share in 1990 because of agglomeration economies based on input linkage. It then dispersed mainly due to increased local production in North America and East Asia. The dispersion to other parts of East Asia was also due to the lower labor costs in those parts. In the meantime, automobile production in China and South Korea underwent remarkable growth. The South Korean automobile industry is oriented strongly toward exporting (70 percent of total production in 2005) and dominated by the HyundaiKia Automotive Group.9 Chinese automobile production, in contrast, is aimed mainly at the domestic market, buoyed by the entry of multinational ﬁrms, including ones from Japan. According to Fan and Scott (2003), the location of the automobile industry in China was inﬂuenced greatly by the government’s initial policy of choosing the northeastern city of Changchun (Jilin Province) to facilitate technological transfer from the former Soviet Union in the 1950s, and then Chonquin and Shiyan (Hubei Province) in the west which were to serve as the growth pole for inland development. However, newly established automobile production plants and their suppliers have tended to concentrate in three coastal industrial agglomerations (Bohai Bay Rim, Shanghai and Guangdong). According to the 2006 Overseas Japanese Companies Data published by Toyo Keizai Inc., there are 445 active Japanese corporate aﬃliates in China registered under the industrial classiﬁcation ‘automobiles and parts’ or who explicitly state that their business area is related to the automobile industry. Ninety percent of these aﬃliates are located in six coastal provinces: 122 in Guangdong, 70 in Shanghai, 81 in surrounding provinces (62 in Jiangsu and 19 in Zhejiang), 54 in Tianjin, and 20 in Shandong.10 Among the ASEAN countries, Thailand emerged as the automobile production hub. The Thai market has a peculiar taste for pickup trucks, which

90%

25

20

World share of three regions 93%

East Asia breakdown

16.2% Korea

0 1990

2005

Sources: Jidosha Nenkan (Automobile Industry Yearbook) 1993 and http://www.oica.net/htdocs/statistics/statistics.htm.

Figure 2.4

Automobile production in the three major regions

ASEAN-4 9.7% (of which Thailand 4.9%)

24.9% China

34.4% East Asia

47.2%

31.3% Europe

Japan

24.6% North America

5.1% ASEAN-4 (of which Thailand 1.8%)

8.1%

3.1% China

Korea

81.7%

34.1% East Asia

Japan

34.8% Europe

5

24.2%

29

10

North America

Millions

15

30

Basic viewpoints on East Asian economic integration

constitute 65 percent of the domestic market. This is the second largest market for pickup trucks in the world after the USA. For the Thai people, the pickup is attractive because maintenance is easier, and spare parts are cheap and easily available because they are mostly produced locally. This situation also reﬂects the fact that agriculture makes up an important part of the Thai economy. The government sales tax of 3 percent on pickups is over ten times lower than the 30–50 percent rate on passenger cars. In 2005, 1.1 million cars were produced, a doubling of production in only three years. Domestic sales in 2005 were 0.7 million units; the remainder were exported. Taking advantage of low labor cost, the Thai automobile sector exports assembled pickup trucks to Australian, ASEAN, Middle Eastern and European markets, earning US$5.5 billion in 2005. This can be seen as a typical example of the HME whereby countries tend to become net exporters of the goods that they tend to consume in large amounts. Meanwhile, parts worth US$4.6 billion are exported to the parent companies’ home countries, such as Japan and the USA, or to other ASEAN countries where these companies have other plants. Enhancing scale economies thus enables a wide variety of parts suppliers to locate in Thailand. In 2005 total exports of the Thai automobile industry reached US$10.1 billion. Japanese automakers dominate the Thai automobile industry, claiming nearly 90 percent of both the domestic market and exports. Since assembly plants tend to rely on close and informal relationships when dealing with parts suppliers, both are concentrated in the Bangkok metropolitan area and the surrounding central and eastern regions within a 200-km highway distance from Bangkok. Many key components are produced by aﬃliates of Japanese auto-parts suppliers. In light of Thailand’s expanding market opportunities, direct investment by global auto-parts makers from Europe and North America is also increasing. 2.4.2

Electronics Industry

In comparison with automobiles, electronic products are generally compact and easier to transport. Therefore the forces of agglomeration tend to exceed those of dispersal, leading to a small number of concentrations of production at a global scale. East Asia, as shown in Table 2.3, claims a substantial share of the global market for leading products such as personal computers, mobile phones, digital cameras and semiconductors. Although technological development in electronics has been equally intense in both Europe and North America, East Asia has an advantageous condition in that innovation centers, ﬁrst appearing in Japan then in South Korea and Taiwan, and cost-reduction production centers, such as China and Southeast Asia, are closely located, constituting a regional production

31

Regional integration in East Asia

Table 2.3 Country share of major electronics products and components, 2005 Notebook PCs

Mobile phones

Semi conductors*

Digital cameras

China South Korea Taiwan Singapore Malaysia Philippines Thailand Indonesia Japan

83.5 3.0 5.3 0.5 1.4 0.5 2.6

35.0 26.0 6.0 2.1 2.9 0.1 6.2

2.9 12.8 14.3 7.0 6.2 2.4 2.0 0.6 18.4

48.0 1.5 0.1

East Asia

94.2

72.1

48.2

100.0

Europe North America South America

1.8 1.3

12.8 3.7 5.2

9.4 24.0

10.4

39.2

Note: * Share calculated from the value of production in 2004. Shares of other products are based on the volume of production in 2005. Source: METI, White Paper on International Trade, 2006, Figures 2-1-5 and 2-1-6.

system. Historically, this production linkage was ﬁrst established as the oﬀshore assembly of standardized ﬁnal goods, while high-quality products and intermediate goods continue to be produced in Japan. This phase can be understood in the context of the ﬂying-geese pattern of spreading industrial development. Later, the production of intermediate goods started to grow in some assembly locations as the scale economy in production became signiﬁcant. The forward and backward input–output linkage led to agglomeration in particular locations. Most recently, the growth of local consumer markets has given rise to further agglomeration, realizing the positive feedback between market magniﬁcation and the increase in the local production of consumer goods. Thus the production of electronic products has resulted in ‘agglomerated dispersion’, or the migration of production from Japan and the NIEs to limited locations in the three Chinese agglomerations and several spots in ASEAN. The concept of agglomeration economies also partly explains some industry localization in developed countries. Table 2.3 shows that while China dominates in the production of most ﬁnal products, technological forerunners have been able to maintain a competitive position. South Korea exhibits strength in the mobile phone industry due to its

32

Basic viewpoints on East Asian economic integration

technological lead in adapting the CDMA (code division multiple access) mobile telecommunication system. South Korea developed a national project in the early 1990s involving public and private cooperation to introduce CDMA as its national standard. The country became the world’s largest digital mobile telecommunication market in the late 1990s, and the HME created a big push for its mobile phone industry (Abe 2006). The same can be said about digital still-camera production in Japan, which requires a wide variety of parts produced with sophisticated technology only available in Japan. Although price competition has forced a shift of production to China in this category, according to METI (2006) 75 percent of digital-camera production in China is done by Japanese aﬃliates. Production of semiconductors, where China’s global share is low, is an exception. Semiconductor production is divided into three stages: R&D and design (knowledge intensive), wafer production (capital intensive), and assembly and packaging (labor intensive). Japanese semiconductor producers are generally vertically integrated. Analyzing the behavior of major Japanese semiconductor ﬁrms in the locating of production, Arita and McCann (2002) and McCann and Arita (2006) characterize the spatial organization of these ﬁrms as follows: the R&D and design stage is located in major metropolitan areas because close communication with their headquarters is required; most wafer production and assembly/ packaging plants are located jointly in peripheral regions within Japan in order to facilitate close information exchange and knowledge secrecy between the two production processes; only assembly/packaging plants are located in China and Southeast Asia to meet the demands of the local plants of Japanese MNFs. South Korea is another major actor in the world semiconductor market. The spatial pattern of South Korean semiconductor producers is similar to that of the Japanese, with strong vertical integration. South Korea’s capacity for semiconductor production has been increasing strongly due to its aggressive investment strategy, while Japanese ﬁrms have lost a substantial share of the global market due to their cautious investment attitude during the prolonged stagnation of the Japanese economy. In contrast, US semiconductor ﬁrms have increased their subcontracting to oﬀshore foundries. The evolution in the locating of their production is summarized in Figure 2.5. US semiconductor ﬁrms initially moved assembly plants to East Asia, not only to reduce cost through labor cost saving, but also to locate production near customers who produced electronic products in East Asia. From the 1980s, the whole production process, including wafer production, started to be contracted to oﬀshore foundries, mainly in Taiwan. Taiwan has become one of the biggest semiconductor producers in the world due to its ﬁnancial strength and government

33

Regional integration in East Asia

1960s to 1980s 3

Assembly abroad Companies intially moved assembly, testing and packaging offshore.

Hong Kong 3

1 2

Taiwan

United States

3

3

Malaysia

1980s to 2000s China

2

Foundries abroad

3

United

1 States

Companies began contracting with offshore fabrication plants to produce water from designs

Taiwan

2

3 Philippines Malaysia 3

1 Canada

2000s to 2005 1

China 2

Design abroad

Complex global production chains developed as designs may be fabricated in different locations, and water then sent to still other locations for assembly, testing and packaging

Multicountry design teams

3

Some design services were offshored, or a part of global teams operating in many countries

Taiwan

United

1 States

1 2 3 Philippines Malaysia 3

Design teams 2 Fabrication 3 Assembly testing and packaging

1

Source: USGAO (2006), p. 10.

Figure 2.5 The outsourcing of US semiconductor manufacturing, 1960–2005 support. Thus US semiconductor ﬁrms can avoid costs related to huge investment, which enables them to concentrate their resources in R&D and design, while the Taiwanese foundries, in turn, obtain scale economies through subcontracting for many customers, including some Japanese semiconductor ﬁrms which have partly adapted the US model to achieve cost reduction. Most recently, US semiconductor ﬁrms’ R&D and design processes have also become dispersed spatially, while multi-country design teams work concurrently using advanced information technology. The semiconductor industry in China is still underdeveloped although some foundries set up using local capital are growing rapidly. This can be understood as risk-averse behavior because the ﬁnancial requirements for state-of-the-art vertically integrated plants tend to be huge. However, since China has become a major market for electronic devices, some MNEs have made large-scale investments.

34

2.4.3

Basic viewpoints on East Asian economic integration

Summary

To summarize the discussion in this section, intensifying market integration on a global scale has led to the concentration of industrial production in North America, Europe and East Asia. Some diﬀerences in the geographical distribution of activities have emerged due to the characteristics of transport costs. In the automobile industry, whose ﬁnal products are relatively harder to transport, production is distributed somewhat proportionally to the three core economies. Within East Asia, the degree of concentration in Japan has been reduced, but the dispersion process has not been ubiquitous; rather it has spread to certain particular locations under the inﬂuence of agglomeration economies. Within China, the automobile industry is growing in three locations (Tianjin, Shanghai and the vicinity of Guangdong), which have attracted foreign investment, in contrast to the traditional auto-production locations designated by the government which brought factories to inland provinces and the northeast. In Southeast Asia, Thailand’s Bangkok and nearby provinces have attracted the production of pickup trucks, for which the Thai domestic market exhibits the HME. Hence the current location pattern of the automobile industry can be summarized as: (1) globally concentrated while being distributed to three core economies; (2) regionally dispersed in East Asia; and (3) highly concentrated within each country. Regarding the electronics industry, the locational pattern at the global level exhibits a concentration in East Asia because of the dominance of agglomeration economies over dispersion forces due to lower transport costs. Dispersion of labor-intensive assembly processes started in the 1980s. Later, agglomeration economies strengthened their concentration in particular locations through input–output linkage and the magnifying eﬀect of a growing variety of consumption. Concentration in China is much more intense in the electronics industry; nevertheless, Japan, South Korea and Taiwan preserve certain market shares in products of the most sophisticated type, maintaining innovation-based agglomeration. Innovationintensive activities tend to localize because of the knowledge externalities created through the interaction of skilled people requiring some face-toface communication. An indication of innovation capacity is shown in Table 2.4, which gives the number of patent applications ﬁled by residents in each country and the share of R&D expenditure as a percentage of GDP. In both categories the table shows Japan’s predominant position in the region in innovation capacity. South Korea has achieved a remarkable increase in innovation capacity. China also shows substantial progress in patent applications, while Taiwan and Singapore maintain high ratios of R&D expenditures.

35

Regional integration in East Asia

Table 2.4

Patent applications filed by residents

China Hong Kong Indonesia Japan South Korea Taiwan Malaysia Philippines Singapore Thailand Vietnam

1991

2004

R&D expenditure (% of GDP**)

7 372 16 54 335 564 13 253 106 147 80 37

65 586 127 226 362 342 105 027 16 747 522 157 509 819 69*

1.31 0.6 0.06 3.15 2.64 2.02 0.69 0.11 2.15 0.24 0.19

Notes: * Data of 2002. ** Average of 2000–2003. Source: WIPO, Vietnam NOIP and World Bank, World Development Indicators, 2006.

Although it is maintaining its leading position in innovation, indications suggest that Japan as the single core of innovation in the region will not continue; rather the Asian NIEs and China will steadily increase their own innovative capacities, implying the consolidation of a regional structure with multiple cores. In contrast, the innovative capabilities of the ASEAN countries are still lagging behind.

2.5

AGGLOMERATION AND REGIONAL INEQUALITY

Figure 2.6 plots the evolution of each country’s GDP per capital (PPP – purchasing power parity) expressed proportionally to that of Japan. The graph shows the tendency of economic catching up, which appears in diversiﬁed ways. Small developed economies such as Hong Kong and Singapore have already achieved the same income level as Japan. South Korea and Taiwan are rapidly catching up, followed by Malaysia and Thailand, although the latter experienced a serious setback during the currency crisis in the late 1990s. This process of income convergence ﬁts with the ﬂying-geese pattern of catching-up industrialization. China has shown steady growth since the beginning of the 1990s, escaping from the lowest income group and already exceeding the level of the Philippines and

36

Basic viewpoints on East Asian economic integration

1.4

Hong Kong

1.2

Singapore 1

Japan = 1 Taiwan

0.8

Korea 0.6 Malaysia 0.4 Thailand China Philippines

0.2

Indonesia Lao PDR Cambodia 2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

0

Source: Penn World Table.

Figure 2.6

Income convergence in East Asia

Indonesia, which, in contrast, have experienced slackening growth over the long term. We may infer from the above that the existence of agglomeration economies helps countries to grow faster. The transformation of East Asia’s economic geography from the Japanese single-core monocentric type to a multi-cored structure has induced the convergence of incomes among some countries, but at the same time, the graph shows that the gap continues to widen between the core and non-core economies such as the Philippines, Indonesia, Lao PDR and Cambodia. As noted in Section 2.2, the concentration of economic activities appears at many geographical levels. Because there is much freer mobility of productive resources intranationally than internationally, the problem of income disparities due to agglomeration forces may be more serious in the context of each country because the core–periphery structure may become clearer.11 In their examination of China, Fujita and Hu (2001) showed that the income disparity between the coastal area and the interior increased

Regional integration in East Asia

37

during the initial stage of economic liberalization from 1985 to 1994, and industrial production showed strong agglomeration toward the coastal area, while within coastal provinces there was a trend toward convergence. Higher growth was related to production agglomeration, prompted by exposure to globalization (exports and foreign direct investment) and economic liberalization (reduction in the share for state enterprise). We can conﬁrm a continuation of this trend from the recent data (1994 and 2004) as depicted in Figure 2.7, which shows the level of per capita gross regional product (GRP) of each province as a ratio of the national average. This shows that the disparity increased as the growth of GRP in the richest areas, i.e. Shanghai, Beijing and Tianjin, was considerably greater than the national average. At the same time, other coastal provinces such as Zhejiang, Jiangsu and Shandong also showed remarkable growth, changing the order. In the previous section, it was pointed out that these provinces have had a high growth of exports as well as a large share of MNE participation in those exports. This implies that, although we can see a narrowing of the gap in the per capita income of less developed provinces when compared to the national average, the diﬀerence between the coastal core and inland periphery has widened. To compare the two extremes, the diﬀerence between the richest (Shanghai) and the poorest (Guizhou) has widened from 10 times in 1994 to 13 times in 2004. Economic growth in Thailand, depicted in Figure 2.8, also shows a clear tendency toward the strengthening of the core – periphery structure. In this case, the ‘Bangkok and vicinity’ core includes provinces in the central and eastern regions. This ﬁgure shows that in 1981 many of the provinces with per capita GRP higher than the national average were in non-core regions (expressed by white bars), including the northeast. However, in 2003 most of the provinces with income higher than the national average were in the core regions, and the number of such provinces decreased from 36 in 1981 to 14 in 2003, leaving the remaining provinces below the average. It should also be noted that the size of the income gap between the poorer provinces and the national average has widened. This core– periphery structure, which is more accentuated than that in China, may be related to the higher mobility of labor in Thailand, which magniﬁes the eﬀect of agglomeration through the backward and forward linkage eﬀect of the core region. It can be concluded from the above analysis that intensifying economic integration and the related structural changes in economic geography can generate a mix of convergence and divergence in income inequality at diﬀerent levels. First, within East Asia, some countries attract industries while others do not: the former tend to grow faster while the latter remain at the economic periphery. Second, within each country, industrial

38

Basic viewpoints on East Asian economic integration

5 4 3

1994

2004

2 1 0

Shanghai Beijing Tianjin Guangdong Zhejiang Liaoning Jiangsu Fujian Hainan Shandong Heilongjiang Xinjiang Jilin Hebei Hubei InnerMongo Qinghai Shanxi Guangxi Hunan Ningxia Anhui Sichuan Yunnan Henan Jiangxi Shaanxi Tibet Gansu Guizhou Chongqing

National average = 1

6

Source: Authors’ calculation based on National Bureau of Statistics of China, China, Statistical Yearbook, 1995 and 2005 volumes.

Figure 2.7

China’s regional income disparity

agglomeration occurs in a limited spatial range, sharpening the regional contrast between the core and the periphery, while the income gap within the core may be narrowed due to the sprawl of agglomeration economies.

2.6

CONCLUDING REMARKS

Spatial economics studies the formation of industrial agglomeration as a consequence of positive feedback based on the interaction between the scale economy and transport costs, in the presence of the diversity of consumption, intermediate goods and people. Its fundamental proposition suggests that a decrease in transport costs may lead to a core–periphery structure of the economy in the location of industrial production. To a large extent the process of economic globalization meets the prediction of the spatial economic perspective. At the global level it coincides with the strong concentration of industrial production in North America, Europe and East Asia, which constitute the global core economies. Within East Asia, economic integration has developed through increasing specialization in each country in diﬀerent technological intensities and through intra-regional trade, taking advantage of the diversity of development stages within the region. This has led to dispersion of production from Japan, which had played the leading role in the traditional ﬂying-geese pattern of economic development. However, due to agglomeration economies, that dispersion spread to some countries and, within them, to a limited number of areas. This resulted in income convergence among regions gaining agglomeration forces, while leaving untouched or even widening the gap between the core

0

Figure 2.8 Samut Bangkok Pathum Phangnga Ranong Chon buri Nakhon Khon kaen Phuket Udon thani Loei Ubon Mukdahan Samut Chaiyaphum Kanchanabu Sakon Nong khai Nonthaburi Buri ram Kalasin Nakhon Roi et Maha Am nat Rayong Surin Yasothon Saraburi Si sa ket Nong bua Phachuap Nakhon Trat Singburi Chachoengs Ratchaburi Phetchaburi Songkhla Chai nat Ang thong Chanthaburi Lop buri Chiang mai Suphan buri Yala Trang Phra nakhon Uthai thani Chumphon Lampang Tak Mae hong Nakhon Surat thani Satun Nakhon Prachinburi Sukothai Krabi Phitsanulok Uttaradit Phichit Narathiwat Kam phaeng Phetchabun Samut Chiang rai Lamphun Nakhon si Phatthalung Phrae Nan Pattani Phayao

0

Rayong Samut sakhon Samut prakan Phra nakhon Chon buri Bangkok Pathum thani Phuket Chachoengsao Saraburi Prachinburi Lamphun Nakhon Ratchaburi Songkhla Phangnga Nonthaburi Surat thani Phachuap khiri Krabi Phetchaburi Lop buri Singburi Kam phaeng Satun Ranong Trang Kanchanaburi Trat Chumphon Chiang mai Nakhon si Samut Pattani Ang thong Nakhon nayok Chai nat Nakhon sawan Suphan buri Yala Khon kaen Phitsanulok Chanthaburi Lampang Tak Phichit Nakhon Uthai thani Uttaradit Phatthalung Narathiwat Sa kaeo Sukothai Phrae Udon thani Phetchabun Loei Nan Chiang rai Phayao Ubon Mae hong son Kalasin Nakhon Chaiyaphum Maha Sakon nakhon Roi et Yasothon Nong khai Mukdahan Buri ram Si sa ket Surin Am nat Nong bua lam

39 Baht

Baht

70000

60000

1981

30000

20000

700000

200000

Thailand’s regional income disparity

Bangkok and vincity

50000

Central and eastern regions

40000 National average

10000

600000 2003

500000 400000

300000

100000 National average

Source: Authors’ calculation based on National Statistical Office of Thailand, National Income Statistics.

40

Basic viewpoints on East Asian economic integration

and the periphery. These developments in the East Asian economy point to the following possible policy implications and future research issues: (1) How to promote further regional economic integration Economic integration has brought about the emergence of multiple economic cores. Local economies can take advantage of economies of scale by integrating themselves into the global market. The current eﬀorts to transform de facto integration into de jure integration by negotiating one regionwide free trade agreement should be prioritized. In this connection, policies on transportation infrastructure for both domestic and international trade should be emphasized, which will require eﬀective international cooperation for underdeveloped regions. (2) How to promote harmonious integration while mitigating negative eﬀects The direct implication of spatial economic models for economic integration is the formation of a core–periphery spatial pattern, involving the ‘sucking eﬀect’ of economic activities toward the core. Allowing the beneﬁts of regional integration to be excessively concentrated in a small number of regions may have negative eﬀects both economically and politically. Economically, it may cause too much congestion, leading to degradation of the quality of life in large metropolitan areas. The widening of income disparity between the core and periphery, in turn, may result in political conﬂicts. Income transfer policies may mitigate such problems, but permanent dependence on subsidies in underdeveloped regions may trigger ﬁscal-burden fatigue and discontent among urban residents. In the long run, technological development will be necessary in peripheral areas to enable self-support. Such a policy analysis is another issue that spatial economic models will have to deal with. (3) How to build up the capacities of underdeveloped countries/regions Underdeveloped countries/regions may fail to be integrated into the global/regional market due to lack of policy-making capacity and eﬃcient institutional management. Regional cooperation among researchers, policy-makers and public oﬃcers to promote capacity-building will contribute to overcoming such constraints and promote self-support.

NOTES 1. 2.

Industrial localization involving ﬁrms with strong technological or informational linkages may give rise to specialized medium-size cities. It is worth noting that this is diﬀerent from the traditional location theory based on neoclassical economics. In the latter, if ﬁrms were to choose the location where labor is

Regional integration in East Asia

3. 4.

5. 6. 7. 8. 9. 10. 11.

41

cheaper, ﬁrms would relocate whenever income growth raised production costs. Recent progress in information and telecommunication technology raises this possibility. Head and Mayer (2004) state that the following spatial economic propositions can be examined in empirical studies: (1) market potential raises local factor prices; (2) market potential induces factor inﬂows; (3) there is a home market eﬀect; (4) trade induces agglomeration; and (4) there is a shock sensitivity (lock-in eﬀect). This part on intra-regional trade is based on Fujita (2005) and Fujita and Hamaguchi (2006). Although the USA and the rest of the world also export their products to East Asia, this is omitted in Figure 2.3 in order to emphasize that East Asia as a whole has developed an integrated export platform that ships to the world. China Statistical Yearbook 2005, National Bureau of Statistics of China. See Chapter 3 for a detailed explanation of the HME. Hyundai acquired Kia in 1998. The two companies together account for more than 70 percent of Korea’s automobile exports as well as domestic market sales. See Chapter 9, which examines how Japanese automobile companies choose locations in China compared to Korean companies, discussed in Chapter 8. Baldwin and Wyplosz (2003) emphasize this problem in European integration.

REFERENCES Abe, Makoto (2006), ‘Kankoku keitai denwa tanmatsu sangyo no seicho’ (‘Growth of Korea’s mobile handset industry’), in Ken-ichi Imai and Momoko Kawakami (eds), Higashi Ajia no IT Kiki Sangyo (The information technology equipment industry in East Asia), Chiba, Japan: Institute of Developing Economies, JETRO, pp. 17–53. Anderson, James E. and Eric van Wincoop (2004), ‘Trade costs’, Journal of Economic Literature, 42, 691–751. Arita, Tomokazu and Philip McCann (2002), ‘The spatial and hierarchical organization of Japanese and US multinational semiconductor ﬁrms’, Journal of International Management, 8, 121–39. Baldwin, Richard and Charles Wyplosz (2003), Economics of European Integration, New York: McGraw-Hill. Baldwin, Richard, Rikard Forslid, Philippe Martin, Gianmarco Ottaviano and Frederic Robert-Nicond (2003), Economic Geography and Public Policy, Princeton, NJ: Princeton University Press. Fan, C. Cindy and Allen J. Scott (2003), ‘Industrial agglomeration and development: a survey of spatial economic issues in East Asia and statistical analysis of Chinese regions’, Economic Geography, 79, 295–319. Fujita, Masahisa (2005), ‘Development of East Asian regional economies’, Institute of Developing Economies, JETRO, mimeo. Fujita Masahisa and Nobuaki Hamaguchi (2006), ‘The coming age of China-plusone’, background paper for the World Bank project, ‘Dancing with Giants’ (available at the World Bank website). Fujita, Masahisa and Dapeng Hu (2001), ‘Regional disparity in China 1985–1994: the eﬀects of globalization and economic liberalization’, The Annals of Regional Science, 35, 3–37. Fujita, Masahisa, Paul Krugman and Anthony Venebles (1999), The Spatial Economy: Cities, Regions and International Trade, Cambridge, MA: MIT Press. Fujita, Masahisa, Tomoya Mori, J. Vernon Henderson and Yoshitsugn Kanemoto

42

Basic viewpoints on East Asian economic integration

(2004), ‘Spatial distribution of economic activities in Japan and China’, in J.V. Henderson and J.F. Thisse (eds) (2004), Handbook of Regional and Urban Economics: Cities and Geography, vol. 4, Amsterdam: Elsevier, pp. 2911–77. Head, Keith and Thierry Mayer (2004), ‘The empirics of agglomeration and trade’, in J.V. Henderson and J.F. Thisse (eds) (2004), Handbook of Regional and Urban Economics: Cities and Georgraphy, vol. 4, Amsterdam: Elsevier, pp. 2609–69. METI (2006), White Paper on International Economy and Trade 2006, Tokyo, Japan: Ministry of Economy, Trade and Industry. McCann, Philip and Tomokazu Arita (2006), ‘Clusters and regional development: some cautionary observations from the semiconductor industry’, Information Economics and Policy, 18, 157–80. United States Government Accountability Oﬃce (2006), Oﬀshoring: U.S. semiconductor and software industries increasingly produce in China and India, Washington, DC: US Government Accountability Oﬃce.

3.

Specialization and agglomeration forces of economic integration Koji Nishikimi

3.1

INTRODUCTION

This chapter, illustrated with graphs and diagrams, makes clear the theoretical basis of our main view as presented in the preceding chapters. As was argued in Chapter 1, economic integration is expected to exert two opposing influences on industrial location. On the one hand, international difference in comparative advantage leads different industries to locate in different countries. As comparative advantage structure changes over time, industries trickle down from one country to another, in conformity with the flying-geese pattern of industrialization as described in Chapter 2. Here we define this influence as the ‘specialization force of economic integration’. As a result of the operation of this force, industries tend to be dispersed over many countries, rather than being localized in a small number of countries and regions. On the other hand, when scale economies of production are significant, firms tend to locate in regions with large markets in order to exploit scale benefits. This is likely to lead to industrial agglomeration in a limited number of regions, leaving other regions vacant. Hence, with significant scale economies, trade liberalization and factor mobilization tend to intensify geographical concentration of industries and regional inequality. We call this influence the ‘agglomeration force of economic integration’. In the following sections, we look more closely at the above two forces. The analysis of the specialization force is carried out on the basis of the neoclassical theory of comparative advantage, in which production processes are assumed to exhibit constant returns to scale. We begin our analysis by reviewing a simple standard model in which two consumer goods are produced from two factors of production. We then introduce the possibility of fragmentation in order to consider situations in which firms separate their production into several blocks and relocate them to other countries. It will be shown that either with or without fragmentation, trade liberalization promotes efficiency in the international allocation of resources and brings about significant gains from trade. By contrast, the 43

44

Basic viewpoints on East Asian economic integration

analysis of the agglomeration force is based on the theory of spatial economics, which differs from neoclassical theory in that it allows for scale economies in production. This small alteration of assumption provides a theoretical mechanism that explains how regional integration creates economic disparities among member countries as well as among regions within each country. Both specialization and agglomeration forces are expected to become more significant as integration progresses further as a result of the future creation of more FTAs. The relative potency of the two forces crucially affects regional aspects of economic growth in East Asia. For example, the strength of the forces will determine how strongly China attracts MNEs, compared with ASEAN countries, and whether CLMV (Cambodia, Laos, Myanmar and Vietnam) countries of Southeast Asia can smoothly catch up with the front-runners.1 It also crucially influences how the development of large inequalities among domestic regions will be accompanied in each country by economic growth brought about by Asian integration. Both forces must be analyzed if there is to be harmonious development of economic integration in East Asia, where much heterogeneity can be seen among countries as regards levels of industrialization, per capita income, and other socio-economic indicators. The remainder of this chapter is organized as follows. The specialization forces are examined in Section 3.2, which consists of three subsections. In Section 3.2.1, we provide a basic framework for analyzing the specialization forces. Our model here is based solely on the neoclassical theory of comparative advantage. In Section 3.2.2 we examine the effects of trade liberalization, while in Section 3.2.3 we introduce the possibility of fragmentation in the manufacturing sector and examine the effects of trade liberalization. Section 3.3 is devoted to the analysis of agglomeration forces. In Section 3.3.1, we first look at the HME, which constitutes a key source of agglomeration and regional disparities. Then, in Sections 3.3.2 and 3.3.3, respectively, we examine the effects on disparities among integrated countries and among domestic regions. Section 3.4 presents a summary.

3.2

SPECIALIZATION FORCE

The lowering of trade barriers that accompanies the process of economic integration alters the terms of trade and thereby affects the pattern of world production and specialization. In this section, we examine the effects of reductions in tariff and non-tariff barriers, and the lowering of international transport costs among integrated countries. After developing an

Specialization and agglomeration forces

45

analytical framework in Section 3.2.1, we look at these effects in Sections 3.2.2 and 3.2.3 respectively, and in respect of both cases, we look at the situation with and without the fragmentation of production processes. 3.2.1

Basic Framework

We begin by building a simple neoclassical model of comparative advantage so that we can examine how the specialization force works in the process of economic integration. Consider a world composed of a large number of small countries. All consumers are assumed to have identical preferences. Each country is endowed with certain amounts of skilled (H) and unskilled (L) labor and produces two kinds of goods, H-intensive (high-tech) goods, M, and L-intensive (traditional) goods, A, using constant-returns-to-scale technologies.2 Both kinds of labor are assumed to be mobile only across industries within a country, and the two types of labor as well as the two products are transacted in competitive markets. In these circumstances, factor endowment should be an essential determinant of the comparative advantage of each country: H-abundant countries such as Japan, Korea and Singapore have comparative advantage in production of H-intensive (high-tech) goods, M, while other L-abundant countries have advantage in L-intensive goods, A. Transport costs Let Countries 1 and 2, respectively, be well endowed with skilled and unskilled labor. In these circumstances, reflecting their comparative advantages, Country 1 exports M goods and imports A goods, while Country 2 exports A and imports M. Suppose that all international trade is transacted on the world market and necessitates transport costs between this market and each country of the importer or exporter. To keep the analysis simple, we assume the iceberg type of transport cost: if a unit of A is exported to (or imported from) the world market, only √TA units arrive. Similarly, only √TM out of a unit of shipped M reach the destination. In other words, (1 √TA ) and (1 √TM ) units melt down during transportation between a country and the world market, where 0 TA 1 and 0 TM 1. Then the c.i.f. price of A import (P1A) in Country 1 is given by 1/TA times the f.o.b. price in Country 2 (PA2), i.e. PA1 PA2/TA . Similarly, the c.i.f. price of M in Country 2 is 1/TM times as large as its f.o.b. price in Country 1. Because the world market is competitive, firms encounter completely elastic demand for their products, and hence domestic prices of exported and importcompeting goods are, respectively, equalized to the f.o.b. and c.i.f. prices of their substitutes. As a result, the relative price of M to A, PM/PA, is higher in Country 2 than in Country 1, at an equilibrium. This difference in

46

Basic viewpoints on East Asian economic integration

H B M1 M2

D

A2 A1

(H/L)

0 Figure 3.1

(w/r)1

C

(w/r)2

E

L

Equilibrium wages

product price is reflected in the relative wage in each country, via competition in domestic labor markets. Figure 3.1 shows the equilibrium difference in relative wage between Countries 1 and 2. In the figure, the curves Ai and Mi (i 1,2) show the required inputs of skilled and unskilled labor (isoquants) for producing a unit value (e.g. one dollar) worth of A and M respectively, where the values are measured in the domestic prices of A and M in each country. Note that A1 and M2 lie below and to the left of A2 and M1 respectively, because the c.i.f. price is always higher than the f.o.b. At an equilibrium, the relative wage for unskilled labor, (w/r), is given by the slope of BC in Country 1 and that of DE in Country 2.3 Note that even in equilibrium, differences in factor prices are not totally eliminated by international trade, because of the operation of transport costs. Since unskilled labor is relatively scarce in Country 1, the relative wage of unskilled labor becomes more expensive in Country 1 than in Country 2. Figure 3.2 depicts the equilibrium production and consumption in Country 1. The curve BPD represents the production possibility frontier (PPF), while the curve uCu provides an indifference curve of the representative consumer.4 The equilibrium production is given by point P, where the PPF is tangent to the trade possibility line, EF, whose slope represents the

47

Specialization and agglomeration forces

M u

E B Mp

P transport cost

export

I

Mc

C

2

1

pA /pM J u

import

0 Figure 3.2

Ap

D Ac

1 1 pA /pM

F

A

Equilibrium production and consumption

terms of trade for Country 1, PA1 /P1M. The equilibrium consumption is shown by point C, which is the tangent point of the trade possibility line and the indifference curve. Hence Country 1 exports Mp Mc of manufactured goods and imports Ac Ap of traditional goods. If we draw a line (PIJ) that passes through P and follows the slope of the f.o.b. prices, P2A P1M , then the distance between C and I represents the total transport cost incurred in the importation of good A. The equilibrium production and consumption in Country 2 can be drawn in a similar way.5 Import tax When Country 1 imposes an import tax, the price of traditional goods in that country diverges from the c.i.f. price by the amount of the tax rate, i.e. P1A Pcif tA, where Pcif A A and tA represent, respectively, the c.i.f. price and the tariff rate on A in Country 1.6 This rise in the domestic price of A creates a larger demand for unskilled labor, which is the factor that the industry producing A uses intensively. Consequently, the relative wage of unskilled labor rises when Country 1 levies an import tax on A. Figure 3.3 illustrates this effect of import tax on wage rates. First, a rise in the price of A makes the unit-value-equivalent quantity smaller than before, which shifts the unit

48

Basic viewpoints on East Asian economic integration

H B B' M1

A'1 A1

0

Figure 3.3

(w/r)'1

C'

C (w/r)1

L

Import tax and equilibrium wages

value isoquant of A toward the origin from A1 to A. 1 Then, the new equilibrium wage, (w/r)1, in the competitive market is shown by the slope of the common tangent BC to the new set of unit value isoquants, M1 and A1. Clearly, this new relative wage is higher than the wage in the case without import tax, (w/r)1, which is shown by the slope of the line BC.7 Figure 3.4 shows the welfare effects of import tax imposition. In the figure, the slope of II (and JJ) represents the relative price, P1A P1M , when an import tax is levied, while the slope of the segment PC shows the price P1M . If the tax revenue is evenly redistributed over all conwithout tax, Pcif A sumers in Country 1, the equilibrium production is given by point P where the PPF and II are at a tangent to each other, and the consumption is given by point C, the tangent point of the indifference curve uCu and JJ. Since the imposition of a tax on imports of A raises the domestic price of A, it stimulates domestic production of A (from P to P in Figure 3.4) and discourages consumption of A (from C to C), which results in decreases in both import of A and export of M. Eventually, economic welfare in Country 1 worsens (from u to u) because the tariff forces consumers to buy expensive A goods merely to allow less productive farmers (in Country 1) to survive the competition with inexpensive imports.

49

Specialization and agglomeration forces

M u'

E

u

PI J

M P' export

P'

C'

M C'

I

C J

import

0 Figure 3.4 3.2.2

A'P

A'C

u'

u

F A

Welfare eﬀect of import tax

Effects of Economic Integration

Now suppose that Countries 1 and 2 decide to promote economic integration between themselves by lowering import tax, simplifying procedures for customs clearance, and harmonizing international differences in practices of contract and payment. Major consequences of elimination of these barriers have been extensively studied in the field of international economics. In this subsection, we demonstrate their essence by using the framework presented above. Reduction of import tax First, consider the effects of a reduction of import tax. Suppose that the government of Country 1 eliminates the tax on imports of A from Country 2, while retaining the taxes on imports from other countries.8 This discriminatory tariff reduction stimulates A imports from Country 2. If the export capacity of Country 2 is so large that Country 1’s imports of A is totally substituted by imports from Country 2, then the domestic price of A in Country 1 declines to Pcif A . The equilibrium points for production and consumption shift back to P and C, respectively, in Figure 3.4.9 In such circumstances, the economic welfare of the representative consumer is

50

Basic viewpoints on East Asian economic integration

improved to the level of uCu. At the same time, the fall of A price brings about a decline of the relative wage of unskilled labor from (w/r)1 to (w/r)1 in Figure 3.3, without affecting the world prices as well as the domestic prices and wages in Country 2. This implies that the relative wage tends to converge among the integrated countries as a result of trade liberalization since w/r was higher in Country 1 than Country 2 before the tariff reduction. If Country 2’s export capacity is not large enough to substitute all A imports of Country 1, the domestic price should not fall to Pcif A but should stay at the level of the import price from countries other than Country 2. In this equilibrium, positive profits are obtained by importers in Country 1.10 If the profits are distributed over consumers (via the stock market), production and consumption in Country 1 can be given by P and C respectively, and hence economic welfare remains at the same level as the point of equilibrium before the tariff reduction. In this case, factor prices also remain unchanged. Effects of a tariff reduction in Country 2 can be examined similarly; thus a reduction of Country 2’s import tax on M from Country 1 brings about better or constant welfare for the representative consumers in Country 2. The relative wage of unskilled labor in Country 2 rises toward the wage level in Country 1 if all M imports to Country 2 are substituted by imports from Country 1. If the substitution is only partial, the tariff reduction exerts no influence on income distribution in Country 2, as in the case of a tariff reduction in Country 1. In general, we can summarize the above results as follows: if a discriminatory tariff reduction in a country causes a substantial change in the domestic product prices in that country, it reduces the difference in the factor prices among the integrated countries and makes the representative consumer in that country better off than before. Reduction of NTB and transport cost Next, let us consider the effects of lowering non-tariff barriers (hereafter NTBs). An NTB, like any complicated procedure for customs clearance, entails substantial non-pecuniary costs but yields no government revenue. Hence, in our framework, its influences should be analyzed in the same way as an examination of the effects of international transport costs, rather than the effects of import taxes. Yet, as in the case of an import tax, the result of an NTB reduction depends on the export capacity of the trading partners with which the NTB reduction is to be implemented. Figure 3.5 shows the result of an NTB reduction in Country 1 when Country 2 has sufficient export capacity. In this case, the domestic price of A in Country 1 declines as the import price drops, reflecting the lowering of the NTB.

51

Specialization and agglomeration forces

M E E'

u

B

u'

P' P X2 C' C u

u'

Y2 0 Figure 3.5

D

F

F'

A

Reduction of NTB (1)

Equilibrium production and consumption then shift to P and C, respectively, as shown in Figure 3.5, and hence representative consumers obtain a higher level of utility, uCu. The decline in A price causes a decrease in the production of A and a fall in the demand for unskilled rather than skilled labor. As in the case of tariff reduction discussed in the preceding subsection, this results in a decline of the relative wage of unskilled labor. When the export capacity of Country 2 is too limited to dominate the import market in Country 1, the domestic price of A in Country 1 is not influenced by the NTB reduction. In these circumstances, the equilibrium production remains unchanged at P in Figure 3.6. The difference between the c.i.f. price of A imported from Country 2 and that of A from the rest of the world (ROW) is obtained by importers as profit and eventually spent by consumers in Country 1. In Figure 3.6, the slope of the segment PG represents the import price from Country 2, and the distance Y2 and YR, respectively, equal the imported quantities of A from Country 2 and ROW. In equilibrium, consumption in Country 1 is given by C , and accordingly economic welfare in Country 1 improves in this case, too. However, the income distribution between skilled and unskilled labor remains the same as before. In other words, income and expenditure in Country 1 grow proportionally as a result of the NTB reduction.

52

Basic viewpoints on East Asian economic integration

M E

u u'

B P

G

X2 XR

C'' C u' u Y2 YR

0 Figure 3.6

D

F

A

Reduction of NTB (2)

In sum, promotion of economic integration either in the form of tariff decline or in the form of NTB reduction improves or at least retains the economic welfare of representative consumers in the integrated countries. At the same time, the integration causes a convergence in relative wages among the integrated countries. 3.2.3

Fragmentation

So far, we have examined the effects of economic integration, focusing on gains from trade in final products. However, in Asia, international division of labor has recently made substantial progress, particularly in the production of intermediate goods. Many enterprises in the automobile and electronics industries, for example, separate several processes of production and relocate them to different countries, according to the market conditions prevalent in each country. This phenomenon is often called fragmentation and has been intensively studied since the late 1990s. In this subsection, we incorporate fragmentation in our analysis of economic integration, following Arndt (2001), Deardorff (2001), and Jones and Kierzkowski (2001).

53

Specialization and agglomeration forces

H B M1

0

M1 D I

Z1

F

Q2

0

R Q1 Z2 J

A1 G

C (w/r)1

Figure 3.7

(w/r)2

E L

Fragmentation in M sector

Fragmentation and factor prices Suppose that a whole process of M production is divisible into two partial processes, production of components (z1) and assembly of them (z2). In Figure 3.7, vector M01 represents the input requirements to produce onedollar-equivalent M at the initial equilibrium, and this overall process can be divided into two processes, z1 and z2 (say, component production and assembly processes, respectively), which can be separately relocated to other countries. The line DE shows the unit isocost line for Country 2 and is identical to the one shown in Figure 3.1. If the M firm relocates its assembly process, z2, to Country 2, 0z2 units of z2 will cost 0Q2 0R dollars, but will cost 0Q1 0R if produced in Country 1. In other words, the firm can save Q1Q2 0R dollars of production cost by relocating the unskilledlabor-intensive process of assembly (z2) to Country 2, where unskilled labor is cheaper than in Country 1. Firms will relocate the assembly process to Country 2 if this cost-saving effect exceeds the extra costs of fragmentation, such as costs for transport of parts and components, and intra-firm communications. Let us assume that the total cost of M production with fragmentation, including costs of transport, communication and so on, is 0F0B in Figure

54

Basic viewpoints on East Asian economic integration

H

B F B'

~ Z

1

Z1

M1

A1 0 Figure 3.8

(w/r)1

G C C ' (w/r)'1

L

Fragmentation and wages in Country 1

3.8, which is less than the total cost without fragmentation, i.e. one dollar. In this case, fragmentation occurs and Country 1 specializes in production of z1 and A, while Country 2 specializes in z2 and A. Since the production process of z1 is more skilled-labor-intensive than M production as a whole, the relative wage of skilled labor rises in Country 1 as a consequence of fragmentation. Figure 3.8 illustrates this change in relative wage. As discussed above, a dollar-equivalent M costs 0F0B ( 1) of a dollar, which is equiv~ alent to z 1 units of z1.11 This yields a positive profit to the firms, which in turn increase the demand for skilled labor, in particular. Then, as a result of competition in labor markets, the relative wage of unskilled labor in Country 1 declines to (w/r)1, that is, the slope of BC. In short, fragmentation in the skilled-labor-intensive sector changes the income distribution in skilled-labor-abundant Country 1 in a direction favorable to skilled labor. In contrast, the effect on the relative wage in Country 2 depends on the factor intensity of the assembly process, z2. If z2 is skilled-labor-intensive ~ compared to A, and if z 2 in Figure 3.9 represents the M1-equivalent input ~ vector (corresponding to z 1 in Figure 3.8), then the relative wage of unskilled labor in Country 2 decreases to (w/r)2 at the equilibrium with

55

Specialization and agglomeration forces

H A M2

C A2 (w/r)'2

A ~ Z

2

0 Figure 3.9

~ Z 2' L (w/r) 2 D

Fragmentation and wages in Country 2

fragmentation. The opposite applies in the case where the M1-equivalent input vector is given by ~z 2, which is more unskilled-labor-intensive than A; then relative wage rises to (w/r)2. Welfare effects of fragmentation When the total cost of M1 under fragmentation is less than a dollar, as illustrated by Figure 3.8, the final good M can be produced more efficiently with fragmentation. Accordingly, the production possibility frontiers (PPFs) for both Countries 1 and 2 are expanded vertically as a result of the fragmentation. For example, Figure 3.10 shows the case of Country 1, which has an abundant supply of skilled labor. When fragmentation occurs, the PPF shifts from BD to BD, but this change does not affect the world prices of M and A as, according to our assumption, Country 1 has a small presence in the world market. Consequently, equilibrium production shifts from P to P, while consumption shifts from C to C. The representative consumer becomes better off as a result of the fragmentation. Similarly, the fragmentation vertically expands the PPF of Country 2 and improves the economic welfare of the representative consumer in that country, too.

56

Basic viewpoints on East Asian economic integration

M E B' B

E'

P' P C' C u' u

0 Figure 3.10

D

F

F'

A

Fragmentation and welfare

Tariff imposition and fragmentation A firm’s decision on whether or not to fragment its production process is influenced by various economic policies in each country, such as tax exemptions, government support and other attractive policies. In particular, it should be noted that an import tax on the non-fragmentable sector, A, in Country 1 may excessively encourage the fragmentation of M firms in that country. In Figure 3.11, the broken line BC represents the isocost line for one dollar in Country 1 in a situation where there is no import tax and no fragmentation in any industry. It rotates to the solid line BC when the government levies an import tax on A. If the M1-equivalent input vector under fragmentation in M industry is contained in the shaded triangle BBD, as ~ shown by z 1 in Figure 3.11,12 then the fragmentation becomes profitable under this scheme of import tax. In such a case, firms will decide to accomplish the fragmentation even though it does not, in fact, realize the efficient allocation of production factors. Such a misallocation occurs because the import tax on A depresses the relative wage of skilled labor, which in turn encourages specialization in the skilled-labor-intensive process, Z1. Similarly, an import tax on M in Country 2 might exert an undesirable effect on a firm’s decision on whether or not to employ fragmentation. The imposition of the tax makes an efficient fragmentation unprofitable if the

57

Specialization and agglomeration forces

H B' Z1 B ~

D M1

A1 A1' 0 Figure 3.11

(w/r)'1 C' (w/r) 1 C

L

Import tax on A and fragmentation

vector ~z 2 falls in the lightly shaded area, BBE, of Figure 3.12, whereas it encourages an inefficient fragmentation if the vector falls in the dark area, EDD. Hence, in general, elimination of an import tax in the process of economic integration removes these distortions in incentives for fragmentation and improves the efficiency of factor allocation in the integrated countries. Figure 3.13 depicts the welfare effects of the excessive fragmentation motivated by the imposition of an import tax on A in Country 1. When fragmentation does not occur in the M sector, points P and C respectively represent the production and consumption at the level of equilibrium without import tax, while P and C are those at the equilibrium with import tax. As in the case of Figure 3.4, imposition of an import tax lowers the economic welfare of the country. If fragmentation is achieved in the situation depicted in Figure 3.11, the PPF drops downwards to EF because in fact, the fragmentation impairs efficiency in the production of M. As a result, the production and consumption points shift to P and C respectively, which magnifies the welfare-worsening effect of the import tax. Similarly, we can show that the introduction of an import tax on M in Country 2 may cause economic welfare to deteriorate because of the distorting influence of the tax on the incentives for firms to undertake fragmentation.

58

Basic viewpoints on East Asian economic integration

H M2 M'2 B B'

E

A2 D' DL

0 Figure 3.12

Import tax on M and fragmentation

The above results assert that an import tax imposes a large burden of inefficiencies, including forgone gains from trade in final products as well as misallocation of production factors related to fragmentation. In particular, an import tax on traditional goods in industrialized countries, which often causes persistent obstruction of economic integration, might also inflict serious damages on those industrialized countries by stimulating excess fragmentation and relocation of their industries.13 To put it the other way round, lowering of trade barriers by economic integration would remove all these causes of economic inefficiency and would make representative consumers better off.

3.3

AGGLOMERATION FORCE

In the preceding section, we examined the effects of trade liberalization in the process of economic integration, focusing on efficiency in specialization and factor allocations. The main result, from this point of view, is that trade liberalization is beneficial for consumers in the integrated countries, while it creates different effects on income distribution in different countries. Basically, trade liberalization tends to reduce income differ-

59

Specialization and agglomeration forces

M

B

E P

E' P' P''

C C'

C''

u u' u''

0 Figure 3.13

F

D

A

Welfare effects of tariff and fragmentation

ences across countries, by encouraging mutual use of the advantages of individual countries. Accordingly, in this approach, there is no systematic force whereby economic integration creates or enlarges income differences among integrated countries and among regions within each country. In reality, however, regional disparities in production and income are often reported to be accelerated by integration, as was discussed in Chapter 1. In this section, invoking the framework of spatial economics, we examine the main mechanism whereby integration magnifies regional disparity.14 3.3.1

Home Market Effects

The origin of the disparity-enlarging force can be found in scale economies of production, which were set out during the explanation of our analytical view in the preceding section. Let us now introduce the factor of scale economies in the production of manufactured goods. We then assume that domestic transportation carries no cost, while international transportation entails significant cost. In these circumstances, if M firms produce the same homogeneous good, all domestic production of M should be supplied by a single firm as a result of competition. In this case, M firms operating in

60

Basic viewpoints on East Asian economic integration

countries with large markets tend to be more productive and hence more competitive in the world market. Exports bring more production and higher efficiency to those firms, and this, in turn, lowers the domestic price of M, which makes the consumers in the country concerned better off. In short, the size of the home market affects the real income (utility) of the consumers in that market. This mechanism is often called the HME. The above account of a natural monopoly with homogeneous M products gives a simple and clear picture of HMEs, but the result might seem to heavily rely on the monopolistic structure of the M market, something that is seldom observed in the real economy. Readers may wonder if HMEs can be obtained even when the market is not purely monopolistic. When manufactured goods are not homogeneous but slightly differentiated from each other, there should be multiple (or even many) producers at an equilibrium. Such situations are lucidly described by the monopolistic competition model of Dixit and Stiglitz (1977). In this model, each firm exhibits increasing returns to scale, and hence each variety of manufactured good, Mi, becomes cheaper as its home market becomes larger. When the price of Mi declines, however, consumers may spend their spare money to buy a new variety of M, rather than to increase their purchase of the existing varieties. In fact, at the equilibrium of monopolistic competition in Dixit and Stiglitz (1977), market size does not affect the production quantity of each firm but increases the number of varieties of manufactured good. In this case, however, consumers enjoy higher utility by consuming a wide variety of the M good. Accordingly, the size of market affects the real income, and the HMEs are therefore at work for many M firms, too. 3.3.2

Disparity among Integrated Countries

Let us now modify the model presented in Section 3.2.1 so as to consider the specific mechanism whereby the HMEs create regional disparities. Baldwin et al. (2003) carry out such modifications and provide several variations of models with HMEs. Nishikimi (2006) extends their footloose capital model to a three-country case and examines how integration affects production and income disparities across countries.15 Here we incorporate the essential features of these analyses in a three-country footloose-capital model.16 Basic settings Consider a world composed of three countries, 1, 2 and ROW, of which Countries 1 and 2 are forming an economic integration as shown in Figure 3.14.17 Each country is endowed with given amounts of labor, L, and capital, K, which is owned by L. Suppose that K is internationally mobile without cost, while labor is immobile across countries. There are two industries, tra-

Specialization and agglomeration forces

61

ROW T M'

T M' > TM

T M'

Country 2

Country 1 TM

Economic integration Figure 3.14

Three-country model of economic integration

ditional agriculture (A) and manufacturing (M) with scale economies. The A sector produces homogeneous goods, while the M sector supplies differentiated products, Mi. For the sake of simplicity, we assume that all varieties of M are produced using the same technology which requires labor in proportion to the output and a fixed amount of capital. Note that this type of technology exhibits scale economies: average cost decreases as production increases, due to the existence of a fixed input. Production of the homogeneous A good, by contrast, is assumed to require only the proportional input of labor and hence exhibits constant returns to scale. As in Section 3.2, we assume the iceberg type of cost structure for international transport whereby M: (1TM) 100 percent of the shipment melts during transport between Countries 1 and 2, while (1TM) 100 percent melts between ROW and Country 1 or 2, as shown in Figure 3.14. Here we assume TM TM since integration is progressing between Countries 1 and 2. To keep the analysis simple, costs of international transport of A and any domestic transport, in contrast, are assumed to be zero. Markets of K, L and A are in perfect competition, whereas the M market is characterized by Dixit–Stiglitz monopolistic competition. Following Baldwin et al. (2003), we assume that the world capital market is completely integrated, so that total profit is distributed among countries in proportion to their endowment share of capital.18 It should be noted here that the three specific characteristics of the M industry, namely scale economies,

62

Basic viewpoints on East Asian economic integration

monopolistic competition and significant transport costs, create HMEs, which lead M firms to agglomerate in a country with a large market, as briefly described in the previous section. Equilibrium Let us now look at the equilibrium location of M firms among the three countries. Let us first consider a symmetric case in which the three countries are endowed with identical quantities of labor and capital. For this world economy, there exists a stable equilibrium.19 At a diversified equilibrium,20 all the prices of products (f.o.b. prices) and factors are equalized across countries.21 Note especially that despite the international immobility of labor, wages in the three countries are equalized since the agricultural wage is internationally equalized through competition in the A sector. Figure 3.15 shows the equilibrium for the three countries with symmetric factor endowment. Under the simplified settings presented above, we can separately draw the figures of the trade equilibrium within the integrated economy (right-hand side of Figure 3.15) and between it and ROW (lefthand side). In the left figure, the horizontal axis represents the total share of the two integrated countries, Countries 1 and 2, in world expenditure (s12 E ), while the vertical axis measures the share of the M varieties pro22 duced in those countries (s12 n ). The kinked line NN shows the values of 12 12 sE and sn that equalize the profit rates in the three countries, while line EE shows the equilibrium income share for Countries 1 and 2. Note that EE is drawn vertically at s12 E 23 because, by assumption, the two countries are endowed with two-thirds of the world’s capital and labor, and both wages and capital revenues are equalized across countries. Curve NN appears in linear form, again because of the setting of independent national income noted above. The curve slopes upward as capital is attracted to the countries with large income and markets. Moreover, it should be noted that because of HMEs, the share of M production disproportionately increases as the relative size of the domestic markets increases; hence the slope of NN is larger than 45. The right side of Figure 3.15 illustrates the relation between the expenditure share (s1E ) and the M-production share (s1n) for Country 1. As in the left side of the figure, the kinked line nn shows the values of s1E and s1n which equalize the operating profits in Countries 1 and 2,23 and the line ee depicts the expenditure share of Country 1.24 The equilibrium is given by points E and E. Because ROW is relatively separated from Countries 1 and 2 (TM TM) by assumption, its equilibrium share of M production should be smaller than one-third (1 s*12 13). Countries 1 and 2, by contrast, hold equal shares (s*12 2 13) of M production since they are completely symmetric.

63

Specialization and agglomeration forces

s 1+2 n

s 1n

1

EE s

ROW n

ee

E

a

nn

A

s 2n a/2

E'

NN 0

45º

Figure 3.15

b 2/3

1

s 1+2 E

0

1/3

2/3

s 1E

Equilibrium for the symmetric case

Influences of trade liberalization Let us now examine how trade liberalization between Countries 1 and 2 affects the equilibrium obtained above. In Section 3.2, we have seen that trade barriers and international transport costs exert similar effects on equilibrium prices. Here, we focus on the analysis of the influences of changes in transport costs. Suppose that the transport cost between Countries 1 and 2, TM, falls because of progress in economic integration. This lowers the c.i.f. price of each variety, Mi, and expands the M market in the integrated countries. As a result, the line NN0 in Figure 3.16 shifts up to NN1, and this increases the total share of Countries 1 and 2 in the world production of M from s*12 to s** 12. By contrast, the effects on the relative share of M production within the integrated economy are a little more complicated. On the one hand, a fall of TM intensifies the HME, which makes the nn0 line steeper. On the other hand, the new nn line has to pass the point (1/3, s** 12 2) because Countries 1 and 2 are completely symmetric when s1E s2E 13. Therefore, the line nn0 rotates anticlockwise to nn1 in the right side of Figure 3.16. The equilibrium points E0 and E0 shift up to E1 and E1 respectively, and hence a fall of the transport cost between Countries 1 and 2 stimulates relocation of M firms to Countries 1 and 2. Figure 3.17 depicts the relationship between the degree of trade liberalization and the distribution of M production among the three countries, in circumstances where all the countries have an identical factor endowment. The degree of trade liberalization is measured by TM, where 0 TM 1 and

64

Basic viewpoints on East Asian economic integration

s 1+2 n

1

sn

1 a1 a0

ee'

ee''

EE E1

nn 1 nn 0

E0 a 1/2 a 0/2

NN 1 NN 0 0 Figure 3.16

1+2

2/3

1

sE

0

E 1' E 0'

E 0'' E 1''

1

1/3

2/3

sE

Effects of a fall in TM

TM approaches to 1 as trade costs decrease. When TM TM (in the left side of Figure 3.17), the three countries are situated in identical conditions and hence produce the same number of M varieties (s1n s2n sROW ). As trade n liberalization progresses between Countries 1 and 2, these two countries gain larger shares of M production because of the HME, and consumers in these countries enjoy higher economic welfare because they can purchase their domestic products at lower prices (without bearing transport costs). Now let us consider a case where the two integrated countries differ in size. Suppose that Country 1 is much smaller than Country 2, while the total share of the two countries remains at 2/3. Then s1E may be given by the line ee in the right side of Figure 3.16, while the left side is unchanged. In this case, trade liberalization between Countries 1 and 2 shifts the equilibrium E0 down to E1, and thus Country 1 loses its share of M production. As integration progresses, the difference in the share of M production within the integrated economy becomes wider, as shown in Figure 3.18. We can verify that economic welfare is higher in Country 2 than Country 1 because consumers in Country 1 purchase a larger number of imported M at higher prices (the Dixit–Stiglitz version of the HME). Finally, let us consider the effects of an import tax on agricultural goods. An import tax on A in Country 1 raises the domestic price of A and the wage in that country. This wage rise on the one hand increases nominal

65

Specialization and agglomeration forces

sn 1

2/3

s nROW

s 1+2 n

s 2n

1/2

s 1n

1/3

s 1n 0 T M' Figure 3.17

1

TM

Transport cost and production shares (symmetric case)

income in Country 1, but on the other hand pushes up the price of Mi and reduces the country’s share of M production. Consumers in Country 1 encounter rises in the prices of A and domestic products of Mi, so that economic welfare deteriorates. In sum, the model in this subsection shows how economic integration produces disparity among countries in the situation where the HME works significantly. Under the influence of the HME, producers at a site near a large market can enjoy a greater profit, which attracts more capital to that site. This implies that economic integration is likely to attract world capital to the countries forming integration, by creating a large integrated market in their vicinity. However, the magnitude of this attractive force is positively (and disproportionately) related to the size of the domestic market (the real home market) of the integrated countries. Accordingly, if the difference in the domestic market size is too big, economic integration will result in contraction of the manufacturing sector (with scale economies) in the small member countries and disproportionate growth in the manufacturing sector of the large members. This phenomenon is sometimes called the ‘straw effect’ since it is as though the large countries are sucking advantages through a straw from the small.

66

Basic viewpoints on East Asian economic integration

sn 1

s nROW

s 1+2 n

2/3

s 2n

0 Figure 3.18

3.3.3

s 1n T M'

TM 1

Transport cost and production shares (small s1E)

Disparity among Domestic Regions

We now move the focus of our discussion to disparity among domestic regions within each country. Our main question here is whether a reduction in trade costs among the integrated countries intensifies the uneven distribution of industries between urban and rural areas within each country. To examine this, we extend the model in the preceding section so that Country 1 is composed of two regions. Extension of the model Let us assume that three countries, 1, 2 and ROW, have identical endowments of labor and capital, as shown in Figure 3.15. It is further assumed that Country 1 is composed of two regions, U and R, and that all international ports are located in U, while R is situated deep inland. In these circumstances, all of Country 1’s international trade is carried out through region U, as depicted in Figure 3.19. Transport of M between U and R entails the iceberg type of cost, TM, as well as the international transport costs, TM and TM. In contrast, to keep the analysis simple, we assume that domestic transport in the other two countries is negligible, with no segregation of regions in those countries.25 The transport cost of A good is

Specialization and agglomeration forces

67

ROW

T M'

T M'

T M' Region U1 Country 2

T M'' Region R 1 Country 1

Economic integration Figure 3.19

Two-region extension

assumed to be zero, as was assumed in Section 3.3.2. As regards factor mobility, as in Section 3.3.2, we assume that capital moves without cost across countries, while labor is mobile only within each country. In Country 1 in particular, labor moves across industries (A and M) and across domestic regions (U and R), pursuing higher utility. Together with the assumption of zero transport cost on A good, this ensures that at the level of equilibrium, Country 1 produces A good only in the region where the market wage is lower. As previously discussed, region U has good accessibility to international markets and hence attracts both M firms and consumers. If there is no centrifugal force, all M firms and M labor are concentrated at U at an equilibrium. As a force that diminishes the attractiveness of region U, we introduce congestion diseconomies on consumers’ preference: consumers gain less satisfaction as population increases in their region of residence.26 If this effect is sufficiently large, there is an equilibrium distribution of population with which consumers in U and R attain the same level of utility. Figure 3.20 shows an equilibrium population distribution between U and R in Country 1. In the figure, the horizontal axis represents region U’s share

68

Basic viewpoints on East Asian economic integration U

R

w /w

V1

D0

D1

V0

1

sn

0

1

s1n G0 D0

G1

1

V1 0 Figure 3.20

V0

U

½

1 0

1

U

R

L /(L +L )

Equilibrium with regional disparity

of population in Country 1, LU/(LU LR), and the vertical axis shows the wage ratio of region U to R, wU/wR. Curve V0V0 shows the supply of M labor under a given share of capital invested in Country 1, s1n. Note that when U and R have the same population [LU/(LU LR) 1/2], VV passes below wU/wR 1, because of the good accessibility of U to international markets. An increase of M labor (and M firms) in region U, on the one hand, is accompanied by an increase of M production in U, which lowers the average retail price of M and raises the real income there. On the other hand, it lowers the real income due to congestion diseconomies. Hence, if the latter effect is sufficiently large, labor requires a higher wage in U than in R, which causes curve V0V0 to slope upward, as shown in Figure 3.20. In contrast, curve D0D0 depicts demand for M labor in region U, which shows the wage and labor demand that equalize the operating profits in U and R. The curve has a negative slope as a lower wage in U increases profit and labor demand in that area, but it becomes flat at a wage level higher than wR because of the location advantage of region U over R. 13, Suppose that V0V0 and D0D0 represent the case of s1E s2E sROW E which corresponds to the equilibrium given by E0 and E1 in Figure 3.15.27 The short-run equilibrium, which applies to all domestic markets in Country 1 under a given allocation of international capital (s1n a3), is

Specialization and agglomeration forces

69

given by the intersection of the two curves, G0 in Figure 3.20.28 At the equilibrium, more than half of the population resides in U, and accordingly the nominal wage in U exceeds that in R, i.e., w1U w1R. Since A good is produced in the low-wage region, R, and labor is freely mobile across industries and regions, there exist following relations among equilibrium regional wages. w1U w1R w2 wROW This wage difference makes the M price higher in U and reduces demand for U’s M products, which causes a decrease in Country 1’s share of M production. As a result both D0D0 and V0V0 shift leftwards, as shown in Figure 3.20. In consequence of this capital reallocation, the long-run equilibrium is obtained at G1, where profits in all countries and regions are equalized. Note that in our model, uneven distribution of M firms between U and R does not mean income inequality between them because, at an equilibrium, consumers in the two regions enjoy the same level of utility (real income). However, a rise of nominal wage in U causes capital to escape from Country 1, which, ceteris paribus, makes all consumers in both regions of Country 1 worse off. Influences of trade liberalization Suppose the transport cost between Countries 1 and 2 falls. This brings about an expansion of the export potential of Countries 1 and 2, which attracts more M firms to the integrated economy and raises s12 n . If the market size in Country 1 is not as small as ee in Figure 3.16, this expansion of the export potential causes capital inflows to Country 1. As a result, both VV and DD shift to the right, and more M firms and labor come to locate together in region U. In this case, a decrease in international transport cost leads to the sharpening of the polarization of manufacturing within Country 1. On the other hand, if the market size of Country 1 is very small, as shown by ee in Figure 3.16, the same fall in the transport cost results in dispersion of M firms and labor within Country 1, but at the same time it leads more M firms to relocate from Country 1 to 2. In other words, industrial polarization is moderated within a country but intensified across countries. Finally, let us examine the effects of an import tax on agricultural goods in Country 1. An import tax on A raises the equilibrium wage in both R and U, which raises the price of M products in the two regions. This causes outflows of capital from Country 1 and moves VV and DD curves to the left. As a result, the import tax moderates the polarization in Country 1, by increasing agricultural income in the rural area. However, at

70

Basic viewpoints on East Asian economic integration

the same time, it dampens the competitiveness of the M sector and stimulates de-industrialization of Country 1. This makes all consumers in Country 1 worse off. To summarize, when domestic regions are situated with different degrees of accessibility to international markets, urban agglomeration tends to be formed at a location with good accessibility. In this situation, a decrease in international trade cost brings greater attractiveness to the urban locations, relative to the less accessible rural areas. Accordingly, it is expected that economic integration, which accompanies trade liberalization, intensifies the economic disparity among domestic regions.

3.4

CONCLUSION

In this chapter, we have examined the effects of trade liberalization on aggregate gains from trade and distribution of the gains among countries and domestic regions. The analyses from the viewpoint of comparative advantage show that trade liberalization, by lowering tariff and/or nontariff barriers, generally encourages the efficient use of resources in the integrated countries, both among and within countries, with or without fragmentation. The imposition of a tariff causes economic welfare to deteriorate, not only by raising retail prices but also by wrongly motivating the fragmentation. In particular, an import tax on traditional (nonfragmented) industry tends to stimulate excessive fragmentation in Hintensive manufacturing sectors. Protection of traditional sectors (such as agriculture) in developed countries harms the competitiveness of their advanced manufacturing sector as well as hampering the progress of economic integration. We have also examined how economic integration affects regional disparities across countries and domestic regions, using simple models of spatial economics. In both models, the HMEs provide the main forces that create inequalities. Industries with scale economies exhibit higher production efficiencies when they can gain access to larger markets. Hence, if the capital market is efficient enough, investment tends to be concentrated in the regions/countries that have good access to a large market. If several countries form an integrated economy, they come to share a large common market, and this entices more capital and business activities to the member countries. However, this same mechanism might cause problems by increasing inequalities across domestic regions. Accessibility to international markets is generally different among regions, due to the geographic nature and/or the structure of transport networks (such is often the case, for example, between coastal and inland regions). In such cases, interna-

Specialization and agglomeration forces

71

tional economic integration might magnify the disparities among those regions. Regional integration has to be promoted with special attention to the need to achieve harmonization with the national interests of individual countries.

NOTES 1. 2.

3. 4.

5.

6. 7. 8. 9. 10.

11. 12. 13. 14. 15.

The CLMV countries are Cambodia, Republic of Laos, Myanmar and Vietnam, which joined ASEAN in the second half of 1990s. Their average per capita GDP in 2003 was US$680, whereas the average for all ASEAN countries was US$3166. Our model includes skilled labor (human capital) rather than physical capital, which has been traditionally treated as a typical production factor. This is because physical capital becomes highly mobile across countries, so that it hardly determines the comparative advantage of a country. In the model of agglomeration forces presented in Section 3.3, we introduce physical capital as a production factor. Note that the lines BC and DE represent isocost curves, i.e., any combination of L and H on the line incurs the same costs. Hence, at an equilibrium, the unit-value isoquants, Ai and Mi, must be tangent to the same isocost line. The PPF shows possible combinations of products which can be produced under full employment of all production factors. An indifference curve shows the consumption bundles which give the consumer the same level of satisfaction (utility). For details, see various standard textbooks on microeconomics. Since Country 2 has an abundant supply of unskilled labor, the PPF of that country is somewhat flattened toward the A axis. As a result, for Country 2, the production point P should lie to the lower right of the consumption point C, which implies that Country 2 exports A and imports M, paying a certain amount of transport cost. This is because Country 1 is sufficiently small relative to the size of the world market, by assumption. This result is usually presented as the Stolper–Samuelson theorem. Here we limit our examination to the effects of a tax reduction on Country 1’s imports of A. But we can easily show that a reduction of tariff on M in Country 2 causes similar effects on production, consumption and economic welfare. Economic integration is not necessarily formed by two countries. The price effect of a tariff reduction depends on the total export capacity of all the partner countries in the integrated economy. This results from intensive competition among A producers in Country 2. If they exercise market power in the market of Country 1, a part of the benefits from the tariff reduction should be transferred to Country 2. If the export price rises greatly, economic welfare in Country 1 might worsen. Fragmentation does not affect the world prices because both Countries 1 and 2 are sufficiently small. Accordingly, with no change in trade costs, the domestic price of M in Country 1 also remains unchanged. Such a situation is likely to appear if the additional requirements for the fragmentation (e.g. communication and coordination works) are skilled-labor-intensive. The discussion about the effects of import tax in this subsection gives the analogical conclusions on the effects of NTB and international transport costs which affect the domestic prices of M and A. For details of spatial economics, see for example Fujita et al. (1999), Baldwin et al. (2003), and Fujita and Thisse (2003). The footloose capital model of Baldwin et al. (2003) is presented as a simplified version of the core–periphery model by Fujita et al. (1999). In both models, the main force causing disparity is exerted by the HME, but the former model features no circular

72

16. 17. 18. 19. 20. 21.

22. 23. 24. 25. 26. 27. 28.

Basic viewpoints on East Asian economic integration causality and hence never yields catastrophic agglomeration as in the latter model. See Baldwin et al. (2003) for details. In this chapter, we do not consider fragmentation caused by scale economies, an aspect that Krugman and Venables (1995), Harris (2001) and Amiti (2007) examine in detail. See Chapter 8 for a literature survey on fragmentation by MNEs. For a model with HMEs, the number and size of countries must be predetermined, unlike the linear model in Section 3.2, because market size crucially affects regional distribution of production and income at equilibrium. This assumption may be too extreme for the current status of capital market in Asia, even though globalization and integration have rapidly progressed in that market. Technically speaking, stability of equilibrium depends on the assumption of adjustment dynamics. See Baldwin et al. (2003) and Nishikimi (2006) for details. Diversified equilibrium is defined as equilibrium at which both A and M are produced in all countries. Throughout Section 3.3, we assume the existence of the interior equilibrium. Since transport of A is undertaken at no cost, perfect competition makes the price of A equalized across countries, which equalizes the wages in all countries. This, in turn, leads M firms in the Dixit–Stiglitz type of monopolistic competition to charge the same prices. Finally, the profit rate must be equalized across countries; otherwise, investors will shift their capital to the country with a higher profit rate. By assumption, each country’s share of M varieties is equal to its share of capital investment. The slope of the line nn also shows that HME works. By assumption, lines NN and EE can be drawn independently of nn and ee. For details, see Baldwin et al. (2003) and Nishikimi (2006). The difference in domestic transport costs may be caused by differences in the development level of domestic transport systems and by the geographical configuration of the country. In fact, there are several causes of congestion diseconomies, such as rise of rents and extension of commuting distance. Strictly speaking, we cannot draw such simple figures as Figure 3.16 for the case of a country composed of two domestic regions. We assume a particular process of market adjustment where domestic labor migration occurs first and then international capital moves. It should be noted that stability of equilibrium depends on the dynamics, as discussed in note 19.

REFERENCES Amiti, M. (2005), ‘Location of vertically linked industries: agglomeration versus comparative advantage’, European Economic Review, 49, 809–32. Arndt, S.W. (2001), ‘Offshore sourcing and production sharing in preference areas’, in S.W. Arndt and H. Kierzkowski (eds), Fragmentation: New production patterns in the world economy, Oxford: Oxford University Press, pp. 76–87. Baldwin, R., R. Forslid, P. Martin, G. Ottaviano and F. Robert-Nicoud (2003), Economic Geography and Public Policy, Princeton, NJ: Princeton University Press. Deardorff, A.V. (2001), ‘Fragmentation across cones’, in S.W. Arndt and H. Kierzkowski (eds), Fragmentation: New production patterns in the world economy, Oxford: Oxford University Press, pp. 35–51. Dixit, A.K. and J.E. Stiglitz (1977), ‘Monopolistic competition and optimum product diversity’, American Economic Review, 67, 297–308. Fujita, M. and J.-F. Thisse (2002), Economics of Agglomeration: Cities, industrial location, and regional growth, Cambridge: Cambridge University Press.

Specialization and agglomeration forces

73

Fujita, M., P. Krugman and A. Venables (1999), The Spatial Economy: Cities, regions and international trade, Cambridge, MA: MIT Press. Harris, R.D. (2001), ‘A communication-based model of global production fragmentation’, in S.W. Arndt and H. Kierzkowski (eds), Fragmentation: New production patterns in the world economy, Oxford: Oxford University Press, pp. 52–75. Jones, R.W. and H. Kierzkowski (2001), ‘A Framework for fragmentation’, in S.W. Arndt and H. Kierzkowski (eds), Fragmentation: New production patterns in the world economy, Oxford: Oxford University Press, pp. 17–34. Krugman, P.R. and A.J. Venables (1995), ‘Globalization and the inequality of nations’, Quarterly Journal of Economics, 110: 857-880. Nishikimi, K. (2006), ‘Economic integration and regional disparities’, paper presented at the International Workshop on East Asian Economic Integration from the Viewpoint of Spatial Economics, at the Institute of Developing Economies, JETRO, 13–14 December.

PART II

General Views on East Asian Economic Integration

4.

Industrial clustering and MNE management in East Asia: recent progress and prospects for the Asian Triangle Akifumi Kuchiki

4.1

INTRODUCTION

A growing number of industrial clusters have emerged in East Asia in the process of (de facto) economic integration. Multinational enterprises (MNEs) widely play a key role as anchor ﬁrms in forming industrial clusters and creating logistic links among them. These enterprises make signiﬁcant eﬀorts to organize their aﬃliated plants spread over Asia into an intra-ﬁrm global production network and to manage the network as eﬃciently as possible (value chain management). In this chapter, we overview the development of industrial clusters in the growing economies of Asia and examine how Asian economies are being regionally integrated on the basis of fragmentation and value chain management by MNEs. Among all growing economies in East Asia, China, India and Vietnam, in particular, have been achieving remarkable economic growth. As shown in Table 4.1, their average GDP growth rates from 2000 to 2004 are as high as 8.5 percent (China), 5.8 percent (India) and 6.6 percent (Vietnam). These three economies are also expected to grow continuously into the latter half of the 2000s. Recently, the Global Strategy Research Institute of Matsushita Electric Industrial Co. Ltd has forecast that a triangle with boundaries formed by China, India and ASEAN, as shown in Figure 4.1, will be a prosperous manufacturing industry production site for the world market around 2030. We designate this area the Asian Triangle. Several highway networks are now being developed to connect various cities in this triangle. For example, the Delhi–Mumbai industrial corridor is growing in India and a number of foreign ﬁrms are agglomerating along the corridor. In the Greater Mekong Subregion (GMS), there are three programs of international economic corridor creation in progress, including East–West, North–South and Southern corridors. Among these corridors, a 1450-km 77

78

Table 4.1

Japan China Malaysia Myanmar Thailand Vietnam Bangladesh India Pakistan

General views on East Asian economic integration

Growth rate of GDP and per capita GNP (% per year) 2000

2001

2002

2003

2004

Average

Per capita GNP 2004, US$

2.9 8.0 8.9 13.7 4.8 6.1 5.9 4.4 3.9

0.4 7.5 0.3 11.3 2.2 5.8 5.3 5.8 1.8

0.1 8.3 4.1 12.0 5.3 6.4 4.4 4.0 3.1

1.8 9.3 5.3 13.8 6.9 7.1 5.3 8.5 5.1

2.3 9.5 7.1 12.6 6.1 7.5 5.5 6.5 6.4

1.5 8.5 5.1 12.7 5.1 6.6 5.3 5.8 4.1

37 180 1 100 3 780 – 2 190 480 400 530 470

Sources: Asian Development Outlook 2005; World Development Indicators 2005; Economics Affairs Bureau Research Division, Major Economic Indicators 2006.

East–West corridor from Mawlamyine in Myanmar to Danang in Vietnam opened in December 2006, and a Japanese ﬁrm, Canon, plans to deliver its parts and components of ink-jet printers from Bangkok in Thailand to Hanoi in Vietnam via that corridor. Similarly, various trading corporations are using these new corridors to facilitate their logistics, from Hanoi to Guangzhou in China, where several automobile ﬁrms, including Toyota, have agglomerated. In this way, an increasing number of industrial clusters in the Asian Triangle are being connected more closely, and this may help the evolution of the triangle into a world growth core in the coming few decades. Hence, in this chapter, we pay special attention to the Asian Triangle in studying the progress of economic integration in East Asia. Regional integration in Asia refers to economic partnerships which will promote the linking of clusters with other clusters by lowering the barriers created by national borders. This chapter explains the relationship between the triangle, industrial clusters, cluster-to-cluster linkages and economic partnerships in Asia. The remainder of this chapter is organized as follows. Section 4.2 explains industrial clusters in the triangle. Section 4.3 examines the key factors for successful industrial clusters. Section 4.4 clariﬁes the formation of an Asian network fueled by the value chain management of MNEs. Section 4.5 explains the linkages between clusters in Asia from the point of view of the value chain management of MNEs. Section 4.6 shows how Asian regional integration promotes linkages between the clusters of MNEs in the Asian Triangle. The ﬁnal section draws conclusions.

79

Industrial clustering and MNE management Cluster Nagoya Japan

China C2C Tianjin Beijing Shanghai Guangzhou Cluster

India

Figure 4.1

4.2

Delhi Mumbai [Bombay] Bangalore

ASEAN

Fukuoka

Bangkok Leamchabang (Thailand) Penang (Malaysia)

The Asian Triangle and cluster linkages

INDUSTRIAL CLUSTERS IN THE ASIAN TRIANGLE

China introduced an open door policy in 1979. By 2005, two decades later, it had become an economic power and was said to be a threat to its neighbors. The Asian Triangle is emerging, and industrial clusters are expanding within the triangle as MNEs link a part of their value chain in one cluster with a part of a diﬀerent value chain in another. Moreover, the Indian economy is expected to grow continuously, and in 2030 it will be more inﬂuential especially as regards the Asian economy. The potential of the economy of the Asian Triangle seems certain to grow in the future. In this section, we discuss details of the Asian Triangle. In 2005, the Global Strategy Research Institute of Matsushita Electric Industrial Co. Ltd projected that China, India and ASEAN would form an Asian triangle that would be the core of growth in the manufacturing sector in Asia. As shown in Table 4.1, the per capita GNPs of India, China and Vietnam, all in the triangle, were US$530, US$1100 and US$480, respectively, in 2004. The average growth rate of the countries of the region, as shown in Table 4.1, is as high as 6.1 percent.1 In this area, there are many clusters of high-tech industries, such as automobiles, electronics and die-casting industries, as well as other laborintensive industries. For example, the automobile industry is clustered in Bangkok, Thailand and several regions in China. In 2005, China was the world’s fourth-largest car producer. In that same year, Toyota laid the cornerstone for a third factory in Tianjin, China. Nissan planned to produce the latest version of the Bluebird in China beginning in 2006. Also, Mazda, Suzuki and Mitsubishi were scheduled to make additional investments in

80

General views on East Asian economic integration

China within the year. The Chinese government intends to promote sales of small cars such as the Honda Fit (Jazz) in order to promote energy conservation. Three large automobile industry clusters are forming in the Bohai Economic Circle, Changjiang Delta Economic Area and Pearl River Delta Economic Zone. Hyundai, a Korean company, has established a plant in Beijing, and Toyota, a Japanese ﬁrm, has constructed a plant in Tianjin in the Bohai Economic Circle. General Motors, Ford and Volkswagen have established plants in Shanghai. Multinational corporations use value chain management to select locations for their research and development, procurement, assembly, and marketing facilities, and these locations will link clusters together. It is important to establish linkages among clusters within the Asian Triangle as well as linkages between these clusters and those outside the triangle, such as those in Japan. We call these cluster-to-cluster linkages. Below, we describe the detailed characteristics of the individual clusters in the Asian Triangle. 4.2.1

The Pearl River Delta Economy

The east side of the Pearl River Delta began its economic development in Shenzhen and Dongguang in the 1980s and 1990s. Foreign companies in the textile and electronics industries outsourced the manufacturing and assembly of their products to factories on the east side of the Delta. At the beginning of the 2000s, the west side of the Delta began to develop as Honda, Nissan and Toyota responded to the industrial cluster policy of Guangzhou municipality, creating an automobile industry cluster based around plants constructed by the three Japanese ﬁrms in Guangzhou, Guangdong Province, which borders on Hong Kong. Guangzhou City, in the Pearl River Delta, is the only city in the world with plants operated by all three of Japan’s major automobile ﬁrms. Hyundai, a Korean company, and Isuzu, a Japanese ﬁrm, also constructed plants. A Honda staﬀ member informed us that Honda can procure parts and components from its tier 1, tier 2, tier 3 and tier 4 suppliers. Nissan can procure many of the necessary parts and components in China. Tier 1, tier 2 and tier 3 suppliers of Toyota were still constructing plants in 2006. The annual production capacity of automobiles in Guangzhou in 2010 is expected to be about 1.3 million vehicles. In this way the automobile industry cluster centered on Guangzhou is expanding. Recently, a joint venture between Nippon Yusen and Mitsui O.S.K. Lines Ltd announced that it would construct the largest port facility in China in the Pearl River Delta to ship cars. Firms in the electric and electronics industry have constructed plants in Dongguang City on the east side of the Pearl River Delta, and ﬁrms that produce parts and components for the automobile industry are also moving into Dongguang.

Industrial clustering and MNE management

4.2.2

81

The Yangtze River Delta Economy

The Yangtze River Delta is centered on Shanghai City, and incorporates Jiangsu and Zhejian provinces. The GDP growth rate of Jiangsu Province in 2003 was 13.8 percent, higher than the national average of 9.1 percent. Jiangsu Province was ranked ﬁrst in the amount of foreign direct investment implemented. The Shanghai area is expanding as an agglomeration of the automobile component, die-cast, electric and electronics, and other industries. In 2001, the Shanghai municipality established Shanghai International Automobile City in 2001 as a base for auto production, designed after the model of Detroit in the USA. It has a total area of 680 000 km2, divided into six districts for trade, research and development (R&D), education etc. A college of automotive engineering was established at Tongji University, to develop human resources for the automobile industry. The city is well endowed in terms of physical infrastructure, human resources, institutions and living conditions. The Shanghai Automobile Group and Volkswagen launched a joint venture in 1984, and constructed three factories in the city under the joint venture. About 200 companies had moved in as of August 2004, approximately half of them automobile component manufacturers. Shanghai’s industrial zones are characterized by the intention to form clusters of companies from speciﬁed countries or in speciﬁc industries. In April 2000, Shanghai City announced that it would develop an industrial zone targeting Japanese companies, not only in high-tech industries but also including various industries in the manufacturing and service sectors. The Shanghai Taiwan Commerce and Industrial Park in Shanghai was established to attract Taiwanese companies. In 2002, the Shanghai government announced that it would establish an industrial park for Taiwanese companies, which would target companies in the information technology (IT) industry. In the above cases, local governments have taken the lead in creating clusters of Japanese and Taiwanese companies. Another case is the Wuxi–Singapore Industrial Zone, which was established by SembCorp Industries, a Singaporean conglomerate, in cooperation with the Wuxi City government. The zone has good infrastructure and an abundance of capable human resources, from unskilled labor to engineers. 4.2.3

The Bohai Rim Economy

The National People’s Congress (NPC), China’s parliament, endorsed the development of the Tianjin Bohai New Area for Social and Economic

82

General views on East Asian economic integration

Development from 2006 to 2010. Its development follows that of the Shenzhen Special Economic Zone in the Pearl River Delta in the 1980s, and Shanghai’s Pudong New Area in the Yangtze River Delta in the 1990s. Within the Tianjin Bohai New Area, the Tianjin Economic Development Area is crucial to attracting foreign investors to Tianjin. Toyota began to manufacture Vios-type automobiles in Tianjin in 2002, and related ﬁrms began to agglomerate there before and after the investment. Kuchiki (2005b) calls this agglomeration the ‘Toyota eﬀect’. There were approximately 6000 industrial zones in China in 2004. Some are sites of industrial agglomeration. We now illustrate this by looking at the industrial cluster in Dalian. 4.2.4

Japan–China Joint Venture Dalian Industrial Zone: Bohai

China has various kinds of industrial zones that target companies of speciﬁc countries such as Japan or Singapore, or a speciﬁc industry such as the high-tech industry. In commemoration of the twentieth anniversary of the normalization of diplomatic relations between Japan and China, the two countries cooperated on a project to establish the Japan–China Joint Venture Dalian Industrial Zone and began to sell land in lots. At ﬁrst they expected it to become a cluster of labor-intensive industries, but in consideration of a rise in wage levels, the mayor of Dalian changed the target from labor-intensive industry to high-value-added industry. The zone has stimulated investments by Japanese companies. In 2001, Tostem Construction Materials, Yoshida Fastener and Fuji Plastic decided to invest, and by 2002, 2054 Japanese companies had made investments, for a total ﬁgure of US$5.4 billion in contract terms. Of these companies, 77 percent were in the manufacturing industry. 4.2.5

Hong Kong’s IT Industry: The Free Trade Zone Scheme

A free trade zone between the Chinese mainland and Hong Kong was proposed in December 2001. The mainland and Hong Kong Closer Economic Partnership Arrangement (CEPA) was signed in June 2003. Hong Kong is proceeding with its Cyber Port and Science Park projects with the aim of creating bases for the IT industry. The Cyber Port is part of Hong Kong’s policy to establish an IT hub for the formation of an IT industry cluster. It is located on the western part of Hong Kong Island, on a plot of land owned by the government. The businesses located in the port are in information services, multimedia and contents manufacturing. The Science Park, for its part, attaches importance to the ﬁelds of IT, electronics, biotechnology, precision instruments and high-tech manufacturing.

Industrial clustering and MNE management

83

These projects show that Hong Kong, which has placed priority on economic liberalization, is aiming at creating a cluster in the IT industry by establishing government-led industrial zones. However, the following facts show that it will not be easy for Hong Kong to cluster the IT industry, since the city’s main role is to act as a gateway to the Chinese economy: the number of bases of foreign companies in Hong Kong hit a historical high in June 2001. It hosts 3237 regional head oﬃces, with 40 percent in services related to wholesale, retail and trade businesses. Its shares of ﬁnancial services, commercial and professional services, and manufacturing of electronics products are 10 percent, 9 percent and 7 percent, respectively. Hong Kong is one of China’s ﬁnancial centers. 4.2.6

China’s Die-casting Industry

There are clusters in die-casting industries, which are crucial to the development of parts and components industries, in China’s coastal areas. In particular, Zhejiang Province is famous for its die-casting industry. The skilled labor in that province is also contributing to the development of clusters in the die-casting industry in Guangdong Province and Shandong Province. The die-casting industry in China has developed through three major routes: ﬁrst, it has been developed by the many small and medium private companies that have long existed in Shandong Province. Second, it has been developed by foreign companies in Guangdong Province. And ﬁnally, state-owned enterprises such as Hiaer and Hisense, electronics companies, have led the die-casting industry in Qingdao, Shandong Province. These diﬀerent regions are planning to form clusters in the die-casting industry by making use of industrial zones. The Ninhai Prefecture government plans to establish the Ninhai die-casting zone along a highway near Taizhou City. Ninhai City, which is in the same province, is home to a lightindustry die-casting zone, where there are more than 150 enterprises involved in plastic die-casting. The Taizhou City government, with the intention of establishing a high-tech industrial cluster, started to produce press die-casting for the machine industry in the 1950s. At that time, there were more than 20 townships and village enterprises involved in diecasting. One of the state-owned enterprises mentioned above spun oﬀ its department involved in die-casting and relocated it to the Qingdao economic technology development zone. 4.2.7

The High-tech Industrial Development Zone

Each of the high-tech industrial development zones is a cluster of Japanese high-tech companies, including Sharp Electronics Components, Panasonic

84

General views on East Asian economic integration

and Toshiba. The zones are located in Suzhou and Wuxi, which are linked to Shanghai via a highway, and have wage levels that are 30 percent to 50 percent lower than those in Shanghai. The Shanghai government has adopted a policy of shifting labor-intensive industry from the coastal areas to inland areas. The purpose of the policy is to concentrate the high-valueadded industries in Shanghai and to foster service industries such as ﬁnance. This will make it easy for foreign companies to move into the zone, taking advantage of one-stop services to complete their contracts. Preferential tax rates are also given to foreign companies. In concrete terms, high-tech companies are exempted from corporate tax for two years after they make their ﬁrst proﬁts, and taxes are reduced by 50 percent from the third to the eighth year. The zone is characterized by the presence of many Japanese companies and by the variety of industries, such as electronics, pharmaceuticals and packaging materials. 4.2.8

The Automobile Industry Clusters in Chennai and Delhi, India

The development of industrial zones in India has proceeded as follows: KIADB (Karnataka Industrial Areas Development Board), the Tata group and the Singaporean government established the International Technology Park Ltd (ITPL) in Bangalore in 1994. Following that, the Infrastructure Development Authority of the state government and Hyundai, a Korean corporate group, established an industrial zone in Chennai, Tamil Nadu. In 2004, India began to establish special economic zones similar to those in China. It is now trying to increase industrial zones for the IT industry and promote investment by foreign ﬁrms in them. State governments have been the main players in implementing the policy, together with private companies. 4.2.9

Software Technology Park of India (STPI)

The STPI plan was established in 1991 by the Ministry of Information Industry to promote the export of software and information services. There were 164 STPIs in March 1992, and the number increased up to about 6500 by March 2000. Foreign investors in the parks, who have the right to be 100 percent foreign-owned, are exempted from import tariﬀs on capital goods. They also have access to one-stop services, and are exempted from corporate taxes until March 2010. India’s Software Export Promotion Organization is inviting Japanese companies to enter into joint projects in the software industry. India’s export of software has increased by 31 percent thanks to the expansion of new markets such as China, Japan and the EU. Indian software companies

Industrial clustering and MNE management

85

have established branches in Shanghai, China, and are developing and selling software for ﬁnancial institutions and manufacturing companies. 4.2.10

Malaysia’s Automobile Industry: The Proton City Scheme

Proton, which was the ﬁrst national automobile company in Malaysia, is planning to build an industrial city in Tanjung Malim, Perak State, in the northern part of the Malay Peninsula. Former Prime Minister Mahatir promoted the development of Proton in the 1980s as part of the ‘Look East Policy’. At ﬁrst the policy was not successful, but it began to produce proﬁts during the high economic growth period in the late 1980s. A second national automobile company, Perodua, began to produce automobiles in 1994. However, the Malaysian government gave approval for foreign investors to own more than 50 percent of capital in Malaysian ﬁrms in order to boost competitiveness with the companies of ASEAN and China. As a result, Mitsui Corporation and Daihatsu, a Japanese car company, purchased 51 percent of the capital of the company, with Perodua retaining 49 percent. Malaysia was well known as an electronics industry cluster, but lagged behind Thailand in the automobile industry. In response to this situation, Mahatir launched a plan to make his country’s automobile industry competitive with those of other Asian countries by developing the national automobile manufacturers Proton and Perodua. For that purpose the government implemented tax reform to attract foreign investors. For example, the rate of the domestic value-added tax on the machine industry and capital equipment industry was lowered from over 30 percent to the 20 percent range. In addition, target companies were exempted from income and investment taxes. 4.2.11 Thailand’s Automobile Industry: The Investment Encouragement Law Thailand enacted the Investment Encouragement Law in 1977, revising it in 1992. Later, in a second revision, the decision-making powers of the Board of Investment of Thailand (BOI) concerning reduction of the corporate income tax for foreign investors were enlarged, making it possible for the BOI itself to decide how by much it would reduce the income tax. The reduction period was extended from between three and eight years to up to eight years. The zone system for awarding preferential tax incentives was also altered. Each zone under the system would now be able to determine its own corporate tax reduction rate. It is noted that the revision placed an upper limit on the reduction of corporate tax. The total amount

86

General views on East Asian economic integration

of the reduction had to be less than 100 percent of the total investment of a foreign investor, excluding investment in land and operating costs. In 2001, foreign investment in Thailand recorded nearly the same level as the previous year, although many economists had expected it to be lower due to the terrorist attacks in the USA. However, Taiwanese investment in 2001 was directed more toward China than Thailand. Although former President Li Denghui had called on Taiwanese companies to invest in ASEAN, in keeping with a policy of developing the southern part of Asia by controlling their investment in China, most invested in China in 2001. Japan became the largest investor in Thailand in both the amount of investment and number of projects, with most of the new investment coming from small- and medium-sized enterprises in automobile parts and components, and with the share of the electric and electronics industry dropping drastically. Bangkok New International Airport, whose capacity is three times that of Narita International Airport in Japan, was opened in 2006. The Thai government has launched a new project to develop human resources for the automobile industry in order to strengthen the economic relationship between Japan and Thailand. Honda established an automobile research center in Thailand after the entry of Toyota and Isuzu. The Thai government aims to transform Thailand into the ‘Detroit of Asia’ as an automobile industry cluster. Thailand has a dual policy of stimulating domestic demand and introducing foreign investment. Thailand Toyota set a target for 2010 to assemble cars using 100 percent local parts, following a request by the Thai government to use locally produced parts. Some economists insist that Thailand is on a par with China in terms of its competitiveness in commercial vehicles. 4.2.12 Vietnam’s High-Tech and Software Industries: Task Force for Attracting Foreign Investment Vietnam’s Ministry of Planning and Investment established a task force to promote attraction of foreign investment and gave responsibility to the industrial zone bureau, permission bureau and statistics bureau to cooperate eﬃciently in administrating foreign investment. Under a trade treaty which Vietnam entered into with the USA in 2001, it faces the task of abolishing non-tariﬀ barriers and opening its service markets. Vietnam is also opening its domestic markets as part of its preparations to become a member of the WTO and the ASEAN Free Trade Agreement. In 2001, Vietnam gave approval to foreign investors to own 100 percent of capital as a measure to attract them. Hasegawa was the ﬁrst Japanese company to receive approval.

Industrial clustering and MNE management

87

The Vietnamese government hopes to develop clusters in the software and motorcycle industry. It will nurture supporting industries, which are not well developed at present. The import tariﬀs on motorcycle parts were raised, largely to foster domestic industries by preventing the importation of Chinese parts. Hoa Lac Hitech Park in Hanoi and Saigon Software Park in Ho Chi Minh City were established with the aim of becoming clusters for the software industry. The government expects that US companies will invest in these parks, outsourcing components or product assembly. Some of the US companies may be run by ethnic Vietnamese in the USA. The Vietnamese methods of inviting foreign investors into industrial zones and creating industrial clusters are typical examples of how things are done in Asia, as we explain in the next section.

4.3

KEYS TO SUCCESSFUL INDUSTRIAL CLUSTERS

Figure 4.2 shows the ﬂowchart approach to industrial cluster policy.2 Many industrial clusters in East Asia share features: (1) industrial zones, (2) capacity-building, and (3) anchor ﬁrms. First, a local government constructs an industrial zone to attract foreign investors. Next, the government builds capacity in order to improve business and living conditions for foreign investors. The elements of capacity-building include: (i) constructing physical infrastructure, (ii) institution-building, (iii) developing human resources, and (iv) creating living conditions amenable to foreign investors. Physical infrastructure refers to roads, ports, communications and so on. Institutionbuilding, which is also crucial to success in luring foreign investors, includes streamlining investment procedures through one-stop services, deregulation and the introduction of preferential tax systems. The human resources include unskilled labor, skilled labor, managers, researchers and professionals. The living environment includes the provision of hospitals and international schools in order to attract foreign ﬁrms. Finally, an anchor ﬁrm initiates plans to invest after capacity-building is carried out. The typical situation in Asia was for central governments to establish industrial zones in the early stages of industrialization. In the early stages of development of Thailand and Malaysia in the 1980s, actors in the semipublic sector were responsible for establishing two types of industrial zones; export processing zones and free trade zones. Japanese trading corporations such as Sumitomo Corporation and Mitsubishi Corporation also established many industrial zones in the ASEAN countries, particularly in the 1990s.

88

General views on East Asian economic integration

Step I: Agglomeration

Industrial zone

(a)

Capacity-building (I)

(b) 1. Infrastructure 2. Institutions

3. Human resources 4. Living conditions

Step II: Innovation

(c)

Anchor firm

(d)

Related firms

(a)

Universities/research institutes Capacity building (II)

(b) 1. Infrastructure 2. Institutions

3. Human resources 4. Living conditions

Source:

(c)

Anchor persons

(d)

Cluster

Kuchiki (2008), reproduced with permission of Palgrave Macmillan.

Figure 4.2

Flowchart approach to industrial cluster policy

An electric and electronics industry cluster was formed in northern Vietnam as a result of Canon’s response to the industrial cluster policy of the Vietnamese government. Canon, as an anchor ﬁrm in Figure 4.2, moved into an industrial zone in Hanoi, Vietnam, in 2002. Canon is a manufacturer of printers, which are composed of more than 600 parts and components. Its suppliers then moved into industrial zones in Hanoi and Haiphong. Consequently, Canon and these related ﬁrms agglomerated in northern Vietnam. Thus an industrial cluster in northern Vietnam was born (Kuchiki 2007). In 2006, a highway was completed between Guangzhou in China and Hanoi in Vietnam, making it possible for a car to travel between the two cities in 14.5 hours. Sumitomo Corporation, located in Thanglong Industrial Park in Hanoi, is planning to form a distribution network of logistics to link Hanoi and Haiphong, and Guangzhou’s automobile industry cluster will expand to northern Vietnam through this distribution

Industrial clustering and MNE management

89

network. As we explain below, the network will expand through the Asian Triangle with the facilitation of the highway traﬃc network. The GMS Program of the Asian Development Bank (ADB) incorporates an international framework that covers the countries of Vietnam, Laos, Cambodia, Thailand, China and Myanmar along the Mekong River. ADB has launched an economic scheme to promote trade and investment by means of a network of roads in Asia. There are 11 projects, including the East–West Corridor, the second East–West Corridor, and the North–South Corridor. The East–West Corridor is a road across Indochina linking Danang in Vietnam, Savannakhet in Laos, Khon Kaen in Thailand and Mawlamyine in Myanmar. The second East–West Corridor is a route linking Ho Chi Minh City, Phnom Penh and Bangkok. The North–South Corridor is a route linking Bangkok and Kunming in China. ASEAN is taking the lead in developing this corridor. China and ASEAN are aiming to strengthen the economic linkages between these nations through cooperation in infrastructure, commerce, sightseeing and human development of the GMS Cooperation Program. For example, the Asian Highway Project of the Economic and Social Commission for Asia and the Paciﬁc (ESCAP) is a project involving 15 countries and will link Singapore to Europe. In addition, a scheme for a railway passing through Southeast Asia from north to south, linking Singapore to Kunming in China, was proposed by former Prime Minister Mahatir. In addition, the China ASEAN Business Port was constructed in Kunming as a trading town in 2005, with commodity exhibition halls. The Yuan economic region is expanding over southern China, and the development of the GMS will contribute to its expansion and promote Guangzhou’s automobile industry cluster to expand within the Asian Triangle.

4.4

FORMATION OF A NETWORK OF MULTINATIONAL VALUE CHAIN MANAGEMENTS

Under free competition and globalization, private companies cannot survive without competitive advantages. Value chain management and the establishment of core competence are ways to attain a competitive advantage. We explain the mechanism of these in Appendix 4A1. As Kuchiki (2000) and Chapter 3 explained, fragmentation of the process enables various forms of production and a new method of management called value chain management. A company that wishes to maintain a competitive advantage in the long run must have a core competence as a part

90

General views on East Asian economic integration

of its value chain. No company can have competitive advantages in the entire value chain, so each must select a core competence and focus on it. Here we summarize the deﬁnition of a value chain as is shown in Appendix 4A2, Figure 4A2.1: Design & planning → Procurement of parts & components → Assembly → Marketing (logistics & after-sales services). The strategies for value chain management are to optimize the value chain to maximize customer satisfaction, as shown in Appendix 4A2, Figure 4A2.2. The measures are (1) Outsourcing, (2) Mergers & acquisitions (M&A), and (3) Alliances. For example, IBM places emphasis on marketing and R&D in its value chain, and outsources the assembly process as shown in Appendix 4A2, Figure 4A 2.3. Ford and GM began Internet business for marketing, starting from the main process of assembly and production (see Appendix 4A2, Figure 4A2.4). The development of IT has encouraged the formation of networks in three regions in the world: the USA, Europe and Asia. Asian MNEs must network their activities by introducing value chain management in order to remain competitive internationally (see Appendix 4A2, Figure 4A2.5). Regional integration in Asia is closely related to the links between clusters in Japan, India and ASEAN. Regional integration will help link clusters by reducing tariﬀ and non-tariﬀ barriers among countries.

4.5

DEVELOPMENT OF CLUSTER-TO-CLUSTER LINKAGES

When they ﬁrst began to invest in China, many Korean ﬁrms selected Qingdao and Yantai in Shandong Province for proximity with Korea, as well as Liaoning Province, which is also near Korea. LG, a Korean ﬁrm, invested in Shenyang, the capital city of Liaoning Province. Recently Korean ﬁrms have changed their strategy, choosing to invest in the main cities of China. Hyundai invested in Beijing and has increased its sales in China. Many large Korean ﬁrms have invested in the Shanghai region. Hyundai decided to invest in Huadu, Guangzhou, Guangdong Province. Korea is now planning to open the new Pusan Port as a national project, at a cost of about US$10 billion. The aim is to make the port a hub for cargo in Northeast Asia. The Korean government has adopted a policy to create a cluster in the LCD (liquid crystal display) industry through legal deregulation to allow the location of LCD plants in Seoul. Korean ﬁrms are now trying to create networks between industrial clusters in Korea and China. Next, we describe the linkage between Japan’s industrial clusters and the Asian Triangle, which involves cluster-to-cluster linkages. Most MNEs

91

Industrial clustering and MNE management R&D • Design

Procurement

Assembly

Marketing

Guangzhou

Guangzhou

Guangzhou

Guangzhou

Location

Nagoya

Figure 4.3

Shanghai

China

Nagoya

Japan

Guangzhou Camry value chain

R&D • Design

Procurement

Assembly

Marketing

Nagoya

Nagoya

Nagoya

Japan

Location

Figure 4.4

Shanghai

Shanghai

Guangzhou

Guangzhou

Nagoya Prius value chain

have adopted value chain management, as explained in the last section. In the past, Japanese ﬁrms focused on the processes of procurement and assembly, but they are now adopting the goal of maximizing their proﬁts by maximizing consumer satisfaction. Figures 4.3, 4.4 and 4.5 illustrate a case involving cluster-to-cluster linkages. Suppose that there are automobile clusters in Guangzhou and Nagoya, and that Toyota is operating in both clusters. Here Nagoya means ‘the zone covering Aichi Prefecture centered in Nagoya’. Figure 4.3 shows that it assembles Camry-type cars in Guangzhou. Meanwhile, R&D activities are carried out in Nagoya and the results are sent to Guangzhou. The plant in Guangzhou procures some of its parts and components from Nagoya, and the ﬁnal products are marketed in both China and Japan. This is the value chain for Camry-type cars in Guangzhou. Similarly we can describe the value chain for Prius in Nagoya. Figure 4.4 shows that Toyota assembles Prius-type cars in Nagoya, which is the center of R&D for Camry, Prius and other Toyota models. Each of Toyota’s companies throughout the world procures parts and components from all over the world, including Guangzhou. Next, we look at the linkages within Toyota between the cluster in Guangzhou and the cluster in Nagoya. Figure 4.5 shows that linkages take

92

General views on East Asian economic integration

Value chain

R&D • Design

Procurement

Assembly

Marketing

(China • India • ASEAN) Guangzhou cluster

Guangzhou

Guangzhou

C2C Nagoya cluster

Nagoya

Guangzhou

C2C

C2C

Nagoya

Nagoya

Crucial components

Mother factory

The Asian triangle C2C Japan

Note: C2C: → Trans-border links based on ‘cluster-to-cluster’ movements.

Figure 4.5 Links between Japan and the Asian Triangle based on ‘clusterto-cluster’ movements place in a borderless fashion, between clusters in diﬀerent countries. The results of R&D carried out in Nagoya are applied at Toyota’s Guangzhou factory. Key parts such as engines are exported from Nagoya to Guangzhou and used for the cars assembled there. Some parts, conversely, are produced in Guangzhou and used in Nagoya. The Nagoya factory plays the role of mother factory, by manufacturing prototypes for new car models. If these cars are successful in Japan, the Guangzhou factory begins to manufacture them as well. Thus the manufacture of new models takes place in Japan, while the factories in foreign countries manufacture standardized models. Under this system, the factories in Japan have relationships with Asia. The company allocates diﬀerent models of cars to diﬀerent production sites in an industrial cluster. The Guangzhou factory assembles Camry-type cars and sells them in Japan. The Nagoya factory assembles Prius-type cars and sells them in China. Figure 4.5 shows the linkages between clusters through the exchange of value chains. In order to be competitive in the world market, Japanese ﬁrms should have as many linkages as possible.

4.6

CLUSTERS IN ASIA AND ASIAN ECONOMIC INTEGRATION

There are many papers on fragmentation and trade costs, such as Markusen and Venables (2005) and Davies (2005). For example, Fujita and Gokan (2005) examine how the decline of communication costs between management and production facilities within ﬁrms and the decrease in trade costs of manufactured goods aﬀect the spatial organization of a tworegion economy with multi-unit/multi-plant ﬁrms. The development of IT

Industrial clustering and MNE management

93

Table 4.2

Customs clearance at borders of Indochinese countries

Country

Point of entry

Time required*

Customs inspection

Ingress of other countries’ vehicles

Thailand

Border

Cambodia

Phnom Penh customs (imports whose import tariff is more than US$300 and all exports) Border

A few hours – 1 day 1.5–3 days

Only specified cargo All

Vehicles from Laos can enter (prior registration required) By cargo transship area

0.5–1 day

All

2–3 days

All

• Vehicles from Thailand can enter (prior registration required) • Vehicles from Vietnam can enter (Destinations are limited; prior acceptance required) By cargo transship area

1–2 days

All

By cargo transship area

Laos

Myanmar

Vietnam

Border (export records from competent border customs are required) Border

Note: * If importers/exporters declare in advance, time for customs clearance can be shortened (only cargo inspection, completed in a few hours). Source: Field Studies, Sankyu Inc., December 2004.

reduces the costs of communication and trade, and thus promotes the fragmentation of ﬁrms. In a broad sense, trade costs include transport costs, tariﬀs, non-tariﬀ barriers and customs clearance to transport a commodity from one place to another. The trade costs constitute major obstacles to the strengthening of cluster-to-cluster linkages in the Asian Triangle. In Asia today, as economic integration progresses, these trade costs are falling, and the industrial clusters in this region are becoming more closely connected. Table 4.2 summarizes how high the barriers at borders are and how long it takes to clear customs in Thailand, Cambodia, Laos, Myanmar and

94

General views on East Asian economic integration

Vietnam. By and large, the border barriers of these countries are signiﬁcant. All cargoes are inspected at the borders in Cambodia, Laos, Myanmar and Vietnam, while Thailand inspects only speciﬁed cargoes. The process takes from a few hours to a day in Thailand, from a half-day to a day in Laos, from a day to two days in Vietnam, from a day and a half to three days in Cambodia, and from two to three days in Myanmar. The time required for customs clearance should be shortened in order to expand trade between the countries of Asia. These border barriers can be lowered if these countries enter into FTAs with each other. FTAs can be eﬀective for strengthening cluster-to-cluster linkages in the Asian Triangle by lowering tariﬀ rates, abolishing non-tariﬀ barriers and shortening the time required for customs clearance.

4.7

CONCLUDING REMARKS

In this chapter, we have examined economic change in Asia, particularly with respect to the behavior of MNEs, and have examined the importance of industrial clusters in the Asian Triangle. It is naturally expected that an Asian triangle of growth will be formed in the coming few decades, bordered by China, India and ASEAN. There are many industrial clusters within the triangle. One is the automobile industry cluster in Guangzhou, China, where Honda, Nissan and Toyota are located. The automobile industry clusters will expand over ASEAN along with the growth of the Yuan economy. MNEs use value chain management to select their locations for R&D, procurement, assembly and marketing. These locations will link diﬀerent clusters together. The Asian Triangle is developing and industrial clusters within it are expanding. MNEs link value chains in one cluster with those in another. Regional integration means Asian economic partnership, and this will promote links between clusters by lowering the barriers at national borders. This chapter has explained the necessity of regional integration for clusterto-cluster linkages in the Asian triangle of growth.

ACKNOWLEDGMENT I would like to express my thanks to K. Nishikimi and S. Kumagai for their helpful comments. Remaining errors are mine.

Industrial clustering and MNE management

95

NOTES 1. Island countries such as Indonesia, Japan and the Philippines are located outside of the triangle, as shown in Figure 4.1. 2. See Appendix 4A1 for details of the ﬂowchart approach to industrial cluster policy.

REFERENCES Davies, R.B. (2005), ‘Fragmentation of headquarter services and FDI’, North American Journal of Economics and Finance, 16 (1), 61–79. Fujita, M. and T. Gokan (2005), ‘On the evolution of the spatial economy with multi-unit-multi-plant ﬁrms’, Portuguese Economic Journal, 4 (2), 73–105. Johnston, R.B. and V. Sundararajan (eds) (1999), Sequencing Financial Sector Reforms: Country experiences and issues, Washington, DC: International Monetary Fund. Kuchiki, A. (2000), ‘Value chain network of Paciﬁc Basin countries’, in the New Zealand Japan Center (ed.), New Zealand and Japan: What next?, Palmerston North: Massey University, pp. 14–19. Kuchiki, A. (2005a), ‘Theory of a ﬂowchart approach to industrial cluster policy’, IDE-JETRO, Discussion Paper No. 36. Kuchiki, A. (2005b), ‘A ﬂowchart approach to Asia’s industrial cluster policy’, in A. Kuchiki and M. Tsuji (eds), Industrial Clusters in Asia, London: Macmillan, pp. 169–99. Kuchiki, A. (2007), ‘Agglomeration of exporting ﬁrms in industrial zones in Northern Vietnam’, in M. Kagami and M. Tsuji (eds), Industrial Agglomeration, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 97–138. Kuchiki, A. (2008), ‘The flowchart model of cluster policy’, International Journal of Human Resources Development and Management, 8(1/2), 63–95. Markusen, J.R. and A.J. Venables (2005), ‘A multi-country approach to factorproportions trade and trade costs’, CEPR Discussion Papers No. 4872. Markusen, A. (1996), ‘Sticky places in slippery space: a typology of industrial districts’, Economic Geography, 72, 293–313. Porter, M.E. (1998), The Competitive Advantage of Nations, New York: The Free Press. Tsuji, M., S. Miyahara, Y. Ueki and Somrote Komolavanij (2008), ‘An empirical examination of the ﬂowchart approach to industrial clustering’, in A. Kuchiki and M. Tsuji (eds), The Flowchart Approach to Industrial Cluster Policy, London: Palgrave Macmillan, pp. 194–261. World Bank (1998), World Development Report, Washington, DC: World Bank.

96

General views on East Asian economic integration

APPENDIX 4A1

A FLOWCHART APPROACH TO CLUSTER POLICY

The World Bank (1998) examined what sequencing of reforms is best to attain a policy target. Johnston and Sundararajan (1999) discussed the sequencing of monetary policies. Our ﬂowchart approach examines the ordering and prioritization of policy measures and decides ‘yes’ or ‘no’ at each step of policy measures by judging indices of the achievement of policy measures. We cannot proceed to the next step if an index indicates ‘no’ at any one step. We must then ﬁnd actors who can clear the critical level of the index and solve the bottlenecks. For example, we must ﬁnd actors, or economic agents, to construct roads when roads are insuﬃcient in order to implement industrial cluster policy. Candidates for the economic agents are the central government, local governments, the quasi-public sector and the private sector. Porter (1998) constructed a diamond model that shows that four factors are a condition of an industrial cluster. However, it is not easy to satisfy the four conditions at the same time. Our ﬂowchart model will prioritize the four factors in order, not in a diamond form, but through the linearity of ﬂowcharts. Markusen (1996) classiﬁed industrial districts into three types: Marshallian ID, hub-and-spoke, and satellite platform. He found a relationship between anchor ﬁrms and their related ﬁrms in the case of huband-spoke. However, he did not show any conditions for the formation of an industrial cluster. Our ﬂowchart model proposed a practical method for cluster policy by ordering and prioritizing policy measures in a linear way. This purpose of this appendix is to propose the hypothesis of a ﬂowchart approach to make ﬁrms in the high-tech and automobile industries agglomerate and innovate. We hypothesize a ﬂowchart for the agglomerated ﬁrms to innovate at step II of innovation. Our hypothesis is one of suﬃcient conditions through which cluster policy leads ﬁrms to agglomerate as step I and innovate as step II. (1) A Prototype Model of a Flowchart Approach to Cluster Policy We cannot strictly prove our hypothesis of a ﬂowchart approach by inductive and deductive methods. We propose a suﬃcient condition for success in industrial cluster policy. That is, we can form an industrial cluster if we follow a ﬂowchart that satisﬁes the suﬃcient condition. We can illustrate cases where our hypothesis of a ﬂowchart approach holds. We can increase the number of cases, but cannot prove our hypothesis as a

Industrial clustering and MNE management

97

1. Find factors

A ,

B , C , D , E

2. Select some factors from all factors, and order them into a flowchart Figure 4A1.2, Appendix Figure 4A1.3. 3. Apply the flowchart approach to a region and find actors if the answer at one step of the flowchart is 'No' Appendix Figure 4A1.4.

Figure 4A1.1

Overview of flowchart approach

suﬃcient condition by inductive and deductive methods. Our ﬂowchart cannot deny other ﬂowcharts whose ordering of factors or ingredients is diﬀerent from that of ours. We can show, by increasing the number of cases, that our ﬂowchart may be generally applied to industrial cluster policy in other regions. Our hypothesis is practical since we can form a cluster if we follow the following four steps: 1. 2. 3. 4.

Find ingredients such as industrial zones, capacity-building, joint actions and an anchor ﬁrm (Figure 4A1.1). Select the minimum number of factors from the ingredients found above for a ﬂowchart (Figure 4A1.2). Order them into a ﬂowchart (Figure 4A1.3). Specify actors to proceed at each step of the ﬂowchart if the step proceeds not to ‘yes’ but to ‘no’. (Figure 4A1.4).

Our ﬂowchart of the automobile industry cluster policy can be described as follows: First, a local government constructs an industrial zone as a receptacle for the invitation of foreign investors. Second, the government builds up capacity to improve business and living conditions for foreign investors. The elements of capacity-building include (i) constructing physical infrastructure, (ii) building institutions, (iii) developing human resources, and (iv) creating living conditions amenable to foreign investors. Physical infrastructure refers to roads, ports, communications and so on. Institution-building, also crucial to success in inviting foreign investors, includes streamlining investment procedures through one-stop services, deregulation of laws and introduction of preferential tax systems. Human resources, which are usually an

98

General views on East Asian economic integration

C

A

E

Cluster Figure 4A1.2

Selection Example Industrial zone

Capacity-building

Anchor firm

Cluster Figure 4A1.3

Ordering

initial condition for foreign investors, include unskilled labor, skilled labor, managers, researchers and professionals. The living environment includes, for example, the provision of hospitals and international schools in order to attract foreign ﬁrms. An anchor ﬁrm will be ready to invest after the capacity-building mentioned above has been carried out.

Industrial clustering and MNE management

99

C Yes

Actor Local government

A No

E

Cluster

Figure 4A1.4

Specify actors

We explain our ﬂowchart approach with reference to Figures 4A1.5–4A1.7. First, we identify whether industrial zones have been established. If they have not, we must decide which actors should establish such zones. Then we return to the mainstream of the ﬂowchart. Next we apply the ﬂowchart’s second step, capacity-building, which takes place after the establishment of industrial zones. We examine whether water supply is suﬃcient for the industrial zones (Figure 4A1.6). We proceed down the ﬂowchart to examine power supply, communication and transportation. After the facilitation of physical infrastructure, we examine whether institutions are in place. The central government must institutionalize national tax systems and the local government must institutionalize local tax systems. It is well known that one-stop services of investment procedures are crucial to success in attracting foreign investors. Concerning human resource development, an abundance of unskilled labor with a high literacy rate is the necessary condition to attract foreign investors, whose purpose is to employ cheap labor. On the other hand, it sometimes happens that an industrial cluster faces a shortage of skilled labor after industrialization has progressed to some degree; universities and on-the-job training centers for innovation are then needed for further development.

Find actors

Industrial zone No

Local gov. Central gov. Semi-gov. NPOs Private companies

Priority of actors to bottlenecks No

Yes Return

Return (after finding actors)

No

Local gov. Central gov. Semi-gov. NPOs Private companies

No Return

Find actors

Infrastructure: electricity No

Local gov. Central gov. Semi-gov. NPOs Private companies

No

Yes Return

Return

No

Local gov. Central gov. Semi-gov. NPOs Private companies

1

No 2

Yes Return

Return

Find actors

Infrastructure: transport

100

No

Local gov. Central gov. Semi-gov. NPOs Private companies

No

Yes Return

Return

No

Local gov. Central gov. Semi-gov. NPOs Private companies

Yes

No Return

Return

4

2

Find actors

Human resources No

Local gov. Central gov. Semi-gov. NPOs Private companies

No

Yes Return

Return

2

Find actors

Living conditions No Anchor firms

3

Find actors

Institutions

Local gov. Central gov. Semi-gov. NPOs Private companies

Return

Source: Kuchiki (2008), reproduced with permission of Palgrave Macmillan.

Figure 4A1.5

2

Find actors

Infrastructure: communication

(4)

NPOs Private companies

1

2 Return

(3)

1

Find actors

Infrastructure: water Yes

(2)

Local gov. Central gov. Semi-gov.

Flowchart approach

No Return

1

Step I: Agglomeration

Prioritization Does industrial zone exist?

No

Find actors

Assign local govt priority

1

Assign central govt priority Assign semi govt priority

1

Assign NPOs priority

Yes

Assign private companies priority

Capacity-building Does infrastructure exist?

No

Assign local govt priority Find actors

(water, electricity, communication, transport)

Assign central govt priority

2,3 1

Assign semi govt priority Assign NPOs priority

Yes

Assign private companies priority Assign local govt priority

No Do institutions exist?

Find actors

4 2

Assign central govt priority Assign semi govt priority

101

Assign NPOs priority

Yes

Assign private companies priority Assign local govt priority Do human resources exist?

No

Find actors

Assign central govt priority Assign semi govt priority

2

Assign NPOs priority Assign private companies priority

Yes

Assign local govt priority Are living conditions sufficient?

No

Find actors

Assign central govt priority Assign semi govt priority

Specify core competence

No Feedback

Are conditions sufficient for anchor firms

Yes

Source: Kuchiki (2008), reproduced with permission of Palgrave Macmillan.

Figure 4A1.6

Assign NPOs priority Assign private companies priority

Flowchart approach – Step 1: Agglomeration

1

102

General views on East Asian economic integration

Local government Priority 1 = industrial zone Priority 2 = water, electricity & communication Priority 3 = transport Priority 4 = institutions

Central government Priority 1 = electricity Priority 2 = institutions

Semi-government Priority 1 = industrial zone Priority 2 = human resources

NPOs Priority 1 = living conditions

Source: Kuchiki (2008), reproduced with permission of Palgrave Macmillan.

Figure 4A1.7

Priorities of actors

Good living conditions are crucial for attract foreign investors. Investor company researchers will have incentives to work hard if they can enjoy their lives. Satisfactory conditions must be created with regard to housing, schools, hospitals and so on. These are the last conditions that must be satisﬁed in order to attract anchor ﬁrms. We explain the priorities of each actor in Figure 4A1.7. Local governments play the main role in constructing industrial zones, supplying electricity, facilitating transport and forming institutions. The ﬁrst priority of local government in Figure 4A1.7 is to construct industrial zones to accept foreign investors. The second priority during the same stage is to supply electricity, facilitate transport and form institutions. The central government supplies electricity and builds institutions at an equal level of priority. The typical industrial cluster policy was theorized by deﬁning an industrial zone as ‘quasi-public goods’, and it was shown that the policy enhances economic growth under a production function of ‘increasing returns to scale’ of an anchor company. Critical amounts of the production of ‘scale economies’ that are used by the related companies to decide whether or not to invest in clusters were also shown (Kuchiki 2005a).

Industrial clustering and MNE management

103

(2) Demand Condition Our ﬂowchart approach does not explicitly discuss the demand side of the manufacturing industry since we suppose that the demand conditions are satisﬁed as follows: Case 1: (i) An anchor ﬁrm locates in an export processing zone and exports its products. There is in this case little constraint on its demand since its factory can attain its minimal optimal level of production by exporting products to the world. In this case the logistics of the anchor ﬁrm are crucial to attaining the minimal optimal level. (ii) Suppliers to the anchor ﬁrm can thus attain their minimal optimal levels of production. The demand for the suppliers is ‘demand derived from their anchor ﬁrm’. Case 2: (i) The anchor ﬁrm sells its products locally. In this case, the market size in the region where the anchor ﬁrm locates should be large enough for the anchor ﬁrm to attain the minimal optimal level of production. An anchor ﬁrm decides to invest in a location once it judges that the local demand satisﬁes this condition. (ii) The suppliers’ condition is the same as that of Case 1. (3) Patterns of Flowchart Models This section explains three industry-speciﬁc patterns of ﬂowchart models, one each for the IT industry, the biotechnology industry and the automobile industry. The ﬂowchart model for the IT industry is almost the same as that for the biotechnology industry. Knowledge-intensive industries include the IT industry, the biotechnology industry and the nanotechnology industry. The ﬂowchart model for the knowledge industry is diﬀerent from that for the automobile industry. The existence of universities is a precondition for the knowledge industry. Local governments thus play important roles in implementing cluster policy. A large diﬀerence in the ﬂowcharts between the automobile industry and the knowledge industry is that anchor ﬁrms play an important role in implementing the automobile industry cluster policy, while superstars play an important role in implementing the knowledge industry cluster policy. The reason is that, while the knowledge industry requires partnership between intellectuals, a car assembler as an anchor ﬁrm is unavoidable in the automobile industry cluster policy since a car is composed of more than 22 000 components. Superstars are needed for the

104

General views on East Asian economic integration

knowledge industry cluster policy since knowledge is embodied in human resources. Tsuji et al. (2008) pioneered the development of an econometric method to support the prioritization of the ﬂowchart approach, except for the fact that infrastructure was not found to be a signiﬁcant factor.

105

Industrial clustering and MNE management

APPENDIX 4A2

VALUE CHAIN MANAGEMENT

Design & Planning

Procurement

Assembly

Marketing

D

P

A

M

Figure 4A2.1

D

Value chain from design & planning to marketing.

P

Figure 4A2.2

A

Customer satisfaction

M

Optimization for customer satisfaction

D

P

A

(3)Alliance

(1)Outsource

(1)Outsource

Figure 4A2.3

M

(2)Mergers & acquisitions

Measures of value chain management: an example

D

P

A

M

Internet business Ford, GM Figure 4A2.4

Core Competence (selection and concentration)

Automotive company —> consumer company (e-business, finance company) D

Japan – – – = Domestic

P

A

M

Japan Tianjin

– Tianjin

– Shanghai

–

– – China

–

–

= M&A

= Alliance

= Outsourcing (cross-dock logistics)

Value chain network of Paciﬁc Basin countries; core competence D Source: Kuchiki (2005b), reproduced with permission of Palgrave Macmillan.

Figure 4A2.5

Toyota’s value chain network in China

5.

Evolution of institutions and policies for economic integration in East Asia: history and prospects Yoshihiro Otsuji and Kunihiko Shinoda

5.1

INTRODUCTION

As discussed in Chapter 4, de facto economic integration in East Asia has been substantially accelerated by the development of industrial clusters, a phenomenon that has accompanied the increasing investment of MNEs in the region since the Plaza Accord of 1985. Moreover, in terms of economic systems and policies, great eﬀorts have been made to promote liberalization and rule-making in trade and investment to facilitate economic integration in East Asia. Most of the East Asian countries are enthusiastically concluding FTAs and EPAs with each other. Accordingly, from now on, economic integration is expected to progress further on a de jure as well as on a de facto basis. In this chapter, we shall study the development process of the institutional and policy framework for economic integration in East Asia, with special reference to the policy issues that each of the ASEAN member states, as well as China and Japan, currently encounters. Individual institutions and policy measures are not necessarily based on considerations of spatial economics. However, close examination and analysis of these aspects are essential in order to develop a clear picture of regional economic integration. We thus attempt to examine the historical progress of East Asian economic integration and analyze the institutional and policy issues that have to be overcome for deepening the integration among the countries of the region. The remainder of this chapter is organized as follows. Section 5.2 presents a historical overview of developing regional systems in East Asia in comparison with those of Western Europe. Section 5.3 touches upon the qualitative change in the East Asian economy caused by the Asian currency crisis at the end of 1990s. Section 5.4 traces the movement towards initiating FTA negotiations in East Asia after the currency crisis. Section 5.5 analyzes the 106

Evolution of institutions and policies

107

economic conditions underlying the move toward economic integration after the crisis, including the emergence of a middle class and the formation of industrial networks. Section 5.6 identiﬁes and explains the impediments to ongoing regional integration in East Asia. Section 5.7 discusses possible thrusts to promote economic integration. Following on from these discussions, Section 5.8 indicates the future direction of economic integration in East Asia.

5.2

REGIONAL SYSTEMS, REGIONALIZATION AND REGIONALISM

The East Asian regional system was fashioned in an embryonic form in the early years of the cold war under US hegemony. In those days, the USA confronted two major strategic issues in the region. The ﬁrst was how to counter the threat of international communism and contain China (and the Soviet Union). The second was how to rebuild Japan in economic terms and make it an ally, while at the same time ensuring that Japan would never again pose a threat to the USA. The USA came up with two sets of answers to deal with these issues: ﬁrst, construction of a regional security system with the USA as the hub and bilateral security treaties and military-base agreements (such as those between the USA and Japan, South Korea, and the Philippines) as the spokes, and second, the creation of a triangular trade relationship linking the USA, Japan and Southeast Asia. In the early years of the cold war, the USA also faced two similar strategic challenges in Western Europe, one being how to counter the communist threat and contain the Soviet Union, and the other being how to rebuild West Germany as an economic power and make it an ally, while making sure that it also would never again be a threat to the USA and its allies. However, the answers the USA found to these questions were signiﬁcantly diﬀerent from those in East Asia: so far as security was concerned, the answer in Europe was the creation of the North Atlantic Treaty Organization as a collective security institution, whereas in respect of economic reconstruction, the response was the formation of a European common market built on a partnership between France and West Germany, an institution that has now evolved into the European Union. The fact that the USA made diﬀerent strategic choices in East Asia and Western Europe thus had a profound impact on the structure of the two regional systems. In Western Europe, Europeanism, underpinned by a widespread tendency for people to see themselves as Europeans, informed the political will to build a peaceful and prosperous Europe and led to the formation of a community anchored in the larger North Atlantic collective

108

General views on East Asian economic integration

security system. In other words, regionalism has been the driving force in the making of Western Europe. In East Asia, by contrast, there was neither the political will to create a regional community nor a sense of a common identity among Asians to serve as the basis for ‘Asianism’ as a regionalist ideology. Instead, East Asia has emerged as a region through a process of economic integration supported by the tremendous economic development of the East Asian countries, a process that was substantially powered by foreign direct investment by Japanese, South Korean and overseas Chinese businesses. This was underpinned by the region’s US-led security and triangular trade systems. In other words, it was market forces, and not political will informed by Asianism, that led to the regionalism that emerged in East Asia in the 1980s and 1990s. Until the 1997–98 economic crisis, it was Japan and the ASEAN countries that formed the core of East Asian regionalization. After World War II, diametrically opposite historical lessons were drawn about nationalism in East Asia and Western Europe. In Western Europe, nationalism was seen as a demonic force that needed to be contained and domesticated by European regionalism. In East Asia, by contrast, nationalism was viewed in a positive light as having powered the movements for national liberation from colonial rule and as providing the basis for the legitimacy of the region’s newly independent nation-states. The ASEAN countries, in line with the spirit of nationalism, initially sought to pursue their national economic development while placing thoroughgoing restrictions on foreign investment and imposing high tariﬀs to encourage import substitution. This strategy, however, soon reached a dead end. The ASEAN member states began to shift to a strategy of export-oriented industrialization, accompanied by moves to encourage foreign direct investment. Around the same time, the 1985 Plaza Accord led to a sharp appreciation in the value of the yen. To remain internationally competitive, Japanese manufacturers, above all electronics ﬁrms, made major investments in Southeast Asia and moved many of their production facilities there.1 This led to the emergence of what has been described as a ﬂying-geese pattern of industrial development in East Asia, together with the growth of cross-border business networks and other ties among the countries of the region. As ASEAN emerged as a base for export production in the 1990s, Japanese automobile manufacturers, along with their parts manufacturers and other supporting ﬁrms, moved their production facilities there.2 ASEAN, which was originally established as a conﬁdence-building mechanism, has thus evolved into a framework for economic cooperation, and since the end of the cold war, it has grown into a ten-nation association with the addition, to the previous six members, of Burma (Myanmar), Cambodia, Laos and Vietnam; it is now aiming for the

Evolution of institutions and policies

109

formation of an ASEAN Free Trade Area, or AFTA, with a population of 500 million.

5.3

CHANGES IN THE EAST ASIAN ECONOMY CAUSED BY ASIAN CURRENCY CRISIS

Following the hardships experienced in ASEAN due to the Asian currency crisis that occurred in 1997, negotiations on FTAs (free trade agreements) were realized at the beginning of the twenty-ﬁrst century. Consideration of the factors which brought about this major change will provide important clues to the question of how best to overcome the obstacles to future development in this region. The ﬁrst factor leading to the change was the exposure of the structural weakness of the ASEAN economy. The ASEAN economy underwent a bubble-like expansion before the currency crisis by inviting a large amount of foreign investment under a dollar-pegged currency regime. However, the currency crisis, which was caused by rapid movements of international short-term funds, revealed the structural weakness of the ASEAN economy, which was over-dependent on foreign direct investment and exports, and could not realize self-sustaining growth. In other words, the currency crisis exposed many problems relating to the lack of self-sustaining economic growth in ASEAN, including an insuﬃcient level of domestic demand from a macroeconomic viewpoint, and underdeveloped economic and ﬁnancial systems from a microeconomic viewpoint.3 It also became apparent that ASEAN had not substantially realized its own economic integration as an economic institution. For example, it did not implement suﬃcient common economic countermeasures against the currency crisis, and some ASEAN countries took a negative stance toward policy measures for accelerated economic integration, such as tariﬀ reductions, believing that such measures would exert a harmful eﬀect on domestic industries.4 The second factor that caused the change was the rapid development of the Chinese economy (Figure 5.1). Although speculation spread that China was not serious about joining the WTO in the latter half of 1990s, China’s entry into the WTO in 2002 raised expectations of business expansion there and led to rapid inﬂow of foreign direct investment (Figure 5.2). This inﬂow of foreign capital brought about an expansion of domestic production bases and the growth of production in manufacturing industries in the iron and steel sector, as well as in electronics and electric appliances. Accordingly, China became the world’s third-largest exporter, accounting for 7.3 percent of world exports in 2005. Thanks to the rapid expansion of investment and exports, China experienced double-digit growth of real

110

General views on East Asian economic integration

Rank 1 2 3 4 5 6 7 8 9 10

Country USA Japan Germany China UK France Italy Canada Spain Korea

Nominal GDP (US$ billion)

Share (%) 28.1 10.3 6.3 5.0 5.0 4.7 4.0 2.5 2.5 1.8

12486 4571 2797 2225 2202 2106 1766 1130 1127 793

(a) GDP of major countries (2005) 50000

USA, 42101

Japan, 35787

40000 30000 World average 20000 6968 10000 0

Russia, 5369 Brazil, 4316 China, 1703 India, 714

(b) Per capita GDP (2005, unit: US$)

Note: China became the fourth largest economic power in the world as measured by GDP in 2005. China’s per capita GDP exceeded US$1000 in 2004 and reached US$1703 in 2005. Source: IMF.

Figure 5.1

China’s rapid economic growth

GDP for three consecutive years, with rates of 10.0 percent in 2003, 10.1 percent in 2004 and 10.2 percent in 2005. It was obvious that China’s economic performance would have a substantial impact on the world economy and that China, aiming to strengthen its political inﬂuence among the Asian countries, would advance a proactive foreign policy in the region by taking advantage of its economic power. For instance, in 2001, China oﬀered to begin negotiations on a free trade agreement with ASEAN ahead of other ASEAN dialogue partners, and concluded the China–ASEAN FTA on trade in goods in 2005.5 China also played a major role in proposing the start of the track-one study on the East Asian FTA among ASEAN, China, Japan and South Korea in 2004. On the other hand, rapid economic development over the past 20 years has exposed many domestic economic problems in China, including signiﬁcant economic disparities. In recent

Evolution of institutions and policies 180000 160000

111

ASEAN-6 China

US$ million

140000 120000 100000 80000 60000 40000 20000 0

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Note: ASEAN-6 Thailand, Malaysia, Indonesia, Phillippines, Singapore, Vietnam China (Jan.–Nov.), Singapore and Malaysia (Jan.–Sept.), Philippines (Jan.–June) in 2005. Source: Investment statistics in each country.

Figure 5.2

Expansion of FDI into China, 1996–2005

years, the Chinese economy has continued to grow rapidly, incurring a high risk of economic overheating ending in a hard landing, together with a great surplus in production capacity caused by excess investment. Despite China’s impressive economic growth, conspicuous economic disparities have emerged between urban areas and rural areas, and between coastal areas and inland areas. For instance, per capita GDP in Shanghai is 9.9 times larger than that in Guizhou.6 The third factor is the change in Japan’s regional strategy. At the end of the 1990s, there was a growing debate in Japan on how to address the regional integration movements following the Asian currency crisis. Japan changed its WTO-oriented multilateral trade policy, and starting with negotiations on the Japan–Singapore EPA, it has attached high priority to the promotion of regional economic integration. During the currency crisis, Japan provided large-scale ﬁnancial assistance to relieve the pressure on ASEAN, where many Japanese manufacturing industries had set up production bases. However, Japan’s representation in East Asia weakened somewhat because of the delay in its own economic recovery. Once the Japanese economy broke out of its deﬂationary cycle, Japanese businesses again put a great deal of eﬀort into their foreign operations. Japanese companies have been in the process of rebuilding their regional strategy in Asia, and while doing so, they have taken into consideration the movements toward regional integration within the region.

112

5.4

General views on East Asian economic integration

DEVELOPMENT IN FREE TRADE AGREEMENTS IN EAST ASIA

After the outbreak of the economic crisis of 1997–98, the ASEAN countries began to search for ways to overcome the eﬀects of the ﬁnancial disruption that the crisis had caused. It was this search for a way out that led them, in 1997, to align ASEAN with Japan, South Korea and China in the form of the ASEAN+3 framework. China was lukewarm about this framework initially, but as the time for its entry into the WTO approached, its position changed. Joining the WTO had been China’s dream for many years; once China was part of the WTO, however, it would surely face demands to open its markets from the USA and from the European countries. China recognized the need for partners in dealing with these demands. It also felt it necessary to ﬁnd new markets for products in sectors in which it held a comparative advantage, such as textiles and electronics, as a means of coping with the painful adjustments in its domestic industrial structure that were expected to result from entry into the WTO.7 Then, late in 1999, Japan and Singapore started talks aimed at a free trade agreement. This encouraged the Chinese government to make its own independent decision to pursue a regionalist policy for East Asia and led to Premier Zhu Rongji’s proposal in 2000 to start negotiations on the formation of an FTA between China and ASEAN. A year later, the leaders of China and ASEAN agreed at the China–ASEAN Summit in Brunei to begin talks aimed at concluding a free trade pact within ten years. What about Japan? As has often been noted, the Japanese government traditionally stressed multilateral talks, primarily under the auspices of the WTO, as the basis for its trade policy, and shied away from regionalism. This is why there was virtually no progress in institutionalizing regional integration in East Asia even though regionalization proceeded, driven by Japanese and overseas Chinese cross-border direct investment in the 1980s and early 1990s. In recent years, however, with the rapid acceleration of moves toward regional economic integration worldwide, Japan has come to recognize that the advantages of expanded trade and investment from such integration can outweigh the drawbacks of the division of the world into economic blocs. This awareness led to the start of talks with Singapore in 1999 aimed at the conclusion of Japan’s ﬁrst free trade agreement, which marked an important turning point for Japanese trade policy.8 The talks were completed in October 2001, and the resulting agreement dealt not just with trade liberalization but also with measures to promote trade and investment ﬂows, such as investment rules, mutual recognition agreements, and online handling of customs procedures, as well as cooperation in areas such

Evolution of institutions and policies

113

as IT and human resource development. For this reason, the agreement with Singapore was named the ‘Economic Agreement for a New-Age Partnership’.

5.5

NEW MOVEMENTS TOWARD REGIONAL INTEGRATION

The Asian currency crisis was followed by two new developments that had important implications for the future of the East Asian economy in the twenty-ﬁrst century. The ﬁrst of these has been the rise of a new middle class in East Asia. It is estimated that there are about 80 million middle-class consumers in China, in addition to those in Japan and South Korea. The new middle class in China enjoys shopping on the Internet with their own money earned in free economic activities.9 During the 1990s, a new middle class was created in the ASEAN countries amid rapid economic growth that was the result of a post-independence developmentalism that gave top priority, above all else, to economic development. The members of the new middle class, who supported the gradual promotion of democracy in ASEAN, experienced an economic shock during the currency crisis because they had turned heavily to the use of ﬁnance loans, particularly in Thailand. After the currency crisis, the new middle class comprised about 100 million highlevel consumers in ASEAN. This means that there is a market of 300 million consumers in the rising new middle class in the East and Southeast Asian countries. They have much in common with each other across borders in lifestyles, consumption patterns and business practices, and they all take advantage of globalization and information technology. They are likely to constitute the social basis for an emerging East Asian regional identity in which the people of the region see themselves as Asians, even with their diversity, in contrast to Europe, where religion forms a social basis for common identity. The second transformation is the search for spontaneous development in East Asia, a movement that is closely related to the rise of the new middle class. Economic development in ASEAN and China since the currency crisis has continued to depend on the promotion of foreign direct investment and exports. However, the emergence of the new middle class has exposed a gap between urban areas and rural areas and between the rich and the poor, and this has become a major political issue. East Asian countries have been obliged to solve this politically diﬃcult problem, as Japan was obliged to do in the 1970s. For instance, Thai Prime Minister Thaksin devised slogans calling for the eradication of poverty in rural villages, promoting SMEs and

114

General views on East Asian economic integration

endorsing the One-Village, One-Product movement in order to realize spontaneous and self-sustaining economic development.10 At the Fifth Plenary Session of the 16th CPC Central Committee in October 2005, China oﬃcially decided to narrow the gap between the rich and the poor under the slogan ‘harmonious society’.11 China’s engine for economic growth is changing from export demand to domestic demand, and this means that, before long, China will move from the stage of playing the role of world factory to the stage of new development. The development of a single market in East Asia will eventually be led by economic growth driven by domestic demand in the respective Asian countries, in addition to exports to the USA and Europe. The third movement is the establishment, by MNEs, of important manufacturing networks in East Asia. Economic growth driven by foreign direct investment has contributed to the development of industrial clusters of MNEs in East Asia’s manufacturing sector. East Asia has no equal as a production base because from the 1970s onwards, various clusters in electronics and automobile industries were established by MNEs from Japan, the USA, Europe, South Korea and Taiwan. In recent years, the proportion of Japan’s trade and investment with East Asia has been increasing because the MNEs have expanded their network of production, sales and procurement in East Asia (Figure 5.3). Against the background of the wide network of MNE activity in East Asia, trade in intermediate materials, and especially in general machinery and electrical machinery, has been increasing among the countries of the region (Figure 5.4). Japanese companies made investments in ASEAN long before the currency crisis and have intensively invested in China in the twenty-ﬁrst century. They are now promoting a pattern of advanced international specialization throughout the East Asian region (including Japan) using the catchphrase ‘China plus one’.12 They are expanding investment to produce low-value-added components and products in East Asia while producing high-value-added components such as high-precision engine parts and liquid crystal devices in Japan.13 The MNEs tend to constantly maintain and enhance their networks of production, distribution and procurement in order to keep ahead in the ﬁerce international competition in the region. It is not time for countries to select which investing companies they would prefer, rather it is time for companies to select which countries and which regions they would prefer as investment destinations. The same principle applies even to rapidly developing China. It can be expected that Asia-speciﬁc regional integration will be accelerated in such a way that the economic circumstances of the MNEs’ expansion into the region will strongly encourage institution-building measures such as the creation of new FTAs. For instance, based on the economic conditions

EU

Direct Investment to EU

Direct investment to NAFTA

38%→36%

39%→15%

Imports from EU 14%→13%

Japan Japan Changes in the last 5 years in the proportions of total trade and investment amount for Japan

(25 (25 countries) countries)

Exports to EU 18%→16%

115

Population: Approximately 450 million Real growth rate: Imports Approximately 0.9%

Population: 130 million Real growth rate: Approximately 1.8%

from East Asia

Exports to East Asia

36%→47%

39%→44%

Population: Approximately 1.91 billion Real growth rate: Approximately 5.3% Total for the 10 countries in ASEAN, China, South Korea, Taiwan, Hong Kong

East Asia (excluding (excluding Japan) Japan)

Imports from NAFTA 25%→16%

NAFTA

(North (North America AmericaFree FreeTrade Trade Area) Area) United United States, States, Canada Canada and Mexico and Mexico

Exports to NAFTA 33%→25% Direct Investment to East Asia 11%→26%

Population: Approximately 410 million Real growth rate: Approximately 2.8%

Reference: Proportion of foreign direct investment of Japan: 1999→ 2004 changes Proportion of exports and imports to Japan: 1999→ 2004 changes Population/Real Growth Rate: 2003 Document prepared by Ministry of Economy, Trade and Industry

Note: The United States has been Japan’s main partner for trade and investment, but recently the proportion of trade between Japan and East Asia has been increasing. Source: Ministry of Finance, Japan, Statistics Documents.

Figure 5.3

Deepening of mutual economic interdependence in East Asia (1)

52.0

1. The amount of general machinery parts 1990 2001

46.2

Japan

Japan 40.4

Japan 25.4

USA

EU 117.3

0.4

72.2

71.5

34.8

68.3

2.4

ASEAN-4 2.7

61.2

67.3 69.7

0.4

5.0

00.6

0 00.6

China 1.4

ASEAN-4 22.0

2.5 1.1

1.4 1.2

China 14.0

16.3 2. The amount of electrical machinary parts

19.3

116

ASEAN-101 44.4

21.3

1990

2001

Japan 30.8

Japan 63.6 11.0

24.6 1.1

China + Hong Kong

2.8

49.4

S. Korea ASEAN-4 8.7

(a) Trade relationships among Japan, ASEAN, China and Korea (USA, EU) (2003, US$ billion)

2.9

0.5 0.2

00.1 0.1

China 2.7

ASEAN-4 48.5

6.0 10.7 3.7 1.7 1.2

China 19.3

(b) Expansion of intermediate material trade (US$ billion)

Notes: 1 ASEAN-10 refers to Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippiness, Singapore, Thailand and Vietnam; 2 The figure in circles in (b) show the amounts of exports. Source:

IMF, DOT (FOB).

Figure 5.4

Deepening of economic mutual interdependence in East Asia (2)

Evolution of institutions and policies

117

under which Japanese companies have expanded their networks in East Asia, Japan has made eﬀorts to promote regional economic integration including the ASEAN-Japan Comprehensive Economic Partnership (AJCEP) in order to further promote economic development in the region. At the same time, Japanese companies are in the process of rebuilding their regional strategy in Asia, taking into account the movements toward institutional economic integration in East Asia. This means that economic integration in East Asia will be accelerated by the synergy eﬀect of economic conditions and economic institutions within the region (Figure 5.5).

5.6

IMPEDIMENTS TO REGIONAL INTEGRATION IN EAST ASIA

The countries of East Asia have made a variety of eﬀorts to conclude individual FTAs with other countries as a ﬁrst step toward realizing regional economic integration in the twenty-ﬁrst century. During the ﬁrst ﬁve years of this century, many FTAs were created in the region, but several impediments that hamper further regional integration still remain in place. 5.6.1

Growing Nationalism

There are two major impediments, largely absent in the USA and Europe, that stand in the way of eﬃcient economic integration in East Asia. The ﬁrst of these is the growing strength of nationalism in East Asia. Harmonization of economic systems is vital for building a seamless economic structure in East Asia. However, rising nationalism in the region tends to make compromise diﬃcult and threatens to destroy harmonious international relations within the region rather than providing an attractor for a uniﬁed Asian identity. Recently, it has often been the case that rising nationalism has developed as a byproduct of trade and investment liberalization and harmonization of economic systems in the Asian countries. For example, as ASEAN countries accelerate the process of tariﬀ reduction in their drive toward economic integration, some countries win and others lose the international competition for attracting foreign investment. There is a tendency among loser countries for nationalism to grow, led by domestic industries which have become dissatisﬁed with economic circumstances.14,15 Economic integration in East Asia could also suﬀer a setback if respective countries were to intensify the competition for inﬂuence in the region. For example, if China and Japan had a leadership struggle, it would jeopardize the promotion of economic integration in East Asia. There is a growing debate among East Asian countries on whether ASEAN+3 or 117

118 Note: China, Japan, South Korea, Hong Kong and Taiwan, together with the ASEAN-10, account for one-third of the world population (2 billion) and one-ﬁfth of the GDP of the entire world (US$7 trillion). By adding Australia and New Zealand, this group accounts for half of the world population (3 billion) and one-quarter of the world GDP (US$8.3 trillion) (as of 2003). Source: Ministry of Economy, Trade and Industry, Japan.

Figure 5.5

Development of economic partnerships in East Asia

Evolution of institutions and policies

119

ASEAN+6 should play a major role in community-building in this region. It has also been pointed out that such ASEAN countries as Singapore and Thailand could undermine and weaken ASEAN economic integration should they give priority to the formation of bilateral FTAs with countries outside the region.16 Rising nationalism runs the risk of spurring public distrust in some countries and engendering conﬂict rather than regional integration in East Asia. 5.6.2

Regional Diversity

The second impediment is regional diversity within East Asia, a feature that is more clearly apparent than in other regions of the world. Although it does not work against the regional integration in the long term, in the short term regional diversity increases the cost of realizing well-balanced regional development in East Asia. For instance, because the CLMV countries (Cambodia, Laos, Myanmar and Vietnam) of Southeast Asia have disparities in their economic development compared to other ASEAN countries, not only is the process of economic integration with ASEAN being delayed, but a transitional period has to be set for trade and investment liberalization and facilitation in the CLMV countries in FTA negotiations with their dialogue partners.17 In addition, regional diversity in terms of ethnicity and political systems, together with rising nationalism, increases the cost of building up political and security aspects as well strengthening the social and cultural aspects of Southeast Asia, aspects whose role is being discussed in the process of establishing the ASEAN Community and the East Asia Community. For example, because of this diversity, a long period of time was required, in the process of constructing the ASEAN Community, for ASEAN countries to agree upon the ASEAN Charter, which incorporates the concept of democracy and human rights with which the member states should comply.18

5.7

POSSIBLE THRUSTS FOR DEEPER REGIONAL INTEGRATION

In this section, we discuss ﬁve eﬀective measures for promoting deeper regional integration in East Asia: (1) comprehensive support for developing infrastructures; (2) fostering middle-class communities and deepening international exchanges among them; (3) creating a new mechanism for regional economic development; (4) resolving energy and environmental issues; and (5) developing high-level comprehensive FTAs. We shall examine each of these measures with special reference to Japan’s role.

120

5.7.1

General views on East Asian economic integration

Comprehensive Support for Developing Infrastructures

Further economic integration in East Asia requires development of infrastructure, in both physical and ‘soft’ forms. Here we look at each of these in some detail. Cross-border physical infrastructure As regards cross-border physical infrastructure, several major infrastructure development projects have already started. High-speed motorway networks are one example of cross-border infrastructure in the area of logistics. Major highways such as the East–West Corridor and the South Corridor have been developed, and international bridges have been built across the Mekong River with the assistance of Japan’s yen loans for development in the Mekong region. In order to promote electriﬁcation in the Mekong region, cross-border electricity supply grids have been established with the aid of loan programs provided by international organizations and donor countries, under the framework of the GMS Development Program initiated by the Asian Development Bank (ADB).19 It is expected that such examples of cross-border economic cooperation in the transportation and energy sectors will accelerate the development of special economic zones on the borders of Thailand, Cambodia and Laos, will lead to the development of regional industrial and logistical networks, and will eventually contribute to economic development through sub-regional integration within the Mekong Region20 (Figure 5.6). Until recently, cross-border infrastructure developments such as international highway networks and electricity supply grids have seldom been the targets of ODA ﬁnance. For instance, Japan has not tended to prioritize development of such infrastructure projects because individual ASEAN countries were supposed to request provision of yen loans on their own initiative. But now the situation is changing, and as we have seen, development of an ASEAN-wide physical infrastructure is beginning to gain a higher priority. ‘Soft’ infrastructure Development of cross-border physical infrastructure alone is not suﬃcient to bring about a deepening of regional integration. We also need ‘soft’ infrastructure such as logistics, standards and conformance, intellectual property rights and environmental protection.21 ASEAN faces various problems, including a delay in liberalization as well as a delay in facilitation and system harmonization. Harmonization of economic systems such as those for intellectual property rights, standards and conformance, and customs procedures is indispensable for realizing

121

Evolution of institutions and policies

Hanoi

Myanmar Lao PDR Vientiane Yangon

Savannakhet

Mae Sot Myawadi

Thailand

Mukdahan

Bangkok

Poipet Pailin Koh Kong

Lao Bao

Vietnam

Cambodia Phnom Penh

Ho Chi Minh Bavet Sihanoukville Moc Bai

Notes: 1. The logistical networks link the industrial clusters such as Eangkok–Hanoi, Bangkok–Ho Chi Minh and Bangkok–Yangon. They are mainly used for transportation of materials and components for automobile, electric and electronics, and textile industries. 2. Special economic zones include Polpet, Koh Kong, Sihanoukville, Pailin, Bavet (Cambodia), Savannakhet (Lao PDR) and Myawadi (Myanmar). They promote investment in garments and textiles, parts and components industry, agricultural and food processing and cottage industries. Source: Ministry of Economy, Trade and Industry, Japan.

Figure 5.6

Industrial and logistical network in the Mekong region

seamless economic activities in the region.22 Problems such as the lack of system harmonization and delays in the liberalization of service industries have a more serious eﬀect in ASEAN than they do in China, Japan and South Korea. From the viewpoint of spatial economics, this diﬃculty prevents ASEAN from fully exerting the HME, and could become a serious obstacle on the road to continuing economic development as well as economic integration in East Asia.23 Trade facilitation, harmonization of economic systems and liberalization of the service market would oblige ASEAN to commit itself to the acceleration of economic integration. For instance, promotion of logistical eﬃciency is necessary for achieving economic integration and improving the investment environment in ASEAN.24 In ASEAN, improvement of key regional distribution routes in

122

General views on East Asian economic integration

terms of soft infrastructure is a quite important option among many policy measures for reducing logistics costs and lead times. The current land transport system means that cargoes are often held up for long periods for customs procedures to be carried out at border crossings on the routes linking Thailand–Malaysia–Singapore and Thailand–Vietnam, for example. To develop eﬃcient logistics, ASEAN countries ﬁrst need to conduct some pilot projects by sending trucks with cargo along key regional distribution routes and promoting eﬃciency of customs procedures through the use of RFID (radio frequency identiﬁcation).25 So far as practical problems are concerned, however, a quite serious diﬃculty is the shortage of the manager-level human resources that are needed to upgrade logistics industries. ASEAN may well be able to develop human resources in logistics services by employing a national certiﬁcation system for logistics specialists such as the one that is used in Japan. Moreover, ASEAN is scheduled to promote the ASEAN Single Window System, which is aimed at digitizing customs procedures in the member states by 2012. A pilot project should be conducted in the near future to support the establishment of the electronic customs processing system by utilizing its electronic systems technology26 (Figure 5.7). In the case of China, it is widely recognized that China has succeeded in establishing a free economic system but has failed to secure a level playing ﬁeld for foreign investors. China’s entry into the WTO is highly signiﬁcant in this respect. It signiﬁes a promise on China’s part to pursue its development on the same level playing ﬁeld and according to the same rules as are observed by the rest of the world. According to a JETRO survey on the problems experienced by Japanese companies in trying to operate in the trade and investment environment in China, among the priority issues that need to be tackled are the insuﬃciency and unpredictability of the legislation and tax systems, unfair enforcement and non-uniformity in the legislation, infringement of intellectual property rights, and restrictions on currency exchange and remittances (Figure 5.8). In particular, Japanese companies suﬀer from serious problems related to infringement of intellectual property rights, including counterfeiting and piracy of their products. The problems have become more serious quantitatively not only concerning products traded in China but also with regard to goods exported to third countries. Moreover, there are also increasingly serious qualitative problems such as infringements of trademarks and of industrial designs and patents.27 Taking these points into consideration, it is imperative to develop a soft infrastructure, including an antimonopoly law, intellectual property rights, a judicial system, corporate governance and information disclosure, so that

Program Option (A): Logistics efficiency in ASEAN

Advancement of ASEAN logistics • Human resource development • Utilization of RFID Program Option (B):

Cooperation for facilitation of trade-related procedures in ASEAN • Support for implementation of ASEAN single window

Improvement of 6 key wide-area routes Utilization of advanced tools Human resource development for logistics Facilitation of trade-related procedures

123 Program Option (C):

Trial projects in Mekong region to realize improved routes • Pilot-run projects for building best-practice in overland supply chain in ASEAN (including improvement of infrastructure, facilitation of trade-related procedures, IT utilization, etc.)

Hanoi

Bangkok Ho Chi Minh

Source: Drafted by Public–Private Partnership on International Logistics Competitiveness.

Figure 5.7

Action program for logistics eﬃciency within ASEAN

0

100

200

Sufficiency and transparency in the legislation

300

400 (number of respondents)

Sufficiency and transparency in the tax system Fair enforcement and uniformity in the legislation Tariff rates Protection of IPR Export and import licenses MFN, national treatment

124

Liberalization of trade in service Import quota system, import bans Anti-dumping and safeguards Distribution licenses Standards and conformance, SPS Others Note: Multiple answers allowed (783 respondents out of 1330 pointed out speciﬁc problems). Source: Inquiry survey of Japanese companies in China conducted by JETRO (2003a), p. 26.

Figure 5.8

Problems of Japanese companies in the trade and investment environment in China, 2003

Evolution of institutions and policies

125

free business activities may be guaranteed. Regional integration requires support for seamless economic activities through consistent implementation of WTO commitments and the development of a soft infrastructure. In order to improve the business environment in China, it is important to begin talks on a trilateral investment treaty among China, Japan and South Korea, to ensure national treatment rather than investment protection. It is also important to establish a consultation mechanism for improving the business environment among these three countries, focusing on those issues of rule enforcement that cannot be suﬃciently covered by the trilateral investment treaty.28 Meanwhile, it will be important for Japan to change its way of thinking on how to support the development of the East Asian countries. Whereas hitherto Japan has contributed to the development of physical infrastructure projects through economic aid provisions made possible by its outstanding economic strength, it now needs to attach higher priority to the development of a soft infrastructure so as to encourage the East Asian countries to realize spontaneous economic development. In other words, the quality of Japan’s cooperation and human resources should be examined more closely than ever, since, in the future, aid will focus more than hitherto on development of soft infrastructure and institution-building. Japan has comparative advantages in business practices and know-how, including fostering of skilled labor, trade practices and development of supporting industries, and also in economic systems and policy measures including IT skill standards, SME policies, energy conservation, recycling methods and logistics standards.29 It is to be hoped that Japan will conceptualize these practices and systems and share them with East Asian countries as possible regional standards. 5.7.2 Fostering the Middle Classes and Deepening International Exchanges Among Them The creation of transparent and participatory systems in terms of politics, economy and society to foster a new middle class is important. The rising middle class in East Asia plays a political role by supporting democratization in respective countries and an economic role in so far as middle-class people constitute major consumers in the regional market. It should be noted that the ‘development ﬁrst’ strategy, which once served as the basis for stability and prosperity in East Asian countries, has lost its power and legitimacy as a paradigm, and that in many countries, the search is on for fundamentally more liberal political and economic systems, though the details diﬀer greatly from country to country. This is why attempts have been made in many countries to further democratization, decentralization and governance reform.

126

General views on East Asian economic integration

It should also be noted that attempts have been made to create more transparent and participatory social and economic systems that will allow middle-class people to play greater roles. For instance, the development of a sustainable economic society is expected to be achieved through the strengthening of the economic/legal infrastructure including ﬁnancial sector reforms (above all the disposal of non-performing loans) as well as improvements in bankruptcy and antimonopoly legislation for a fair and competitive environment. These have become important issues ever since the Asian currency crisis of 1997. In order to further foster the growth and well-being of the middle class in East Asia, it will be important to develop human and management resources through the promotion of small and medium-sized enterprises, human resources development in the ﬁeld of information technology, and educational reforms. In addition, development of transparent, fair tax and social security systems are likely to lead to the improvement of economic and social stability.30 In order to realize regional integration in East Asia, it is sometimes necessary to change centrifugal forces such as diversity and nationalism into centripetal forces such as respect for other countries. With this in mind, it is more important than ever to deepen international exchanges not only between young people who carry the future of Asia in their hands, but also between intellectual middle-class people who share a similar lifestyle and sense of values. The time has come when it is very signiﬁcant for the nation of Japan, including its corporations and its citizens, to win the trust and respect of middle-class people in East Asia, a requirement that is essential if Japan is to coexist peaceably and constructively with other Asian countries in the long term. It is imperative for Japan (i) to promote globalization of universities, research institutes and personnel, (ii) to enhance international competitiveness in tourism and services to attract foreign visitors, and (iii) to convey Japanese culture and branding to the rest of East Asia.31 5.7.3

Creating a New Mechanism of Regional Economic Development

The recent development of the East Asian economy has been largely brought about by the expansion of trade and investment. It is important for this region to continuously improve the trade and investment environment so as to create a virtuous cycle of economic growth among Asian countries. For that purpose, Asian countries, particularly Japan, ought to encourage regulatory reform and the improvement of legal systems to facilitate free economic activities. They should also support innovation in terms of human resources, production facilities, infrastructure, ﬁnance, technology and management skills in order to create new businesses, including service industries.32 It is vital for Japan (and not only for the growing Asian

Evolution of institutions and policies

127

countries) to drive the economic growth of the other countries in the region through these economic policy measures. It is very much to be hoped that these policy measures will come to constitute best practice, to be shared among developing Asian countries in the future. At the same time, it should also be recognized that the economic development process in East Asia has shifted from the ﬂying-geese pattern led by Japan to a leap-frog pattern created by such emerging Asian countries as China. Since China has become a strong driving force of economic growth in East Asia, it can be assumed that both Japan and China, as the two engines of growth in the region, will drive the region’s economic development. It should be kept in mind that building healthy political and economic relations between China and Japan will substantially contribute to economic integration in the region, which includes, also, ASEAN and South Korea. It is also important to establish a policy coordination mechanism for economic integration in East Asia. Examples include ﬁnancial cooperation schemes such as the Chiang Mai Initiative that were developed to address the emergent ﬁnancial problems caused by large external shocks such as the Asian currency crisis. Although it has not been pursued to its full extent because of regional diversity problems, one of the most important challenges facing East Asia is to establish a policy coordination mechanism to harmonize economic policies and systems. The ASEAN+3 framework, under which various ministerial meetings have been established, should be reorganized to function in practical terms as a policy coordination mechanism. A permanent policy coordination mechanism can be expected to play a useful role in harmonizing economic policies and systems in East Asia in the future33 (Figure 5.9). 5.7.4

Resolving Energy and Environment Issues

Lately, the international energy market has been tight because Chinese demand for energy has risen swiftly as a result of rapid economic growth and the presence in China of ineﬃcient manufacturing plants and equipment. Indeed, the rapid rise in China’s oil imports has come about through an expansion of domestic energy consumption based on an energy eﬃciency that is about one-ninth that of Japan’s. As for environmental issues, China is responsible for a level of SO2 (sulfur dioxide) emissions that is the world’s highest, and more than 20 times that of Japan, a state of aﬀairs that is the ultimate cause of acid rain problems in neighboring countries (Figure 5.10). Japan has proactively implemented energy cooperation with Asian countries in such areas as energy conservation, new energy, clean-coal technology, oil stockpiling and nuclear energy. It has

E E E

128 Source: Ministry of Economy, Trade and Industry, Japan.

Figure 5.9

ERIA (the Economic Research Institute for ASEAN and East Asia)

Evolution of institutions and policies

129

also promoted environmental cooperation aimed at limiting emissions of CO2 and SO2, and at accelerating the creation of a recycling-oriented society. China and Japan should continuously carry out joint cooperative projects in the areas of energy and the environment under the framework of ASEAN+3 and the East Asia Summit. Cooperation in these areas could make an important contribution toward the building of mutual economic and political trust in the region.34 5.7.5

Developing High-level Comprehensive FTAs

Modalities of bilateral FTAs in East Asia may well need to be reviewed. For instance, China has been trying to be proactive and expeditious as regards the negotiations of the ASEAN–China FTA. However, so far, it has mainly focused on trade liberalization, and has taken a longer time to proceed with negotiations in such important areas as trade in services and investment. It has also become apparent that important items related to the automobile and electronics industries are being excluded from normal track negotiations. It has been pointed out that China has promoted FTAs in East Asia for political reasons rather than for economic ones, and thus it has not contributed fully to regional economic integration in any real sense.35 It is certainly important to assess the impact of China’s regional initiatives on the East Asian regional system (Figure 5.11). Considering that the ASEAN–China FTA might be one of the frameworks in which to build up economic integration in East Asia, eﬀorts should now be made to apply the brakes on the proliferation of low-level FTAs in the region in order to realize comprehensive and high-level economic integration in terms of trade and investment liberalization and facilitation.36 In the medium and long term, East Asian countries should explore the possibilities of realizing FTAs under the framework of the (East Asia Summit) EAS and APEC in cooperation with the USA and Australia, both of which are promoters of high-level liberalization and rule-making in trade and investment. For that purpose, and at least in the short term, the countries of East Asia should also employ APEC as a forum in which to develop and share best practices and model measures for regional trade agreements and free trade agreements.37

5.8

FUTURE DIRECTION OF EAST ASIAN ECONOMIC INTEGRATION

The East Asian countries have promoted institutionalization for regional economic integration step by step. There are multiple FTAs which are like

gy c) Self-Sufficiency Ratio of Mineral Resources in China

a) Energy consumption per GDP in major countries 20.0

Change from export country to import country

160

15.0

10.0

- China’s energy -ChinaÕs energy efficiency is oneefficiency is oneninth of Japan’s ninth of JapanÕ s

9.0 7.1

5.0

3.3

2.7 1.0

1.6

Japan

EU

3.8

4.3

Self-Sufficiency Rate

17.6

140 120 100

Zinc Lead

80 60

Copper

Nickel

40 20

0.0 Primary energy consumption / GDP (Japan = 1)

U.S.

South Canada ASEAN Middle China Russia East Korea

0

b) Oil production and consumption in China

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

d) SO2 Emissions in Major Countries (thousand ton)

Million Barrel / Day

25,000

16 14

Net oil importer since 1993

130

12

Consumption

- China rapidly increased oil imports due to increased energy consumption

10 8 Net Import

6 4

20,000

China

15,000

U.S.

10,000 5,000

2

Australia Japan

Production

0 1980

- Demand for -Demand for domestic mineral domestic mineral resources increased resources increased while the utilization while theutilization efficiency is low is lowrate - efficiency Self-sufficiency rate of-Self-sufficiency mineral resources of mineral resources rapidly decreased due rapidly decreased to dependency on due to dependency foreign minerals on foreign minerals

1990

2000

2010

2020

2030

- The -Thelevel levelofofChina’s ChinaÕ s SO is is the 2 emissions SO2 emissions the world’s highest and more worldÕs highest and than 20 times of that of more than 20 times of Japan that of Japan - The acid rain problem acidrain problem in-The China adversely in China affected itsadversely neighboring affected its neighboring countries countries

0 1997

1998

1999

2000

2001

2002

2003

2004

The 11th Five-Year Guidelines: Aim to decrease energy consumption per GDP by 20% by 2010 by 20% by 2010

Aim to realize a resource-saving and environmentally friendly type society

Source: a) IEA Energy Balance, 2004; b) IEA World Energy Outlook, 2004; c) WBMS; d) United Nations Framework Convention on Climate Change (UNFCC), Greenhouse Gas Inventory Database.

Figure 5.10

China’s countermeasures for energy and environmental problems

New Zealand • Signed the China-New Zealand Trade and Economic Cooperation Agreement in May ’04. Negotiations began in December ’04.

ASEAN

ASEAN (overall) • At a summit meeting in November ’04, ChinaASEAN leaders signed an agreement covering the trade of goods. Tariffs will be reduced by 2005. • Next, China will promote negotiations on services and investment, with the objective of concluding an agreement by 2010.

Australia

Thailand • The Early Harvest program of the Framework Agreement on ASEAN-China Comprehensive Economic Cooperation was carried out ahead of schedule. Approx. 200 agricultural items were made non-dutiable from January ’03 onward.

• Signed the Australia-China Trade and Economic Framework in October ’03. • Agreed that current joint studies on a bilateral FTA will be completed by March ’05.

Vietnam • The Early Harvest program of the Framework Agreement on ASEAN-China Comprehensive Economic Cooperation is in effect.

Gulf Cooperation Council (GCC) • Free trade negotiations began in September ’04. (GCC: UEA, Bahrain, Kuwait, Oman, Qatar, Saudi Arabia)

Singapore • Preparations underway to begin negotiations.

131

Southern African Customs Union (SACU) • In June ’04 at a meeting of the China-South African Economic Committee, it was decided to begin negotiations on a China-SACU free trade agreement. (SACU: South Africa, Botswana, Namibia, Lesotho, Swaziland.)

China

South Korea • In September ’04, reached agreement to have a think tank conduct joint studies (date undecided). Japan/China/South Korea • A trilateral research body is conducting joint study on an FTA • At a summit meeting of Japanese, Chinese and South Korean heads of state in November ’04, agreement was reached to begin inter-governmental discussions to sort out a legal framework for investment.

Chile • Joint studies ended in September ’04. In November ’04 at the APEC summit meeting, the leaders of China and Chile agreed to enter into negotiations. The first round of negotiations was held in January ’05. Hong Kong India • Agreement was reached (at a summit meeting in June ’03) to set up a joint study group, composed of government and academic personnel, to look into areas where both countries could complement one another.

• CEPA concluded in June ’03. • In addition to the 374 items (starting January 1, 2004) a further 713 items will be made non-dutiable from January 1, 2005. In addition to the 18 service sectors, access in 8 new sectors is slated to be liberalized.

Source: Ministry of Economy, Trade and Industry, Japan.

Figure 5.11

Current status of FTAs between China and other countries

132

General views on East Asian economic integration

spokes centering on ASEAN as a hub. These include the FTAs of ASEAN–Japan, ASEAN–China and ASEAN–Korea. It will be important in the ﬁrst stage to make the current FTAs eﬀective as soon as possible, and in the second stage, to integrate the current FTAs into one single FTA. In order to facilitate this process, Japan, China and South Korea need to further develop liberalization, facilitation and cooperation among themselves. In addition, discussions are being held among East Asian countries on whether or not it is desirable to aim at developing regional integration through the building-block approach by adding India, Australia and New Zealand to the ASEAN+3, thus creating a framework of ASEAN+638 (Figure 5.12). Furthermore, it is necessary to discuss how East Asian countries should take advantage of other fora such as multilateral trade systems including the WTO and regional economic architectures that involve the USA, APEC being an example, in the strengthening of governance under international rules throughout the region.39 There is a growing debate on how to promote regional integration aimed at the establishment of an East Asian Community.40 The EU provides a good example for establishing a regional community where the respective countries share the same sense of values, including political values, under a common institutionalized framework. We are not opposed to the current discussion on the possibility of creating an East Asia Community, but believe that a more feasible route to community formation is to direct substantial eﬀorts towards upgrading the level of the existing FTAs with a view to integrating these into a pan-East Asian FTA as a ﬁrst step toward.

5.9

CONCLUSION

To summarize, this chapter has presented an overview of the history and prospects of economic integration in East Asia, in terms of institutions and policies. It has shown how East Asian countries underwent structural changes after the Asian currency crisis, and has described the process whereby the countries of the region have moved along the path of economic integration. Finally, we have argued that the East Asian countries are now encountering economic and political issues that must be squarely tackled if steady progress is to be made in the integration process. To go into more detail, Japanese manufacturers, above all electronics and automobile ﬁrms, have expanded their investments in Asian and have established an international network of production, procurement and sales across the region since the 1985 Plaza Accord. While such de facto economic integration has made substantial progress in East Asia, the Asian currency crisis

Basic Concept of CEPEA Comprising ASEAN and its FTA/EPA partners (China, Japan, Korea, India, Australia and New Zealand) (ASEAN+6) Covering a wider agenda than FTA (trade in goods, common rules of origin, services, investment, intellectual property, cooperation, etc.) Further developing regional production networks unique to this region Establishing a Free, Fair and Rule-based Market Economy ASEAN as a driving force of East Asian Economic Integration Keeping the open dynamism of the partners who engage in this region Proposed First Step Commencing track-two study group of CEPEA by experts of ASEAN+6 countries Background

2. Development of FTA/EPAs in the region

1. Developing de-facto integration in East Asia

133

ASEAN is working on FTA negotiations with its dialogue partners that may be concluded by 2007.

< Intra-regional Trade Ratio >

(%) 70

EU: Intra-regional market integration completed

EU: Agree to form single market

EU: Euro introduced

60%

EU15

60

ASEAN -China

East Asia

50

ASEAN-Korea

ASEAN-Japan

57% ASEAN-India

ASEAN:CEPTcommences

38%

45%

40 NAFTA:enters in to force

30 1980

1982

Source: IMF DOT

1984

1986

1988

1990

ASEAN-Australia-NZ

EU15 NAFTA

NAFTA

East Asia 1992

1994

1996

1998

2000

2002

(year)

Board of Foreign Trade, Taiwan, Chinese Taipei “Trade Statistics"(http://eweb.trade.gov.tw/default.asp)

*ASEAN countries: Brunei Darussalam, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand, Vietnam

ASEAN (AFTA)

Note: East Asia includes Japan, China, Korea, Hong Kong, Taiwan, ASEAN-10, Australia, New Zealand and India. Source: Ministry of Economy, Trade and Industry, Japan.

Figure 5.12

The vision of CEPEA (Comprehensive Economic Partnership in East Asia)

134

General views on East Asian economic integration

of 1997–98 constituted a turning point in the evolution of regional economic policy. After the currency crisis, Japan changed its WTO-oriented multilateral trade policy, China came to the economic forefront with its entry into WTO, and both countries proactively developed frameworks for economic integration in East Asia. At the same time, economic integration has been further expanding and deepening in East Asia due to the emergence of a new middle class across the region and as a result of the search for spontaneous development in the Asian countries since ASEAN and China resumed an economic recovery track after the currency crisis. Although this region is still faced with growing nationalism, regional diversity and delay in the development of soft infrastructure related to economic institution and policybuilding, it is widely expected that the ASEAN countries and their dialogue partners will conclude a number of FTAs, and that this will lead to the development of a single East Asia-wide FTA in the future. As mentioned above, the countries that belong to this region are in the process of steadily developing frameworks of de jure economic integration so as to give a boost to de facto economic integration driven by the private sector. From the viewpoint of spatial economics, there is a strong likelihood that trade facilitation, harmonization of economic systems and liberalization of trade in services will enable East Asia as a whole to fully exert the HME, leading to further economic development as well as economic integration across the region.

ACKNOWLEDGEMENT We would like to extend our appreciation to Dr Takashi Shiraishi for allowing us to refer to his research paper.

NOTES 1.

2.

Japanese manufacturers made major investments in Asian NIEs such as Singapore, Taiwan and South Korea and in ASEAN countries such as Thailand, Malaysia and Indonesia following the 1985 Plaza Accord. They expanded their investments in China and Hong Kong during the ﬁrst half of the 1990s. Some Japanese automotive companies in ASEAN have established the following system of international specialization: (Thailand) diesel engines, evaporators (air conditioners) (Indonesia) gasoline engines, horns (Malaysia) engines, condensers (air conditioners) (Philippines) transmissions, combination meters.

3.

It has been pointed out that, at the time of the Asian currency crisis, Asian countries were faced with a variety of issues concerning the need for corporate and ﬁnancial sector

Evolution of institutions and policies

4.

5.

6.

7.

8.

9.

135

reform. These included enhancing the functions of ﬁnancial institutions, promoting small business ﬁnance, developing domestic bond markets, strengthening the enforcement of bankruptcy law, and improving corporate governance. At the Sixth ASEAN Summit in December 1998, ASEAN leaders announced a series of ‘bold measures’ as one of the initiatives aimed at revitalizing the ASEAN economy after the Asian currency crisis. In this connection, they agreed that the six original signatories to the agreements on the Common Eﬀective Preferential Tariﬀ (CEPT) scheme for the ASEAN Free Trade Area (AFTA) would bring forward the implementation of AFTA by one year from 2003 to 2002. However, according to the Leaders’ Statement, by 2002, 100 percent of items in the Inclusion List of the CEPT would have tariﬀs of 0–5 percent with ‘some ﬂexibility’. ASEAN also excluded from the list of products with tariﬀ reduction under this initiative some sensitive products, such as automobiles in the case of Malaysia and rice and sugar in the cases of Indonesia and the Philippines. Using the ASEAN–China Free Trade Area (ACFTA) as a basis, China launched the Early Harvest Programme to reduce tariﬀs on vegetables and fruit at the beginning of 2004 and also implemented the Trade in Goods Agreement to reduce tariﬀs on general goods in July 2005. China is scheduled to complete FTA negotiations with ASEAN on other chapters such as trade in service and investment by 2010. The Chinese economy has grown rapidly thanks to the investment boom that has occurred since 2002. That said, there is a risk that the investment boom will turn into a bubble and that excessive supply in the medium and long term will lead to an economic slowdown and an increase in the quantity of non-performing loans. Since early 2004 the Chinese government and the central bank have strengthened policy measures to restrain the investment, focusing on industries such as iron and steel, aluminum, cement, automobiles and real estate, all of which have suﬀered from excessive investment problems. In response to the widening of economic disparities between urban and rural areas, and between coastal and inland areas, the Chinese government has redoubled its eﬀorts to address the so-called three agricultural problems related to the farming industry, farm villages and farm people. Moreover, the Chinese government has announced key policies aimed at well-balanced economic development, including projects for developing not only the country’s western region but also the northeastern region, where many state enterprises are located. So far, China has implemented several cooperation projects including the Training Program on Practical Technology for Environmental Protection, the Seminar on Economic and Technological Development Zones, the Senior International Engineering Project Management Training Program for ASEAN Countries, and the Agricultural Technology and Management Training Program for ASEAN countries. It has also taken the lead in launching, in 2005, the East Asia Free Trade Area (EAFTA) Expert Group under the framework of the ASEAN Economic Ministers and the Ministers of the People’s Republic of China, Japan and the Republic of Korea (the so-called AEM Plus Three). At the Council of Ministers on the Promotion of Economic Partnership, Japan approved the ‘Basic Policy towards further promotion of Economic Partnership Agreements (EPAs)’, on 21 December 2004. This describes the signiﬁcance of promoting economic partnership agreements in the following terms: ‘Economic Partnership Agreements (EPAs), against the background of growing economic globalization, contribute to the development of Japan’s foreign economic relations as well as the attainment of its economic interests as a mechanism to complement the multilateral free trade system centering on the WTO. Simultaneously, EPAs facilitate promotion of structural reforms of Japan and its partners. These EPAs contribute to the creation of an international environment that is further beneﬁcial to our country from politically and diplomatically strategic views via, among other things, fostering the establishment of an East Asian Community’. ‘Consumer markets are expanding because of rising income levels caused by rapid economic growth in East Asia. In the 1990s, in the coastal cities of China, a higher income group emerged and formed a new middle class that included the managers and engineers

136

10.

11.

12.

13.

14.

15.

General views on East Asian economic integration of foreign companies and ﬁnancial and IT companies, private-sector entrepreneurs, individual corporate managers, and professional personnel such as doctors, lawyers and accountants. Meanwhile a new middle class in the ASEAN-4 (Indonesia, Malaysia, the Philippines and Thailand) is again on the rise following economic recovery from the Asian currency crisis’ JETRO (2003b), p. 37. ‘Thailand’s economic growth strategy, called the “dual track policy”, was introduced by the Thaksin administration and seeks both to promote domestic demand and to attract increased foreign investment. While other Asian countries are adopting similar approaches, Thailand is unique in that the dual track policy is based on an objective analysis of the Thai economy, and diﬀers markedly from the economic development models pursued in Europe and North America, and from the “V formation” [ﬂying-geese pattern] model followed by Japan and Korea. So far as export strategy is concerned, the dual track policy is characterized by the development of “niche export demand”. The Thai authorities have concluded that conventional price competition based on mass production is a diﬃcult way to compete, and instead have decided to analyze and take advantage of Thailand’s own particular strengths and develop a unique character so as to achieve diﬀerentiation. The dual track policy also seeks to spur sustained domestic demand by strengthening the nation’s economic underpinnings (for example rural society) and by tapping into unique local characteristics for the beneﬁt of community revitalization, thereby breaking free of an economic structure that has been hitherto characterized by excessive reliance on external demand. In concrete terms, to spur consumption, the Thai government has developed the necessary legal infrastructure for recognizing the asset value possessed by owners of land, intellectual property, machinery and food stalls, thereby making it possible for the owners to pledge those assets as security and tap into capital markets. As for the provision of funds, the government is lending support in this regard through public-sector ﬁnancial institutions’ (Ministry of Economy, Trade and Industry (2004), p. 268. China has coined a national slogan, ‘harmonious society’, to describe its aim of achieving well-balanced economic development among its regions and economic strata. ‘Harmonious society’ means building up society for the people, and signiﬁes the alleviation of negative features of economic development. The purpose is to enable more people to feel and beneﬁt from aﬄuence. (See the website of JETRO (Japan External Trade Organization) Beijing.) With regard to countries and regions where Japanese ﬁrms hope to expand their business functions in terms of sales, production, research and development, there has been a fall in the proportion of Japanese ﬁrms aiming to expand the production of generalpurpose goods and sales in China since the previous survey. There are increases in the proportion of Japanese ﬁrms that aim to expand their production functions in Thailand and sales in India and Vietnam since the previous survey. It is thought that Japanese ﬁrms are taking a ‘China plus one’ approach in order to reduce the risks of excessively concentrating their production in China. Source: JETRO (2006b), Annex. Many Japanese companies produce high-value-added parts and components in Japan, such as panels for LCD and PDP televisions, hard disks for portable music players, steel and plastic products for auto parts and forged parts for motorbike engines. These highvalue-added parts and components are then assembled into ﬁnal products elsewhere in East Asia. According to the report of the Japan Automotive Manufacturers Association (JAMA) on the domestic sales volume of automobiles in major ASEAN countries in 1996 (before the Asian currency crisis) and in 2005, the following increases in sales occurred: in Indonesia from 332 000 in 1996 to 553 000 in 2005; in Malaysia from 365 000 in 1996 to 551 000 in 2005, in Thailand from 589 000 in 1996 to 703 000 in 2005. During the same period, there has been a decrease of sales in the Philippines, from 162 000 in 1996 to 97 000 in 2005. Malaysia delayed the introduction of tariﬀ reductions on completed cars and the Philippines delayed tariﬀ reductions on petrochemicals under the Common Eﬀective Preferential Tariﬀ (CEPT) scheme for the AFTA.

Evolution of institutions and policies 16.

137

Apart from regional FTAs between ASEAN and its dialogue partners (Japan, China, South Korea, India and CER (Australia, New Zealand)), Singapore and Thailand have (substantially) agreed on bilateral FTAs with the following countries and regions: Singapore with Japan, South Korea, India, Australia, New Zealand, the USA, EFTA, Jordan, and P4 (Singapore, Chile, New Zealand and Brunei). Thailand with Australia, New Zealand, Japan (substantially agreed), Peru (agreed on protocol on early harvest program in some products) China, and India (tariﬀ reduction through early harvest scheme).

17.

18. 19.

20.

21.

According to the Agreements on the Common Eﬀective Preferential Tariﬀ (CEPT) scheme for the ASEAN Free Trade Area (AFTA), in principle, the ASEAN-6 countries (Brunei, Indonesia, Malaysia, the Philippines, Singapore and Thailand) will eliminate tariﬀs on all goods by 2010 and the new member countries (Cambodia, Lao PDR, Myanmar and Vietnam) will eliminate tariﬀs on all goods by 2015. The ASEAN countries will eliminate tariﬀs on goods in priority integration areas three years earlier than the above-mentioned schedule. In general, as for the implementation of ASEAN’s FTA with China, India and Japan, the time frame for the new member countries is ﬁve years later that that for ASEAN+6 countries. For further details of the ASEAN Charter, refer to the ‘Kuala Lumpur Declaration on the Establishment of the ASEAN Charter’ (http://www.aseansec.org/18030.htm), which was adopted at the ASEAN Summit in December 2005. The Greater Mekong Subregion (GMS) comprises Cambodia, the People’s Republic of China, the Lao People’s Democratic Republic, Myanmar, Thailand and Vietnam. In 1992, with the ADB’s assistance, the six countries entered into a program of subregional economic cooperation, designed to enhance economic relations among the countries of the group. The program has contributed to the creation of an infrastructure to enable the development and sharing of the resource base, and to promote the freer ﬂow of goods and people within the subregion. It has also led to the international recognition of the subregion as a growth area. (Excerpt from ADB website: http://www.adb.org/ gms/) Special economic zones (SEZs) are planned for Poipet, Koh Kong and Sihanoukville in Cambodia and Savannakhet in the Lao PDR, which are located close to the border with Thailand. In this connection, the Japanese government has supported the development of physical infrastructure in transportation and electricity supply, and a soft infrastructure in the form of SEZ laws and oﬃcials in charge of SEZ management. Japan implemented the following cooperation on the development of soft infrastructure toward ASEAN countries: Intellectual Property Rights: a training program for IPR-related examination practices and enforcement, support for developing systems to provide information on IPRrelated examination; Logistics: enhancement of the ASEAN logistics network, development of human resources in logistics, facilitating the informatization of logistics, promoting the eﬃciency and rationalization of customs systems; Standards and Conformance: the ASEAN Standard and Conformance Cooperation Program centres on capacity-building in the areas of standardization, conformity assessment, and metrology); Energy and Environment: support for introducing systems of qualiﬁed personnel for pollution prevention management and energy management; Industrial Human Resource Development: small and medium enterprise management consultant system, automotive industry cooperation (dispatch of roving automotive experts, establishment of an automobile skill certiﬁcation system), certiﬁcation systems for information-processing engineers, academic–industrial alliance cooperation.

138

General views on East Asian economic integration

22. ASEAN aims to build up the ASEAN Economic Community after achieving tariﬀ reductions and tariﬀ elimination to some extent through AFTA and the ASEAN Industrial Cooperation (AICO) scheme. At the Ninth ASEAN Summit in Indonesia in October 2003, ASEAN leaders signed the Declaration of ASEAN Concord II to establish an ASEAN Community by 2020. In order to eﬀect this, ASEAN countries agreed to accelerate their economic integration process in 11 priority sectors and to promote economic integration in the areas of trade in goods and services, investment, rules of origin, customs procedures, standards and conformance, intellectual property rights and the movement of naturalized persons. At the 12th ASEAN Summit in January 2007, ASEAN leaders agreed to accelerate the establishment of the ASEAN community by 2015 as envisioned in the ASEAN Vision 2020 and the ASEAN Concord II. For example, liberalization of investment and trade in services is expected to strengthen the competitiveness of service industries, including logistics, telecommunications and ﬁnance, and will also beneﬁt manufacturing industries which utilize these service industries. It is hoped that reform of customs procedures and promotion of logistics eﬃciency will decrease lead times and the costs of physical distribution for manufacturing industries. Mutual recognition agreements and standardization of products will reduce lead times for marketing products and the costs of inspecting them. 23. Given the existence of a home market eﬀect (HME), the integration of the Asian market improves the growth potential of the integrated ASEAN region against the rest of the world. At the same time, an intra-ASEAN income gap may increase according to the diﬀerence in the size of the domestic market (the home market) of each country. Some policy measures may be needed to balance these two home market eﬀects. 24. ‘Although the proportion of logistics costs to service, general and administration expenses in 2004 was approximately 10 percent in developed countries (11.6 percent in North America, 9.8 percent in EU and 8.3 percent (as of FY2002) in Japan), the proportion in Asian countries was 19.7 percent, about twice as much as that of developed countries. As examples of the problem of long logistics lead times, to complete exportrelated procedures, China and Malaysia take about three to four times as long to complete as developed countries do. Transit times in Asian countries are generally long partly because sea routes are usually used as major transportation routes due to the delay in development of an overland transportation infrastructure in the region. For instance, cargo transport between Bangkok and Hanoi takes about 10–15 days by sea, but with appropriate infrastructure improvements, there is a possibility that it may take only 3–4 days by land’ Ministry of Economy, Trade and Industry and Ministry of Land, Infrastructure and Transport (2006), pp. 10–11. 25. In December 2006, the Japanese Ministers of Economy, Trade and Industry and of Land, Infrastructure and Transport convened the Committee on Partnership for International Logistics Competitiveness and decided on the ‘Action Plan to Strengthen Japan’s Competitiveness in International Logistics’. Based on the action plan, Japan is planning to support the development of a transportation infrastructure including sea ports and airports, as well as road and railway-related improvements of a logisticsrelated legal infrastructure in order to facilitate transportation along the following six priority logistical routes in ASEAN: 1. 2. 3. 4. 5. 6. 26.

Thailand–Malaysia–Singapore Thailand–Indonesia Thailand–(Lao PDR/Cambodia)–Northern Vietnam Thailand–(Lao PDR/Cambodia)–Southern Vietnam Thailand–Philippines Thailand–Myanmar

‘At their meeting in December 2005, the ASEAN Economic Ministers signed a protocol to implement an ASEAN single window system. This system aims to enable shippers to submit uniﬁed electronic application forms and get approval of licenses from all relevant

Evolution of institutions and policies

27.

28.

29.

30.

31.

139

government agencies related to export- and import-related procedures. Japan will support transnational interconnection of ASEAN countries with advanced electronic management assistance, and will conduct surveys to identify issues and provide support for developing national single window systems for other ASEAN countries. Japan will also implement a demonstration pilot project to examine the possibility of rationalizing export- and import-related customs procedures using information and communication technology such as RFID (Radio Frequency Identiﬁcation)’ JETRO (2006a), p. 26 and Ministry of Economy, Trade and Industry and Ministry of Land, Infrastructure and Transport (2006), pp. 49–51. According to a report on a survey of the damage caused by counterfeited goods conducted by Japan Patent Oﬃce in FY2004, Asian countries constitute a major area for producing, transshipping, selling and consuming counterfeited goods. According to the survey, in 2004, some 93.1 percent of the world’s reported counterfeited goods were being produced in Asian countries. China accounted for 52.3 percent of the production sites of counterfeited goods in Asian countries. At a summit meeting of China, Japan and South Korea in January 2007, the three countries’ leaders agreed to begin negotiations on a trilateral investment agreement as soon as possible before the end of 2007. The agreement aims to protect investment operations and the assets of companies that have already invested in these countries, and to improve the transparency and predictability of laws and regulations which lead to new entries and growth in the number of potential investors in these countries. In addition, the three countries began trilateral consultations on the improvement of the business environment in May 2005. These consultations aim to encourage the three countries to implement feasible voluntary policy measures for improving the business environment based on requests from private sector interests, before agreeing on the ﬁnal trilateral investment agreement. ‘After the Thai government adopted Japan’s small and medium enterprise management consultant system, public and private oﬃcers in Thailand conducted 1,200 instances of management consultancy, and often use the word Shindan, which means ‘management consulting’ in Japanese. As for IT cooperation, ASEAN countries adopted Japan’s certiﬁcation systems model for information processing engineers, and promoted mutual recognition of certiﬁcates among the member states, aimed at achieving common skill standards for information processing engineers in East Asia. As for recycling methods, Japan disseminated its knowledge and know-how on a wide range of recycling systems including a mechanism to share the costs and set up a cooperative framework among governments, municipalities, producers and recyclers in East Asian countries’ Ministry of Economy, Trade and Industry (2006a), p. 56. Japan has implemented cooperation projects on small and medium-sized enterprises, environmental policy, ﬁnancial sector reforms and human resources development in the ﬁeld of information technology under the framework of ASEAN+3 and the ASEAN–Japan Ministerial Meetings. Japan has also put into eﬀect bilateral cooperation projects for capacity-building in the areas of tax systems, pension systems, improvements in bankruptcy and antimonopoly legislation, and educational reforms. The Global Economic Strategy (Ministry of Economy Trade and Industry, April 2006) proposed the following policy measures: (i)

(ii)

Globalization of universities / research institutes / personnel ● Formation of education and research bases that are open to the world ● The ‘Young Asian Leaders Development Fund’ ● Expansion of venues for foreign student activity ● Expanding the entry of talented technical and skilled labor Enhancement of the international competitiveness of tourism and visitor-attraction services ● Advertising the appeal of Japan ● Forming internationally competitive sites to attract tourists and visitors ● Preparation of foreigner-friendly tourism and visitor-attraction systems

140

General views on East Asian economic integration (iii) Conveying Japanese culture and Japanese branding to the world ● Further conveying the ‘Japan brand’ and fostering ‘soft’ power ● Synergistic expansion of soft power and hard power.

32.

33.

34.

The Ministry of Economy, Trade and Industry, Japan, introduced a New Economic Growth Strategy in June 2006. To achieve ‘new growth’ in the context of a declining population, the strategy will promote the enhancement of international competitiveness (global industrial strategy), and the stimulation of regional economies (regional industrial strategy) and will implement a cross-sectoral policy for innovation in the ﬁeld of human resources, capital, money, skills and knowledge. For further details, refer to http://www.meti.go.jp/english/report/downloadﬁles/New Economic GrowthStrategy(outline).pdf. In order to achieve sustainable economic growth in East Asia, it is important to address a wide range of common issues in the region as well as trade and investment liberalization through the establishment of FTAs and/or EPAs. The Economic Research Institute for ASEAN and East Asia (ERIA) will be established in 2008 as a ﬁrst step toward this goal. ERIA will provide intellectual assistance such as making policy recommendations for economic integration in East Asia centering on ASEAN. In the future, ERIA is expected to evolve into an ‘East Asian OECD’ by way of collaboration among East Asian countries. ERIA will deal not only with trade and investment but also with human resource development, small and medium enterprises and supporting industries, infrastructure development in transportation and IT, the narrowing of the economic development gap, energy and environmental issues, intellectual property rights, standards and conformance, and capacity-building. ASEAN leaders signed the Cebu Declaration on East Asian Energy Security at the Second East Asian Summit in January 2007. This agreement has the following aims: (i) (ii)

to improve the eﬃciency and environmental performance of fossil fuel use; to reduce dependence on conventional fuels through intensiﬁed energy eﬃciency and conservation programs, hydropower, expansion of renewable energy systems and biofuel production/utilization, and for interested parties, civilian nuclear power; (iii) to encourage the development of open and competitive regional and international markets geared towards providing aﬀordable energy at all economic levels; (iv) to mitigate greenhouse gas emissions through eﬀective policies and measures, thus contributing to global climate change abatement; and (v) to pursue and encourage investment on energy resource and infrastructure development through greater private sector involvement. 35.

36.

Under the terms of the ASEAN–China Free Trade Area (ACFTA) China included in the highly sensitive list color televisions, passenger cars, and subsets of trucks as items for tariﬀ reduction to below 50 percent in 2015. Some ASEAN countries included air conditioners, refrigerators, washing machines and color televisions in the sensitive list as items for tariﬀ reduction to 0–5 percent in 2018, and also included passenger cars and motorbikes in the sensitive list or highly sensitive list. As for the liberalization of trade in services and investment, China and ASEAN agreed on the liberalization of trade in the subset of services such as transportation and tourism at the ASEAN–China Summit of January 2007. Article XXIV of GATT allows FTAs and customs unions if: (1) (2) (3)

trade barriers after integration do not rise on average (Article XXIV:5); all tariﬀs and other regulations of commerce are removed on substantially all intraregional exchanges of goods within a reasonable length of time (Article XXIV:8); and they are notiﬁed to the WTO council.

Developing countries may, if they wish, invoke the provisions of the GATT that allow them to establish agreements that do not meet the conditions of Article XXIV. The 1979

Evolution of institutions and policies

37.

38.

39.

40.

141

Decision on Diﬀerential and More Favorable Treatment of Developing Countries (the so-called Enabling Clause) essentially removes the ‘substantially all’ test and allows for preferences between developing countries. (Hoekman and Kostecki, 2001.) While the number of RTAs/FTAs has increased in the Asian Paciﬁc region, it is important to make them as comprehensive and high-quality as possible in terms of liberalization and rule-making and to contribute to the Bogor goals, which aim to realize free and open trade and investment in the region. In 2004, the APEC economies agreed on the FTA Best Practice guidelines, which laid down the principles of consistency with WTO, as well as comprehensiveness and transparency as a reference for RTA/FTA negotiations by individual countries. In 2005, the APEC economies started to develop FTA model measures which are composed of important elements for inclusion in the appropriate chapters of RTAs/FTAs. In 2005, the APEC member countries agreed on FTA model measures on trade facilitation and in 2006 they negotiated another six chapters including trade in goods, technical barriers to trade (TBT), transparency, government procurement, dispute settlement and cooperation. In January 2007, the Chairman’s Statement of the Second East Asia Summit stated ‘To deepen integration, we agreed to launch a Track Two study on a Comprehensive Economic Partnership in East Asia (CEPEA) among EAS participants. We tasked the ASEAN Secretariat to prepare a time frame for the study and to invite all our countries to nominate their respective participants in it. We welcomed Japan’s proposal for an Economic Research Institute for ASEAN and East Asia (ERIA). In November 2006, the Ha Noi Declaration of the 14th APEC Economic Leaders’ Meeting stated ‘We instructed Oﬃcials to undertake further studies on ways and means to promote regional economic integration, including a Free Trade Area of the AsiaPaciﬁc as a long-term prospect, and report to the 2007 APEC Economic Leaders’ Meeting in Australia.’ In December 2005, the ASEAN+3 Summit Declaration stated that the ASEAN+3 process would continue to be the main vehicle for achieving the goal of realizing an East Asia community, while the East Asia Summit Declaration stated that the East Asia Summit could play a signiﬁcant role in community-building in the region.

BIBLIOGRAPHY Hoekman, Bernard M. and Michel M. Kostecki (2001), The Political Economy of the World Trading System, New York: Oxford University Press. Japan External Trade Organization (JETRO) (2003a); Inquiry Survey of Japanese Companies in China, February (in Japanese). Japan External Trade Organization (JETRO) (2003b), White Paper on International Trade and Foreign Direct Investment, August (in Japanese). Japan External Trade Organization (JETRO) (2006a), The Current Situation of ASEAN Economic Community (AEC) and Changes in the Business Environment, February. Japan External Trade Organization (JETRO) (2006b), Fy 2005 Survey of Japanese Firms’ International Operations, March (in Japanese). Ministry of Economy, Trade and Industry (2004), White Paper on International Economy and Trade, June (in English). Ministry of Economy, Trade and Industry (2005), ‘Indications of self-supporting, sustainable growth in East Asia–business opportunities and risks, in East Asia’, White Paper on International Trade, July, Chapter 2. Ministry of Economy, Trade and Industry (2006a), Global Economic Strategy, April (in Japanese).

142

General views on East Asian economic integration

Ministry of Economy, Trade and Industry (2006b), New Economic Growth Strategy, June. Ministry of Economy, Trade and Industry and Ministry of Land, Infrastructure and Transport (2006), Action Plan to strengthen Japan’s Competitiveness to International Logistics, December (in Japanese). Mori, Kazuko (2005), ‘Higashi Asia Community ni Do Approach suru ka’ (‘How should we approach East Asian Community?’), Asia Kenkyu, 51 (2), 1–5. Otsuji, Yoshihiro and Takashi Shiraishi (2002), ‘Nihon-ASEAN no Kakudai FTA o Teisho suru’ (‘Building closer ties with ASEAN’), Chuo Koron, 117 (2), 68–76.

6.

Economic integration in Asia: European perspectives Anthony J. Venables, L. Alan Winters and Linda Yueh

6.1

INTRODUCTION

Europe’s integration project has now been running for half a century, a period spanning the postwar birth of economic cooperation and the more recent enlargements and deepening of the union. The project has been enormously successful in both political and economic terms, although there have been frequent tensions and undoubted failures. This chapter draws out some of the main messages from the European experience of integration. We look at both the political and institutional development of the European Union, and at its economic development. What have been the driving forces behind the integration process? What institutions have developed to manage integration? What has been the impact of integration on trade ﬂows and income levels across European countries? We then endeavor to draw out some of the lessons that the European experience may have for integration in Asia. Evidently, the two continents are very diﬀerent in both political and economic terms. On the economic side, integration has had a large impact on European trade and incomes, both through trade creation and through intensifying competition. One might argue that the heterogeneity of countries in Asia oﬀers an even greater potential for trade creation, and also for using integration to facilitate the development of production networks. On the political level, European experience suggests that achieving the economic gains has required continuing and far-reaching policy measures. These, in turn, require a deep political commitment to integration and the existence of institutions to promote integration and protect it from the inevitable inter-member frictions and preoccupations with national goals. In Europe progress has been driven largely by the Franco-German partnership and by the Brussels institutions. It is hard to see what their equivalents in Asia might be. For Asia, therefore, while the economic arguments 143

144

General views on East Asian economic integration

are compelling, the lack of political commitment suggests that trade integration will not necessarily be followed by deeper economic integration. The remainder of the chapter comprises four sections. Section 6.2 explores the political economy of European integration, considering the history of, and commitment to, integration among its members, the roles of the institutions that it has created and the particular nature of its intermember relations. We suggest that the key driving forces in Europe do not have any close parallels in Asia. Section 6.3 deals with economics, arguing that Europe has seen both trade creation and trade diversion, and that integration has generally been a force both for promoting eﬃciency through specialization, and for increasing competition and industrial eﬃciency. With its greater diversity between members, Asian integration may generate greater trade creation, investment ﬂows and competitive pressures than did European integration, but possibly at the expense of greater divergence between members. Reaping the economic beneﬁts, however, will require ongoing integration, gradually rolling back the various barriers and frictions on intra-regional trade. Section 6.4 considers the dynamics of the integration. Regional integration creates its own dynamic as ‘domino eﬀects’ come into play. In Europe this took the form of continuing enlargement of the EU, but in Asia the dynamic seems to be leading to countries competing to gain hub status and to a proliferation of agreements. Section 6.5 concludes.

6.2

POLITICAL ECONOMY

This section reviews the history of European integration and institutions to see what lessons it contains for Asia. It will become plain that the two cases are fundamentally diﬀerent and that the casual drawing of parallels could be very misleading. 6.2.1

Europe

A grand vision European integration is an ancient aspiration, although its current manifestation arises from the geo-politics of the mid-twentieth century: the desperate need, following World War II, to ﬁnd a way of preventing future Franco-German conﬂict, coupled with a strong sense of internationalism that saw the future in terms of institutionalized cooperation between countries.1 Perhaps the most important factor in understanding the history of postwar European integration is to see that it was essentially a political-ideological phenomenon. It was not driven by the careful

European perspectives

145

calculation of economic costs and beneﬁts, still less by trade negotiators, but by a grand vision which had fortunate economic side eﬀects. This fact has had fundamental eﬀects on Europe’s evolution, for the grand vision helps to move internal debates beyond mercantilism and the calculation of beneﬁts issue by issue. It induces a generalized reciprocity, whereby every party gains in the end, but where everyone recognizes the value of the system as a whole and is prepared to accept losses on some deals. The dayto-day compromises necessary to achieve cooperative outcomes become easier to make, or, which is basically the same thing, easier to sell at home. Political institutions The ﬁrst major step in modern European integration was the European Coal and Steel Community (ECSC), founded in 1951, whose origins illustrate the political motivation for integration. Its purpose was to stimulate the recovery of heavy industries in (West) Germany while making it impossible for their output ever to be used to wage war again. The proposal – due to Jean Monnet and Robert Schuman – was that, by establishing a truly common European market in coal, iron and steel, countries would become so interdependent that war would be not only ‘unthinkable, but materially impossible’. The customs union was supplemented by a ‘High Authority’, which had the power to dictate national output quotas, establish maximum and minimum prices, and enforce competition. The High Authority was an administrative body, controlled in policy but not day-to-day matters by a Council of the Community on which the separate governments were represented, and also by a European Parliament. A Court of Justice was established to oversee the legal aspects of the Community. Following the ECSC, attempts were made to establish both a defense community (the EDC) and a political community (the EPC). Both failed, so the ‘integrationists’ were thrown back on to economic integration in the form of the European Economic Community (EEC), and the Atomic Energy Community (Euratom), which were created in the Treaties of Rome in 1957. At ﬁrst, the EEC and Euratom existed separately but parallel to the ECSC, but in 1967 the three bodies were merged, to form the European Communities (EC) with one Commission (successor to the High Authority), one Council, one Parliament and one Court. The Maastricht Treaty, in 1992, turned the EC into the European Union (EU), creating European citizenship, some cooperation in foreign and security aﬀairs, and paving the way for monetary union. These institutions of integration have evolved and expanded, but the basic structures remain as they always were. Thus although the EU now has a common currency (introduced for the ‘Eurozone’ countries in 2002) and (limited) powers to make common political and foreign policies, it is in

146

General views on East Asian economic integration

essence just a continuation of the old EEC, with institutions designed primarily for deep microeconomic integration. The Commission comprises commissioners appointed by member states for four-year terms, one for each member. It initiates Union policy and executes it, but it cannot actually make policy – that falls to the Council. The Commission is explicitly supranational, and is charged with preserving and promoting the European ideal. The Council formally comprises the foreign ministers of all member states, although much business is conducted by ministers concerned with speciﬁc issues, e.g. agriculture ministers discuss the Common Agricultural Policy (CAP).2 The Council shares executive power with the Commission. It may adopt the latter’s policy proposals, in which case they become law, but it may not generally amend them. Decisions are theoretically taken by qualiﬁed majority vote, where votes are allocated to member states according to size. Until the 1990s, however, all countries informally had a right of veto on issues of declared fundamental national interest (under the ‘Luxembourg Compromise’). As a result, decisions had to be reached by trading compromises (often on unrelated issues) to obtain a unanimously acceptable package. Recently strong eﬀorts have been made to re-establish majority voting in most spheres (but not, for example, ﬁscal policy and various ‘pet areas’ such as audio-visual policy) and there is hope that this will reduce the horse-trading. Nonetheless, the tradition of consensus remains strong within the EU. The Court of Justice interprets Union law. Its ﬁndings are binding even on member governments. The judges are appointed by member states, but they are required to be wholly independent of national interests and cannot be removed by member governments. The European Parliament has a small but growing role in the Union. It must be consulted by the Commission and the Council before they decide many issues, and it has some power over the Union budget. Its greatest power is to dismiss the Commission en masse, although this is such an unwieldy weapon that it is of little practical use.3 These institutions form a constitutional structure just as complex and delicately balanced as the US Constitution, but without, of course, its democratic legitimacy. Like the latter, they have to balance ‘states’ rights’ against the center and rely on powerful legal bodies for enforcement. Arguably such balances are necessary to create the conﬁdence that allows member governments to proceed with deeper aspects of integration that impinge directly on issues of sovereignty and internal distributions of income. Although it is fashionable, and to some measure warranted, to decry Brussels’ bureaucracy and interventionism, one should not lose sight either of its origins or of its role in the integration process. The institutions stem

European perspectives

147

from a period when there was much greater faith in governments than now, when governments were much more heavily involved in economic management than now, and when the essential task was political. Arguably, the subsequent diﬃculties were not due to the original structures per se, but their inability to evolve as circumstances changed. Such ﬂexibility is an important lesson for today’s would-be integrators. Inﬂexibility is similarly the problem with agriculture. The CAP stems from a period when agriculture provided a substantial part of employment in all six original members, and was strongly protected. The error should be seen less in the original policies, which were thought to make sense at the time, but in the danger of giving particular sectors special constitutional standing (agriculture is singled out in the Treaty of Rome) and/or their own bureaucracies. Each makes reform very diﬃcult when circumstances change. As integration occurs, it is important to avoid institutionalizing the special cases that are bound to arise. Recognize them as explicit failures and exceptions so that they can be addressed later. Guardians of the vision European integration has always been a rather ‘on and oﬀ’ aﬀair with periods of enthusiasm and rapid advance followed by periods of doubt and retrenchment. The former are, understandably, associated with economic booms and the latter with recessions. Thus the early 1980s found the EU very much down in the dumps. After the severe anti-inﬂationary policies at the beginning of the decade, the US and Japanese economies began to recover, but those of the EU seemed ﬁrmly stuck in the mire. Moreover, the rapid increase in intra-EU trade that had characterized the early stages of integration seemed to have halted or even gone into reverse. The cry was frequently heard that ‘the steam had gone out of integration’ and doubts were expressed about the viability of the EU as an institution, let alone any further progress. During such ‘depressions’, the Commission’s role as the guardian and champion of the European ideal has been vital to the goal of integration. While member governments, and thus the Council of Ministers, are focusing on their local problems, the Commission is constitutionally required to take a broader, longer, and more European view. In the mid-1980s its response to the lethargy of the European economy was dramatic and imaginative. It had long been recognized that the actual integration of the EU economies fell short of the aspirations of the Treaty of Rome. Recalling the stimulus that the initial creation of the EEC had induced, and following the prevailing intellectual trend towards economic liberalism, the Commission proposed a bold step towards complete economic integration with the launch in 1986 of the Single Market Initiative.

148

Table 6.1

General views on East Asian economic integration

GDP shares, EU and Asia Europe (EU-15)

Germany France UK Italy Other EU Other non-EU (future members of EU-15)

Asia (ASEAN+3)

1958

1973

1998

20.1 21.2 23.2 11 7.3 17.2

26.6 19.7 14.1 12.9 11.7 15

25.1 17 16.7 14 27.2 0

2003 Japan China South Korea ASEAN

62.8 19.9 7.5 9.8

Note: ASEAN: Brunei, Cambodia, Indonesia, Lao PDR, Malaysia, Myanmar, Philippines, Singapore, Thailand, Vietnam.

Similarly, the Commission was the driving force behind the Maastricht Treaty of 1992, which formally created the EU and extended the competences of the Union to foreign aﬀairs and justice. This was far from popular, being rejected by a referendum in Denmark and very nearly so in France. It illustrates a further cycle in the dynamics between the Commission and the states: ﬂushed with one success (in this case the Single Market), the Commission attempts to follow it by further deep integration and centralization, only to ﬁnd it rejected by governments and electorates. A further cycle followed, with the adoption of monetary union (2002), and then the rejection of the proposed EU constitution by the voters of France and the Netherlands (2005). These rejections, however, do not threaten the basic fabric of the common market: a tribute to its deep foundation in European perceptions, and to its pragmatic and non-confrontational mode of progress. The balance of power The preceding section showed the importance to EU development of political will, and following from that, institutional depth. A key driver of both has been, and continues to be, the Franco-German relationship. It is at this fundamental level that Asia looks most diﬀerent from the EU. Not only is the postwar political imperative absent, but so too is the balance of power that obtains within the EU. Table 6.1 gives GDP shares of countries in Asia and in the EU. The motivating force in Europe has been the need for two roughly equal-sized powers to cooperate and create something new that is diﬀerent from both of them. The third power, the UK, is large enough to be taken seriously but not to derail the whole enterprise. As Table 6.1 shows, France, Germany and the UK accounted for 19.7 percent, 26.6

European perspectives

149

percent and 14.1 percent respectively of EU output in 1973. The Asian predicament, on the other hand, is that China and Japan together would account for 82.7 percent of output in an ASEAN+3 bloc, dominating its output. And of course, these two countries are at diﬀerent stages of development. The rise of China is viewed with suspicion in Japan and there is not yet a recognition that the two economies may one day become roughly equivalent in size and should cooperate. The third power, Korea, is not nearly the size of Japan and China, and would not necessarily have the inﬂuence of the UK. As China’s economy expands at nearly 10 percent per annum, it and Japan will be the key to the success of an ASEAN+3 FTA. The same is true when population is taken into account, as China alone is more than double the population size of the ten countries of ASEAN. Overall, the diﬀerent size distributions of members in Asia and the EU make it diﬃcult to perceive strong parallels in the two groups’ political dynamics. Decision-taking in the EU We have seen that institutional reform in the EU has proved diﬃcult, but day-to-day decision-taking has been broadly successful. The key features have been the initial and continuing European vision, the balance of power, and the role of the Commission. To this must be added the fact that it has generally proved possible to buy oﬀ dissent. This is partly because the broad agenda of EU competencies creates scope for deal-making. There is a ‘generalized reciprocity’ – losses on some issues can be accepted in expectation of gains on other issues. It is also because direct, if limited, ﬁscal transfers are available. The role of ﬁscal transfers increased greatly from the 1980s onwards. The original ‘Six’ EEC members were fairly homogeneous in terms of income levels, but later enlargements began to introduce a wider spread, especially the ‘southern enlargement’ to Greece (1981), Spain and Portugal (1986) and more recent enlargements to Eastern Europe (2004, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia, as well as Cyprus and Malta). This widening membership raised serious issues of intra-EU distribution, not only in helping the new poorer members to catch up, but also within existing members. Regional policy of one form or another – the ‘structural funds’ and ‘cohesion funds’ – now account for more than one-third of the budget of the EU and 0.37 percent of total EU income. In fact, distribution is a major factor in much EU decision-making, and the existence of institutions to address it helps to prevent it from becoming a barrier to progress and an impediment to eﬃciency-enhancing decisions. The transfer mechanism – small as it is compared with those in federal and unitary states – has been essential to the running of the EU since the southern accession.

150

General views on East Asian economic integration

As a direct consequence of its relatively consensual nature, policymaking in the EU is patchy, inconsistent and ragged. Compromise and pragmatism have been the watch-words, rather than eﬃciency and elegance, with particular members being granted derogations from some measures and the enforcement of others relying heavily on turning a blind eye. One might regret this, in principle, but it is notable that despite a number of shocks, European integration has avoided serious set-backs for half a century. 6.2.2

Asia

We have argued that key features underlying the success of the EU are a grand vision; the development of central institutions; a balance of power; and, arising from these, a general ability to reach compromise agreements. How does Asia measure up against these features? It is interesting that statements of the motivation driving Asian integration are almost entirely defensive. For example, an overview of cooperation in Asia (Lamberte 2005) gives ﬁve driving forces: a defensive response to the rise of regionalism elsewhere; the slow progress in multilateral trade liberalization; competition with other regions of the world for FDI; concern to tidy up bilateral agreements; and institutionalizing the de facto increase in economic interactions. The leading countries in the region do not present integration as a means to heal historical conﬂict – on the contrary, discussions of integration tend to inﬂame old wounds and highlight political tensions. As for institutional development, there is no impetus for a pan-Asian framework, along the lines of Europe in the postwar period. The largest economy in the region, Japan, was closely tied to the USA after World War II and its development was accordingly Western-oriented. China in the postwar period turned inward, and political turmoil and communism kept its doors closed until the 1980s. The ASEAN countries ﬁrst established their regional alliance in 1967. The Association of South East Asian Nations was started by Indonesia, Malaysia, the Philippines, Singapore and Thailand. Later on, Brunei joined in 1984, Vietnam in 1995, Laos and Myanmar in 1997, and Cambodia in 1999. The original ﬁve members joined together to secure themselves against communist expansion in Vietnam. In 1976, it increased its scope to economic cooperation and in 1991 embarked on developing a free trade area known as the ASEAN FTA (AFTA). However, the extent of trade among ASEAN countries has always been limited despite the intent to create an FTA. Approximately 23 percent of exports from ASEAN countries are sold to other ASEAN countries. This is in contrast to intra-Europe trade, which accounts for three-quarters of all European exports. The institutional

European perspectives

151

framework of the ASEAN trade bloc is also limited. ASEAN has made some headway in developing a dispute settlement system, but the system is considered to lack transparency. In general, there is no supranational institutional structure to bring together the ASEAN+3 nations. The nations have vastly diﬀerent political and legal systems, and two of the countries (China and Vietnam) are transition economies which have mixed state/market economies. Fundamentally, the structures found in the EU – notably the Court and Parliament – are lacking in Asia. Moreover, the institutional and political diﬀerences among these nations make it unlikely that there is a premise for deep integration. The balance of power has already been discussed. The major economies of Japan and increasingly China make the region appear more similar to the dominance of the USA in discussions of the FTAA. The next-largest economy is South Korea, which is considerably smaller than the two major economies, while the total GDP of ASEAN only just exceeds that of Korea. The highly unbalanced size of the economies in Asia will make the role of China and Japan critical in any FTA, and by the same reasoning means that their agreement will be essential in any successful regional economic agreement. However, these two nations have considerable political diﬀerences, manifested in public rebukes of the Japanese prime minister’s visits to the controversial Yasukuni shrine, among others. What are the implications of this for the day-to-day decision-taking that has to take place in a regional integration agreement? The current ASEAN arrangements are rife with exemptions and exceptions. Without vision, or institutions, or breadth of agenda or transfer mechanisms, this is unsurprising. The challenge for ASEAN+3 will be to overcome political diﬀerences and adopt a proactive stance toward an FTA. At the moment, the ASEAN–China discussions are under way to establish an FTA by 2010. This is causing Japan to act to establish an ASEAN–Japan FTA, and similarly South Korea. This defensive stance does not create the potential of the EU to compromise and work toward an integrationist agenda. Instead, Asia looks more likely to be characterized by multiple bilateral agreements driven by an economic agenda; i.e. to realize the gains from trade from rising intra-regional trade in Asia, which already accounts for 50 percent of all trade and is increasing due to China’s fast growth. Therefore, in contrast to Europe, Asia lacks the features that have underpinned regional integration in the EU. There is no grand vision, no impetus to develop central institutions; there is a disparate set of economic powers and no evidence of ability to reach compromise agreements. Asian integration is characterized by a defensive motive in reaction to regionalism elsewhere and the slowness of multilateral trade agreements, as well as a

152

General views on East Asian economic integration

recognition of the rapid development of regional trade. The moves toward integration are thus motivated by economic gains without a corresponding set of political will and institutional vision. The balance of power further suggests that one or two countries will dominate any arrangement, moving Asia more toward a hub-and-spoke system than an integrated region.

6.3

TRADE AND PRODUCTION

Since Asian integration is focused on free trade, as opposed to an attempt at deeper political integration, we now turn to the economics of integration. The EU is far from being a perfect parallel, being more compact geographically, more homogeneous in income levels, and more intent on deeper integration, but it provides our best view of the long-run economic eﬀects of regional integration. 6.3.1

Europe

Enshrined in the Treaty of Rome are the four basic economic principles underlying European economic integration: freedom of trade in goods; freedom of trade in services; free mobility of capital; and free mobility of labor. We discuss progress in these areas (leaving labor mobility to one side), highlighting both the obstacles that have been encountered and the outcomes that have been attained. We look ﬁrst at trade, where several quite distinct mechanisms have been at work. The ﬁrst is that integration causes reorganization between sectors of the economy: sectors expand or contract in line with eﬃciency diﬀerences between countries, although there are also concerns about trade diversion. The second is that trade changes the nature of competition between ﬁrms and induces an industrial reorganization within sectors of the economy; this is often associated with foreign direct investment. Trade may also cause large changes within ﬁrms, as production networks develop; we postpone discussion of this aspect of trade until Section 6.3.2, on Asia. Following discussion of these issues we turn to look at outcomes, in particular the extent to which there has been convergence of incomes within the EU, and then to some of the policy questions that have arisen. The growth of trade European integration had a dramatic eﬀect on the geographical patterns of members’ trade. Every member has seen a strong reorientation of its trade towards other members following accession or the formation of the EEC. Moreover, this is as true of manufacturing (and probably services) as of the

153

European perspectives % 35 30 25 20 15 10 5

19 50 19 52 19 54 19 56 19 58 19 60 19 62 19 64 19 66 19 68 19 70 19 72 19 74 19 76 19 78 19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96

0

Spanish share

Figure 6.1

Time UK share

German share

Shares of members’ imports coming from the EC-6, 1950–96

grotesquely distorted agricultural trade. Figure 6.1 plots the shares of three EU members’ imports coming from the original EC-6. As an original member, Germany experienced increasing integration with the remaining ﬁve from 1957, with duty-free and quota-free access from 1968. The UK acceded on 1 January 1973 and Spain on 1 January 1986. The pattern is very clear: starting slightly before the formal date of the integration, the trade share starts to rise. It rises for 10–12 years and then stabilizes. For Spain the growth is still continuing at the end of the period. Freund and McLaren (1999) have explored the dynamics of regionalism more formally using both trade shares and trade intensity indices.4 For the latter – the more appropriate measure analytically – they ﬁnd some evidence of anticipation eﬀects, starting on average 2.5 years before formal integration, followed by 9.5 years of higher growth before achieving a new steady state.5 On average, EU countries increased their intra-bloc trade intensity by 53 percentage points over this process. Sectoral specialization The growth of trade can lead to economic gains through increased specialization. The specialization may arise in response to technological or factor endowment diﬀerences between countries, or may occur because of the beneﬁts of clustering related activities in one place. In either case, market integration should promote an increase in specialization, as the cost of trade is reduced. EU integration has been associated with a rather modest increase in manufacturing specialization. Measures of the diﬀerence between the industrial structures of EU countries have been computed at the level of 36 industrial sectors (Midelfart-Knarvik et al. 1999). All EU countries except the Netherlands have, since the late 1970s, seen their industrial structure becoming more dissimilar from that of other EU countries. For the

154

General views on East Asian economic integration

initial six members of the EU there is a more or less steady increase in specialization throughout the period. Later entrants experienced increasing specialization starting from around their date of entry. However, despite this, EU countries and regions remain very much less specialized than comparable-size geographical units in the USA. So far at least, integration has not generated specialization and clustering of activity of the order that US experience suggests would be expected in a single country. Econometric analysis of these changing patterns of specialization indicates that it is largely in line with intra-union comparative advantage. For example, skilled-labor-intensive activities have tended to relocate towards skilled-labor-abundant countries, and R&D-intensive activities have relocated towards scientist-abundant countries (Midelfart-Knarvik and Overman 2002). However, reallocations in line with intra-union comparative advantage are not necessarily welfare increasing as they could be at variance with countries’ comparative advantage with the rest of the world. This is the phenomenon of trade diversion. Trade creation and trade diversion Increased intra-bloc trade is frequently taken as indicating successful economic integration, especially in popular debate, but, of course, it shows no such thing. The traditional economic question hinges on whether the share increases as a result of trade creation or trade diversion. There is no doubt that European integration has been accompanied by a good deal of trade creation, both internal and ‘external’, in which imports from outside the bloc displace members’ domestic production and/or expand consumption. Thus Truman’s (1975) decomposition of the apparent consumption of manufactures into shares due to imports from partners, imports from nonpartners and domestic supplies, shows both sources of imports growing strongly at the expense of the domestic share. Truman ﬁnds that out of 53 country–sector combinations observed over 1960–68, 31 display such ‘double trade creation’ while a further 13 display internal creation and external diversion. Over 1975–82 Jacquemin and Sapir (1989) ﬁnd roughly similar proportions of ‘double trade creation’ and less evidence of trade diversion, while Sapir (1992) ﬁnds ‘double trade creation’ for aggregate EC9 trade over 1980–91. The predominant pattern of ‘double trade creation’ does not imply absence of trade diversion, as external trade should be compared with what it would have been in the absence of integration. There is an unavoidable need to specify the anti-monde when estimating integration eﬀects. Two approaches exist to modeling the anti-monde more explicitly. First, one can model trade ﬂows in terms of prices and incomes and explicitly allow for the diﬀerent tariﬀs faced by diﬀerent suppliers. This requires considerable information

European perspectives

155

and some eﬀort to model the determinants of trade ﬂows through time in a theoretically coherent fashion. Winters (1983) takes this approach to UK manufacturing trade following its accession to the EEC in 1973. He ﬁnds relatively little trade diversion, but certainly some evidence of it. The second approach is to use a gravity model, which essentially uses trade between other (unrelated) countries to identify the anti-monde for partners’ trade. The gravity model explains trade between two countries in terms of their incomes, populations, location and geographical characteristics, plus at least two sets of dummy variables to capture the eﬀects of each regional arrangement: one on intra-bloc trade and one on trade between partners and non-partners. The coeﬃcients on such dummy variables reﬂect a huge variety of eﬀects and can be highly signiﬁcantly diﬀerent from zero at any point in time. Hence to measure integration eﬀects one needs to observe not their levels but their changes over periods when regional integration has occurred. Within Europe, Bayoumi and Eichengreen (1997) ﬁnd strong signs of EEC–EFTA trade falling below expected values as the EEC was formed, and some evidence of the acceding countries’ trade with non-members similarly falling below par as they joined.6 Sapir (1998) similarly ﬁnds EU–EFTA trade penalized by EEC formation and enlargement. Soloaga and Winters (2001) use a much wider range of countries than just Europe to deﬁne their anti-monde, but at the expense of considering only the period 1980–96. They use three dummies to capture trade eﬀects, breaking the extra-bloc trade eﬀect into an export and an import eﬀect. In 1980 the EU shows unusually strong trade with non-partners and lower than expected trade within the bloc. (This is a common result in gravity models based on large samples of countries.) As integration deepens and Iberia enters the Union, however, these eﬀects decline absolutely – that is, intra-trade grows relative to expected and extra-trade falls. Moreover, Soloaga and Winters show that these changes are statistically signiﬁcant, suggesting the presence of trade diversion. The overall message from these studies is that there has been rapid trade growth and trade creation, but there is some evidence of trade diversion as well. Of course, the extent of trade diversion varies across sectors, and the grossly distorting Common Agricultural Policy has certainly been trade diverting for members who would otherwise have had a more liberal agricultural trade regime. Industrial reorganization: competition, scale and market integration A central feature of EU trade is that a large part of it has been intraindustry, rather than inter-industry. Such trade is not (necessarily) based on cost or comparative advantage diﬀerences between countries, but can

156

General views on East Asian economic integration

instead arise simply from competition between ﬁrms in imperfectly competitive markets. Small, national, segmented markets are liable to be dominated by a few national producers, possibly operating at sub-national scale and exploiting considerable monopoly power. Market integration should remove this segmentation, allowing ﬁrms to compete more eﬀectively in other national markets and permitting expansion of relatively eﬃcient ﬁrms. Market integration permits ﬁrms to be larger (and better exploit economies of scale) and competition to be more intense. EU experience has indicated that, in some sectors at least, achieving truly integrated markets can be quite diﬃcult. Even when tariﬀs have been eliminated, markets appear to remain segmented, with substantial price diﬀerentials between countries, and borders still having a strongly negative eﬀect on trade ﬂows. These observations were among the motivations for the Single Market Initiative. The Single Market Initiative (SMI) was launched in 1986 for completion in 1992, with the objective of eliminating market segmentation and ‘completing the internal market’. The economic policy measures introduced fall into four main categories: (1) the simpliﬁcation or removal of frontier formalities, facilitating and speeding the ﬂow of goods across borders; (2) the simpliﬁcation of product standards, in particular the adoption of the ‘mutual recognition principle’, whereby goods approved for sale in any member state are deemed acceptable in all; (3) the deregulation of transport sectors, allowing for improved eﬃciency in the internal distribution of goods; and (4) the opening up of public procurement to supply from all member states. Although individually small, these measures were estimated collectively to reduce the costs of trade across borders by an amount equal to several percent of the gross value of goods traded. More importantly, their indirect eﬀects were predicted to lead to gains equivalent to several percent of EU GDP, as markets became more competitive and ﬁrms reorganized, increasing their scale to that of the larger integrated market. Ex post evidence on actual gains is patchy, however. The SMI was accompanied by a burst of merger activity, and there is some evidence of further trade creation (Pelkmans 2001). Griﬃth (2001), in a study of UK manufacturing, ﬁnds a signiﬁcant increase in both labor productivity and total factor productivity in establishments in sectors that were particularly aﬀected by the SMI. Increased scales of operation have been attributed to the SMI, particularly in sectors where liberalization of public procurement was important, although the size of ﬁrms in the EU remains generally smaller than their US counterparts. The Single Market Initiative left countries with diﬀerent national currencies, until monetary union was introduced for 12 core (Eurozone) currencies

European perspectives

157

in 2002. Part of the motivation of this initiative was to promote price transparency and achieve further market integration, although most of the analysis has surrounded its macroeconomic impact. What messages come from this experience of market integration for Asia? The European experience points to the importance of ‘deep integration’. The pro-competitive and scale-economy gains of market integration can be impeded by frontier frictions that individually appear quite minor, but collectively allow ﬁrms to retain dominant positions in their home markets. The list of such frictions is long. A free trade area, as opposed to a customs union, is bound to retain border formalities as well as rules of origin. Contingent protection has been widely used both by the USA and within Asia, and its ‘trade chilling’ eﬀects are well known (ADB 2005). Meeting national product standards is costly, and harmonization of standards almost impossible. Europe took the mutual recognition route, but this involves a level of acceptance of foreign standards and a willingness to delegate product approval to foreign institutions that is inconceivable in Asia. Despite the success of the Single Market Initiative, it has not reached all – or even many of the largest – areas of economic activity. Opening up of service sectors to competition was one of the objectives of the Single Market Programme, imposing on member states the obligation to abolish restrictions on the free movement of services and extend mutual recognition to professional qualiﬁcations. However, progress remains slow, with diﬀering legal standards and regulatory regimes still impeding cross-border investments and competition. This remains an area where the European Commission is still performing its role of trying to secure further market integration, while encountering stiﬀ opposition from a number of member states. Foreign direct investment Accompanying the rapid growth of trade in the EU there has been expansion of foreign direct investment (FDI). World FDI stocks have grown faster than both income and trade in recent decades, and the EU-15 holds around one-third of the global stock of inwards FDI. This share surged to over 40 percent at the time of the Single Market Initiative, driven by a crossborder merger wave. The importance of FDI for EU economies is illustrated by the fact that 47 percent of Irish manufacturing employment is in foreign-owned ﬁrms, and this share is substantial even the larger EU countries (France 26 percent, UK 16 percent). Much of the growth of FDI within the EU has been intra-EU investments, which accounts for a majority of the total. Investments from outside the region have also been important as economic integration has allowed outside ﬁrms to supply the entire European market from a single plant. Indeed, for many suppliers FDI is a much more important means of

158

General views on East Asian economic integration

reaching the European market than is foreign trade. For manufacturing as a whole, sales of goods by US subsidiaries in the EU were, in 1998, 3.75 times larger than EU manufacturing imports from the USA. There is also considerable evidence that some of the inward Japanese investments of the 1980s were driven largely by EU tariﬀ and non-tariﬀ barriers. These investments are perceived to have important positive eﬀects. Productivity is generally higher in ﬁrms that are multinational than in ﬁrms that supply only the domestic market (Martin and Criscuolo 2001). Particular importance attaches to FDI in services, as this may be the only means through which foreign competition can enter the domestic market. Consequently both the entrenched interests of incumbent ﬁrms and the potential economic gains from liberalization are large. Opening up of service sectors to competition was one of the objectives of the Single Market Programme, imposing on member states the obligation to abolish restrictions on the free movement of services and extend mutual recognition to professional qualiﬁcations. However, progress remains slow, with diﬀering legal standards and regulatory regimes still impeding cross-border investments and competition. Income convergence Some of the most important economic questions concern the eﬀects of regional integration on growth. Is regional integration likely to be good for growth in the region as a whole, and for poorer countries in particular? Standard neoclassical trade theory presumes that integration should lead to convergence of factor prices and incomes – with the limit being factor price equalization. However, a number of qualiﬁcations need to be made to this benchmark case. Even if integration brings convergence of per capita income, it need not bring steady convergence of all factor prices. Feenstra and Hanson (1997) show how the relocation of production activities to low-wage countries can reduce the wages of unskilled labor in these countries. The argument is that the activity that relocates may be unskilled labor intensive relative to other activities in the high-wage country, but skilled labor intensive relative to activities in the low-wage host country. In a more general model in which comparative advantage stems both from endowments and from location (with remote regions having a comparative disadvantage in high-transport-costs goods), reducing trade barriers brings peripheral countries into the trading system and raises their real incomes. However, changes in the prices of individual factors can go either way, depending on both the location and the endowments of individual countries (Venables and Limao 2002). Some analyses of wage diﬀerences across Europe have focused less on factor endowment diﬀerences across countries, and more on the relationship

European perspectives

159

between locations with good market access (the center) relative to those with worse access (the periphery). Empirically, European cross-country wage diﬀerentials follow a strong ‘center–periphery’ wage gradient, and there has been concern about the possibility that integration might draw activity out of peripheral regions and into the center. Theory suggests that this gives rise to a U-shaped relationship between the ratio of wages in the periphery to those in the center as the degree of integration changes (Krugman and Venables 1990). When trade barriers are high, local manufacturing is protected, allowing higher wages to be maintained; at the other extreme, perfectly free trade brings factor price equalization. It is at intermediate levels of trade barriers that ﬁrms are drawn into ‘central’ regions which oﬀer large markets and from which they can supply the periphery. Peripheral regions are poor locations for manufacturing, and as a consequence have lower wages in equilibrium. In the European context it has generally been argued that barriers are low enough that countries are on the upward-sloping section of the U. Further reductions in trade barriers cause ﬁrms to relocate to lowerwage peripheral regions, thus ﬂattening wage gradients. Empirically, the evidence on convergence suggests that this has – to a limited extent – happened. Turning to the evidence, the EU has experienced signiﬁcant, although by no means steady, convergence of per capita income across member states. The outstanding features are the rapid catch-up of Ireland, Spain and Portugal, and at the same time the continuing poor performance of Greece. The overall experience of convergence has been analyzed by many authors and can be summarized in many diﬀerent ways. Summary measures of the cross-country dispersion of per capita income in the EU indicate signiﬁcant convergence through the 1960s and 1970s, but no further aggregate change during the 1980s. There was some resumption of convergence between countries from the late 1980s, although this was accompanied by divergence at the sub-national level (Puga 2002). The role of EU policy A continuing message is that the economic beneﬁts outlined above have not been achieved simply by removing tariﬀs, but have required continuing leadership from the EU institutions and from some of the member states. The best examples are the Single Market Initiative and monetary union, as discussed above. However, in addition to these positive interventions, there have also been repeated interventions by the EC to prevent the use of national policies to distort competition. While increased mobility of ﬁrms can reduce the incentives to use burdensome regulation, it can also increase incentives to use distortionary subsidy policies, and this has been an issue in the EU. National interventions can take many diﬀerent forms. At one extreme are direct state aids to

160

General views on East Asian economic integration

industry, which amounted to some 4 percent of EU manufacturing value added in 1986–88, a ﬁgure that had declined to below 3 percent by the late 1990s. The bulk of this goes to R&D support and to meet regional policy objectives. Of the part that goes to speciﬁc industries, aid is highly concentrated on a few sectors, particularly shipbuilding and steel. Other national policies include general infrastructure and training schemes, and use of corporate taxation; low corporate taxes in Ireland have been viewed as highly eﬀective in attracting mobile FDI projects to Ireland from other potential locations in the EU. Aware of the possible distortions to competition that would arise if countries were free to subsidize industry, Articles 92 and 93 of the Treaty of Rome explicitly prohibit such subsidies. These articles are policed by active monitoring and intervention. For example, between 1998 and 2000 more than 1500 cases were reviewed by the Commission and in 7 percent of these cases negative decisions were reached requiring recovery of aid (European Commission 2001). As for corporate taxation, Ireland has had several instances of conﬂict with the EU. Negotiations with the Commission led to termination of a complete corporate tax holiday on proﬁts related to export sales and to an increase of the basic rate of corporate income tax from 10 percent to 12.5 percent. The weakness of these policies lies in the number of loopholes. For example, state aids are allowed in order to reduce regional disparities, and can take the form of regional incentives to enterprises in selected (but large) regions. Total expenditures to an enterprise are capped, and aids to new investments are preferred to ongoing subsidies. While policy in these areas is still developing, the broad conclusion of the research literature is that these policies have done little to distort the location of industry. Midelfart-Knarvik and Overman (2002) show that specialization is taking place according to comparative advantage, despite the use of state aids. Braunerhjelm et al. (2000) conclude that competition for activity generally takes the form of measures that count as good economic management, rather than wasteful tax or subsidy competition. 6.3.2

Asia

We have argued that the EU has received gains from trade creation and specialization, and perhaps more important gains from market integration and the associated pro-competitive eﬀects and industrial reorganization. However, achieving market integration has required continuing policy eﬀort from the EC to secure liberalization. What might the experience of Asia look like compared to this? The share of emerging Asia in world trade has doubled, to nearly 20 percent, over the last 20 years, and trends in Asian trade and investment are

161

European perspectives

Table 6.2 Intra-industry trade as a percentage of total trade growth (averaged over five years)

East Asia South Asia

1986–90

1991–95

1996–2000

42.5 31.2

46.9 21.8

75.0 34.5

Source: IMF (2002).

widely documented elsewhere (IMF 2002; Ng and Yeats 2003; Zebregs 2004). Two points are particularly noteworthy. First, there has been particularly rapid growth of intra-regional trade, with the share of emerging Asia as a destination for exports from other countries in emerging Asia doubling to 40 percent over a 20-year period. And second, much of the growth is in intra-industry trade. The rise in intra-industry trade in Asia can be seen in Table 6.2, which gives the estimates since the 1980s (IMF 2002). It shows that for East Asia, three-quarters of total trade growth can be accounted for a rise in intra-industry trade. Superﬁcially, this looks similar to experience in the EU, but there are important points of diﬀerence. The ﬁrst is that the greater heterogeneity of Asian economies creates more scope for gains from comparative advantage than was the case in EU, particularly with its initial membership. Second, in Asia much of the growth in trade is through vertical specialization, so intra-industry and intra-regional trade both reﬂect a rise in cross-border production chains and supply chains. Whereas in Europe high levels of intra-industry trade reﬂected ‘horizontal’ trade in similar products, in Asia they reﬂect high levels of ‘vertical’ trade in products at diﬀerent stages of the production process in a particular sector. Similar comments apply to FDI. In the EU most FDI is ‘horizontal’, while much Asian FDI is ‘vertical’. The former occurs mainly in order to serve the local market, and involves making investments that duplicate investments in the home country, as when an assembly plant is built in each market. The latter are made to minimize production costs, and involve moving stages of the production process to lowest-cost locations, such as the relocation of unskilled-labor-intensive stages of production to low-wage economies. A measure of the extent of vertical specialization is the ratio of merchandise trade to merchandise value-added. Increasing numbers of parts and components that travel across borders for further processing would result in a higher trade-to-value-added ratio. In spite of the diﬃculty of distinguishing ﬁnal from intermediate goods, the rising ratios of total merchandise to value-added indicate the growing presence of

162

Table 6.3

General views on East Asian economic integration

Ratio of merchandise trade to merchandise value-added (%)

Asian region China Taiwan, Singapore, Hong Kong, South Korea Pakistan, Bangladesh, Malaysia, the Philippines, Thailand, Indonesia

1980

1990

2000

93.8 12.1 216.5

115.6 23.7 259.3

168.5 32.9 365.5

39.4

52.4

84.3

Source: IMF (2002).

cross-border production chains in Asia, as seen in Table 6.3. This evidence is also consistent with the export-oriented growth strategy undertaken by Asian nations. Local capacity-building would be much slower than plugging into an existing global production chain of a multinational ﬁrm. If much of the growth in Asian trade has been of a diﬀerent sort of trade from that in the EU – ‘vertical’ rather than ‘horizontal’ intra-industry trade – what implications does this have for policy-making? The ﬁrst point is that trade frictions and trade costs are particularly important, as vertical specialization means that products cross borders multiple times during the various stages of the production process. The second point is that reducing these costs may require much less institutional eﬀort in Asia than in the EU. Horizontal intra-industry trade involves existing ﬁrms in each country becoming more exposed to import competition (as well as having more export opportunities). There is therefore always a constituency of incumbent ﬁrms that will seek to retain import barriers, and it is this that has made market integration in the EU so diﬃcult, and required repeated policy interventions, such as the SMI. By contrast, vertical intra-industry trade involves much of a plant’s output being exported. It therefore poses no threat to existing ﬁrms in the same line of business – indeed, there may not be any such ﬁrms. Thus, while the institutional mechanisms that Europe has needed to secure ‘deep integration’ are absent in Asia, their absence may be relatively unimportant, at least for the development of production networks. The historical record already indicates that the lack of deeper integration has seemingly not hampered the growth of intra-industry trade in Asia. It would however be incorrect to see trade and FDI in Asia as entirely ‘vertical’ in nature. For example, the opening of the Chinese economy is attracting ‘horizontal’ investment in China, given the current higher costs of accessing the Chinese market through trade. As China itself is a large market potentially two to three times larger than the USA and EU, the economies of ASEAN would beneﬁt from integration with such a market

163

European perspectives

Table 6.4

China ASEAN Sources: years.

Foreign direct investment inflow (in billions), 1996–2004 1996

1997

1998

1999

2000

2001

2002

2003

2004

41.7 29.9

45.3 33.9

45.5 22.2

40.3 27.3

40.7 23.4

46.9 19.4

52.7 13.5

53.6 19.4

60.6 25.6

China Statistical Yearbook, various years; ASEAN Statistical Yearbook, various

given their small domestic economies. In other words, the opening of China’s market has attracted FDI for production reasons on account of its low-cost labor but also increasingly due to the high trade barriers in China’s domestic markets. For ASEAN in particular, the trend of inward FDI moving to China raises questions and concerns about the location of industry. Economic integration, rather than just reduction in tariﬀs, would allow Asia to beneﬁt not only from trade creation and specialization, but also the pro-competitive eﬀects of market integration and industrial reorganization. However, the beneﬁts from integration are likely to be uneven, given the diﬀerence in size and income of the ASEAN+3 countries. With the EU institutional framework, there is scope for redistribution and transfers, particularly seen with the accession of ten new countries mainly from Central and Eastern Europe in 2005. In Asia, there would be no similar framework and less room for bargaining and trades. Therefore, although the heterogeneity in income levels and endowments of Asia suggests that traditional comparative advantages can be exploited with the lowering of tariﬀs, the lack of deeper integration could limit the gains from trade. The shift of FDI to China and the development of a potentially large domestic market which is beginning to transform the nature of investment suggest that an integrated market would be potentially beneﬁcial. However, the lack of a vision or a supranational body to implement economic policies as well as the historical rifts among these nations make such deep integration unlikely. The lessons from Europe should be heeded, though, including the evidence on trade diversion and the entrenchment of special cases such as agriculture, which is a concern for Japan and South Korea.

6.4 6.4.1

INTEGRATION DYNAMICS Europe

The refrain throughout this chapter is that European integration has been successful because it has been a continuing process of steps to achieve

164

General views on East Asian economic integration

deeper integration, going far beyond the removal of tariﬀs. In addition to this internal dynamic, the development of a large regional trading arrangement also creates powerful incentives for non-members. The point is essentially that in addition to beneﬁting insiders, regional integration may well harm countries left outside. Trade diversion is one mechanism. Countries come to source their imports from other countries in the bloc rather than from outsiders, reducing the demand for outsiders’ exports and depressing their terms of trade. Another mechanism arises from ﬁrms’ location decisions. It will be proﬁtable to relocate plants inside the bloc, in order to get the beneﬁts of duty-free access to a large market. These forces create ‘domino eﬀects’ (Baldwin 1993), whereby a free trade area creates incentives for outsiders either to join the free trade area, or to create an alternative competing area. In Europe these forces played out in two ways. First, the European Free Trade Area (EFTA) was established as an alternative to the EC in 1960, with a membership including the UK and Scandinavian countries. But second, the attraction forces created by the larger EC meant that a stream of countries left EFTA to join the EC, with the residual EFTA countries (Norway, Switzerland, Iceland and Liechtenstein) forming the European Economic Area with the EU in 1994. In addition, countries that were in neither EFTA nor the EU sought to join the EU, taking it to its current membership of 27 countries, with further applications for membership being negotiated. 6.4.2

Asia

There has not been a political will behind a free trade area in East Asia. The major economies in the region also have a history of avoiding preferential trade agreements, namely, Japan and China. Even bilateral trade agreements are rare. South Korea signed its ﬁrst ever FTA with Chile in 2003. The driving force behind integration in Asia is therefore considerably diﬀerent from that in Europe. Economic motives are the predominant ones. The dynamics would thus predict that countries will enter into FTAs to secure trade relationships, which could generate a ‘domino eﬀect’. This can result in a proliferation of bilateral trade agreements or a regional FTA (characterized by a hub-and-spoke system) that capture these bilaterals, but not necessarily an FTA that is characterized by deeper economic integration. Countries would aim to join an FTA if others did, but there is no mechanism (political or institutional) that would push the FTA to become more integrated economically. The rise of China has the potential to transform the dynamics toward more economic integration in a way that is not dissimilar to the USA in the proposed FTAA, but the same concerns about

European perspectives

165

the dominance of the USA would apply to China/Japan. Moreover, having two dominant and historically diﬃdent countries would add a further dimension of tension. Thus, unlike the European experience, the recent momentum behind the development of regional trade in Asia is based on defensive motivations rather than positive aims and objectives. The development of regional agreements elsewhere in the world, such as EU and NAFTA, alongside the slow progress of the development of the multilateral WTO framework, provided some of the impetus behind the drive to form ASEAN+3. The fragility felt in the region after the Asian ﬁnancial crisis of 1997–98 also contributed to a desire to increase the gains from trade. However, the remarkable growth of China in the past two decades since it implemented its ‘open door’ policy has undoubtedly contributed to the momentum behind ASEAN+3. Since 1992 China has increased its market share of global manufactured exports to an impressive 6 percent from just over 1 percent and joined the WTO in 2001, becoming a part of the global trading system. It has also become the world’s third-largest exporter and importer, as well as its fourth-largest economy. The initial reaction in the Asian region was dominated by the perception that China would adversely aﬀect the trade position of its neighbors. However, the evidence suggests that China is a magnet for foreign direct investment and has integrated into Asian production chains. Intra-industry trade during the 1990s increased to account for 75 percent of intra-regional trade growth in Asia (IMF 2002). China’s abundant, low-cost labor and considerable domestic market are behind the increase. In 2002, China accounted for 14 percent of all exports from developing countries in Asia, of which approximately 80 percent was intermediate and capital goods for production and the remainder was goods for consumption (Zebregs 2004). This ﬁgure is set to grow as trade barriers such as the Multi-Fiber Agreement are removed, which would allow China to sell more textiles and clothing with the elimination of the worldwide quota system in 2005. A free trade agreement between an ASEAN country and China could fuel the impetus for other ASEAN countries to pursue an FTA. An FTA between Japan and South Korea or between China, Japan and South Korea, could similarly start a domino movement among ASEAN countries, whose ﬁrms would not wish to lose their trade positions in these large economies. Baldwin (2005) believes that a Northeast Asian FTA (NEFTA) comprising China, Japan and South Korea could provide a trigger since their collective size would jeopardize between 35 and 45 percent of ASEAN exports. These bilateral trade agreements or a NEFTA could generate momentum toward ASEAN+3. However, the reluctance for deeper integration among Asian countries as well as existing divisive political

166

General views on East Asian economic integration

relationships make the prospects for regional integration less likely than in Europe. The development of bilateral agreements could, however, result in a ‘hub-and-spoke’ system where the large economies of China and Japan are the main hubs and the others form the spokes (Baldwin 2005). Baldwin (2005) argues that this arrangement has the eﬀect of marginalizing the spoke economies because industrial ﬁrms tend to locate in the hub economies. In this system, the ﬁrst-best gains from trade are unlikely to be realized. If Asia, on the other hand, did adopt a regional FTA, then the eﬀects on the economic performance of the region would be positive and indeed considerable. The Asian economies have diﬀerent endowments and historically diﬀerent specializations, which allow for considerable gains from trade. Most of the ASEAN countries are small in size, the sum of their GDP equaling that of South Korea, and vulnerable to ﬂuctuations in the USA, the EU and Japan, their major export destinations. With the rise of China, regional patterns of trade could be dramatically aﬀected. Although aggregate consumption is comparatively low in China, it has a population in urban areas that dwarfs that of the USA and the EU. With economic growth exceeding 9 percent per annum, incomes are expected to double approximately every eight years. Economic integration with an economy the size of China’s would generate considerable gains from trade for ASEAN, increasing the consumption/production possibilities of these economies as well as increasing eﬃciency. Moreover, China and Japan are major trading partners at diﬀerent stages of development, and the gains from accessing this integrated market would again be notable. And South Korea would extend this market further, particularly given its industrial development. East Asia has the potential to become a formidable economic region. The gains from an FTA are signiﬁcant for the developing countries in ASEAN as well as for the developed countries of Japan and South Korea. For ASEAN, economic integration would provide the market that they individually lack. For Japan and South Korea, utilizing the diﬀerent endowments of ASEAN and China would allow their economies to sustain economic performance. For China, economic integration would improve the eﬃciency of the proliferating production chains which already exist and are growing in the region, which would complement its economic development. In short, the potential gains from an FTA in Asia are considerable. However, the political will and institutional structures in ASEAN+3 are much less favorable to eﬀective integration than in Europe. The lack of a historical perception of Asia as a viable political entity and the remaining tensions in the region make the likelihood of economic integration more remote than in Europe.

European perspectives

6.5

167

CONCLUSIONS

This chapter has discussed the economics and politics of economic integration. In Section 6.2 it discussed the deep institutional structure of the EU, with its Council, Commission, Court and Parliament and its subtle consensual decision-making processes. It showed how these reﬂected a deep political commitment to integration and how they provided ballast in the stormy waters of achieving integration in practice. Section 6.3 summarized the economics of integration in terms of trade creation and diversion, economies of scale, competition, agglomeration, FDI and production networks. It observed the very special nature of Asia in terms of the last even without formal integration agreements. Section 6.4 discussed the dynamics of integration, showing that one step towards eﬀective integration may create incentives for the next. Such domino eﬀects are arguably evident in Europe and would likely also operate in Asia. Turning to the prospects for integration in Asia, the refrain throughout the chapter is that European integration has been successful because it has been a continuing process of steps to achieve deeper integration, going far beyond the removal of tariﬀs. The EU experience shows how the inevitable stresses can be handled institutionally and points to the importance of deep integration in achieving the full potential of a regional agreement. However, the EU performance is grounded in the deep political commitment of its members and in the creation of a political and institutional framework that can pursue integration and regional reform independently of national governments. It is in these dimensions that Asia is most fundamentally diﬀerent from the EU, and the possibility of following the European model is limited. While complementarities between Asian economies mean that there are beneﬁts from regional integration, if ‘deep integration’ cannot be achieved, then the beneﬁts of multilateral liberalization exceed those of regional integration, and should perhaps be the focus of trade reform in Asia.

NOTES 1. The same internationalism produced the UN, the IMF, the World Bank and the GATT. 2. The meeting of heads of government is known as the European Council. It has regular bi-annual meetings. 3. In 1999 the entire Commission resigned, under parliamentary pressure, over allegations of fraud and mismanagement. 4. The trade intensity index for i’s trade with j is (Tij /Tiw)/[(TwjTij)/Tww], where T represents trade in both directions and where subscript w represents the world. 5. Anticipation eﬀects have been noted previously – e.g. in Winters (1983) for the UK. 6. Just as with the apparent consumption exercise, these exercises are colored by the reduction in the accedants’ tariﬀs on other countries as they adopted the common external

168

General views on East Asian economic integration tariﬀ. In this case, however, the external trade changes may reasonably be attributed to integration.

REFERENCES Asian Development Bank (2005), Asian Development Outlook, Manila: Asia Development Bank. Baldwin, R.E. (1993), ‘A domino theory of regionalism’, in R.E. Baldwin, P. Haaparanta and J. Kiander (eds), Expanding Membership of the European Union, Cambridge: Cambridge University Press, pp. 25–53. Baldwin, R.E. (2005), ‘Asian regionalism: promises and pitfalls’, in Choong Yong Ahn, Richard E. Baldwin and Inkyo Cheong (eds), East Asian Economic Regionalism: Feasibilities and challenges, Berlin: Springer, pp. 157–74. Bayoumi, T. and B. Eichengreen (1997), ‘Is regionalism simply a diversion? Evidence from the evolution of the EC and EFTA’, in T. Ito and A.O. Krueger (eds), Regionalism versus Multilateral Trade Arrangements (NBER East Asia Seminar on Economics, vol. 6), Chicago: The University of Chicago Press, pp. 141–68. Braunerhjelm P., R. Faini, V.D. Norman, F. Ruane and P. Seabright (2000), ‘Integration and the regions of Europe: how the right policies can prevent polarization’, Centre for Economic Policy Research, London. European Commission (2001), Case decisions found at: http://europa.eu/ institutions/decision-making/index_en.htm. Feenstra, R. and G. Hanson (1997), ‘Foreign direct investment and relative wages: evidence from Mexico’s maquiladoras’, Journal of International Economics, 42, 371–93. Freund, C. and J. McLaren (1999), ‘On the dynamics of trade diversion: evidence from four trade blocks’, International Finance Discussion Paper no. 637, Board of Governors of the Federal Reserve System, Washington, DC. Griﬃth, R. (2001), ‘Product market competition, eﬃciency and agency costs: an empirical analysis’, Working Paper 01/12, Institute for Fiscal Studies, London. International Monetary Fund (IMF) (2002), World Economic Outlook, September. Jacquemin, A. and A. Sapir (1988), ‘International trade and integration of the European Community’, European Economic Review, 32, 1439–49. Krugman, P. and A.J. Venables (1990), ‘Integration and the competitiveness of peripheral industry’, in C. Bliss and J. de Macedo (eds), Unity with Diversity in the European Community, Cambridge: Cambridge University Press, pp. 56–77. Lamberte, M.B. (2005), ‘An overview of economic cooperation and integration in Asia’, in Asian Development Bank (ed.), Asian Economic Cooperation and Integration: Progress, Prospects, and Challenges, Manila: Asian Development Bank, pp. 3–41. Martin, R. and C. Criscuolo (2001), ‘A note on ownership and productivity in UK businesses’ mimeo, Centre for Economic Performance, LSE. Midelfart-Knarvik, K.-H. and H.G. Overman (2002), ‘Delocation and European integration: is structural spending justiﬁed?’, Economic Policy, 35, 321–59. Midelfart-Knarvik, K.-H., H.G. Overman, S. Redding and A.J. Venables (1999), ‘The location of industry in Europe’, London: CEPR.

European perspectives

169

Ng, F. and A. Yeats (2003), ‘Major trade trends in East Asia: what are their implications for regional cooperation and growth’, Policy Research Working Paper Series No. WPS 3084, World Bank, Washington, DC. Pelkmans, J. (2001), European Integration: Methods and economic analysis, London: Pearson. Puga, D. (2002), ‘European regional policy in the light of recent location theories’, Journal of Economic Geography, 2 (4), 372–406. Sapir, A. (1992), ‘Regional integration in Europe’, Economic Journal, 102, 1491–506. Sapir, A. (1998), ‘The political economy of EC regionalism’, European Economic Review, 42, 717–32. Soloaga, I. and L.A. Winters (2001), ‘Regionalism in the nineties: what eﬀect on trade?’, Centre for Economic Policy Research, Discussion Paper Series, No. 2183. Truman, E.M. (1975), ‘The eﬀects of European economic integration on the production and trade of manufactured products’, in B. Balassa (ed.), European Economic Integration, Amsterdam: North Holland, pp. 3–40. Venables, A.J. and N. Limao (2002), ‘Geographical disadvantage: a Heckscher–Ohlin–von Thunen model of international specialization’, Journal of International Economics, 58, 239–63. Winters, L.A. (1983), ‘British imports of manufactures and the Common Market’, Oxford Economic Papers, 36, 103–18. World Trade Organization (2005), International Trade Statistics 2005, Geneva: WTO. Zebregs, H. (2004), ‘Intraregional trade in emerging Asia’, IMF Policy Discussion Paper No. PDP/01/1.

PART III

Case Studies of East Asian Economic Integration

7.

Structure and determinants of intra-regional trade in East Asia Satoru Kumagai

7.1

INTRODUCTION

The ﬁrst East Asian Summit was held in Kuala Lumpur, Malaysia, on 14 December 2005. This summit heralded the advent of the era of de jure regional integration in East Asia. However, de facto economic integration of the region unfolded much earlier, in the latter half of the 1980s. In East Asia, de facto regional integration has been pushed forward primarily by extensive networks of multinational enterprises (MNEs) and so far has been leading de jure regional integration. This chapter has two goals: the ﬁrst is to provide an overview of the structure of intra-regional trade in East Asia during the last two decades. The second is to analyze the determinants of intra-regional trade in East Asia by industrial sector. Speciﬁcally, the determinants of trade are analyzed by estimating a variant of the gravity equation. This equation is well known for its successful use in empirical research related to international trade. Knowing the determinants of trade makes it possible to predict the impact of further progress in regional economic integration. Further, this knowledge may also conﬁrm that spatial economics is relevant and indispensable for understanding intra-regional trade in East Asia. This chapter is structured as follows: an overview of the structure of intra-regional trade in East Asia is provided in Section 7.2. This serves as background information to assist in understanding the following sections. In Section 7.3, a theoretical framework for analyzing the determinants of intra-regional trade is proposed. Section 7.4 includes an explanation of data and empirical models used in the analysis. A report of the results of the estimation and subsequent interpretation is then presented in Section 7.5. The last section includes discussion of prospects and policy implications for regional integration in East Asia.

173

174

7.2

Case studies of East Asian economic integration

STRUCTURE OF INTRA-REGIONAL TRADE IN EAST ASIA: AN OVERVIEW

In the era of de facto regional integration, intra-regional trade in East Asia has increased steadily despite the lack of regional institutional arrangements for trade liberalization. This seems to be a consequence of incentives that have been given unilaterally to MNEs so that they may realize ‘virtual’ free trade by governments that are trying to attract foreign direct investment (FDI). The following analysis shows that intra-regional trade in East Asia is driven by extensive networks of MNEs and (as opposed to the EU and NAFTA) is heavily skewed toward production-oriented rather than consumption-oriented. 7.2.1

Volume of Intra-regional Trade

The volume of intra-regional trade in East Asia has increased tremendously over the last two decades. In East Asia,1 it has increased from US$120 billion in 1985 to US$1033 in 2004, a factor of 8.6 (Table 7.1). During the same period, intra-regional trade in NAFTA and the EU-25 has increased by a factor of 4.9 and 4.6 respectively. As a consequence, Table 7.1

International trade matrix (1985 and 2004; US$ billion)

1985 East Asia

NAFTA

EU-25

ROW

Total

East Asia NAFTA EU-25 ROW

120 139 49 4

59 153 67 48

34 89 440 120

80 70 220 72

293 450 776 281

Total

348

327

683

442

1800

East Asia

NAFTA

EU-25

ROW

Total

East Asia NAFTA EU-25 ROW

1033 617 445 215

226 703 237 124

216 343 2174 459

389 327 669 359

1865 1990 3524 1159

Total

2311

1289

3193

1744

8538

2004

Source: COMTRADE.

Structure and determinants of intra-regional trade

175

the volume of intra-regional trade in East Asia has surpassed that of NAFTA and is equivalent to half that of the EU-25 in 2004. Table 7.1 shows important characteristics of three trade blocs. East Asia’s share of intra-regional exports in total exports was 44.7 percent in 2004, the lowest among three trade blocs; those of NAFTA and the EU-25 were 54.5 percent and 68.1 percent respectively. On the other hand, NAFTA’s share of intra-regional imports in total imports was 35.3 percent in 2004, the lowest among three trade blocs; those of East Asia and the EU-25 were 55.4 percent and 61.7 percent respectively. This implies that East Asia is ‘open’ to export to other regions, while NAFTA is ‘open’ to import from other regions. Actually, exports from East Asia to NAFTA are quite large, and these two trade blocs complement each other in terms of trade structure. This complementary state is primarily due to the fact that a large number of Japanese and US MNEs have been using developing economies in East Asia as lowcost production and export bases targeting US markets. In contrast, even considering the diﬀerence in the number of member countries, the EU-25 seems to be a self-suﬃcient trade bloc when compared with the other two. 7.2.2

Changes in the Composition of Traded Goods

Table 7.2 compares the composition of intra-regional trade in East Asia, NAFTA and the EU-25 by the single-digit SITC classiﬁcations of 1985 and 2004. For East Asia, the largest share of trade in the region was SITC-3 (mineral fuels, lubricants and related materials, 26 percent), followed by SITC-7 (machinery and transport equipment, 23 percent) and SITC-6 (manufactured goods classiﬁed chieﬂy by materials, 17 percent) in 1985. Intraregional trade in East Asia was dominated by primary goods (SITC-0 to SITC-4) composed of over 40 percent of all intra-regional trade at that time. In 2004, primary goods comprised only 12 percent of all intra-regional trade. The largest share of trade was SITC-7 (51 percent) followed by SITC-8 (miscellaneous manufactured articles, 15 percent) and SITC-6 (13 percent). For NAFTA, there was no drastic change in the composition of intraregional trade between 1985 and 2004. For the EU-25, trade of SITC-7 increased, and the trade of SITC-3 decreased during the same period. However, the change was not as signiﬁcant as that of East Asia. Table 7.2 shows that East Asia experienced a signiﬁcant change in the composition of intra-regional trade during the past two decades. The dominant sector of intra-industry trade in East Asia changed from primary goods to manufactured goods. This may be due to the massive inﬂow of FDI into the region, especially after the Plaza Accord of 1985. This accord promoted the industrialization of host countries and thus changed trade structures in East Asia drastically.

176

Case studies of East Asian economic integration

Table 7.2 The composition of intra-regional trade in East Asia, the EU-25 and NAFTA (1985 and 2004, SITC 1-digit; percentages) SITC

0 1 2 3 4 5 6 7 8 9 Total

1985

2004

East Asia

NAFTA

EU-25

East Asia

NAFTA

EU-25

7.7 0.3 6.8 25.8 0.9 4.9 17.3 23.1 9.9 3.4

4.3 0.5 5.9 13.8 0.1 4.7 12.1 48.4 5.4 4.8

9.7 1.4 5.1 10.4 0.6 12.1 20.3 29.0 10.7 0.8

2.8 0.2 2.2 6.7 0.4 8.1 13.0 50.7 14.8 1.2

4.9 0.5 3.4 11.6 0.2 7.4 14.1 44.5 9.6 3.9

7.6 1.4 2.8 5.0 0.4 15.5 16.6 38.5 10.2 1.8

100.0

100.0

100.0

100.0

100.0

100.0

Note: SITC 0food and live animals chiefly for food; 1 beverages and tobacco; 2 crude and inedible materials, except fuels; 3 mineral fuels, lubricants and related materials; 4 animal and vegetable oils, fats and waxes; 5 chemicals and related products not elsewhere specified; 6 manufactured goods classified chiefly by materials; 7 machinery and transport equipments; 8 miscellaneous manufactured articles; 9 commodities and transactions not classified elsewhere in the SITC. Source: COMTRADE.

7.2.3

Changes in the Composition of Trade in Manufactured Goods

This subsection includes an examination of the detailed composition of intra-regional trade in machinery and transport equipment (SITC-7) in each region using the two-digit SITC classiﬁcation. Table 7.3 includes a comparison of the composition of intra-regional trade in machinery and transport equipment in East Asia, the EU and NAFTA in 1985 and 2004. For East Asia in 1985, the largest share was SITC-77 (electrical machinery, 26 percent) followed by SITC-72 (machinery specialized for particular industries, 18 percent) and SITC-76 (telecom and soundrelated equipment, 14 percent). In 2004, the largest share was SITC-77 (48 percent) followed by SITC-75 (oﬃce machines and automatic data processing machines, 17 percent) and SITC-76 (15 percent). These three largest categories, speciﬁcally the electric and electronics (E&E) industry, comprised about 80 percent of the intra-regional trade in East Asia. SITC-78 (road vehicles) comprised only 4 percent of intra-regional trade.

177

Structure and determinants of intra-regional trade

Table 7.3 The composition of intra-regional trade in East Asia, the EU and NAFTA (1985 and 2004 within SITC-7; SITC two digit; percentages) SITC

71 72 73 74 75 76 77 78 79 Total

1985

2004

East Asia

NAFTA

EU-25

East Asia

NAFTA

EU-25

5.5 17.7 2.5 12.8 5.7 13.8 27.0 8.7 6.3 100.0

9.3 5.5 1.0 4.9 5.6 4.2 8.4 57.1 4.0 100.0

6.6 11.9 2.4 13.0 11.1 4.9 14.8 31.4 3.9 100.0

2.6 5.3 1.7 5.2 17.4 14.9 48.6 3.7 0.5 100.0

9.1 3.5 0.7 9.1 5.4 9.3 16.0 43.7 3.2 100.0

7.1 6.3 1.1 11.5 9.1 8.5 14.7 37.8 3.7 100.0

Note: SITC 71 power generating machinery and equipment; 72 machinery specialized for particular industries; 73 metalworking machinery; 74 general industrial machinery and equipment, nes, and parts of, not elsewhere specified; 75 oﬀice machines and automatic data processing equipment; 76 telecommunications, sound recording, and reproducing equipment; 77 electric machinery, apparatus, nes, and parts, not elsewhere specified; 78 road vehicles; 79 other transport equipment. Source: COMTRADE.

From Table 7.3, it appears that the composition of intra-regional trade in manufactured goods in East Asia is signiﬁcantly diﬀerent from that of NAFTA and the EU-25. The E&E industry dominates intra-regional trade in machinery and transport equipment in East Asia, but the automotive industry dominates trade in the other two regions. Reasons for this may include the fact that, until recent times, automotive markets in most East Asian countries were heavily protected in order to develop ‘national’ automotive industries. On the other hand, the E&E industry in East Asia has enjoyed virtually free trade since the 1970s due to various incentives given by host countries (such as free trade zones or FTZs). 7.2.4 The Structure of Intra-regional Trade by BEC (Basic Economic Category) Classiﬁcation Figure 7.1 shows the structure of intra-regional trade in each region by economic function.2 Intra-regional trade in East Asia is characterized by a large volume of trade in parts (33 percent), while the volume of trade

178

Case studies of East Asian economic integration East Asia Consumption goods 15%

Primary goods Processed 4% goods 29%

Nes

Capital goods 18%

Parts 33% NAFTA Nes

Primary goods 11%

Consumption goods 22%

Capital goods 16%

Processed goods 28%

Parts 20%

EU-25 Nes Consumption goods 30%

Primary goods 4%

Processed goods 33%

Parts 17% Capital goods 14%

Source: COMTRADE.

Figure 7.1 The composition of intra-regional trade (2004; classified by economic function)

179

Structure and determinants of intra-regional trade

Table 7.4 Destination of exports from East Asia by economic function (1985 and 2004, percentages) 1985 Primary goods Processed goods Parts Capital goods Consumption goods

Intra-regional

NAFTA

EU-25

ROW

57.6 44.8 25.0 22.1 21.7

10.9 10.0 30.9 32.8 30.4

13.1 14.1 15.8 20.8 19.7

18.4 31.1 28.3 24.3 28.2

Intra-regional

NAFTA

EU-25

ROW

56.9 46.8 40.8 25.9 17.8

8.6 14.1 18.1 30.0 36.3

9.7 11.8 14.4 24.0 24.0

24.7 27.2 26.6 20.1 21.9

2004

Primary goods Processed goods Parts Capital goods Consumption goods Source: COMTRADE.

in consumption goods (15 percent) is relatively small. The trade in parts within NAFTA and the EU-25 is 20 percent and 17 percent respectively; the trade in consumption goods is 22 percent and 30 percent respectively. It is clear that, compared with the other two regions, intra-regional trade in East Asia is characterized by a larger volume of trade in parts. This means that intra-regional trade in East Asia is driven by international production networks rather than reciprocal exports in consumption goods. 7.2.5

The Structure of Exports by Destination and BEC Classiﬁcation

Table 7.4 compares the destination for exports from East Asia by BEC classiﬁcation in 1995 and 2004. The ﬁrst and second rows of both years show that the majority of exports of primary goods and processed goods from East Asia went to the same region,3 while the exports to NAFTA and the EU were relatively small. This is because East Asia contains countries rich in natural resources such as some in ASEAN, as well as countries scarce in natural resources such as Japan, Korea and China.

180

Case studies of East Asian economic integration

The third rows for both years show that the share in exports of parts from East Asian countries to other East Asian countries has increased dramatically during the last decade and has reached over 40 percent, while the share for NAFTA has decreased signiﬁcantly. This is because MNEs operating in East Asia, especially ﬁrms in the E&E industry, have developed dense production networks in the region during the last decade. The fourth rows for both years show that the share of exports of capital goods from East Asian countries to other East Asian countries has increased to 27 percent, while the primary destination of the exports is still NAFTA. This is partly because the automotive industry is included in capital goods. Exports of the Japanese and Korean automotive industries contribute to the high share of exports of capital goods to NAFTA. The last rows of both years show that the primary destination for exports of consumption goods from East Asia is still NAFTA, and this is followed by the EU-25. Despite the strong economic growth of the region, the share of exports from East Asian countries to other East Asian countries has not changed signiﬁcantly during the last decade. These data show that East Asia still needs the extra-regional demand for consumption goods, and intra-regional trade in parts and capital goods has increased signiﬁcantly. This again supports the view that intra-regional trade in East Asia is driven by international production networks of MNEs rather than reciprocal exports in consumption goods.

7.3

THEORETICAL FRAMEWORK

Heretofore, the structure of intra-regional trade in East Asia has been illustrated and compared with NAFTA and the EU-25. This section includes explanation of the theoretical framework used to identify determinants of intra-regional trade in East Asia. 7.3.1

A Brief History of International Trade Theory

Before going into the detailed theoretical framework used in this chapter, I brieﬂy describe an outline of the historical development of international trade theory. Basically, there are three types of theories that have been developed to explain determinants of international trade. The oldest is David Ricardo’s theory of comparative advantage (Ricardo 1817). This was developed in the ﬁrst half of the nineteenth century and is still valid. The Ricardian theory presumes diﬀerence in technology or productivity to be a source of comparative advantage. The theory further asserts that a

Structure and determinants of intra-regional trade

181

country exports goods for which it has a comparative advantage in production technology. The Heckscher–Ohlin (H–O) theory presumes that diﬀerences in factor endowment of each country are a source of comparative advantage. The H–O theory was developed in the 1930s and elaborated by Paul Samuelson during the 1950s. It asserts that a country exports goods that are produced by intensive use of the relatively abundant factors in that country (Ohlin 1933; Samuelson 1953). These two ‘conventional’ theories of international trade are intended to explain trade between countries that have diﬀerent characteristics. Conversely, ‘new’ trade theories, which were developed after the end of the 1970s, seek to explain trade between countries that have little or no diﬀerence in technology or factor endowments. New trade theories seem motivated by the fact that world trade is dominated by trade between developed countries (North–North trade) that have little or no diﬀerence in technology or factor endowments. The volume of trade between developed and developing countries (North–South trade) that have signiﬁcant diﬀerences in technology or factor endowments falls short of the prediction of conventional trade theories. There are several variations of new trade theory. For example, Krugman (1979) explains international trade without comparative advantages by constructing a model with increasing returns to scale at the factory level and ‘love of variety’ at the consumer level. Subsequently, Krugman (1980) introduced transport costs and uneven market size into the model and derived the famous home market eﬀect (HME). The HME occurs when a country that has a larger market of goods tends to produce more of those goods than required to meet domestic demand and consequently exports those goods to other countries. Brander and Krugman (1983) present a model of mutual trade in identical goods called the ‘reciprocal dumping’ model. In this model, oligopolistic ﬁrms export or ‘dump’ identical goods to another country ‘reciprocally’. These new trade theories evolved into various models of ‘spatial economics’ or ‘new economic geography’ in the 1990s (Fujita et al. 1999). Another line of explanation for the pattern of international trade in East Asia is known as the ﬂying-geese model. This was originally developed by Kaname Akamatsu in the 1930s. Kojima (2000, p. 385) explains the ﬂyinggeese model by citing Okita (1985) as follows: The division of labor in the Paciﬁc region has aptly been called the FG [ﬂyinggeese] pattern of development . . . Traditionally, there have been two patterns or types of international division of labor: the vertical division of labor such as prevailed in the 19th century to deﬁne relations between the industrialized

182

Case studies of East Asian economic integration

country and the resource-supplying country or between the suzerain and the colony; and the horizontal division of labor typiﬁed by the EEC with its trade in manufactures among industrialized countries, often among countries at the same stage of development and sharing a common culture. By contrast with both of these types, the FG pattern represents a special kind of dynamism. In the Paciﬁc region, for example, the United States developed ﬁrst as the lead country. Beginning in the late 19th century, Japan began to play catch-up development in the nondurable consumer goods, durable consumer goods, and capital goods sectors in that order. Now the Asian NICs and the ASEAN countries are following in Japan’s footsteps . . . Because there is such great variety in the Asian nations’ stages of development, natural resource endowments, and cultural, religious, and historical heritages, economic integration on the EEC model is clearly out of the question. Yet it is precisely this diversity that works to facilitate the FG pattern of shared development as each is able to take advantage of its distinctiveness to develop with a supportive division of labor. (Okita 1985, p. 21).

Although it is not well integrated into mainstream international trade theory, the ﬂying-geese model is very popular in East Asia and actually ﬁts the pattern of economic development and international trade in the region very well. 7.3.2

Determinants of Trade and Policy Implications

Determinants of intra-regional trade are identiﬁed in this section, and each one has diﬀerent policy implications. If the H–O theory of comparative advantage (based on factor endowments) is valid for trade in a sector, then basically no sensible policy measure is available to enhance the exports of that sector, at least in the short term. Because factor endowments (such as land, climate and natural resources) are of ‘the ﬁrst nature’ or precondition in the short run, laissez-faire is a desirable attitude. However, if new trade theory is valid for trade in a sector, it may be possible to enhance exports of that sector by some policy measures. In new trade theory, trade is determined by ‘the second nature’ or the result of interactions of human activities.4 For example, protecting the home market of an industry by tariﬀ may enhance the exports of that industry through the HME. Thus ‘strategic’ trade policy has a role to play (Krugman 1986; Brander 1995). If the ﬂying-geese model is valid in a sector, then less developed countries are in a position of natural advantage for catching up in that sector. Thus less developed countries may do well by implementing some policies to attract FDI in those sectors.

Structure and determinants of intra-regional trade

7.3.3

183

The Gravity Equation and its Underlying Theories

Although it is important to identify the determinants of trade in order to derive policy implications for regional economic integration, empirical research seeking to identify these determinants has not been successful. Despite many eﬀorts since Leontief (1953), studies have been ambiguous at best (see Harrigan 2003). One notable exception is the gravity equation, which explains the volume of trade between any two countries surprisingly well. The original gravity equation in physics explains the magnitude of the gravitation force between two masses as the product of their masses times the gravitational constant divided by the square of the distance between them. Likewise, the gravity equation in international trade predicts that the volume of trade between two countries is proportional to the product of their GDPs and inversely proportional to the distance between them. Deardorﬀ (1995) speciﬁed the ‘standard gravity equation’ as follows: YiYj (7.1) Tij A D , ij where Tij denotes the value of trade between countries i and j, Yi and Yj denote the GDPs of each country, Dij is the distance between them, and A is a constant. In empirical studies, equation (7.1) is usually estimated in its log linear form as: log(Tij ) 1log(Yi ) 2log(Yj ) 3log(Dij ) ij.

(7.2)

The gravity equation usually yields a relatively high R2 in various settings. Despite its exceptional success in empirical work, the gravity equation has been criticized because ‘[it] has had a comparatively shallow (if not shaky) theoretical foundation’ (Harrigan 2003). The gravity equation has ‘many’ strong theoretical foundations, but it is diﬃcult to identify which theory actually contributes to its empirical success. The ﬁrst attempt to formalize the gravity equation with rigid economic theory assuming product diﬀerentiation was made by Anderson (1979). The equation was derived from Cobb–Douglas or CES preferences combined with the Armington assumption that goods are diﬀerentiated by country of origin. Bergstrand (1985) also derived the gravity equation from a general equilibrium model with CES (constant elasticity of substitution) preferences and the Armington assumption. Subsequently, Helpman (1987) showed that the H–O model is not consistent with the gravity equation, but a model of trade in diﬀerentiated goods provides a theoretical explanation of the gravity equation.

184

Case studies of East Asian economic integration

On the contrary, Deardorﬀ (1995) derived the gravity equation from a model based on H–O theory that does not incorporate product diﬀerentiation. He asserted that it is spurious to interpret the explanatory power of the gravity equation as proof of the validity of a monopolistic competition model. 7.3.4

How to Identify the Theoretical Foundation of the Gravity Equation

To the extent that it is not clear which theory is actually behind the empirical success of the gravity equation, its exceptional explanatory power seems to contribute nothing to identifying the determinants of trade. However, if the theory behind the success of the gravity equation were to be identiﬁed, it would certainly help identify the determinants of international trade. Feenstra et al. (2001, p. 455) tried to identify theories behind the gravity equation and arrived at the following conclusion: The home market eﬀect shows up consistently for diﬀerentiated goods in the form of a domestic-income elasticity that exceeds the partner-income elasticity. This eﬀect is much less pronounced for the in-between category and is reversed for homogeneous goods, which is consistent with a reciprocal dumping model with barriers to entry.5

Feenstra et al. (2001) show that the monopolistic competition model (a variant of new trade theory) is the theoretical model behind the empirical success of the gravity equation for diﬀerentiated goods (such as apparel and cars). They further show that the reciprocal dumping model (with barriers to entry and also a variant of new trade theory) is behind the success of the gravity equation for homogeneous goods (such as some natural resources and highly standardized manufactured goods). 7.3.5

The Approach Adopted in this Chapter

This chapter includes identiﬁcation of the determinants of intra-regional trade in East Asia based on Feenstra et al. (2001) and extending this identiﬁcation to incorporate other types of determinants of trade. Basically, variants of the gravity equation are estimated, and the domesticincome elasticity of exports (l) is compared with the partner-income elasticity of exports (j) by industrial sector to see if the HME or the reverse HME exists.6 In addition, two explanatory variables are included: (1) the country dummy for home and partner countries and (2) the diﬀerence in per capita GDPs for home and partner countries. The ﬁrst variable is intended to detect H–O types of determinants of trade, and the second is intended to

Structure and determinants of intra-regional trade

185

detect the validity of the ﬂying-geese model. Detailed explanations of the empirical models are given in the next section.

7.4 DATA AND EMPIRICAL MODEL 7.4.1

Data Used in Estimation

In this study, the Asian Input–Output Table compiled by the Institute of Developing Economies (IDE) was used instead of the COMTRADE database, which is an international trade data set compiled by UNCTAD and commonly used in various studies of international trade. IDE’s Asian I–O Table covers ten countries, speciﬁcally nine countries in East Asia (Indonesia, Malaysia, the Philippines, Singapore, Thailand, China, Taiwan, Korea and Japan) plus the USA. The Asian I–O Table contains as many as 76 sectors (24 sectors if used as consistent time series data) and is published every ﬁve years. In this chapter, the 24-sector Asian I–O Table (excluding the USA7) in 1985, 1990, 1995 and 2000 is used as a pooled data set. There are two main reasons for choosing the International I–O Table: ﬁrst, it contains production, consumption and trade data in an integrated form. It is important to know the exact market size for each sector in order to estimate the HME or the reverse HME accurately. Although it is possible to calculate domestic trade by combining international trade data with a production and consumption data set, converters between diﬀerent data sets are required, and this may lead to inconsistency in the compiled data set. Second, by using the I–O Table, it is easy to distinguish trade of intermediate goods from ﬁnal goods. There may be diﬀerent determinants of trade for intermediate and ﬁnal goods. Although it is also possible to distinguish the trade of intermediate goods from that of ﬁnal goods by using ordinary trade data (if the industrial classiﬁcation is detailed enough), great eﬀort is required to compile the data set so that trade of intermediate goods is separated from that of ﬁnal goods. 7.4.2

Empirical Models

Base model First, the following base model was estimated: log(Xijkt ) Dklog(distij ) Iklog(Dikt ) Jklog(Djkt ) GkGDPPC.difijt ttdt ijkt

(7.3)

186

Case studies of East Asian economic integration

Variables are as follows: Xijkt is the value of exports of goods in sector k from country i to country j in year t; distij is the bilateral transport cost between country i and county j, approximated by the great circle distance between the capital city of country i and that of country j; Dikt is the home country’s domestic market size of sector k in country i in year t; Djkt is the partner country’s domestic market size of sector k in country j in year t; GDPPC.difijt is the diﬀerence in GDP per capita between country i and country j in year t; tdt is a vector of time dummies; ijkt is the error term. Equation (7.3) is a common gravity equation, but some needed explanations are given below. First, the actual market size of sector k in the home and partner countries was used instead of their GDPs. Because the demand structure of a given country is not homothetic to its GDP, the GDP cannot be used as a proxy for the actual market size of each sector in each country. Second, the diﬀerence in GDP per capita between home and partner countries was included as an explanatory variable. This variable is intended to test the validity of the ﬂying-geese model. If the coeﬃcient on this variable is negative (positive) and statistically signiﬁcant, it is possible to say that a relatively less developed country tends to export (import) goods in that sector. As described in the ﬂying-geese model, this means that goods in that sector are matured and suitable to be produced and exported by less developed countries. Third, a vector of time dummies was included to eliminate time-speciﬁc eﬀects from the pooled data set. Model with country ﬁxed eﬀects The base model is quite simple and easy to understand. However, it is very likely to suﬀer from unobserved heterogeneity. It is plausible that the volume of trade is aﬀected by time-invariant characteristics of a country, such as factor endowment, that are constant over time from 20 to 30 years. The following speciﬁcation is intended to control these country ﬁxed eﬀects by including full-set dummies for both home and partner countries:8

Structure and determinants of intra-regional trade

log(Xijkt ) Dklog(distij ) Iklog(Dikt ) Jklog(Djkt ) GkGDPPC.difijt ikctyi jk ptnj ttdt ijkt

187

(7.4)

The new variables introduced here are: ctyi as a home country dummy, and ptnj as a partner country dummy. Full-set dummies are included to control unobserved factors that are constant over time and are also expected to control the ‘multilateral resistance’ addressed in Andersen and Wincoop (2003).9 Testing the existence of home market eﬀects To test the existence of the HME or the reverse HME statistically, the following hypotheses were examined: H0: I J (neither the HME nor the reverse HME) H1: I J (the HME) H2: I J (the reverse HME) H0 is a null hypothesis, meaning that there are neither the HME nor the reverse HME. H1 indicates the existence of the HME, and H2 indicates the existence of the reverse HME. To test these hypotheses, the third model derived from equation (7.4) was used.10 log(Xijkt ) Dklog(distij ) 1klog(Dikt ) 2k{log(Dikt ) log(Djkt )} GkGDPPC.difijt ikctyi jk ptnj ttdt ijkt

(7.5)

Using equation (7.5) makes it possible to test the above hypotheses directly by determining if lk are statistically signiﬁcant. H0, H1 and H2 are replaced by the following hypotheses: H: 0 1 0 H1: 1 0 H: 2 1 0 Other estimated coeﬃcients should be identical to those in equation (7.4). Testing the signiﬁcance of country-speciﬁc ﬁxed eﬀects So far, three models have been proposed to test the existence of the HME and the reverse HME as well as the validity of the ﬂying-geese model. In addition, the statistical signiﬁcance of country-speciﬁc ﬁxed eﬀects

188

Case studies of East Asian economic integration

was tested for home country. F-tests were conducted by sector comparing the unrestricted model of equation (7.4) and the following restricted model: log(Xijkt ) Dklog(distij ) Iklog(Dikt ) Jklog(Djkt ) GkGDPPC.difijt jkptnj ttdt ijkt.

(7.6)

In equation (7.6), the home country dummy is omitted from equation (7.4). If the F-test for a sector shows the home country dummy to be statistically signiﬁcant, then time-invariant characteristics of the home country (climate, abundance of land, and available natural resources, for example) may be said to aﬀect the volume of trade in that sector. This means that the H–O type of comparative advantage aﬀects the volume of trade in an important way.

7.5 7.5.1

RESULTS OF ESTIMATION Base Model

Table 7.5 shows the result of the estimation of equation (7.3). The results are reported only for 19 traded goods sectors out of 24 sectors. The ﬁrst ﬁve columns show results for ﬁnal goods and the last ﬁve columns those for intermediate goods. There are three key results found in Table 7.5. First, the distance elasticity of trade diﬀers signiﬁcantly by sector, but as expected, the signs of the coeﬃcients are negative for almost all. Second, the elasticity of trade to the home and partner market size diﬀers signiﬁcantly by sector but is mostly positive. Third, the coeﬃcients of GDPPC.dif are both positive and negative. The coeﬃcients for primary goods sectors tend to be negative, and those for manufacturing sectors tend to be positive. This means that primary goods (manufacturing goods) tend to be exported from less developed (more developed) countries to more developed (less developed) countries. Overall, the explanatory power of the model seems to be rather low. Adjusted-R2s are between 0.018 at the lowest and 0.521 at the highest. This means that there may be explanatory variables other than the distance between trade partners, the market size of each country, and the diﬀerence in per capita income. The next subsection addresses this problem.

Structure and determinants of intra-regional trade

7.5.2

189

Model with Country Fixed Eﬀects

Table 7.6 provides results of estimation for equation (7.4). This equation includes country dummies for both home and partner countries in order to address the possible existence of country ﬁxed eﬀects. There are two key results found in Table 7.6. First, adjusted-R2s show marked improvement over the base model. They range between 0.037 and 0.704. This means that the country ﬁxed eﬀects are very important for explaining the volume of trade for almost all sectors. Second, the coeﬃcients of GDPPC.dif show signiﬁcant change in most sectors compared with those in Table 7.5. This means that what seems to be an understandable relationship found in the previous subsection between GDPPC.dif and volume of trade in primary goods sectors and manufacturing sectors is probably not a ‘true’ relationship but is heavily aﬀected by country ﬁxed eﬀects. The relationships between GDPPC.dif and the volume of trade by sector in Table 7.5 are interesting and not always consistent with intuition. For example, a positive and statistically signiﬁcant relationship between GDPPC.dif and the volume of trade is observed for the crude petroleum and natural gas (F) and ﬁshery (I) sectors. On the other hand, a negative and statistically signiﬁcant relationship between GDPPC.dif and the volume of trade is observed for the other agricultural products (F), forestry (F), pulp, paper and printing (F), machinery (F), other mining (I), pulp, paper and printing (I), non-metallic mineral products (I), machinery (I) and transport equipment (I) sectors. This result suggests that some manufacturing goods such as pulp and paper, non-metallic mineral products, machinery, and parts of transport equipment are likely to be produced and exported by less developed countries. These countries may be described as ‘following geese’. 7.5.3

Testing Home Market Eﬀects and Reverse Home Market Eﬀects

Table 7.7 reports results for equation (7.5), which is intended to directly test the existence of the HME or the reverse HME. If lk for a sector is positive and statistically signiﬁcant, then HME exist in that sector. Conversely, if lk for a sector is negative and statistically signiﬁcant, then reverse HME exist in that sector. Table 7.7 shows that HME exist in the other mining (I), pulp, paper and printing (F,I), chemical products (F, I), rubber products (I), and transport equipment (F) sectors. It also shows that reverse HME exist in the ﬁshery (F, I), crude petroleum and natural gas (F, I), other mining (F), timber and

190

8 Food, beverage and tobacco 9 Textle, leather and products thereof 10 Timber and wooden products 11 Pulp, paper and printing

6 Crude petroleum and natural gas 7 Other mining

5 Fishery

4 Forestry

0.174*** (0.054) 0.046 (0.064) 0.228*** (0.066) 0.083 (0.074)

GDPPC. dif

0.147 0.035 0.276*** (0.180) (0.078) (0.078) 0.571*** 0.669*** 0.239*** (0.210) (0.087) (0.087) 0.085 0.071 0.872*** (0.212) (0.106) (0.106) 0.787*** 0.957***0.157 (0.209) (0.109) (0.109)

log(Dj) 0.049* (0.027) 0.450*** (0.071) 0.243*** (0.082) 0.513*** (0.080) 0.586*** (0.093) 0.354*** (0.069) 0.352*** (0.100)

log(Di)

0.193** 0.017* 0.006 (0.086) (0.010) (0.010) 0.105 0.256***0.015 (0.228) (0.091) (0.091) 1.001*** 0.167*** 0.124** (0.244) (0.060) (0.060) 0.115 0.157*** 0.312*** (0.252) (0.035) (0.035) 0.540* 0.015 0.175 (0.290) (0.143) (0.143) 0.089 0.094*** 0.082*** (0.214) (0.026) (0.026) 0.059 0.056* 0.045 (0.250) (0.032) (0.032)

log(dist)

Final goods

Estimated results for equation (7.3)

2 Other agricultural products 3 Livestock

1 Paddy

Table 7.5

log(Di)

log(Dj)

0.231 0.097 (0.181) 0.307 0.230 (0.188) 0.381 0.099 (0.238) 0.358 0.288 (0.182)

0.112 0.597***0.168*** (0.091) (0.091) (0.055) 0.719*** 0.128 0.107* (0.079) (0.079) (0.057) 0.144 0.888***0.341*** (0.100) (0.100) (0.073) 0.535*** 0.171** 0.182*** (0.075) (0.075) (0.060)

0.427

0.335

0.356

0.247

0.297

0.173

0.240

0.519

0.150

0.300

0.077

GDPPC. Adj.R2 dif

0.018 0.474*** 0.054** 0.076***0.198*** (0.174) (0.025) (0.025) (0.055) 0.219 0.050 0.265*** 0.277***0.551*** (0.198) (0.083) (0.083) (0.064) 0.118 1.210*** 0.366*** 0.600***0.212** (0.265) (0.105) (0.105) (0.082) 0.334 0.117 0.387*** 0.520***0.525*** (0.266) (0.038) (0.038) (0.088) 0.143 0.692** 0.241** 0.738***0.449*** (0.278) (0.105) (0.105) (0.085) 0.134 0.304 0.429*** 0.401***0.810*** (0.459) (0.106) (0.106) (0.146) 0.084 0.412* 0.520*** 0.647***0.293*** (0.222) (0.089) (0.089) (0.068)

Adj.R2 log(dist)

Intermediate goods

191

0.469** 0.478*** 0.008 (0.228) (0.114) (0.114) 13 Petroleum and petro 0.609** 0.005 0.149** products (0.309) (0.060) (0.060) 14 Rubber products 0.806*** 0.078 0.044 (0.283) (0.063) (0.063) 15 Non-metallic mineral 0.045 0.142*** 0.088** products (0.230) (0.043) (0.043) 16 Metal products 0.381 0.170*** 0.461*** (0.232) (0.048) (0.048) 17 Machinery 0.440** 0.657*** 0.309*** (0.200) (0.091) (0.091) 18 Transport equipments 0.404* 0.976***0.007 (0.239) (0.117) (0.117) 19 Other manufacturing 0.412** 0.837*** 0.594*** products (0.188) (0.096) (0.096)

Calculations based on the IDE Asia International I–O Table.

0.398*** 0.309*** 0.378*** (0.070) (0.070) (0.051) 0.245 0.720*** 0.101 (0.152) (0.152) (0.083) 0.608*** 0.416***0.009 (0.102) (0.102) (0.063) 0.629*** 0.266*** 0.070 (0.076) (0.076) (0.055) 0.620*** 0.337*** 0.112* (0.073) (0.073) (0.058) 0.612*** 0.447*** 0.155** (0.091) (0.091) (0.072) 0.994*** 0.363*** 0.259*** (0.092) (0.092) (0.076) 0.689*** 0.339*** 0.140** (0.074) (0.074) (0.061)

Source:

0.238 0.033 (0.160) 0.201 0.606** (0.262) 0.097 0.092 (0.207) 0.169 0.412** (0.177) 0.384 0.187 (0.176) 0.466 0.466** (0.199) 0.476 0.542** (0.225) 0.417 0.414** (0.181)

*** Significant at 1 percent level, ** significant at 5 percent level, * significant at 10 percent level. Standard errors are in parentheses.

0.291*** (0.070) 0.267*** (0.099) 0.012 (0.089) 0.022 (0.072) 0.344*** (0.073) 0.324*** (0.068) 0.400*** (0.078) 0.116* (0.064)

Notes:

12 Chemical products

0.521

0.496

0.484

0.388

0.409

0.254

0.211

0.433

192

log(dist)

log(Di)

8 Food, beverage and tobacco 9 Textle, leather and products thereof 10 Timber and wooden products 11 Pulp, paper and printing

1 Paddy

0.087 (0.213) 1.131 (0.608) 0.296 (0.491) 1.393*** (0.502) 0.791 (0.500) 1.282*** (0.537) 0.157 (0.548)

GDPPC. dif

0.210 0.407 0.508 0.328 (0.154) (0.456) (0.456) (0.385) 0.554*** 0.825 1.706*** 0.209 (0.162) (0.438) (0.438) (0.377) 0.030 0.029 0.730** 0.502 (0.180) (0.299) (0.299) (0.472) 0.562*** 0.751**0.208 1.289*** (0.161) (0.347) (0.347) (0.429)

0.011 (0.012) 0.817 (0.531) 0.009 (0.067) 0.141** (0.055) 1.305*** (0.380) 0.126*** (0.032) 0.158*** (0.037)

log(Dj)

Final goods

Estimated results for equation (7.4)

0.146 0.022 (0.089) (0.012) 2 Other agricultural 0.211 0.064 products (0.209) (0.531) 3 Livestock 1.420*** 0.006 (0.211) (0.067) 4 Forestry 0.320 0.066 (0.221) (0.055) 5 Fishery 0.869*** 0.262 (0.219) (0.380) 6 Crude petroleum and 0.217 0.010 natural gas (0.215) (0.032) 7 Other mining 0.107 0.077** (0.222) (0.037)

Table 7.6

0.643

0.568

0.598

0.446

0.352

0.219

0.539

0.522

0.403

0.361

0.037

Adj.R2

0.066 (0.160) 0.221 (0.161) 0.097 (0.177) 0.246 (0.154)

0.665 (0.162) 0.123 (0.182) 1.354*** (0.226) 0.292 (0.216) 0.890*** (0.201) 0.398 (0.318) 0.558*** (0.195)

log(dist)

log(Dj)

0.808 0.410 (0.476) (0.476) 1.167*** 0.289 (0.429) (0.429) 0.025 0.988** (0.414) (0.414) 1.818*** 0.295 (0.518) (0.518)

0.065 0.037 (0.035) (0.035) 0.377 0.232 (0.619) (0.619) 0.849 2.145*** (0.700) (0.700) 0.264 0.578 (0.366) (0.366) 0.480 1.341*** (0.321) (0.321) 0.135 0.422*** (0.105) (0.105) 1.961*** 0.842** (0.391) (0.391)

log(Di)

0.335 (0.366) 0.036 (0.379) 0.657 (0.423) 1.144*** (0.356)

0.310 (0.388) 0.406 (0.430) 0.391 (0.515) 0.746 (0.493) 1.078** (0.466) 0.874 (0.723) 0.800* (0.450)

GDPPC. dif

Intermediate goods

0.609

0.643

0.547

0.429

0.479

0.634

0.616

0.704

0.406

0.431

0.261

Adj.R2

193

0.462** 0.593 0.190 (0.198) (0.332) (0.332) 13 Petroleum and petro 0.643** 0.047 0.003 products (0.270) (0.065) (0.065) 14 Rubber products 1.121*** 0.007 0.008 (0.249) (0.069) (0.069) 15 Non-metallic 0.358* 0.023 0.011 mineral products ((0.197) (0.045) (0.045) 16 Metal products 0.571*** 0.037 0.417*** (0.220) (0.054) (0.054) 17 Machinery 0.169 0.531 0.035 (0.166) (0.392) (0.392) 18 Transport equipment 0.316 1.279**0.512 (0.211) (0.615) (0.615) 19 Other manufacturing 0.303** 0.381* 0.394* (0.153) (0.219) (0.219) products

(0.251)

(0.251)

1.070*** 0.070 (0.400) (0.400) 0.610 2.235*** (0.395) (0.395) 1.216*** 0.041 (0.386) (0.386) 0.775* 1.442*** (0.436) (0.436) 1.728*** 1.009** (0.509) (0.509) 0.539* 0.473* (0.282) (0.282) 0.581** 0.240 (0.334) (0.334) 0.189 0.263

(0.343)

0.334 0.540 (0.351) 0.234 0.539 (0.483) 0.727 0.483 (0.467) 1.224*** 0.604 (0.397) 0.340 0.543 (0.509) (0.411) 0.775** 0.703 (0.373) 1.415*** 0.676 (0.515) 0.352 0.689

Calculations based on the IDE Asia International I–O Table.

(0.151)

0.054 (0.148) 0.427** (0.204) 0.048 (0.176) 0.409*** (0.149) 0.151 (0.157) 0.153 (0.157) 0.442** (0.185) 0.257*

Source:

0.635

0.611

0.655

0.501

0.450

0.370

0.445

0.445

*** Significant at 1 percent level, ** significant at 5 percent level, * significant at 10 percent level. Standard errors in parentheses.

(0.360)

0.508 (0.473) 0.788 (0.643) 0.725 (0.617) 0.420 (0.462) 0.263 (0.501) 1.202*** (0.436) 0.883 (0.660) 0.507

Notes:

12 Chemical products

194

11 Pulp, paper and printing

9 Textile, leather and products thereof 10 Timber and wooden products

8 Food, beverage and tobacco

6 Crude petroieum and natural gas 7 Other mining

5 Fishery

4 Forestry

3 Livestock

2 Other agricultural products

1 Paddy

0.012 (0.017) 0.881 (0.794) 0.003 (0.090) 0.075 (0.073) 1.044** (0.504) 0.116*** (0.044) 0.235*** (0.052)

0.146 (0.089) 0.211 (0.209) 1.420*** (0.211) 0.320 (0.221) 0.869*** (0.219) 0.217 (0.215) 0.107 (0.222) 0.210 0.102 (0.154) (0.634) 0.554*** 0.881 (0.162) (0.586) 0.030 0.759* (0.180) (0.425) 0.562*** 0.958* (0.161) (0.499)

1

0.328 (0.385) 0.209 (0.377) 0.502 (0.472) 1.289*** (0.429)

0.087 (0.213) 1.131* (0.608) 0.296 (0.491) 1.393*** (0.502) 0.791 (0.500) 1.282** (0.537) 0.157 (0.548)

0.643

0.568

0.598

0.446

0.352

0.219

0.539

0.522

0.403

0.361

0.037

GDPPC.dif Adj.R2

Final goods log(dist)

Table 7.7 Estimated results for equation (7.5)

0.066 (0.160) 0.221 (0.161) 0.097 (0.177) 0.246 (0.154)

0.665*** (0.162) 0.123 (0.182) 1.354*** (0.226) 0.292 (0.216) 0.890*** (0.201) 0.398 (0.318) 0.558*** (0.195)

log(dist) 0.310 (0.388) 0.406 (0.430) 0.391 (0.515) 0.746 (0.493) 1.078** (0.466) 0.874 (0.723) 0.800* (0.450)

GDPPC.dif

0.398 0.335 (0.632) (0.366) 0.878 0.036 (0.578) (0.379) 0.964* 0.657 (0.562) (0.423) 2.113*** 1.144*** (0.692) (0.356)

0.028 (0.048) 0.609 (0.837) 1.297 (0.926) 0.314 (0.484) 0.860** (0.429) 0.287** (0.139) 1.118** (0.521)

1

Intermediate goods

0.609

0.643

0.547

0.429

0.479

0.634

0.616

0.704

0.406

0.431

0.261

Adj.R2

195

0.462** 0.783* (0.198) (0.451) 13 Petroleum and petro products 0.643** 0.044 (0.270) (0.088) 14 Rubber products 1.121*** 0.015 (0.249) (0.095) 15 Non-metallic mineral products 0.358* 0.035 (0.197) (0.061) 16 Metal products 0.571*** 0.381*** (0.220) (0.072) 17 Machinery 0.169 0.566 (0.166) (0.558) 18 Transport Equipment 0.316 1.791* (0.211) (0.949) 19 Other manufacturing products 0.303** 0.014 (0.153) (0.294)

Calculations based on the IDE Asia International I–O Table.

0.054 1.000* 0.334 (0.148) (0.541) (0.351) 0.427** 1.624*** 0.234 (0.204) (0.534) (0.483) 0.048 1.257** 0.727 (0.176) (0.553) (0.467) 0.409*** 0.667 1.224*** (0.149) (0.627) (0.397) 0.151 0.719 0.340 (0.157) (0.724) (0.411) 0.153 0.067 0.775** (0.157) (0.382) (0.373) 0.442** 0.341 1.415*** (0.185) (0.490) (0.515) 0.257* 0.452 0.352 (0.151) (0.332) (0.343)

Source:

0.635

0.611

0.655

0.501

0.450

0.370

0.445

0.445

*** Significant at 1 percent level, **significant at 5 percent level, * significant at 10 percent level. Standard errors are in parentheses.

0.508 (0.473) 0.788 (0.643) 0.725 (0.617) 0.420 (0.462) 0.263 (0.501) 1.202*** (0.436) 0.883 (0.660) 0.507 (0.360)

Notes:

12 Chemical products

0.689

0.676

0.703

0.543

0.604

0.483

0.539

0.540

196

Table 7.8

Case studies of East Asian economic integration

F-Tests for equation (7.6) against equation (7.4) Final goods

1 Paddy 2 Other agricultural products 3 Livestock 4 Forestry 5 Fishery 6 Crude petroleum and natural gas 7 Other mining 8 Food, beverage and tobacco 9 Textile, leather and products thereof 10 Timber and wooden products 11 Pulp, paper and printing 12 Chemical products 13 Petroleum and petro products 14 Rubber products 15 Non-metallic mineral products 16 Metal products 17 machinery 18 Transport equipment 19 Other manufacturing products

Intermediate goods

F

p-value

F

p-value

1.254 2.361** 6.522*** 4.696*** 1.636 3.969*** 2.149** 7.115*** 4.896*** 5.534*** 6.429*** 3.870*** 5.396*** 6.077*** 9.543*** 4.5000*** 5.069*** 1.478 5.720***

0.268 0.018 0.000 0.000 0.115 0.000 0.032 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.165 0.000

2.121** 3.731*** 8.907*** 10.597*** 2.604*** 38.039*** 7.165 7.175*** 7.572*** 20.056*** 10.516*** 3.972*** 10.943*** 8.627*** 4.718*** 2.814*** 4.960*** 4.881*** 2.574**

0.034 0.000 0.000 0.000 0.009 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.005 0.000 0.000 0.010

Note: *** p 0.01, ** p 0.05, * p 0.10. Source: Calculations based on the IDE Asia International I–O Table.

wooden products (F, I), petroleum and petro products (I), and metal products (F) sectors. HME seem to be observed primarily in the sector that exhibits economies of scale, and reverse HME are observed mainly in primary goods sectors. 7.5.4 Testing the Signiﬁcance of Country Fixed Eﬀects for the Home Country Table 7.8 displays the results of the F-test comparing equation (7.6) against equation (7.4) for the dummy variable of the home country. If the F-test for the home country dummy is statistically signiﬁcant for a given sector, then it can be said that time-invariant characteristics of a home country aﬀect the volume of exports in that sector. Table 7.8 shows that the home country dummy is statistically signiﬁcant in almost all sectors regardless of whether the goods are ﬁnal or inter-

Structure and determinants of intra-regional trade

197

mediate. The home country dummy is not statistically signiﬁcant in the following three sectors: paddy (F), ﬁshery (F) and transport equipment (F). However, statistical signiﬁcance is observed in the crude oil and natural gas (I), timber and wooden products (I), petroleum and petro products (I) and forestry (I) sectors (F-values are all above 10). This result makes intuitive sense. In resource-based sectors, the export of intermediate goods, which are processed less than ﬁnal goods, is strongly aﬀected by the endowment of given natural resources which is captured as country ﬁxed eﬀects.

7.6 7.6.1

CONCLUSION Summarizing the Results

Estimations conducted in the previous sections reveal that the determinants of bilateral trade in East Asia diﬀer signiﬁcantly by sector. A summary of results may be found in Table 7.9. In the ‘Market’ column, ‘H’ refers to existence of HME, and ‘R’ refers to the existence of reverse HME. In the ‘GDPPC’ column, ‘’ indicates that developed countries tend to export those goods referred to while ‘’ indicates that less developed countries tend to export those goods. In the ‘Fixed’ column, ‘’ refers to the existence of country ﬁxed eﬀects for the home country, and ‘ ’ is shown if strong country ﬁxed eﬀects for the home country are observed. Three key results emerge from Table 7.9. First, country ﬁxed eﬀects for the home country were signiﬁcant for almost all sectors. In particular, country ﬁxed eﬀects were strong for the crude oil and natural gas (I), timber and wooden products (I), petroleum and petro products (I) and forestry (I) sectors. This means that the trade of those resource-based sectors is strongly inﬂuenced by the country characteristics that have been constant for at least 20 to 30 years. Second, the diﬀerence in GDP per capita plays a certain role in determining trade direction. Some goods are suitable for production and export by less developed countries (‘following geese’ as described in the ﬂyinggeese model). Last but not least, both the HME and the reverse HME were observed in some sectors. Some are quite easy to understand. For example, the transport equipment (F) and chemical product (F, I) sectors are industries that exhibit strong economies of scale, and HME may well play a signiﬁcant role in promoting exports. This suggests that the monopolistic competition model is suitable for explaining trade in those sectors. Reverse HME were

198

Table 7.9

Case studies of East Asian economic integration

Summary of estimated results Final goods

Intermediate goods

Market GDPPC Fixed Market GDPPC Fixed 1 Paddy 2 Other agricultural products 3 Livestock 4 Forestry 5 Fishery 6 Crude petroleum and natural gas 7 Other mining 8 Food, beverage and tobacco 9 Textle, leather and products thereof 10 Timber and wooden products 11 Pulp, paper and printing 12 Chemical products 13 Petroleum and petro products 14 Rubber products 15 Non-metallic mineral products 16 Metal products 17 Machinery 18 Transport equipments 19 Other manufacturing products

R R

R

R R

H

R H H

R

R

H H R

H

H

observed in some resource-based sectors such as crude oil and natural gas (F, I) and timber and wooden products (F, I). This may indicate that the reciprocal dumping model with restricted entry is suitable for explaining trade in those sectors. These results seem to be compatible with Feenstra et al. (2001). Overall, the determinants of trade diﬀer signiﬁcantly by sector, and multiple determinants appear jointly to aﬀect trade in each sector. Further, HME play an important role in some sectors. In Chapter 11, the HME in ASEAN countries are examined using a theoretically rigid framework.

Structure and determinants of intra-regional trade

7.6.2

199

Prospects of Intra-regional Trade in East Asia

What are the implications of these results for future structure of intraregional trade in East Asia? First, trade patterns for sectors in which country ﬁxed eﬀects play a signiﬁcant role will probably not change over the next few decades. For example, the trade of crude oil and natural gas is determined primarily by a country’s natural endowment. It is quite unlikely that this will change in the foreseeable future. Second, less developed countries (such as China or certain ASEAN countries) may attract production and exports in sectors where the diﬀerence in GDP per capita plays a certain role. The machinery sector (both ﬁnal and intermediate goods) is an example. Third, countries that have larger populations and higher per capita GDP (such as Japan and developed China) may attract production and exports in sectors where the HME play a certain role. Pulp, paper and printing (both ﬁnal and intermediate goods), chemical products (both ﬁnal and intermediate goods) and transport equipment (ﬁnal goods) are examples. 7.6.3

Policy Implications

What policy implications may be derived from this analysis for the coming era of de jure regional integration in East Asia? First, countries that have rich natural resources are naturally in a good position to develop related resource-based industries, especially upstream industries for intermediate input. Investments in those sectors will be justiﬁable because their favorable position is likely to be retained over the long term. Second, governments of less developed countries may implement industrial policies that assume the ﬂying-geese model of economic development, but they should be cautious in selecting their favorite industrial sectors. This analysis indicates that the machinery sector (rather than the apparel sector as conventional wisdom predicts) could be recommended for promotional activity in these countries. Production of parts for the automotive industry is also promising. Third, some industrial sectors in which economies of scale seem to play a signiﬁcant role are not suitable for all less developed small countries. Examples are auto assembly and plants for chemical products. However, further regional economic integration may change the situation for those countries that are located in an area having higher market potential such as those surrounded by large markets.

200

Case studies of East Asian economic integration

7.6.4

Remaining Problems

This chapter included description of the structure of intra-regional trade in East Asia and an explanation regarding some of its determinants. Signiﬁcantly, research reported in this chapter showed that determinants of trade are complex, diﬀer signiﬁcantly by sector, and cannot be explained by a single factor. Home market eﬀects also play a certain role in some sectors, and spatial economics is important for analysis of regional integration in East Asia. However, some qualiﬁcations regarding these ﬁndings should be noted. The testing procedure for HME and reverse HME adopted here was based on Feenstra et al. (2001), and other determinants of trade were added to the model in a somewhat ad hoc manner. Although it is not an easy task, in order to further conﬁrm the analysis it is necessary to develop a fully integrated general equilibrium model that includes many diﬀerent possible determinants of trade. Further, the analysis presented in this chapter covers only HME and reverse HME, not supplier access. With its absence of analysis on backward linkages, this chapter is incomplete as a contribution on spatial economics. This is because the size of the home market of a given sector and the size of the domestic supply of that sector are highly correlated in most sectors of most countries. This makes it impossible to use domestic supply as an explanatory variable and thus avoid multicollinearity. This problem should also be addressed in future research. The status of the USA is also controversial in ‘regional’ integration literature related to East Asia. As described in Section 7.2.5, East Asia has been heavily dependent on the demand for consumption goods in the USA. Thus a ‘regional’ analysis excluding the USA may well be incomplete. However, to include the USA in East Asia would no doubt aﬀect the ‘regionality’ of the analysis because the USA is an ‘outlier’ in multiple dimensions. This is not only an academic question but also a practical question for forming a suitable framework of ‘regional’ integration in the future.

NOTES 1. 2.

In this section, ‘East Asia’ refers to the following 16 countries and regions: Japan, China, Hong Kong, Macao, Korea, Taiwan, Indonesia, Malaysia, the Philippines, Thailand, Singapore, Brunei, Cambodia, Laos, Myanmar and Vietnam. Trade data by economic function are compiled from the Basic Economic Category (BEC) classiﬁcation. BEC is a derived classiﬁcation compiled from SITC (Rev. 3). It is intended to categorize traded goods according to their degree of processing and economic function. In this section, the BEC is aggregated as follows: ‘primary goods’ include BEC 111, 21 and 31, ‘processed goods’ include BEC 121, 22 and 32; ‘parts’

Structure and determinants of intra-regional trade

3. 4.

5.

6.

7.

8.

9.

10.

201

include BEC 121, 42 and 53; ‘capital goods’ include BEC 122, 41 and 521; ‘consumption goods’ include BEC 112, 122, 51, 522, 61, 62 and 63; and ‘nes’ includes BEC 7. This classiﬁcation is used in METI (2005). As a destination, ‘East Asia’ excludes Singapore and Hong Kong. These two locations work as ‘hubs’ for export to third countries, so they cannot be considered ﬁnal destinations. Ricardian theory of comparative advantage based on ‘technology’ or ‘productivity’ is somewhere between H–O and new trade theory in that it allows for policy intervention. The productivity of a sector in a country is determined by both ﬁrst and second natures. For example, higher productivity of the agricultural sector in a country may be determined by its climate (ﬁrst nature) or by economics of scale supported by a larger domestic market (second nature). Feenstra et al. (2001) used the classiﬁcation developed by Rauch (1999), who classiﬁed goods into three categories at the SITC four-digit level: (1) goods of organized exchanges; (2) goods that have referenced prices; and (3) diﬀerentiated goods. Oil and gold are examples of goods traded on organized exchanges such as the Chicago Mercantile Exchange. Examples of goods that have reference prices are some chemical products that are not traded on organized exchanges but are highly standardized. They have ‘reference prices’ in industry magazines and newspapers. ‘Diﬀerentiated goods’ are not traded on organized exchanges and have no reference prices. Feenstra et al. (2001) mentioned other combinations of diﬀerentiated/homogeneous goods and HME / reverse HME. If HME are observed in a sector that is considered to be producing homogeneous goods, the reciprocal dumping model with free entry is a plausible theory for explaining the trade. If reverse HME are observed in a sector that is considered to be producing diﬀerentiated goods, the Armington (perfect competition, national production diﬀerentiation) model is a plausible theory to explain the trade. The USA is excluded from the data used in the analysis in this chapter. In standard trade analysis, trade between East Asia and the USA is very important and inseparable. However, if the USA is included in the analysis of East Asian intra-regional trade, it is regarded as an ‘outlier’ in terms of market size and distance from other countries in East Asia. This speciﬁcation is known as a ‘ﬁxed eﬀects estimator’ in empirical literature. The ‘random eﬀects estimator’, which assumes that country-speciﬁc eﬀects are not correlated with any other explanatory variables, may be more suitable than the ﬁxed eﬀects estimator. To determine which was most suitable for this study, a Hausman test was conducted with the result that the ﬁxed eﬀects estimator was suitable in a majority of both ﬁnal and intermediate goods sectors. Statistically, it may be more desirable to use the more suitable estimator by sector, but the ﬁxed eﬀects estimator was used uniformly for simplicity. Andersen and Wincoop (2003) assert that it is necessary to control ‘multilateral resistance’ in order to estimate the gravity equation accurately. However, calculation of multilateral resistance is rather diﬃcult. Redding and Venables (2004) show that by including a full-set country dummy for both the home and partner countries, multilateral resistance can be controlled. Thus, if the relative multilateral resistance of each country is assumed to be constant over time, the inclusion of a full-set country dummy in this study eliminates the eﬀects of multilateral resistance. See Wooldridge (2003, p 141).

REFERENCES Anderson, James E. (1979), ‘A theoretical foundation for the gravity equation’, American Economic Review, 69 (1), 106–16. Anderson, James E. and Eric van Wincoop (2003), ‘Gravity with gravitas: a solution to the border puzzle’, American Economic Review, 93 (1), 170–92.

202

Case studies of East Asian economic integration

Bergstrand, Jeﬀery H. (1985), ‘The gravity equation in international trade: some microeconomic foundations and empirical evidence’, The Review of Economics and Statistics, 67 (3), 474–81. Brander, James (1995), ‘Strategic trade policy’, in G.M. Grossman and K. Rogoﬀ (eds), Handbook of International Economics, Amsterdam North Holland, pp. 1395–455. Brander, James and Paul Krugman (1983), ‘A “reciprocal dumping” model of international trade’, Journal of International Economics, 15, 313–21. Deardorﬀ, Alan V. (1995), ‘Determinants of bilateral trade: does gravity work in a neoclassical world?’, NBER Working Paper Series, no. 5377. Feenstra, Robert C., James R. Markusen and Andrew K. Rose (2001), ‘Using the gravity equation to diﬀerentiate among alternative theories of trade’, Canadian Journal of Economics, 34 (2), 430–47. Fujita, Paul Krugman and Anthony J. Venables (1999), The Spatial Economy: Cities, Regions, and International Trade, Cambridge, MA: MIT Press. Harrigan, James (2003), ‘Specialization and the volume of trade: do the data obey the laws?’, Harrigan in James and Kwan Choi (eds), Handbook of International Trade, Oxford: Basil Blackwell, pp. 85–118. Helpman, Elhanan (1987), ‘Imperfect competition and international trade: evidence from fourteen industrial countries’, Journal of The Japanese and International Economies, 1, 62–81. Kojima, Kiyoshi (2000), ‘The “ﬂying geese” model of Asian economic development: origin, theoretical extensions, and regional policy implications’, Journal of Asian Economies, 11, 375–401. Krugman, Paul (1979), ‘Increasing returns, monopolistic competition and international trade’, Journal of International Economics, 9, 469–79. Krugman, Paul (1980), ‘Scale economies, product diﬀerentiation, and the pattern of trade’, American Economic Review, 70, 950–59. Krugman, Paul (1986), Strategic Trade Policy and the New International Economics, Cambridge, MA: MIT Press. Leontief, W. (1953), ‘Domestic production and foreign trade: the American capital position re-Examined’, Proceedings of the American Philosophical Society, 97, 332–49. Ministry of Economy, Trade and Industry (METI) (2005), White Paper on International Economy and Trade. Ohlin, Bertil (1933), Interregional and International Trade, Boston, MA: Harvard University Press. Okita, Saburo (1985), ‘Special presentation: prospect of Paciﬁc economies’, the Fourth Paciﬁc Economic Cooperation Conference, 29 April 1 May, Seoul, Korea: Korea Development Institute, pp. 18–29. Rauch, J. (1999), ‘Networks versus markets in international trade’, Journal of International Economics, 48, 7–35. Redding, Stephan and Anthony J. Venables (2004), ‘Economic geography and international inequality’, Journal of International Economics, 62, 53–82. Ricardo, David (1817), The Principles of Political Economy and Taxation, London: The Electric Book Company. Samuelson, Paul (1953), ‘Prices of factors and goods in general equilibrium’, The Review of Economic Studies, 21 (1), 1–20. Wooldridge, Jeﬀrey M. (2003), Introductory Econometrics: A Modern Approach, Cincinnati, OH: South-Western.

8.

Location choices of Korean MNEs in East Asia: escaping the nutcracker Ho Yeon Kim

8.1

INTRODUCTION

The increasing role of FDI in global economic development is now widely recognized. FDI emanates from location decisions of MNEs that roam across national borders to maximize their proﬁts. Multinational production, investment and trade are becoming ever more important. As Markusen and Venables (1999) point out, the general view in the 1970s was that multinational investment was detrimental to the welfare of host economies, creating monopoly situations. In the 1990s, however, views became much more optimistic, suggesting that multinationals may stimulate development in host economies. Indeed, FDI represents the cutting edge of globalization; it is a unique bundle of capital and managerial/ technological knowledge. Multinational ﬁrms are constantly in action reshaping production and trade patterns around the globe. East Asia is no exception. Following the collapse of the Bretton Woods regime in the 1970s, as Mason (1999) notes, the value of the yen appreciated sharply and escalated production costs in Japan. A coincident increase in real wage rates encouraged the migration of labor-intensive manufacturing industries to Korea, Taiwan, Hong Kong, and later to China and the countries of the ASEAN. In the late 1990s, the consultant ﬁrm of Booz, Allen and Hamilton (1997) warned that Korea was now caught in a nutcracker, being crushed by China’s low costs and Japan’s technical excellence. It is thus interesting to examine how Korean ﬁrms responded to the daunting challenge. Against the above backdrop, this chapter opens with a brief survey of the related literature. The development of Korean FDI in East Asia is then traced in terms of amount, sector, location and motivation. Following a descriptive account of past and present operations of Korean MNEs, major factors of overseas expansion are identiﬁed, and a 203

204

Case studies of East Asian economic integration

distinction is made between the behaviors of small/medium enterprises and large conglomerates in their location choice. Following the spirit of the ﬁrst part of this book, focus is placed on a given ﬁrm’s motivation to maximize proﬁt. The rest of the chapter is organized as follows: in Section 8.2, major theories related to FDI are brieﬂy reviewed. Section 8.3 provides the main characteristics and peculiarities of Korean FDI with special emphasis on the great role of China as a low-cost site and a vast market. Section 8.4 includes discussion of the global operation of Korean conglomerates in the electronics, automobile and steel sectors. A monopolistic competition model is employed in Section 8.5 to provide an explanation of determinants for location of small and medium enterprises. Conclusions are presented in Section 8.6.

8.2

THEORIES ON FDI1

8.2.1

Product Life-cycle Theory

Vernon (1966) proposed the concept of the ‘product cycle’ to help explain the activities of US MNEs abroad. Initially, the product is produced for the home market in the home country. At a later stage of the product cycle, it is exported to other countries similar to the home country in demand patterns. As demand becomes more price-elastic, as labor becomes a more important ingredient of costs, and as foreign markets expand, the incentives for locating value-adding activities on a foreign soil increase. Eventually, the subsidiary can replace exports from the parent company. The overseas aﬃliates acquire the goods produced at home and soon begin to export back to the home country. The product life-cycle model of FDI combines Vernon’s original model with the internalization theory of the multinational ﬁrm in that foreign production cannot take place unless the ﬁrm holds some monopolistic advantage. Using a sample of 36 countries, Graham (1999) found a positive relationship between Japan’s FDI and both its imports and exports in the early 1990s. This suggests that FDI does not necessarily give rise to hollowing out or deindustrialization of the home country. 8.2.2

Flying-geese Theory

While Vernon’s product cycle theory takes the perspective of developed countries where new products are invented, the ﬂying-geese model takes the perspective of developing countries. Originally proposed by Akamatsu

Location choices of Korean MNEs

205

(1962), it illustrates how a new product is introduced to less developed countries via imports and how they acquire production techniques and become exporters. Inspired by early observations, Kojima (1985) argued that FDI should occur when a country’s comparative advantage in some product is eroded. FDI can deliver technology, management skills and movable capital to foreign sites where total production costs are lowest. Instead of replacing exports, FDI can then generate new exports to third countries or even to the home country. Dowling and Cheang (2000) observed a ﬂying-geese type development in East Asia during the period of 1970–95, and their ﬁndings were supported by Kim and Jin (2003) with more recent data. Asian countries are usually considered to have been involved in a multi-layered catching-up process of industrialization. Japan has supplied capital and technology through the expansion of trade and FDI. In return, the NIEs have exported commodities to Japan in which it no longer has comparative advantage. The same process has been repeated with respect to ASEAN countries. Thus FDI is an eﬀective channel for developed countries to transfer their comparative advantage to less developed countries. 8.2.3

Eclectic Theory

Dunning (1993, p. 90) argues that Kojima’s theory fails to explain the kind of trade ﬂows that are based less on the distribution of factor endowments and more on the need to exploit economies of scale, product diﬀerentiation, and other manifestations of market failure. Dunning’s (1977) eclectic theory is a consolidation of various FDI literature. In short, an ‘ownership’ advantage allows a ﬁrm to overcome the disadvantages of foreign location. This may be a product or a manufacturing process to which other ﬁrms have no access (such as a patent or a trademark). ‘Location’ advantages, such as low input costs or protective trade policy, make it more proﬁtable to produce in a country than to export to it. Finally, ‘internalization’ advantages make it more desirable for a ﬁrm to undertake foreign production by itself rather than dealing with a foreign partner. Dunning (1993, p. 80) observes that at any given moment, the more a country’s enterprises possess ownership advantages, the greater the incentive they have to internalize their use. They are then more likely to ﬁnd it in their interest to exploit them from a foreign location. Markusen (1998) also notes that MNEs are found in industries in which knowledge capital is important. Knowledge capital includes human capital of employees, patents, proprietary knowledge on procedures, and marketing assets such as trademarks and brand names. It often has a ‘public-good’ property within the ﬁrm. Blueprints, chemical formulae or reputation

206

Case studies of East Asian economic integration

capital are very costly to produce, but they can be supplied at a relatively low cost to foreign production facilities without reducing the value or productivity of those assets in existing facilities. Internalization advantages arise from the same property of knowledge; knowledge easily transferred to foreign locations is also easily dissipated. Conversely, Markusen (1991) argues, a developing country may possess many of the ingredients necessary to produce a particular class of goods, but it may lack key technical and managerial skills. The entry of the multinational ﬁrm into such a country provides missing resources and leads to the production and export of goods that were previously imported. 8.2.4

New Trade Theory

The ideas of Dunning (1977) and Markusen (1998) are formalized with the approach used in new trade theory. The main characteristics of this theory are summarized in Navaretti and Venables (2004). As they become multinational, ﬁrms face trade-oﬀs between costs and beneﬁts. Costs are foregone economies of scale and economies of integration. Beneﬁts are better market access and cheaper production factors. With the scale economy, larger output of a ﬁrm or plant leads to more proﬁtability. The magnitude of scale economies also depends on the size of ﬁxed costs. As ﬁxed costs at the plant attached to the headquarters become smaller, the beneﬁt of producing a large amount fades. Thus there is a greater probability of becoming multinational. When production activities are integrated, production eﬃciency increases. Providing counterexamples, Navaretti and Venables (2004) cite the case of reheating steel products or delayed transportation between plants. Further, the lack of adequate communication between plants and headquarters may reduce the productivity of plants when production activities are physically separated, although this kind of ‘diseconomy’ can be somewhat mitigated by the progress of information technologies. The producer operates plants in another country to save transport costs and avoid trade barriers. Better market access from the home country may then actually slow down the evolution of multinational ﬁrms. Markusen (1984) argues that multinational activity arises with moderate to high trade costs. Production factors also play an important role in dictating the location of multinational ﬁrms. The costs of inputs, such as wages, tend to diﬀer across countries. Using cheaper inputs, ﬁrms can supply more aﬀordable products. When transport costs are low, diﬀerences in input costs emerge as the decisive factor for attracting multinational ﬁrms.

Location choices of Korean MNEs

8.2.5

207

Spatial Economics

During the 1990s, renewed interest in the spatial dimension of economic activities gave rise to the so-called ‘spatial economics’ or ‘new economic geography’. Dunning (1993, p. 77) indicates that the failure of the traditional factor endowment approach to explain international production arises because it assumes the existence of perfect markets. As Ensign (1995) observes, location theory has emphasized that production generally moves from decentralization to agglomeration as market imperfections arise. Beneﬁts of agglomeration include economies of scale in production and localization economies due to shared inputs. By combining these concepts with transportation costs, Krugman (1991) constructed what is now known as the core–periphery model. Brülhart (1998) succinctly contrasts competing theories on economics of space as follows: neoclassical theory is characterized by perfect competition, homogeneous products, and non-increasing return to scale. Location is determined exogenously by spatial distributions of natural endowments, technologies and other factors. New trade theory dispenses with all exogenous elements except for market size, and this is determined by the size of the immobile labor force. Inter-industry specialization leads to sectors clustering in locations that oﬀer best access to markets, while intra-industry specialization leads to each ﬁrm producing a unique variety. Spatial economics, on the other hand, treats location endogenously. Both production factors and ﬁrms are mobile, and even market size is explained within the model. Market-size externalities and input–output linkages produce selfreinforcing agglomeration processes. As pointed out by Krugman and Hanson (1993), this spatial economic approach can, for example, explain the eﬀects of trade liberalization better than international trade theory, because ‘the issue is not how countries will specialize, but where production will locate’. According to Baldwin et al. (2003), the spatial conﬁguration of economic activities is shaped by two diﬀerent eﬀects that lead to concentration and one eﬀect leading to dispersion. First, concentration can result from market access eﬀects. An increase in demand makes the market larger. Suppliers locating in a large market can sell more, since goods that are not transported between regions are cheaper. Obviously, this eﬀect becomes weak when transport costs are low. More importantly, under increasing-returnsto-scale production technology, the increase in the number of suppliers in a larger market is more than proportional to the expansion of the home market. As a result, excess goods over local demand are exported. The second force causing a concentration is cost-of-living eﬀects. The price index of goods becomes lower in a region where many suppliers

208

Case studies of East Asian economic integration

gather. Prices of a large share of goods do not include transport costs because they are produced locally. The lower price of goods creates more demand on the region. This eﬀect is stronger when transport costs are high and the factory price is low. Market access and cost-of-living eﬀects reinforce each other. Because the former attracts demand and the latter supply, these two eﬀects trigger a circular causality in which economic activities agglomerate in a region. On the other hand, market-crowding eﬀects can help disperse economic activities. Due to the decrease in the general price index following concentration, the price charged by a speciﬁc ﬁrm becomes relatively high and results in lower demand for its products. This eﬀect naturally becomes weaker as transport costs among regions decrease. Summing these three eﬀects, Amiti (2005) shows that low trade costs can lead to agglomeration of both upstream and downstream ﬁrms in one country due to the beneﬁts of close proximity, even when they diﬀer in factor intensities. This is because the weakening of the crowding eﬀects is greater than the weakening of the market-access and cost-of-living eﬀects. Robert-Nicoud (2006) ﬁnds the same results with multinational ﬁrms. Furthermore, adding the dispersion eﬀects stemming from decreasing returns to labor in agriculture, the author shows that industrial agglomeration emerges when transport costs take intermediate values. Markusen and Venables (1999) further ﬁnd that competition from FDI tends to reduce proﬁts of local ﬁrms, while linkage eﬀects to suppliers reduce input costs. As Ottaviano and Puga (1998) aptly state, the expansion of manufacturing accentuates wage diﬀerences between industrialized and non-industrialized countries. At some point this gap becomes unsustainable, and industry starts to spill over to low-wage economies. Navaretti and Venables (2004) raise a possibility that when there are no local ﬁrms, multinational ﬁrms act as the catalyst for industrial agglomeration. Following the entry of multinational ﬁrms, there can be: (1) a large number of only domestic ﬁrms engaged in production, or (2) a large number of domestic ﬁrms in upstream industry with a large number of multinational ﬁrms downstream. Relocating ﬁrms begin to beneﬁt from the forward and backward linkages, and a critical mass is reached by some countries which then undergo rapid industrialization. Further growth causes the process to repeat itself. Industry thus spills over in a series of waves from one country to another.

Location choices of Korean MNEs

8.3 8.3.1

209

EMERGENCE AND EXPANSION OF KOREAN MNES Overview

Dunning and Narula (1996) stress that FDI is primarily a microeconomic or ﬁrm-speciﬁc activity; economic development is a macroeconomic or country-speciﬁc phenomenon. The outward and inward direct investment position of a country, relative to the rest of the world, is systematically related to its economic development. This implies that countries go through stages of development, and that these stages can be classiﬁed according to the propensity of those countries to be outward and/or inward direct investors. Within this framework, Urata (1999) describes the process in which rapid and steep appreciation of the yen in the second half of the 1980s led to a surge of Japanese FDI. Along with the price eﬀect which made Japanese products more expensive, the wealth eﬀect made Japanese ﬁrms richer in terms of increased collateral and liquidity, and enabled them to ﬁnance FDI more cheaply than their foreign competitors. Korea stands out as a unique case of a country that quickly transformed itself into a major economic power that combined an overarching government and strong entrepreneurship of business people. Hanink (1994, p. 344) looks at the end of the Korean War in 1953 when Korea was devastated with no capital stock. After a brief and unsuccessful attempt at an import substitution policy, the country began to adopt an outward-oriented growth policy. The government encouraged the formation of a small number of large business enterprises called chaebol. In addition, the government kept wages low by suppressing labor unions. These came in tandem with exchange rate adjustments and the removal of tariﬀs on inputs to be used by exporters. Thanks to all these eﬀorts, real per capita income began to take oﬀ. The rest of the phenomenal and compressed development is a well-known story. Korea is now a major producer of automobiles, large vessels, semiconductors and consumer electronics. Korean FDI is a fairly recent phenomenon. It can be traced back to 1968 when public investment was made to secure Indonesian forest products. Private investments overseas were strictly regulated until the late 1980s when increases in trade surplus caused rapid appreciation of the currency. Following the massive democratization movement of 1987, ﬁrms soon found themselves struggling to survive amid sharp increases in labor costs as well as dwindling productivity and proﬁts. During the mid-1990s, the business environment was not very favorable for Korean ﬁrms in general, and small and medium ones in particular. They were facing challenges on all fronts with rising labor and capital costs, high rent and logistics costs,

210

Case studies of East Asian economic integration

and a myriad of government regulations. Figure 8.1 shows changes in manufacturing wages of major competitors in East Asia. During the period 1985–94, wages in Korea increased more than threefold. At the same time, wages in China, Indonesia, Thailand and the Philippines showed few or no changes at a very low level. To make matters worse, delays in the provision of an adequate infrastructure caused severe bottlenecks in logistics management.2 In an eﬀort to overcome growing regionalism and trade barriers, the government gradually lifted regulations on investing abroad, and FDI was virtually liberalized by 1997. Korea acceded to the OECD in 1996 and allowed free capital movements. These added fuel to the ﬁre. Often without the government’s knowledge, ﬁrms started borrowing excessively from the international capital market. Due to mismanagement of foreign currencies and exchange rates, an economic crisis erupted. As depicted in Figure 8.2, Korea’s FDI suﬀered a temporary setback in 1998, but it then soon rebounded.3 The ﬁgure shows that the case count of FDI has been skyrocketing since 2000. By 2005, the outward investment ﬂow reached US$6 billion, with over 4400 cases. Since the size of the total investment was relatively small, a few large projects had a big impact on the general trend. This is reﬂected in discrepancies in the amount invested and the number of cases registered. For example, a real-estate ﬁrm invested US$1.4 billion in Central America through Bermuda in 2000, and LG Electronics invested US$1.5 billion in the Netherlands in 2001. Table 8.1 shows that for its size and compared to China, Korea is quite active in outward investment. Figures for 2005 indicate that China hosted ten times as much investment as Korea but dispatched only three times the amount of Korean FDI abroad. In the ﬁrst half of 2006, outward investment of Korea amounted to US$7.1 billion, and this surpassed the inward ﬂows of foreign fund for the ﬁrst time. 8.3.2

Characteristics of Korean FDI

Lall (1996) oﬀers an interesting assessment of the economic achievements made by Japan, China and Korea. First, Japan developed its ownership advantage by a deliberate strategy of: (1) keeping foreign ﬁrms out; (2) importing technology by licensing, copying and reverse engineering; and (3) developing its own research. This helped protect national enterprises and build up indigenous technology. Japan went international relatively late, and this was when its enterprises started to invest abroad to seek lowcost locations. Over time, the pattern shifted to technologically driven FDI; activities in which Japan was losing competitiveness were relocated overseas or phased out.

US$ 30 000

25 000 Singapore 20 000

Korea Taiwan Hong Kong

15 000

Malaysia

211

Philippines Thailand

10 000

Indonesia China

Source: ILO, Yearbook of Labour Statistics.

Figure 8.1

Changes in manufacturing wages

1994

1993

1992

1991

1990

1989

1988

1987

1986

0

1985

5 000

US$ mil. 7000

5000 4500

6000 4000 5000

3500 3000

4000

Cases 2500

212

3000

2000 1500

2000

1000 1000

Source:

Korea Export/Import Bank.

Figure 8.2

Overall trend for Korean FDI

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

0

1988

500 0

Amount

213

Location choices of Korean MNEs

Table 8.1

Korea China East Asia Asia and Oceania World

Source:

Foreign direct investment flows (US$ million)

Inward Outward Inward Outward Inward Outward Inward Outward Inward Outward

1990–2000 (Annual average)

2002

2003

2004

2005

3 004 3 101 30 104 2 195 48 777 29 473 76 616 37 352 495 391 492 566

3 043 2 617 52 743 2 518 67 350 27 556 96 244 34 726 617 732 539 540

3 892 3 426 53 505 152 72 174 14 441 110 489 18 995 557 869 561 104

7 727 4 658 60 630 1 805 105 074 59 211 157 328 83 446 710 755 813 068

7 198 4 312 72 406 11 306 118 192 54 189 199 951 83 584 916 277 778 725

UNCTAD World Investment Report (2006).

China’s initial isolation was robust, but the country has opened rapidly. The pattern of inward FDI has been shifting from simple labor-intensive operations to heavy industry aimed at the domestic market and to more complex export-oriented activities. Local ﬁrms are now more concerned about learning from MNEs and gaining access to their proprietary technologies and markets than being content to be passive recipients of capital and know-how. On the other hand, Lall (1996) notes that Korea promoted its hand-picked local ﬁrms to become giant conglomerates and forced them into high-technology industries with minimal reliance on FDI. In the process, it developed the deepest and broadest base of technological capabilities anywhere in the developing world and has had the largest investment in R&D. Figures 8.3 and 8.4 reveal the sector-based distribution of outward investment from Korea in terms of, respectively, the amounts and cases recorded. Except for the period of economic crisis, FDI has been generally increasing. If the previously mentioned cases of unusually large projects are removed, the trend for amounts invested resembles that of the number of cases. Manufacturing clearly dominates the scene, but the sales (trading oﬃces) sector has traditionally also been quite strong. Next, Figures 8.5 and 8.6 show the regional distribution of, respectively, the amounts and cases of outbound FDI. In amount, North America closely followed Asia until 2001 but has remained a distant second since then. It is noteworthy that per project investments into Europe were conducted on a very large scale. This was due to the opening of electronics complexes by big

US$ mil. 4500 4000 3500 3000 Manufacturing Sales Finance Other

2500 2000

214 1500 1000 500

Source:

Korea Export/Import Bank.

Figure 8.3

Korean FDI by sector (amounts invested)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

0

2500

2000

1500 Manufacturing Sales Other

215

1000

500

Source:

Korea Export/Import Bank.

Figure 8.4

Korean FDI by sector (number of cases)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

0

US$ mil. 4500 4000 3500 3000 Asia 2500

N. America Europe

2000

216

Other

1500 1000

Source:

Korea Export/Import Bank.

Figure 8.5

Korean FDI by region (amounts invested)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

0

1988

500

3500

3000

2500

Asia

2000

N. America Europe

1500

217

Other 1000

500

Source:

Korea Export/Import Bank.

Figure 8.6

Korean FDI by region (number of cases)

2005

2004

2003

2001

2002

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

0

218

Case studies of East Asian economic integration

Table 8.2

Ranking of major destinations for Korean FDI

Total

Ranking

Amount (US$ mil.)

Country

15 024 13 992 2 285 2 275 2 262 1 911 1 598 1 473 1 193 1 164 1 161 920 831 762 756 754 742 726 522 455 444 443 404 396 320 314 309 269 246 239

USA China Netherlands Hong Kong Indonesia UK Vietnam Bermuda Islands Japan Singapore Germany India Australia Canada Philippines Poland Thailand Yemen Malaysia France Brazil Peru Uzbekistan Mexico Slovakia Italy Cayman Islands Russia Romania Ukraine

Manufacturing Country

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

China USA Netherlands Indonesia Vietnam India Poland Hong Kong UK Philippines Thailand Malaysia Canada Uzbekistan Brazil Slovakia Romania Germany Japan Mexico France Singapore Sri Lanka Bangladesh Ukraine Hungary Turkey Panama Virgin Islands Portugal

Amount (US$ mil.) 11 923 6 527 1 732 1 532 1 080 786 636 604 589 555 436 412 399 377 361 308 237 220 215 213 210 206 157 152 150 137 131 128 112 97

Note: Amounts represent cumulative investment during the period 1968–2005. As such, intertemporal changes in the value of investment are ignored. Source:

Korea Export/Import Bank.

conglomerates. Table 8.2 shows rankings of the major countries of destination for Korean FDI. It reveals, among other interesting facts, that Hong Kong, Singapore and Japan have mainly attracted non-manufacturing investment. If focus is placed on manufacturing only, China and ASEAN countries quickly surface.

Location choices of Korean MNEs

219

Figure 8.7 shows Korean FDI in East Asian countries only.4 Since the early 1990s, major destinations for Korean investment have shifted from Indonesia to China. In particular, China is absorbing twice the amount of what ﬂows to all other countries in the region combined. Figure 8.8 includes a map of the cumulative number of cases for manufacturing investment in each country since 1968. As the traditional stronghold for Korean FDI, Indonesia and the Philippines each managed to secure more than 500 projects. They are now surpassed by a new competitor, Vietnam, and yet the three countries are simply dwarfed by China. With more than 11 000 cases, China claims a lion’s share and begs further scrutiny as to just what the country had to oﬀer foreign investors in general and Korean ﬁrms in particular. 8.3.3

Chinese Connection

Often dubbed ‘the factory of the world’, China is now widely regarded as ‘the future market of the world’. It is truly remarkable that the country has sustained an average growth rate of 9.6 percent for the last 25 years. Using its vast market potential as leverage, China has induced foreign capital and technology. Zhang and Van der Bulcke (1996) observe that it gradually opened to foreign investors under a government-controlled investment scenario, that is, from limited geographic locations to countrywide access, from production-oriented to market-oriented operations, and from natural resources to more human- and capital-intensive assets. The shift of resource allocation from direct command by authorities to market transactions has resulted in the creation of a more eﬃcient market system as well as a more competitive business climate. Korea established diplomatic relationships with China in 1992 despite an outcry from its long-time ally Taiwan. The trade amount between the two countries was only US$6.4 billion in that year, but it skyrocketed to over US$90 billion by 2005. China is now both the largest trading partner for Korea and, since 2002, the biggest recipient of Korean FDI. For China, Korea stands as the third-largest trading partner and the second-largest investor, following only Japan.5 The general perception among Korean entrepreneurs has been that China oﬀers wages eight to ten times lower than those in Korea, and that capital and land rent there are comparably cheaper as well. Korean investment in China has taken place primarily in manufacturing activities. Figure 8.9 provides a sector-based breakdown of Korean FDI in manufacturing. The initial investment in the ﬁrst half of the 1990s was mainly in such labor-intensive sectors as textile, leather and shoemaking; the second wave of investment has been characterized by a larger share of

US$ mil. 4500 4000 3500 3000 Other Indonesia

2500

Hong Kong

220

2000

China

1500 1000 500

Source:

Korea Export/Import Bank.

Figure 8.7

Korean FDI in East Asian countries (amounts invested)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

0

RUSSIA (70)

KAZAKHSTAN (19)

UZBEKISTAN (37)

MONGOLIA (22)

KYRGYZSTAN (4)

TAJIKISTAN (2) CHINA (11,148) (11 148)

JAPAN (215)

IRAN (5) PAKISTAN (9)

NEPAL (4)

BANGLADESH (117)

INDIA (132) TAIWAN (50) HONG KONG (285)

LAOS (5) MYANMAR (33)

221

THAILAND (238)

VIETNAM (632)

PHILIPPINES (524)

KAMPUCHEA CAMBODIA (39) (39)

SRI LANKA (117) MALAYSIA (207) SINGAPORE (41)

PAPUA NEW GUINEA (4)

INDONESIA (568)

AUSTRALIA (73)

Source:

Korea Export/Import Bank.

Figure 8.8

Cases of Korean manufacturing FDI (1968–2005)

US$ mil. 2500

2000 Other Raw metal Machinery

1500

Oil & chemical Textile & clothes

222

Transportation

1000

Electronics & telecomm.

Source:

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1988

0

1989

500

Korea Export/Import Bank.

Figure 8.9

Korean FDI in China by manufacturing subsector (amounts invested)

223

Location choices of Korean MNEs

Table 8.3 Recipient

Discrepancies in FDI statistics (US$100 million) Year

Chinese record

Korean record

Cases

Amount (A)

Cases

Amount (B) 6.9 3.5 6.7 5.9 9.6 14.8

China

1998 1999 2000 2001 2002 2003

1309 1547 2565 2909 4008 4920

18.0 12.8 14.9 21.5 27.3 44.9

261 458 770 1034 1353 1662

Shandong province

1998 1999 2000 2001 2002 2003

342 593 1 012 1 251 1 792 2 431

5.98 5.28 5.67 8.84 15.6 28.4

86 184 307 421 524 631

Sources:

1.83 0.63 2.66 1.87 2.26 3.92

A/B

2.6 3.6 2.2 3.7 2.8 3.0 3.3 8.4 2.1 4.7 6.9 7.2

China Statistical Yearbook; Korea Export/Import Bank.

IT and transportation industries.6 Through a survey of 548 Korean manufacturing ﬁrms operating in China, Ha (2004) found that 43 percent were motivated by the cost saving potential, and 33 percent were targeting the domestic market. The Chinese government oﬀers large tax breaks for reinvestment of proﬁts, and this suggests that actual investments made by Korean ﬁrms may far surpass oﬃcial outbound FDI ﬁgures maintained by the Korean government. According to Yang (2004), following the economic crisis, the steep growth of Korean FDI into China has indeed taken the form of reinvestment of proﬁts and expansion of existing facilities. Large discrepancies between Chinese and Korean statistics shown in Table 8.3 conﬁrm the magnitude of the reinvestment being made. Reinvestment for the entire country is approximately twice the size of new investment. For Shandong province, the amount being reinvested by Korean ﬁrms is approximately six times larger than the amount freshly injected. Widening regional disparity has come with fast economic growth. China is suﬀering from a huge inter-regional income diﬀerential between lucrative coastal areas and poor inland provinces. Shanghai’s per capita GDP is 13 times higher than that of Guizhou. The income disparity between the eastern seaboard and the interior has been increasing rapidly, and industrial production has displayed strong agglomeration toward the coastal area. Biased regional policies have been partly responsible, but Hu and

224

Case studies of East Asian economic integration

Fujita (2001) assert that globalization and economic liberalization have had greater inﬂuence on the increasing regional disparity. Zhang and Van der Bulcke (1996) have found that the early opened 12 coastal provinces received more than 90 percent of the entire inward FDI throughout the 1980s, and that FDI of any particular region was positively related to its gross regional product. Figure 8.10 shows the geographical distribution of Korean FDI in China. Perhaps due to close proximity, Shandong has been the major recipient of the funds. However, the post economic crisis years have seen Jiangsu and Beijing rise as new favored destinations for Korean investors. Figure 8.11 contrasts Korean FDI against the total investment each province has received. Koreans appear to have a pronounced bias toward these three regions and relatively less attraction to Hunan, Guangdong and Fujian. This kind of tendency for location of manufacturing is also apparent from Figure 8.12; which displays an overwhelming preference for Shandong Province.7 What makes China so attractive? Firms may have diﬀerent motives and objectives. Jee et al. (2004) conducted interviews with 231 Korean ﬁrms doing business in Beijing, Tianjin and Shandong provinces. When asked about their original motives for investing in China, 26 percent mentioned market penetration, and 36 percent pointed to the low-cost labor. In contrast, for large companies, market factors received 45 percent of responses, and labor considerations accounted for 27 percent. The results for production pattern are quite illuminating. On the whole, ﬁrms procure 50 percent of intermediate inputs within China; 42 percent is imported from Korea. Of products, 38 percent are sold in China, 27 percent are exported to Korea, and the remainder goes to other countries. This strongly suggests that Korean ﬁrms are fully using China as an input source, a processing base and a market. Interestingly, information regarding China was acquired through existing ﬁrms in China (42 percent), acquaintances (19 percent), Chinese authorities (17 percent), and only 7 percent from various support organizations in Korea. Figures 8.13 and 8.14 show who makes investments in China. Except for the period in which business was particularly bad due to economic turmoil, small and medium enterprises with fewer than 300 employees (SME) appear on the whole to have been investing about the same amount as that of large ﬁrms. Although the amount registered pales in comparison, investments made by private companies and individual investors have been quite strong. They now exceed SMEs in terms of cases recorded. From these ﬁgures, the average size of investment per ﬁrm size can be calculated and is reported in Table 8.4. It is not surprising that large ﬁrms are associated with projects of much bigger scale. The unusual jump in 1998 may reﬂect a pentup supply of funds which were depressed during the economic crisis.

US$ mil. 3000

2500

2000 Other Beijing Jiangsu Shandong

1500

225 1000

Source:

Korea Export/Import Bank.

Figure 8.10

Korean FDI in China by province (amounts invested)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

0

1988

500

500

12 000

450 10 000

400

Total FDI

300 6 000

250 Total FDI received Korean FDI

4 000

200

226

150 100

2 000

50 0 Shandong Jiangsu Beijing Tianjin Liaoning Shanghai Zhejiang Guangdong Jilin Hunan Heilongjiang Hebei Guangxi Fujian Sichuan Anhui Hubei Hainan Henan Jiangxi Shaanxi Shanxi Guizhou Yunnan Gansu Qinghai

0

Sources:

China Statistical Yearbook; Korea Export/Import Bank.

Figure 8.11

Korean versus total FDI in China by province (amounts invested in 2003, US$1000)

Korean FDI

350

8 000

227

Location choices of Korean MNEs

Heilongjiang (273)

Jilin (644)

Nei Mongol

Liaoning (1569) Hebei (266)

Xinjiang

Beijing (486) Tianjin (1065) KOREA Shanxi (15)

Ningxia

Shandong (4318)

Qinghai (1) Gansu (6) Shaanxi (8)

Jiangsu (963)

Henan (31)

Xizang Anhui (32)

Shanghai (477)

Hubei (23) Sichuan (32) Jiangxi (18) Hunan (24) Guizhou (5)

Yunnan (5)

Guangxi (29)

Zhejiang (397)

Fujian (84)

Guangdong (364)

Hainan (13)

Source:

Korea Export/Import Bank.

Figure 8.12

8.4 8.4.1

Cases of Korean manufacturing FDI in China (1968–2005)

LOCATION CHOICE OF KOREAN CONGLOMERATES IN EAST ASIA Overview

In 2004, ten Korean ﬁrms were listed in the 100 largest foreign companies in China. Four of them were aﬃliated with the LG group, and three were subsidiaries of Samsung. According to Jones and Sakong (1980, pp. 304–5), the Korean chaebol are similar to the early Meiji zaibatsu of Japan. Control is centralized in a single dynamic individual who founds one enterprise, gets it running, passes management to a relative or associate, and then moves on to something new. It is true that underpriced resources, notably credit, were channeled into a small number of chaebol throughout

1200

1000

800 Large SME

600

Private

228 400

200

Source:

Korea Export/Import Bank.

Figure 8.13

Korean FDI in Chinese manufacturing by firm size (amounts invested)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

0

1200

1000

800 Large 600

SME Private

229

400

200

Source:

Korea Export/Import Bank.

Figure 8.14

Korean FDI in Chinese manufacturing by firm size (number of cases)

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

0

230

Case studies of East Asian economic integration

Table 8.4 Average amount of Korean manufacturing investment in China by firm size (US$1000) Year

Large

SME

Private

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

1 995 4 494 5 198 6 787 9 039 51 348 20 942 23 485 16 021 12 481 14 073 16 341 20 799

589 422 601 717 415 532 322 348 428 512 720 1 169 1 480

215 187 210 142 139 101 106 120 144 153 175 189 203

the development era. However, Jones and Sakong argue that this is not necessarily due to political favoritism but may simply reﬂect the fact that they have a proven track record that shows them to be capable of using the resources most eﬃciently. Early development and current operations of four large Korean MNEs are reviewed below. 8.4.2

LG Electronics

Lucky Chemical Inc. was established in 1947, and Gold Star Electronics started making radios in 1958 as a subsidiary of Lucky. The two companies merged in 1995 to become the LG group. Their electronics division has ten plants in Korea and employs 30 000 workers. Overseas plants employ another 38 000 at 32 facilities. LG Electronics started investing in China in 1993. The original intention was to create a production base near home for export using low-cost labor rather than making a pre-emptive move to take a huge domestic market. LG now has 16 plants and employs 18 000 workers in China. All of the plants are located in the eastern part of the country. They produce a full range of consumer electronics, and the company plans to expand its markets from coastal areas into inland provinces. From 1997, LG also began production in India, where it manufactures household appliances and mobile phones. An extensive localization policy is being pursued in order to stay competitive.

Location choices of Korean MNEs

8.4.3

231

Samsung Electronics

Samsung Trading Co. was established in 1948 and grew substantially during the Korean War. Its initial growth was in import substitution of consumer goods such as sugar, ﬂour and textiles. When followers began to enter one market, Samsung moved on to something new, one step ahead of the crowd. It then went abroad to seek more business opportunities. The group moved into the electronics business in the early 1970s, and Samsung Electronics is now a world-class manufacturer of displays, memory chips and mobile phones. Its eﬀorts for global operation show in the high share of parts procured from abroad.8 In addition to nine plants at home, Samsung maintains 29 manufacturing facilities in 14 countries. As an example of its reputation, Samsung is known as the most successful foreign manufacturer in Malaysia. It comprises a large complex of assembly plants and 40 suppliers employing 8000 workers. There are 12 Samsung plants along the coastal area of China. According to Yang (2004), generic parts are supplied by local ﬁrms in China. More sophisticated parts come from the global sourcing center in Hong Kong. Samsung has ﬁve regional distribution centers in China. Final products are sold in China or exported worldwide. Samsung invests heavily in research and development, and R&D centers try to localize their products to better serve the Chinese market. Actually, Samsung operates ten R&D centers in eight diﬀerent countries, and in 2005, it ranked ﬁfth in the number of patents registered in the USA. 8.4.4

Hyundai Automotive

Hyundai’s initial growth was also seen during the Korean War, ﬁrst in construction of army barracks and airports, and later in the housing construction boom and international construction markets in Southeast Asia and the Middle East. While Samsung has concentrated on consumer goods, Hyundai has focused on heavy producer goods such as construction, shipbuilding, machinery, and one consumer durable, automobiles. Hyundai has been the leader in a nationwide eﬀort to move the economy into a stage of industrialization. The uncertainty of success seems to have led Hyundai to take more risks than Samsung and to be more innovative. Hyundai began assembling cars under license in the 1960s. A Canadian assembly plant erected in 1989 represents its early eﬀort in global expansion. Hyundai acquired KIA Motors in 1999 to become the third-largest conglomerate in Korea. Its Ulsan facility is one of the world’s largest auto complexes and is responsible for 1.6 million cars per year. With two more plants in Korea, the Hyundai–KIA group annually exports over 2.8 million vehicles

232

Case studies of East Asian economic integration

worldwide. The company’s primary concern is the labor union, which is notorious for its militancy, annual strikes and demand for pay hikes well above productivity increases. Recent plant openings abroad are making workers more nervous. Assembly plants opened in Turkey in 1997 and in India in 1998. The US Alabama plant was completed in 2005, and another plant is under construction in Georgia with the prospect of sharing parts suppliers. The Beijing plant opened in 2002 and is a 50–50 joint venture. It employs 4000 workers. Hyundai chose Beijing to avoid the turf of other foreign competitors in Shanghai (GM and Volkswagen), Wuhan (Citroen) and Changchun (Volkswagen). According to Yang (2004), Hyundai has 49 Korean suppliers in China. They import raw materials from Korea and supply ﬁnished modules to the Beijing plant. More generic parts come from Chinese suppliers. 8.4.5

Pohang Steel

Pohang Steel Co. (POSCO) was established in 1968 as a public company and went through a number of expansion phases. The two large steel mill complexes were completed in 1992 and boasted the world’s largest steelmaking capacity until Mittal Steel and Arcelor merged in 1999. POSCO currently imports most of its iron ore from Australia and Brazil. It imports bituminous coal from Australia, China and Canada. Products are in turn exported to China, Japan, the USA and Southeast Asia. The company is currently trying to make up for sagging domestic sales by increasing foreign production. It has 15 processing plants overseas, and these are mostly in the form of joint ventures. POSCO was privatized in 2000, and the transition seems to have been successful. The company has been keenly interested in securing its own sources of raw material in order to lessen its dependence on foreign vendors. It is also developing new technology that requires less input. In 2005, the company struck a US$12 billion deal to erect a steel mill in a seaport near Puri, India. It is the largest single FDI project ever undertaken by a Korean ﬁrm, and it represents the biggest inward investment for India as well. A prospect for local steel consumption played a role. POSCO also obtained a mining right lasting up to 30 years to secure cheap iron ore from a nearby mine.

8.5

LOCATION CHOICE OF KOREAN SMES IN EAST ASIA

Analyses of FDI can take a variety of forms and approaches. When dealing with large countries, the location choice of ﬁrms within the country may

Location choices of Korean MNEs

233

engender the most curiosity. Meyer and Green (1996) identiﬁed determinants of Canadian FDI toward the USA using various state-speciﬁc attributes such as market size, infrastructure, labor conditions and taxes. China cannot be treated as a dimensionless point. Poor infrastructure, natural barriers and diﬀerent languages all point to segregated markets. Using province-level data, Lee and Kim (2004) have shown that Korean investment in China evolved from a cost-saving approach to one aimed at market penetration. Although this is quite illuminating, an inherent shortcoming of this kind of approach is that it overlooks interregional linkages as well as herd eﬀects often found in reality. For example, Head et al. (1999) concluded that Japanese investors in the USA preferred states chosen by preceding investors, thus exhibiting pronounced agglomeration. The spatial economic models have distinct advantages in this regard, for they explicitly deal with such concepts as backward/forward linkages and agglomeration of manufacturing ﬁrms. Since most SMEs cannot enjoy the luxury of having multiple plants, each ﬁrm may be treated as one that maximizes proﬁt where plants produce slightly diﬀerentiated products. Employing Chamberlin’s (1933) concept of monopolistic competition, Fujita et al. (1994) laid out a basic analytical framework in which various factors make up a proﬁt proﬁle for a ﬁrm. Taking a similar approach in a hierarchical setting, Kim (2003) examined the impact of free trade on automotive ﬁrms in North America. Choi and Kim (2005) studied the eﬀects of the ﬁnancial crisis on attractiveness of East Asian countries as possible production sites. The model explicated in this section retains the same tenets and structures. The underlying idea is that a potential manufacturer of a given product may identify the best place to operate by comparing the proﬁt levels of existing ﬁrms in each region, taking the current industrial conﬁguration as given. A snapshot of the region-by-region proﬁtability of the industry is taken when all goods markets are cleared. Ultimately, a simple measure can be developed which possible entrants can use to compare the short-term proﬁt potential of candidate sites. Appendices 8A1 and 8A2 contain an introduction of the model and a presentation of some basic information on the sample regions. The model is simple yet retains the central features of spatial economic theory such as increasing returns to scale and preference for variety. Due to lack of reliable data encompassing all study regions, however, direct application of the model is quite challenging. For example, published provincial GDP ﬁgures of China tend to be exaggerated.9 Another major problem results from the fact that East Asia is not self-contained. The role of the USA as a big market for Chinese products makes it diﬃcult to apply the model. Furthermore, Deardorﬀ and Stern (1998, p. 126) note that intangible

234

Case studies of East Asian economic integration

Canberra Chittagong Phnom Penh Hong Kong New Delhi Jakarta Tokyo Seoul Vientiane Kuala Lumpur Yangon Wellington Islamabad Manila Singapore Colombo Taipei Bangkok Hanoi Beijing Tianjin Shijiazhuang Taiyuan Hohhot Shenyang Changchun Harbin Shanghai Nanjing Hangzhou Hefei Fuzhou Nanchang Jinan Zhengzhou Wuhan Changsha Guangzhou Nanning Haikou Chongqing Chengdu Guiyang Kunming Lhasa Xi’an Lanzhou Xining Yinchuan Urumqi 0

200000

400000

600000

800000

1000000

1200000

US$

Note: Parameter values: L 5, f 1, 0.7.

Figure 8.15

Estimated profits of Korean SMEs in sample regions

1400000

Location choices of Korean MNEs

235

barriers to FDI, such as restrictions on foreign ownership and the need to go through time-consuming screening processes, can be overwhelming. These and other obstacles notwithstanding, potential proﬁts of suppliers, regarding actual production as the total demand from assemblers, can be calculated. Using the relationships shown in equations (8A1.4) and (8A1.13) of the Appendix, equation (8A1.14) can be rewritten as follows: F y

(1 )LFFD y fFL Wy.

Here, is the assembler’s substitution parameter for a pair of modules. LF and fFL are, respectively, the marginal labor requirement and the ﬁxed labor requirement for module production. Fy is the total demand for a variant of module manufactured in region y where Wy is the prevailing wage level. For a proxy measure of SME production, the cumulative cases of FDI in nonferrous metal, raw metal and processed metal were chosen. Figure 8.15 shows the calculated proﬁts based on the above equation of Korean MNEs at each region in the study area. This approach is admittedly crude, but results nonetheless conﬁrm the popularity of the eastern seaboard of China comprised of Shandong, Jiangsu and Tianjin. Regional ranking remains quite robust against diﬀerent values of parameters for labor requirements and substitutability for parts.10

8.6

CONCLUSION

With the tide of globalization in full swing, borders and barriers in economic activities are crumbling for both production and consumption of goods. In an interview with a Korean newspaper, the chairman of Mercedes Benz remarked that they manufacture cars in China and Eastern Europe not only because these regions oﬀer low costs and potentially large markets, but also because consumers no longer care about where their favorite products are made.11 This chapter included an examination of how Korean MNEs came into existence and how they have evolved over time. Following Japan’s footsteps, and sometimes leaping ahead, Korea now plays an active role in integrating East Asian economies through direct investment. In a struggle to get out of a nutcracker squeezed by Japan and China, Korea’s FDI has grown by leaps and bounds in a relatively short time span. Manufacturing dominates the scene and reﬂects marked preference for the eastern provinces of China. Severe cost pressure has caused ﬁrms to go abroad, creating clusters of related ﬁrms and transferring technology to host countries, just as the theory

236

Case studies of East Asian economic integration

illustrates. However, with intensifying competition, rising costs and a shift of government policy, they must once again ﬁnd a way out. Before their future performance can be predicted, many issues must still be investigated. First, the domestic investment policy of Korea may aﬀect the location pattern of its outward FDI. Since the early 1980s, balanced growth has been the central goal for Korea’s regional development policy. While the province surrounding Seoul is still the most preferred location, ﬁrms have diﬃculty with opening and expanding plants due to prohibitive taxes or outright bans on industrial land development. Foreign ﬁrms in cuttingedge ﬁelds are usually granted locations in the areas surrounding the capital. However, for domestic ﬁrms, permissions are given only in a very reluctant and selective manner. Some argue that this leads only to an exodus of manufacturing ﬁrms to China rather than to growth and proliferation of non-capital regions. Second, current trends toward freer trade will no doubt change investment patterns in East Asia. Korea has an FTA with Singapore. Agreements have been signed with the USA, EFTA and ASEAN. Negotiations are also under way with the EU, Canada, Mexico, India and Japan. An FTA between China and the original ASEAN-6 is to be established by 2010, and by 2015, newer ASEAN members will join it. Japan has an FTA with Singapore and is set to sign another FTA with ASEAN by 2012. An interesting political factor involves North Korea. Currently, 13 South Korean ﬁrms are operating in the Kaesong complex, and they employ over 10 000 North Koreans. Ten more ﬁrms are on their way. Finally, China’s contractionary policy, along with lax protection of intellectual property rights, rising labor costs, and crawling productivity, is making the country much less attractive than before. Reportedly, Korean ﬁrms in China are having increasing problems with high turnover rates and a lack of experienced workers at the managerial level.12 Many SMEs that migrated to China have already lost their competitive edge, and they are now realizing that low wages alone are not suﬃcient to resuscitate them.13 Frequent hikes in the minimum wage and social insurance requirements imposed on all employers may signal the end of the low-wage era. Preferential tax treatments for foreign enterprises are also being slashed. Changing preferences toward high-value-added, high-tech and environmentally friendly industries will make foreign investors more wary. In retrospect, severe wage pressures in the 1980s and the later ﬁnancial crisis in the late 1990s rang a bell; many large and small Korean ﬁrms soon scrambled and made inroads into China and other Asian countries. After a successful march, they seem to have reached major crossroads. Will Korea manage to escape the nutcracker through its FDI? Only time will tell.

Location choices of Korean MNEs

237

NOTES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

This section was written in collaboration with Toshitaka Gokan. In 1996, it cost US$800 to move a container from Busan to Seoul, while it cost only US$400 to ship to Hong Kong and US$500 to ship to Singapore (Joongang Ilbo, 15 November 1996). Unless otherwise noted, the source of data in this section is the Korea Export/Import Bank. The ‘other’ category includes Brunei, Cambodia, Japan, Lao PDR, Macau, Malaysia, Mongolia, Myanmar, the Philippines, Singapore, Sri Lanka, Taiwan, Thailand and Vietnam. Hong Kong and the Virgin Islands are excluded. Appendix 8A.3 provides information on regional distribution of each manufacturing subsector. Shandong leads across the board. Unlike Japanese FDI, Korean ﬁrms display a weak presence in Guangdong. For three provinces in the northeast (Heilongjiang, Jilin and Liaoning), the presence of a large Korean–Chinese population is said to be a major factor shortening cultural distance for Korean investors. In 2004, Samsung plants in Korea dealt with 522 foreign ﬁrms in addition to 1131 domestic suppliers (Maekyung Economy, 23 February 2005). On the other hand, Kim (2005) indicates that published statistics grossly underestimate real purchasing power due to the country’s socialistic distribution system. Values for Japan and Hong Kong are high because large ﬁxed costs in those locations are not reﬂected in the model. Chosun Ilbo, 22 April 2004. Maekyung Economy, 17 May 2006, p. 79. For example, of the 16 ﬁrms that moved to Qingdao, Shandong from 1992 onward, ten have since closed down (Chosun Ilbo, 30 June 2005).

REFERENCES Akamatsu, K. (1962), ‘A historical pattern of economic growth in developing countries’, Developing Economies, 1, 3–25. Amiti, Mary (2005), ‘Location of vertically linked industries: agglomeration versus comparative advantage’, European Economic Review, 49, 809–32. Baldwin, Richard, Rikard Forslid, Philippe Martin, Gianmarco Ottaviano and Frédéric Robert-Nicoud (2003), Economic Geography and Public Policy, Princeton, NJ: Princeton University Press. Booz, Allen and Hamilton (1997), Vision Korea Report: Revitalizing the Korean economy towards the 21st century, Seoul: Maeil Business Newspaper. Brülhart, Marius (1998), ‘Economic geography, industry location and trade: the evidence’, World Economy, 21 (6), 775–801. Chamberlin, E.H. (1933), The Theory of Monopolistic Competition, Cambridge, MA: Harvard University Press. Choi, Yun Ho and Ho Yeon Kim (2005), ‘Measuring direct investment potentials and its applications to Korea’ (in Korean), Kukje Kyungje Yongu, 11 (3), 37–56. Deardorﬀ, Alan V. and Robert M. Stern (1998), Measurement of Nontariﬀ Barriers, Ann Arbor, MI: University of Michigan Press. Dixit, A.K. and J.E. Stiglitz (1977), ‘Monopolistic competition and optimum product diversity’, American Economic Review, 67, 297–308.

238

Case studies of East Asian economic integration

Dowling, M. and C.T. Cheang (2000), ‘Shifting comparative advantage in Asia: new tests of the ﬂying geese model’, Journal of Asian Economics, 11 (4) , 443–63. Dunning, John H. (1977), ‘Toward an eclectic theory of international production: some empirical tests’, Journal of International Business Studies, 11, 9–31. Dunning, John H. (1993), Multinational Enterprises and the Global Economy, Harlow, UK: Addison-Wesley. Dunning, John H. and Rajneesh Narula (1996), ‘The investment development path revisited: some emerging issues’, in John H. Dunning and Rajneesh Narula (eds), Foreign Direct Investment and Governments: Catalysts for economic restructuring, London: Routledge, pp. 1–41. Ensign, Prescott C. (1995), ‘An examination of foreign direct investment theories and the multinational ﬁrm: a business/economics perspective’, in Milford B. Green and Rod B. McNaughton (eds), Location of Foreign Direct Investment: Geographic and business approaches, Aldershot: Ashgate Publishing, pp. 15–27. Fujita, M., G. Abdel-Musik, H. Kim and N. Hamaguchi (1994), ‘A model analysis of trade and FDI diversion eﬀects of NAFTA’, in Koichi Ohno and Yumiko Okamoto (eds), Regional Integration and Foreign Direct Investment: Implications for developing countries, Tokyo: Institute of Developing Economies, pp. 254–69. Fujita, Masahisa, Paul Krugman and Anthony J. Venables (1999), The Spatial Economy: Cities, regions, and international trade, Cambridge, MA: MIT Press. Graham, Edward M. (1999), ‘Foreign direct investment outﬂows and manufacturing trade: a comparison of Japan and the United States’, in Dennis J. Encarnation (ed.), Japanese Multinationals in Asia: Regional operations in comparative perspective, Oxford: Oxford University Press, pp. 87–99. Ha, Byung Ki (2004), An Analysis on the Activities of Korea’s Overseas Aﬃliates (in Korean), Korea Institute for Industrial Economics and Trade. Hanink, Dean M. (1994), The International Economy: A geographical perspective, New York: John Wiley & Sons. Head, Keith, John Ries and Deborah L. Swenson (1999), ‘Attracting foreign manufacturing: investment promotion and agglomeration’, Regional Science and Urban Economics, 29, 197–218. Helpman, E. and P. Krugman (1985), Market Structure and Foreign Trade, Cambridge, MA: MIT Press. Hu, Dapeng and Masahisa Fujita (2001), ‘Regional disparity in China 1985–1994: the eﬀects of globalization and economic liberalization’, Annals of Regional Science, 135 (1), 3–37. Jee, Mansoo et al. (2004), Korean Firms Invested in China: Survey of management practices and implications (in Korean), Korea Institute for International Economic Policy. Jones, Leroy P. and Il Sakong (1980), Government, Business, and Entrepreneurship in Economic Development: The Korean case, Council on East Asian Studies, Harvard University. Kim, Ho Yeon (2003), ‘Impact of trade liberalization on the location of ﬁrms: NAFTA and the automobile industry’, Annals of Regional Science, 37, 149–73. Kim, Ho Yeon and Kyu Sung Jin (2003), ‘A study on sequential changes in the industrial structure of East Asian countries’ (in Korean), Journal of Northeast Asian Economics, 15 (2), 33–62. Kim, Si Joong (2005), ‘How peculiar is the Chinese economy?’ (in Korean), Kyongjehak Yongu, 53 (2), 185–214.

Location choices of Korean MNEs

239

Kojima, Kiyoshi (1985), ‘The allocation of Japanese direct foreign investment and its evolution in Asia’, Hitotsubashi Journal of Economics, 26, 99–116. Krugman, Paul (1991), ‘Increasing returns and economic geography’, Journal of Political Economy, 99 (3), 483–99. Krugman, Paul and Gordon Hanson (1993), ‘Mexico–U.S. free trade and the location of production’, in P.M. Garber (ed.), The Mexico–U.S. Free Trade Agreement, Cambridge, MA: MIT Press, pp. 163–86. Lall, Sanjaya (1996), ‘The investment development path: some conclusions’, in John H. Dunning and Rajneesh Narula (eds), Foreign Direct Investment and Governments: Catalysts for economic restructuring, London: Routledge, pp. 423–41. Lee, Hongshik and Hyukhwang Kim (2004), ‘International investment location decisions: the case of Korean ﬁrms in China’ (in Korean), Journal of International Economic Studies, 8 (2), 257–89. Markusen, James R. (1984), ‘Multinationals, multiplant economies and the gains from trade’, Journal of International Economics, 16, 205–26. Markusen, James R. (1991), ‘The theory of the multinational enterprise: a common analytical framework’, in Eric D. Ramstetter (ed.), Direct Foreign Investment in Asia’s Developing Economies and Structural Changes in the Asia-Paciﬁc Region, Boulder, Co: Westview Press, pp. 11–32. Markusen, James R. (1998), ‘Multinational ﬁrms, location and trade’, World Economy, 21, 733–56. Markusen, James R. and Anthony J. Venables (1999), ‘Foreign direct investment as a catalyst for industrial development’, European Economic Review, 43, 335–56. Mason, Mark (1999), ‘The origins and evolution of Japanese direct investment in East Asia’, in Dennis J. Encarnation (ed.), Japanese Multinationals in Asia: Regional operations in comparative perspective, Oxford: Oxford University Press, pp. 17–45. Meyer, Stephen P. and Milford B. Green (1996), ‘Outward Canadian direct investment and place-speciﬁc attributes’, Geoforum, 27 (2), 225–45. Navaretti, Barba and Anthony J. Venables (2004), Multinational Firms in the World Economy, Princeton, NJ: Princeton University Press. Ottaviano, Gianmarco I.P. and Diego Puga (1998), ‘Agglomeration in the global economy: a survey of the “new economic geography’’ ’, World Economy, 21 (6), 707–31. Robert-Nicoud, Frédéric (2006), ‘Oﬀ-shoring of business services and deindustrialization: threat or opportunity – and for whom?’, CEPR Discussion Paper No. 734. Urata, Shujiro (1999), ‘Intraﬁrm technology transfer by Japanese multinationals in Asia’, in Dennis J. Encarnation (ed.), Japanese Multinationals in Asia: Regional operations in comparative perspective, Oxford: Oxford University Press, pp. 143–62. Vernon, R. (1966), ‘International investment and international trade in the product cycle’, Quarterly Journal of Economics, 80, 190–207. Yang, Pyung Sub (2004), Investment Strategies of Korean Firms in China (in Korean), Korea Institute of International Trade. Zhang, Hai-Yan and Danny Van der Bulcke (1996), ‘China: rapid changes in the investment development path’, in John H. Dunning and Rajneesh Narula (eds), Foreign Direct Investment and Governments: Catalysts for economic restructuring, London: Routledge, pp. 380–422.

240

Case studies of East Asian economic integration

APPENDIX 8A.1 1.

THE MODEL

Basic Structure

The ﬁnal goods (A-goods) are assembled using various modules (F-goods) coming from suppliers. Consumer demand for ﬁnal goods creates subsequent demand for modules. There are J regions, and for reference purposes, suppliers and assemblers are indexed by j and k, respectively. A representative consumer is assumed to consume a numeraire good (N) and a group of diﬀerentiated ﬁnal goods: U N1CA, where CA

J

k1

mkAk

1

.

(8A1.1)

The sub-utility function takes the form proposed by Dixit and Stiglitz (1977). Here, mk is the number of variety for the ﬁnal good produced in region k, and A indicates the quantity of each variant of goods assembled in region k for consumption. The fact that higher m improves utility can be interpreted as a taste for variety in the population at large, with each individual having his or her most preferred variant (see Helpman and Krugman, 1985, p. 116). The is a substitutability parameter (0 1) for a given pair of brands. If is close to 0, A-goods can be substituted with unit elasticity; when is close to 1, the ﬁnal goods are almost perfect substitutes. The production function for assemblers is in the familiar Cobb–Douglas form for the two input categories, labor and intermediate inputs: Ak L1 A CF k

where CF

njF j . J

1

(8A1.2)

j1

Here, nj denotes the number of variety for the modules produced in region j, and Fj is the quantity of each variant of modules manufactured in region j that an assembler demands in region k. The is the assembler’s substitution parameter for a pair of modules. Thus CF represents the subproduction function in terms of diﬀerentiated modules available and the degree of substitution among these modules.

Location choices of Korean MNEs

241

All ﬁrms possess increasing-returns-to-scale technology owing to ﬁxed input requirements. This allows each ﬁrm to specialize in the production of a single variant and implies that the number of ﬁrms is the same as the number of products. The total labor requirement for the assembler x, producing Ax units, is: TLA fAL LAAx, x

(8A1.3)

where fAL is the ﬁxed labor requirement, and LAAk is the variable labor requirement. For the sake of simplicity, suppliers are assumed to use only labor to produce various parts. Analogous to the assembler’s case, the total labor requirement for a supplier y, producing Fy units, is the sum of the ﬁxed and variable requirements: TLF fFL LFFy. y

(8A1.4)

Movement of goods incurs additional expenses, T, in a multiplicative form (on an ad valorem basis). Thus the delivered price in region g for ﬁnal goods assembled in region k is simply computed as PAkTA(k,g). Similarly, the delivered price for modules is PFjTF (j,k). 2.

Demand for Goods

The demand for ﬁnal goods generates subsequent demand for modules. In this section, the demand and price for each class of goods are derived according to the optimizing behavior of the relevant agent. 2.1 Consumer behavior A representative consumer in region g maximizes his or her utility by consuming the numeraire good and a composite group of A-goods, subject to a budget constraint: Max U, s.t. Ig PNN

J

mkPA TA(k, g)Ak.

k1

k

(8A1.5)

The quantity of a variant of ﬁnal good, assembled in region x and demanded by consumers in region g, is then derived as:1

242

Case studies of East Asian economic integration

A(x,g)

Ig{PA TA (x,g)}( 1) x

J

k1

{mkPA TA (k, k

g)}

where 1 .

(8A1.6)

Because 0, it can be shown that the demand for this particular brand increases with a lower price and delivery charge, with a higher income and expenditure share of consumers, and also with higher prices and transport costs faced by competing brands in the market.2 It follows that the total economy-wide demand for this variant is: AD x

J

A(x,g).

(8A1.7)

g1

2.2 Behavior of assemblers The assembler x seeks to minimize the sum of labor and material costs, given the required production level: Min TCA Wx·TLA x

J

njPF TF(j,x)Fj j1

j

1 s.t. AD x LA CF .

(8A1.8)

x

Solving this problem yields the labor requirements for this ﬁrm as: LA x

(1 )AD x Wx bx ,

(8A1.9)

where Wx is the prevailing wage level in region x, and bx is the regionspeciﬁc Lagrange multiplier. Furthermore, the amount for a variant of module produced in region y and in demand by an assembler in region x may be obtained by: F(y,x)

J

j1

bxAD x

njPF TF (j,x)Fj j

1 1

.

(8A1.10)

Apparently, the demand for modules increases, ceteris paribus, with a higher level of A-good production and with lower price and transportation charges for the modules. Demand also soars when the number of competing products (n) drops. Noting that suppliers cater to all regions,

Location choices of Korean MNEs

243

each with m assemblers, the total demand for a particular variant of module is: FD y

J

mkF(y,x).

(8A1.11)

k1

Substitution of these input requirements into the constraint yields the Lagrange multiplier (b) for this problem, and this is binding and thus equal to the marginal cost. All ﬁrms set their prices according to the monopolist’s rule, equating marginal cost to marginal revenue: MCP(11/e) MR, where e is the price elasticity of demand. However, for a CES-type sub-utility function with a substitutability parameter of , the (11/e) term is known to approximate when the number of the variants is large.3 Thus the f.o.b. price of a unit of ﬁnal goods produced in region x can be derived as: PA x

MC bx. (1 1e)

(8A1.12)

This implies that a ﬁrm with greater market power (lower ) can aﬀord to set a higher markup for its products. 2.3 Behavior of suppliers Using labor as their sole input, suppliers in region y each produce FD units of modules that are priced according to the principle: PF LFWy .

(8A1.13)

y

3.

Proﬁt Functions for Firms

The obtained demand functions may now be used to measure the attractiveness of the regions as reﬂected in potential proﬁt. 3.1 Proﬁt of suppliers Since price is higher than marginal cost, it pays to produce as much as demand dictates. The proﬁt function for a particular supplier y serving all assemblers is deﬁned to be the total revenue less the costs associated with labor: F PF FD y Wy·TLF . y

y

y

(8A1.14)

244

Case studies of East Asian economic integration

This reveals, in conjunction with equation (8A1.11), that proﬁt is positively related to market size and the transport costs faced by competitors. On the other hand, proﬁt is inversely related to wage levels, the supplier’s own transportation costs to the market, and the degree of overall competition captured by n. 3.2 Proﬁt of assemblers The proﬁt function for assembler x producing the maximum possible output is: A PA AD x Wx·TLA x

x

x

J

njPF TF(j,x)Fj. j1

j

(8A1.15)

Substituting both AD found in equation (8A1.7) and PA in equation (8A1.12) into the above, an expanded form can be obtained. The ﬁrm’s proﬁtability is shown to depend on the labor price, accessibility to modules, the market size, ease of transportation to the consumption point, and the intensity of the competition.

NOTES 1. 2.

3

See Fujita et al. (1999), pp. 46–8. The magnitude of the impact brought about by changes in prices or transport costs depends on the substitutability of A-goods as perceived by consumers. If is close to one, meaning the goods are not very diﬀerent from other brands, then the price elasticity of demand, as well as w, will be very high. Thus a small reduction in prices or transportation costs will result in a large increase in demand. See Helpman and Krugman (1985), p. 119.

245

Location choices of Korean MNEs

APPENDIX 8A2

No. Country 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

SAMPLE REGIONS AND KEY VARIABLES (2003)

Province

Australia Bangladesh Cambodia Hong Kong India Indonesia Japan Korea Lao PDR Malaysia Myanmar New Zealand Pakistan Philippines Singapore Sri Lanka Taiwan Thailand Vietnam China Beijing Tianjin Hebei Shanxi Inner Mongolia Liaoning Jilin Heilongjiang Shanghai Jiangsu Zhejiang Anhui Fujian Jiangxi Shandong Henan Hubei

Reference city

GDP Per capita Average (US$ bil.) GDP (US$) tariff (%)

Canberra Chittagong Phnom Penh

512.95 52.52 4.21 1 416.60 New Delhi 576.12 Jakarta 238.52 Tokyo 4 299.74 Seoul 605.54 Vientiane 2.09 Kuala Lumpur 103.74 Yangon 9.73 Wellington 78.50

25 615 359 306 22 760 539 1 092 33 705 12 635 366 4 151 183 19 539

5.0 19.5 16.3

Islamabad Manila

73.59 78.14 92.37 18.24 285.83 142.95 39.54 44.26 29.57 85.77 29.68 25.98

500 952 21 765 904 12 530 2 226 489 3 874 3 206 1 270 898 1 084

17.3 4.3 0.5 8.3 5.4 14.0 16.1 10.8

72.52 30.48 53.52 75.52 150.55 113.51 50.00 63.22 34.20 150.25 85.16 65.26

1 723 1 128 1 403 5 644 2 031 2 434 780 1 810 807 1 651 915 1 089

Colombo Taipei Bangkok Hanoi

Shijiazhuang Taiyuan Hohhot Shenyang Changchun Harbin Nanjing Hangzhou Hefei Fuzhou Nanchang Jinan Zhengzhou Wuhan

28.4 7.0 2.6 8.0 8.1 9.6 3.7

246

Case studies of East Asian economic integration

Appendix 8A2 (continued) No. Country

Province

Reference city

37 38 39 40 41 42 43 44 45 46 47 48 49 50

Hunan Guangdong Guangxi Hainan Chongqing Sichuan Guizhou Yunnan Tibet Shaanxi Gansu Qinghai Ningxia Xinjiang

Changsha Guangzhou Nanning Haikou Chengdu Guiyang Kunming Lhasa Xi’an Lanzhou Xining Yinchuan Urumqi

GDP Per capita Average (US$ bil.) GDP (US$) tariff (%) 56.05 164.63 33.05 8.11 27.19 65.92 16.38 29.79 2.23 28.98 15.76 4.71 4.66 22.69

913 2 080 721 1 005 871 775 435 684 830 783 607 879 808 1 172

Source: IMF World Economic Outlook, China Statistical Yearbook, UNCTAD Handbook of Statistics.

APPENDIX 8A3

REGIONAL DISTRIBUTION OF KOREAN FDI BY SECTOR (1968–2006.9)

247

No.

Food

Textile

Shoes

Wood

Paper

Oil

Nonferr.

Raw mtl

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

18 7 3 5 3 28 19

7 57 26 25 12 168 20

1 13 1 15 4 50 3

2 2 4 2 2 30 4

6 3

5 10 3 23 20 71 21

2 12 4 1 9 12 5

3 3

2 2

6 7 15 7

7 9 31 8

1 5 16 1 3 122 1 45

1 1 4 1 3 46

1 17 1 7

1 47 2

8 1

9 1

8

2

8 272 58 125 49

14 1 10 91 11 53 12

1 22 8 21 11

12 2 3 2 8 22 10 17 6

11 392

81

1 102

2 43

4 2 5 26 3 3 14 26 45 29 37 4 150

14

6 20 14 1 8 3

Proc. mtl

1 1 58 6 10 4 27 90 72 86 25 1 1 165

15 1 5 1 5 13 18 23 23 6

17 2 6 1 9 22 12 26 15 1

82

45

23 2 1 15 37 24 77 22 3 1 84

Machinery 7 3 2 30 20 37 42

42 1 3 39 8 3 9 38 48 62 160 28

179

Trans. 5 1 7 47 21 7 1 8 1 1 33 1 2 13 33 88 45 14 1 77

Elec.

Other

8 1 2 135 22 80 59

10 8

1 43 3 1 1 97 13 3 32 75 29 53 381 24 3 1 114

45 10 56 41

14 2 2 40 6 23 2 33 59 68 129 38 1 186

Appendix 8A3

(continued)

248

No.

Food

Textile

Shoes

Wood

Paper

Oil

Nonferr.

Raw mtl

Proc. mtl

Machinery

Trans.

Elec.

Other

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

109 43 18 31 22 7 1 4 360 5 6 1 5 8 4

142 45 98 195 141 3 14 9 987 1 4 3 35 4

12 7 21 45 14

61 33 9 9 11

12 4 10 43 5

31 7 26 44 17

1

74 31 67 110 47 6 10 1 326 6 7 4 41 7 3

35 11 10 26 5

16 1 397

16 4 15 18 13 1 1 2 83

5

2

30 9 35 94 18 11 8

125 1 2 2 4 3 3

82 2

5 2 197 1 2 1 22 1

59 36 84 161 78 3 8 1 457 4 3 2 59 3

34 24 67 196 37 4 9 2 617 2 1 4 145 1 2

61 25 70 119 48 1 8 1 834 10 3 7 42 3 1

3 1 2

6 1

3

1 1 1

6

1

3

24 1

142 1 1 2 7 1 1

2 1

5 1

3

1 1

1 1

2 3

8 1 1

1

5 1

2

187 2 12 2

5 1

1

3

1

1

Note: Figures represent the cumulative number of cases for each sector and region. For regions, refer to Appendix 8A2.

1 1 1

2 1

1

9.

Location choices of Japanese MNEs in East Asia Toshitaka Gokan

9.1

INTRODUCTION

For Japanese MNEs, Asia is the main host of their FDI and production bases. Asia hosts the majority of the overseas subsidiaries of Japanese MNEs both in manufacturing and non-manufacturing.1 Over 90 percent of the Japanese MNEs in Asia were concentrated in East Asia in 2002.2 In East Asia, the share for manufacturing remains larger than that for nonmanufacturing, the reverse for Japanese MNEs in other regions of the world where manufacturing makes up a smaller share than non-manufacturing (Ando and Kimura 2003). This suggests that East Asia is a production base rather than a market for Japanese MNEs. For this reason, in this chapter we focus on Japanese manufacturing MNEs’ choice of East Asia as a place to locate. MNEs bring beneﬁts to host countries via many channels. First, FDI provides employment. In Hong Kong, Singapore and Malaysia, Japanese MNEs employed over 1 percent of the total population in 2002. Second, FDI induces further investments (Bosworth et al. 1999), and these investments are more stable than other types of capital ﬂow such as portfolio investment. Third, FDI makes it possible for the host county to access foreign markets by utilizing the various marketing resources and skills of MNEs. This access to foreign markets helps the host country achieve economies of scale. Finally, MNEs bring technology transfer to host countries.3 Consequently, the growth in FDI brings about signiﬁcant progress in productivity. In fact, Marwah and Tavakoli (2004) estimate that the increase in FDI has provided about one-ﬁfth to one-quarter of the improvement in productivity of the total capital stock in Thailand, Malaysia, Indonesia and the Philippines. In China, the productivity of domestic ﬁrms tends to increase as the number of multinational ﬁrms locating near them increases (Zhou et al. 2000). MNEs also increase the competitiveness of the host country’s domestic markets. The impact of rising competition, however, diﬀers between 249

250

Case studies of East Asian economic integration

consumers and the local ﬁrms in the same industry. The welfare of consumers is increased by the competition among MNEs and local ﬁrms. Zhou et al. (2000) found a negative correlation between the proﬁts of local ﬁrms in an industry and the number of MNEs in the same industry locating near domestic ﬁrms. As a result, some host countries oﬀer incentives for attracting MNEs while others use regulations to direct them toward desirable production activity in the country. Theoretical literature on MNEs has been increasing rapidly. The new theory of trade analyzes intra-industry trade, using the taste for variety, increasing-returns-to-scale technology and non-zero trade costs. This approach formalizes the analysis of multinational ﬁrms. By enabling production factors to move endogenously among regions or sectors, spatial economics adopts the circular causality mechanism to explain the formation of industrial agglomeration. Analyses of vertical relations among ﬁrms have progressed using this approach. The automobile industry is a good example of vertical relations among ﬁrms. Many countries are eager to develop a national automobile industry because the industry requires a large number of inputs produced by many industries. Thus countries tend to protect the industry through regulations and tariﬀs. However, under international agreements, regulations such as local-content requirements are ultimately to be eliminated4 and tariﬀs are also to be lower because East Asia has become a party in international free trade agreements. Japanese automotive MNEs have been investing in the East Asian region, and are expected to be an important player in the coming era of trade liberalization. Thus, later in this chapter we attempt to predict the eﬀects of tariﬀ reductions on the attractiveness of potential locations for the Japanese automobile industry. We do not go into the details of the theory of MNEs in this chapter as these were dealt with in Chapter 8. The remaining sections in this chapter are organized as in the previous chapter, which ﬁrst looked at the characteristics of Korean MNEs, then proceeded to a calculation of their proﬁts. Sections 9.2 and 9.3 of this chapter focus on the characteristics of the locations selected by Japanese MNEs, while Section 9.4 focuses on the eﬀects of tariﬀ reduction on the location of Japanese automobile MNEs. Section 9.2 presents a broad picture of the locations that Japanese MNEs choose for their operations. Section 9.3 then proceeds to examine the characteristics of industries that aﬀect the choice of location for Japanese FDI, the purpose being to clarify the industries that match the theoretical model used for simulations. Based on this examination of the industries’ characteristics, Section 9.4 examines the eﬀects of tariﬀ reduction on the attractiveness of locations for Japanese MNEs in the automobile industry by

Location choices of Japanese MNEs

251

comparing their proﬁts. The simulations are based on Kim (2003). The conclusions of this chapter are set forth in Section 9.5.

9.2

THE DETERMINANTS FOR THE LOCATION OF JAPANESE MNES

This section focuses ﬁrst on East Asia, then on China. The majority of Japanese MNEs are located in East Asia, which in the context of this chapter is made up of China, the NIEs and the ASEAN-4.5 In the case of China, the advantage of location diﬀers greatly among provinces, requiring us to examine the location of Japanese MNEs at the provincial rather than national level. 9.2.1

East Asia

Japanese MNEs have located heterogeneously in East Asia. In 2002 about 30 percent had located in China, about 15 percent in Thailand, and about 10 percent in Taiwan, Malaysia and Indonesia,6 as indicated in Figure 9.1. The number in China in 2002 was larger than that in North America or in Europe.7 The number in the ASEAN-4 was between that in North America and that in Europe. The number of Japanese MNEs in East Asia increased about 1.6 times from 1989 to 2002. Comparing the amount of increase during this period, we see that (1) the number in China increased drastically, (2) the diﬀerences among the ASEAN-4 diminished, and (3) the number decreased in Korea, Singapore and Taiwan.8 We can see, as shown in Table 9.1, that Japanese MNEs have diﬀerent determinants for locating in the NIEs, in the ASEAN-4 and in China.9 In the case of the NIEs, Japanese MNEs expect to have steady or higher local sales in these countries. The reasons for this are the high wage rate and the Generalized System of Preferences (GSP). The high wage rate in the NIEs, pointed out in Chapter 8, implies that consumers have purchasing power. Furthermore, the GSP oﬀered preferential tariﬀs to products exported from the NIEs. Because the NIEs have developed suﬃciently, these countries lost the privilege of GSP from the USA, the EU and Japan. Thus the NIEs lose their advantage as production bases. In the case of China and the ASEAN-4, one determinant is the cost advantage of locating in these countries. Given their low wage rates, we could expect this result. Thus China and the ASEAN-4 can be regarded as production bases for Japanese MNEs. Another determinant of location is the capability to supply other Japanese ﬁrms, which corresponds to the number of Japanese MNEs. This

252

Case studies of East Asian economic integration

Source: RIETI FDI database.

Figure 9.1 The number of Japanese MNEs in East Asia by location in 2002 determinant is more important in the ASEAN-4 than in China. Belderbos and Sleuwaegen (1996) found that horizontal and vertical business groups are more closely connected to Japanese FDI in Southeast Asia than in North America or in Europe. Another determinant for locating in China is the relative importance of exporting to Japan. This may be due to China’s proximity to Japan. Another factor is the lower prospects for steady or higher growth in local sales in China as well as ASEAN, which translated into higher rates of exports to Japan. Finally, local R&D is more important as a determinant for Japanese MNEs to locate in developed countries than in Asia, and in Asia it is a more important factor for locating in China than in ASEAN. Tomiura (2003) found that R&D is more important in Japanese FDI to industrialized countries than to Asia, and Puga and Treﬂer (2005) used US patent data to show that China has displayed a signiﬁcant increase in innovation while ASEAN has not been very active in this respect. From the spatial economic perspective, our interest is in the eﬀects of the number of Japanese MNEs on the location choice of Japanese MNEs. In ASEAN, the eﬀects seems to work well. In China the number of Japanese MNEs is large, but the vertical relations among ﬁrms have not become an advantage. This may be because China is a large country.

253

Location choices of Japanese MNEs

Table 9.1

Determining factors for Japanese FDI: 2003 (%)

Prospects for steady or higher local sales Cost advantage Capability of supplying other Japanese firms Exports to Japan Local R&D

NIEs

ASEAN-4

China

N. America

Africa

32.2

23.7

25.4

35.6

27.9

11.4 10.7

17.9 14.1

17.1 10.3

6 10.9

16.2 1.5

4.1 1.5

6.2 0.6

7.3 1.2

2.8 5.8

5.1 1.5

Note: The figures indicate the percentage of respondents indicating the stated motives. Multiple answers were allowed. Source:

9.2.2

Wagakuni Kigyou no Kaigai Jigyo Katsudo 2004.

China

Japanese MNEs also have located heterogeneously in China. Over 80 percent of the Japanese MNEs in China’s manufacturing sector are located in coastal provinces and cities.10 From the ﬁgures shown in Figure 9.2, in 2002 about 23 percent of the total number of Japanese MNEs in China had located in Shanghai, about 18 percent in Jiangsu, about 14 percent in Guangdong, and about 9 percent in Liaoning.11 The number of Japanese MNEs in the whole of China increased about 7.2 times from 1989 to 2002. The distribution in the location of Japanese MNEs corresponds to the change in the Chinese economy. Shanghai has been the largest producer of many goods in China. Jiangsu was the largest recipient of FDI in 2003. The concentration of economic activity diversiﬁed from Liaoning to Shanghai and Guangdong, as discussed in Chapter 10. Japanese MNEs have also concentrated in Liaoning, Shanghai, Jiangsu and Guangdong. Dai (1998) showed that since China’s economic reforms, the growth rate of regional GDP and that of per capita regional GDP correlate with the share that foreign companies have in the capital formation of enterprise. Japanese MNEs presumably attracted the spatial distribution of economic activity in China. In Figure 9.3 the number in the coastal regions in 1989 is compared with that in 2002. From the ﬁgure, we can see that three core areas have developed centered on Liaoning, Shanghai and Guangdong, and that the central coastal region has grown greatly. The ﬁrst point suggests that the small diﬀerence in the location of Japanese MNEs at the initial stage brought about a core – periphery structure. The second point is similar to the results from the

254

Source:

Case studies of East Asian economic integration

Kaigai Shinsyutu Kigyou Souran 2003.

Figure 9.2

The number of Japanese MNEs in China by location in 2002

Hotelling model in the case of two ﬁrms. The model assumes a segmented linear area with a uniform distribution of consumer demand and shows that ﬁrms locate at or near the center to save transport costs for consumers. Shanghai and Jiangsu have attracted about 40 percent of Japanese MNEs located in China.12 Since Shanghai and Jiangsu share the same border, we can regard them as one region. It would seem that this region has attracted the greatest portion of Japanese FDI because the shipping distance from Tokyo to Shanghai is the shortest for any of Chinese major ports.13 With most intermediate goods being transported from ports on the Paciﬁc Ocean side of Japan, which has the greatest concentration of industrialized areas, Shanghai becomes the best destination in China for saving shipping costs. This is similar to the situation for South Korean FDI in China, where over 70 percent of Korea’s cumulative FDI from 1968 to September 2006 has come to be located in the provinces around the Yellow Sea, and about 40 percent of that in China is located in Shandong.14 As explained in Chapter 8, this concentration is presumably due to the proximity to South Korea. Yentai in Shandong is the closest major port in China to Inchon, South Korea.15 Besides distance, econometric analysis has examined the eﬀects of the size of Japanese FDI. Cheng and Stough (2006) showed that the agglomeration of Japanese MNEs is one of the determinants for the location they

255

Location choices of Japanese MNEs 450 400 350 300 250

1989 2002

200 150 100 50 na n ai H

Li

ao ni ng Ti an jin Be iji n H Sh ebe an i do ng Jia ng Sh su an gh ai Zh ej ia ng F H uji on an g K o G ua ng ng do ng

0

Source: RIETI FDI database.

Figure 9.3 The emergence of a core–periphery structure among Japanese MNEs in China’s coastal region choose in Chinese provinces. Along with this, the authors found that the number of Japanese MNEs locating in an area is also related to the provincial location quotients of Japanese FDI, labor costs, labor quality and the number of national development zones.

9.3

INDUSTRIAL FACTORS AFFECTING THE SPREAD OF JAPANESE MNES

In this section, I examine industries to see which ones match the model used for simulations. The focus will be on industries having the following aspects: (1) the location of Japanese MNEs in the industry is more heterogeneous; (2) local sales are more important than sales in foreign markets; (3) the number of Japanese MNEs attracts more Japanese MNEs into the industry. In this way I examine the characteristics of Japanese MNEs in each industry. As in the previous section, I examine East Asia ﬁrst, then at China alone. 9.3.1

East Asia

In each industry in East Asia, the distribution of Japanese MNEs has been heterogeneous. Of the total number of Japanese MNEs located in the NIEs, in the ASEAN-4 and in China in 2002, about 18 percent were involved in IT and communication equipment, about 14 percent in chemicals,

256

Case studies of East Asian economic integration

about 12 percent in motor vehicles, about 10 percent in ordinary machinery, and about 8 percent in textiles or electronic machinery. The Japanese ﬁrms operating in these industries accounted for about 70 percent of the total number of Japanese MNEs in East Asia. In Figure 9.4, these industries are arranged according to the share of MNEs in each industry. The boxes in the ﬁgure show interquartile ranges, which are the range between the ﬁrst quartile and the third quartile. The bold lines mean the values of the median. The circles show the values of the outliers, which lie more than 1.5 times outside of an interquartile range. The bars above/below the boxes show the largest or the smallest non-outlier values. From the ﬁgure, we see that China attracts the most Japanese MNEs, except in the automobile industry where Thailand along with China are the main host countries for Japanese MNEs. The locations selected by Japanese MNEs diﬀers among the industries. Six industries in Figure 9.4, can be divided into three groups: (1) IT and communication industry; (2) automobile industry and textile industry; (3) chemical industry, ordinary machinery industry, and electric machinery industry. I now examine the characteristics of Japanese MNEs’ distribution in each industry in these three groups. First, the IT and communication industry shows a relatively weak contrast in the location of Japanese MNEs among the countries. Thus this industry does not match the model used for simulations. However, the number of Japanese MNEs in this industry is large. Korean FDI in this industry is also large, as seen in Chapter 8. Firms in this industry seem easily to become MNEs. This suggests that additional ﬁxed costs for additional plants are relatively small in comparison with total ﬁxed costs of a ﬁrm in this industry; i.e., ﬁrms can engage in intensive R&D and duplicate domestic production easily abroad. Moreover, Figure 9.4 shows that Malaysia is the second-largest host for Japanese MNEs in this industry. Penang Island, examined in Chapter 3, is well known for the agglomeration of the IT industry that has developed there. Second, Figure 9.4 indicates that the location of Japanese MNEs in the textile industry and the automobile industry shows a clear contrast between the core and the periphery. Firms in the textile industry have clearly concentrated in China, which in 2002 hosted about 60 percent of the Japanese MNEs in the industry in East Asia. The automobile industry in Thailand has agglomerated in the vicinities of Bangkok and Leamchabang, as explained in Chapter 3. Motor vehicles, chemical industries, ordinary machinery, electronic machinery, and metal products are in automobilerelated industries. The number of Japanese MNEs in these industries in Thailand in 2002 was the second largest in East Asia. These automobilerelated industries seem to match the model used for simulations.

257

Location choices of Japanese MNEs China

China

200 China Thailand 150

China

China

Malaysia

China

Thailand

100

Thailand Indonesia

50

Thailand

0 I1

I2

I3

I4

I5

I6

Note: I1IT and communications equipment industry; I2 chemical industry; I3 automobile and auto-parts industry; I4 ordinary machinery industry; I5 textile industry; I6electric machinery industry. Source: RIETI FDI database.

Figure 9.4 The number and location of Japanese MNEs in East Asia in 2002 Finally, the chemical industry, ordinary machinery industry, and electric machinery industry exhibit a similar pattern of location as that of the textile and automobile industries. While Figure 9.4 indicates that among the countries the core–periphery contrast in this third group is not as clear as that of the second group, it is clearer than that of the ﬁrst group. The ﬁgure shows that the electric machinery industry is concentrated in China more than are the other industries in this group, as shown by the fact that the diﬀerence between China and the country with the next largest number of Japanese MNEs in the electric machinery industry is the largest in this group. Japanese MNEs involved in electric home appliances have concentrated in China and Thailand, while those involved in audiovisual equipment have concentrated in China and Malaysia.16 In spatial economics, the size of the home market is important because ﬁrms become proﬁtable in a large market with increasing-returns-to-scale technology. Therefore, in this study we prefer industries that produce mainly for local markets. Table 9.2 shows how important the local market is for each

258

Case studies of East Asian economic integration

industry. The table divides exports to Japan and exports to third countries by local sales. The industries are arranged in descending order from the smallest to the largest sum of a row, with a smaller value indicating that the local market is more important for the industry.17 The table indicates that transport equipment, food and tobacco, and chemicals depend on local sales. Thus these industries match the model used for simulations. On the other hand, ordinary machinery, electric machinery and precision instruments depend greatly on exports to Japan or other countries. An empirical study by Urata and Kawai (2000) showed the importance of cumulative FDI. The authors examined the determinants of Japanese FDI in developed economies, developing economies and Asia using data from 1980 to 1994 for textiles, general machinery, electric machinery and transport machinery. The authors found that low wage rate is a very important determinant of Japanese FDI in Asia. The authors also found that market size, i.e. GDP, is also important in Asia for large ﬁrms involved in textiles and for those in electric and transport machinery. Cumulative FDI undertaken by Japanese ﬁrms is also important, with the exception of small- and medium-sized enterprises (SMEs) in textiles. Considering that SMEs provide parts and components to large ﬁrms, because (1) large ﬁrms correlate with large markets and (2) large ﬁrms and SMEs correlate with cumulative FDI in the electric and transport machinery industries, we can regard electronic and transport machinery industries as matching the model used for simulations. The authors also pointed out that exchange rate volatility, availability of skilled labor, capability of generating electricity and governance can also be signiﬁcant determinants. In sum, we can see that (1) Japanese MNEs in the automobile equipment industry exhibit a clear core – periphery structure, (2) local markets are important for Japanese MNEs in the automobile industry, and (3) Japanese MNEs in the transport machinery industry are attracted to cumulative FDI. Therefore the transport equipment industry matches the model used for the simulations presented in the next section. 9.3.2

China

The distribution of the number of Japanese MNEs is not even in China in each industry. Of all Japanese MNEs operating in China in 2002, about 23 percent were in electrical and electronic equipment, about 15 percent in chemicals and pharmaceuticals, about 13 percent in textiles, about 12 percent in machinery, about 7 percent in foods and another 7 percent in automobiles.18 These industries account for about 70 percent of all Japanese MNEs in China. The remainder of this sub-section will focus on the location of Japanese MNEs in China and the determinants inducing this location.

259

Location choices of Japanese MNEs

Table 9.2

The Composition of sales for Japanese MNEs in 2003

Transport equipment Food and tobacco Chemicals Non-ferrous metals Metal products Textiles Pottery and clay Ordinary machinery Electric machinery Precision instruments

China (inc. Hong Kong)

ASEAN-4

NIE-3

A

B

A

B

A

B

0.09 0.14 0.59 0.45 0.49 0.84 0.26 2.03 0.82 1.95

0.1 0.2 0.22 0.17 0.14 0.24 0.08 1.27 1.16 0.25

0.08 0.28 0.21 0.47 0.69 0.32 1.04 2.67 1.98 6.38

0.17 0.47 0.46 0.65 0.34 0.9 0.86 0.99 3.22 1.57

0.05 0.1 0.04 0.24 0.14 0.01 0.45 0.35 0.72 5.69

0.09 0.05 0.67 0.35 0.57 0.28 0.62 1.01 0.73 0.71

Note: A is exports to Japan divided by local sales; B is exports to third countries divided by local sales. Source:

Quarterly Survey of Overseas Subsidiaries.

The location and the number of Japanese MNEs in the coastal region at the provincial level are presented in Figure 9.5. The six industries listed along the horizontal axis are divided into two groups. One group consists of electrical and electronic equipment, chemicals and pharmaceuticals, and automobiles and auto parts; these industries do not have outliers expressed by small circles around the location names,19 meaning that the location of these industries is distributed relatively evenly in the coastal region. IT and communication equipment, which fall under electronic equipment, are also distributed relatively evenly. However, Belderbos and Carree (2002) showed that the number of existing Japanese electronics aﬃliates is one of the major determinants for establishing Japanese electronics aﬃliates in China. This suggests that Japanese MNEs in the electronics industry tend to concentrate. Also, the pattern in the location of Japanese MNEs in China’s automobile industry is not as heterogeneous as that in East Asia. The automobile industry will be examined in more detail in the next section. The second group of industries consists of textiles, machinery and foods. These have outliers, meaning that the location of the industries is distributed relatively unevenly. Japanese MNEs in the textile industry have agglomerated in the Shanghai area. Shanghai had an active textile industry historically because cotton could be cultivated in the Shanghai hinterland (Fujii 1979).

260

Case studies of East Asian economic integration

100

Guangdong

Shanghai

Shanghai

80

Shanghai

Shanghai Jiangsu

60

Jiangsu 40 Shandong Shanghai

20

Guangdong

0 I1

I2

I3

I4

I5

I6

Note: I1electrical and electronic equipment; I2 chemicals and pharmaceuticals; I3 textiles; I4machinery; I5 foods; I6 automobiles and auto parts. Source:

Kaigai Sinsyutu Kigyo Soran.

Figure 9.5

The number and location of Japanese MNEs in China in 2002

None of the industries in China examined above show (1) a clear core–periphery structure in the location of Japanese MNEs or (2) that the local market is important. However, it can be seen that textiles, machinery and food industries are located heterogeneously and that the local market is relatively important in electrical and electronic equipment, chemical and transport equipment industries.

9.4

IMPACT OF TARIFF REDUCTIONS ON THE LOCATION OF THE AUTOMOBILE INDUSTRY IN EAST ASIA

From the analysis in the above section, it seems that the transport equipment industry, at least at the country level, matches the model used for simulations. Therefore the automobile industry will be used for simulations. This section will ﬁrst look at the location of automobile production and car parts plants in East Asia, then conduct simulations to examine the impact

Location choices of Japanese MNEs

261

of tariﬀ reduction in attracting ﬁrms to a location. The model used here is very similar in that used in Chapter 8. 9.4.1

Background: The Japanese Automobile Industry in East Asia

The automobile industry has become located heterogeneously in various regions within East Asian countries. The concentrations of automobile production and parts manufacturers have grown over time. Figure 9.6 shows how the volume of automobile production by Japanese MNEs changed from 1992 to 2003. Production increased signiﬁcantly in Thailand and China. During 1997/98 production in Thailand dropped to one-third that of 1996 because of the East Asian ﬁnancial crisis, but it recovered by 2002 and rose to approximately double that of Malaysia, Indonesia and Taiwan in 2003. Meanwhile production in China increased dramatically from 1999. In Taiwan and the countries of Southeast Asia, there was no substantial change in their production volumes. Figure 9.7 shows the volume of automobile production by Japanese foreign aﬃliates in China.20 It can be seen that production increased rapidly in South China from 1999. Guangzhou Honda began production from 1998, and production by Guangzhou Dongfeng Motors also got on track. Guangzhou Toyota began production in 2006, thereby further increasing production in Guangzhou, as pointed out in Chapter 3. Meanwhile, production also continued to increase in China’s northeast. Looking at Japanese MNEs producing motor vehicle parts, these are also located heterogeneously. Figure 9.8 clearly shows that parts plants continued to increase in Thailand and China. Figures 9.6 and 9.8 show that Thailand and China are the two countries that have strongly attracted Japanese auto industries. From Figures 9.7 and 9.8 it also can be seen that the number of parts ﬁrms in Vietnam is large, considering the volume of automobile production. Because parts production is more labor intensive than automobile production, the low cost of labor in Vietnam is attractive to parts manufacturers. Figure 9.9 shows the location of Japanese MNE parts plants in China. Parts plants are concentrated in Shanghai and East China. According to Fourin (1998), the total production of all parts manufacturers in Shanghai in 1997 was twice that in any other province or city. Total automobile sales in Shanghai in 2003 were 600 000 vehicles, making it the largest market in China.21 However, these sales were less than double those in any other province or city. This phenomenon is a good example of the HME in spatial economics. The larger home market has a more than proportionately larger number of parts manufacturers. This shows that the concentration of automobile assemblers is precisely what attracts parts

262

Case studies of East Asian economic integration

1200000 1000000 800000

China Thailand Malaysia Taiwan Indonesia Philippines

600000 400000 200000

96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

94

93

95

19

19

19

19

19

92

0

Sources: Fourin Monthly Kokunai Jidousha Chosa Vol. 2, No. 61; China Automotive Industry Yearbook.

Figure 9.6

Volume of car production by Japanese MNEs in East Asia

manufacturers. Therefore it is natural that Japanese MNEs producing parts are attracted to the agglomeration of the automobile industry in Shanghai. In South China the number of Japanese MNEs producing motor vehicle parts is also large and is increasing annually. This corresponds with the increase in car production in South China. Hence, in South China, as in Thailand, agglomeration has occurred by the interaction between assembly plants and parts plants. In Thailand automobile production together with parts plants increased rapidly, creating the agglomeration of automobile industries. As a result, the typical characteristics of spatial economics emerged. Pickup trucks accounted for 60 percent of new car sales in Thailand in 2001.22 These trucks are used not only as an alternative to passenger cars but also as public transportation. Because the demand in the local market is large, pickup trucks are produced in more than proportionately larger numbers. Thus about 40 percent of the total production is exported23 to North America, South America, Europe, the Middle East and the Asia-Paciﬁc. Toyota started an IMV project centered on Thailand, aiming for worldwide optimal production and supply in more than 140 countries around the world. Toyota manufactures diesel engines in Thailand, gasoline engines in Indonesia, and specialized parts in the Philippines, Taiwan and India. In this project the company produces pickup trucks in Thailand and vans in Indonesia. From

263

Location choices of Japanese MNEs 400000 350000 300000

South China Northeast China North China East China Southwest China Vietnam

250000 200000 150000 100000 50000

Sources:

19

19

92

93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

0

Fourin Monthly Kokunai Jidousha Chosa Nos 2 and 61.

Figure 9.7

Volume of car production by Japanese MNEs in China

Thailand it sends knock-down car kits to Argentina and South Africa. We can presume that Toyota chose Thailand as the center for its projects because of the large number of supporting industries in the country. 9.4.2

Simulations

In this sub section we investigate the impact of tariﬀ reduction on the attractiveness of locations for car assemblers and parts manufacturers. In order to illustrate the theoretical eﬀects in heterogeneous East Asian countries and Chinese provinces in an uncluttered setting, we will use a partial equilibrium model. We compare the short-term operational proﬁts of candidate locations before and after tariﬀ reduction. As used here, operational proﬁt means the remainder after subtracting ﬁxed costs from proﬁts. The model The model used here is very similar to that used in Chapter 8. To avoid redundant information, the explanation here will be limited mainly to the modiﬁed parts of the model. The representative consumer consumes diﬀerentiated automobiles and a numeraire good with Dixit–Stiglitz preferences. The demand for a variety of automobiles is derived by maximizing the consumer’s utility.

264

Case studies of East Asian economic integration

300 250 China Thailand Indonesia Taiwan Philippines Malaysia Vietnam

200 150 100 50

90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

19

19

89

0

Note: The effects of mergers between Japanese firms and acquisitions by foreign firms were adjusted using financial statements and company websites. Sources: Annual editions of Japan Automobile Parts Industry; annual editions of Kaigai Shinshutsu Kigyo Souran.

Figure 9.8 The number of Japanese MNE parts manufacturers in East Asia There are two sectors (car assemblers and parts manufactures) and a numeraire-good sector. There are a number of monopolistically competitive car assemblers and monopolistically competitive parts manufacturers. Each plant is expressed as a ﬁrm. All ﬁrms possess increasing returns to scale technology. The model in Chapter 8 requires that labor be ﬁxed and variable costs in downstream ﬁrms, whereas the model in this chapter has labor only as ﬁxed costs in downstream ﬁrms.24 The inputs for car assemblers are diﬀerentiated parts at variable costs and labor at ﬁxed costs. Thus the production function for assemblers is as follows: AD k

J

j1

njFj 1 ,

where AD k expresses the total quantity of each variety of goods assembled in region k for consumption; nj is the number of the variety for parts produced in region j; Fj is the quantity of each variety of parts manufactured

265

Location choices of Japanese MNEs 120 100 80

East China South China North China

60

Southwest China Northeast China

40

Northwest China 20

91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03

90

19

19

19

89

0

Note: The effects of mergers between Japanese firms and acquisitions by foreign firms were adjusted using financial statements and company websites. Sources: Annual editions of Japan Automobile Parts Industry; annual editions of Kaigai Shinshutsu Kigyo Souran.

Figure 9.9

The number of Japanese MNE parts manufacturers in China

in region j; and is the assembler’s substitution parameter for a pair of parts (0 1). The total labor requirement in a downstream ﬁrm is the same as the ﬁxed labor requirement. The movement of goods incurs additional costs. Transport costs are calculated as follows: T 1 distancead valorem tariﬀ.

(9.1)

The transport costs of parts and automobiles are respectively expressed as TF (j,k) produced in region k and demanded in region g, and as TA (k,g) assembled in region j and consumed in region k. The delivered price in region g for ﬁnal goods assembled in region k is calculated as PA TA (k,g) . k Similarly, the delivered price for parts is PF TF (j,k) . j Although tier-4 suppliers are mentioned in Chapter 4, parts manufacturers are supposed to be a tier-1 supplier. Because assemblers minimize their variable costs, we can calculate the total demand for a diﬀerentiated part. Workers are drawn from a suﬃciently large numeraire-good sector. We

266

Case studies of East Asian economic integration

also assume that the labor supply is perfectly elastic. Thus the wage rates in the assembler sector and parts sector become the same as the wage rate in the numeraire-good sector.25 Assemblers and parts makers set their prices according to the monopolist’s rule. The prices of parts diﬀer due to three types of costs: wage, tariﬀ for parts and distance to assemblers, whereas the prices of cars diﬀer for four types of reasons: the price of diﬀerentiated parts, the number of parts manufacturers in each regions, the tariﬀ on cars and the distance to consumers. The proﬁt functions of parts manufacturers and assemblers are expressed respectively as (8A1.14) and (8A1.15) in Chapter 8. Finally, we suppose that the value of the parameters among regions are the same for the rate of consumption of automobiles and for the marginal labor requirement for parts manufactures. The impacts of tariﬀ reduction To interpret the results of the simulations, we examine four cases, focusing on one region in a multi-regional setting. The four cases are (1) the impact of reducing car tariﬀs on car assembly ﬁrms, (2) the impact of reducing car tariﬀs on auto-parts ﬁrms, (3) the impact of decreasing auto-parts tariﬀs on auto-parts ﬁrms, and (4) the impact of decreasing auto-parts tariﬀs on car assembly ﬁrms. It becomes easier to explain these four cases by following the eﬀects introduced in Chapter 8, Section 8.2; the impacts of tariﬀ reduction are related to market access (market size), cost of inputs and market crowding. If market size is large or market penetration is enough because of better market access, or if products become cheaper because of the lower cost of inputs, or if the price index is large because of mild market crowding, ﬁrms become more proﬁtable. In every case, the impact becomes larger as the scale of tariﬀ reduction increases. Our ﬁrst examination is the impact of a decrease in tariﬀ on cars on the operational proﬁts of car assembly ﬁrms. Improved market access increases operational proﬁts, whereas severe market crowding reduces them. The former eﬀects become stronger as income increases and as assemblers concentrate in the market. The latter eﬀects become stronger if assemblers do not concentrate in the market and the price of cars becomes higher because of milder competition among parts manufacturers. Therefore there is a tendency for the operational proﬁts of assemblers to increase when car tariﬀs are reduced as car assemblers concentrate more in a region. The second examination is the impact of a decrease in car tariﬀs on the operational proﬁts of auto-parts ﬁrms. The impact on the operational proﬁts of these ﬁrms is ambiguous because the ﬂuctuation of their operational proﬁts depends on the demand for automobiles. The eﬀects of a car

Location choices of Japanese MNEs

267

tariﬀ decrease on the demand for cars are ambiguous because of the inverse eﬀects of market access and market crowding. When the demand for cars in the market increases, the operational proﬁt of parts manufacturers also increases. The scale of this increase depends on other factors. The scale increases (1) when many car assembly ﬁrms locate in the market, (2) when market access for parts ﬁrms in the region is better, (3) when parts manufacturers concentrate in the market, and (4) when the wage rate at the production site is lower. The third examination is the impact of a decrease in parts tariﬀ on the operational proﬁts of auto-parts ﬁrms. The eﬀects are the same as in the case where a decrease in car tariﬀ aﬀects the operational proﬁts of car assembly ﬁrms. We only need to replace the car assembly ﬁrms with parts ﬁrms. Our ﬁnal examination is the impact of a decrease in parts tariﬀ on the operational proﬁts of car assembly ﬁrms. Because the price index of car parts and that of cars decreases with a decrease in tariﬀ on parts, improved cost-of-inputs because of the lower price index of car parts makes car assembly ﬁrms more proﬁtable, whereas severe market crowding because of the lower price index of cars reduces operational proﬁts. The eﬀects from cost-of-inputs increase more when the goods are not sold in the home market and transport costs of cars are high. As local assembly ﬁrms grow larger, the negative impact from market crowding becomes weaker. We have examined operational proﬁts in a market. We now need to sum the change in these regional operational proﬁts mentioned above for many regions. To understand the whole picture, we need the following simulations. Results of the simulations We examine which regions become more attractive to Japanese car assemblers and parts manufacturers because of tariﬀ reduction. Suppose that ﬁrms have no incentive to move under the spatial distribution of plants at the baseline, we consider that a region becomes more attractive to Japanese car assemblers and parts manufacturers when their operational proﬁts in that region increase by tariﬀ reduction. When both types of ﬁrms become more proﬁtable after tariﬀ reduction, there is a possibility that their numbers will increase in the region. However, because of the market-crowding eﬀects explained in Chapter 8, Section 8.2, their location patterns may not change. To understand how the automobile industry has come to be located in the diﬀerent regions of East Asia, we need some further examination. In our model we use the provinces and cities of China (excluding Xizang and Qinghai), the NIEs, ASEAN and Japan as proxies for production/

268

Case studies of East Asian economic integration

market points. Distribution costs within each region are ignored. The data and the derivation of the substitution parameters are presented in the Appendix. We use the same parameters in all regions for the consumption share of automobiles and for the marginal labor requirements. Thus these parameters are included in a common term in all regions. Therefore they do not aﬀect the relative amount of operational proﬁts. As a result, we omit these parameters in the simulations. At the baseline, the proﬁts of assembly ﬁrms become positive in all regions. The regions where assembly ﬁrm operational proﬁts are large are (in descending order) Japan, South Korea, Malaysia, Taiwan, Guangdong and Jiangsu. This result can be explained by the fact that Thailand is not such a proﬁtable location at the baseline. This may be because we restrict the automobile trade to East Asia. The regions where the operational proﬁts of parts manufacturers is large are (in descending order) Guizhou, Vietnam, Gansu, Yunnan and Guangxi. This shows that low wage rates bring high operational proﬁts. We examine operational proﬁts using the following three simulation scenarios. In the ﬁrst, we reduce the tariﬀs in ASEAN on both automobiles and auto parts to zero. Under the ASEAN free trade agreements, the ASEAN country will reduce its tariﬀs among member countries to zero until 2015. In the second scenario, we reduce the tariﬀ on parts to zero but leave the tariﬀ on automobiles unchanged in all regions. In the third scenario, we reduce the tariﬀ on automobiles and parts to zero in all regions. The results are summarized in Table 9.3. The results of the ﬁrst scenario, as shown in Table 9.3, indicate that both car assembly ﬁrms and parts ﬁrms have the incentive to move to Cambodia. There are three reasons why Cambodia attracts assembly ﬁrms. First, Cambodia is in the group that sets high tariﬀ rates on parts. Second, the competition in Cambodia is moderate due to the absence of assembly ﬁrms and the small number of countries exporting to Cambodia. Finally, Cambodia is located near Thailand, which has a large agglomeration of parts manufacturers. The results of the second scenario indicate that both car assembly ﬁrms and parts ﬁrms have the incentive to move to Cambodia, the Lao PDR, Malaysia, Thailand, Sichuan, Chongqing, Yunnan and Xinjiang. The results show that operational proﬁts improve when wage rates are not so high and/or parts manufacturers are concentrated in or near the region. The results of the third scenario indicate that both car assembly ﬁrms and parts ﬁrms have the incentive to move to Cambodia, Japan, Indonesia, the Philippines, Singapore, Thailand, Vietnam and Hong Kong. Thailand is attractive because a large number of car assemblers and parts manufacturers exist and wage rates are not so high. But the

269

Location choices of Japanese MNEs

Table 9.3

Effects of tariff reduction on operational profits Tariff on both in ASEAN

Cambodia Indonesia Japan Lao PDR Malaysia Myanmar Philippines Singapore South Korea Taiwan Thailand Vietnam Hong Kong Beijing Jilin Hunan Sichuan Chongqing Fujian Gaungdong Guizhou Zhejiang Heilongiang Anhui Inner Mongolia Shandong Yunnan Gansu Jiangxi Jiangsu Guangxi Shanghai Liaoning Hebei Shanxi Tianjin Hubei Shaanxi Henan Hainan Xinjiang

Tariff on parts in all regions

Tariff on both in all regions

Assembler

Parts firms

Assembler

Parts firms

Assembler

Parts firms

+ + + + + + +

Note: The bold plus signs indicate that operational profits for both assembly firms and auto-parts firms increase with tariff reduction.

270

Case studies of East Asian economic integration

agglomeration of the automobile industry in Thailand has created harsh competition in the domestic market. Under such circumstances the eﬀects on tariﬀ reduction, which face ﬁrms more harsh competition in the local market, are relatively mild in the country where competition is initially intense.

9.5

CONCLUSION

The main purpose of this chapter has been to examine from the spatial economic perspective the location choice of Japanese MNEs in the automobile industry in East Asia. The theoretical simulations calculating the potential operational proﬁts of automobile and parts manufacturers using three scenarios of tariﬀ reduction indicated that Cambodia and Thailand are the two most promising locations for Japanese MNEs in the automobile industry. It might seem odd that Cambodia should attract Japanese MNEs in all three scenarios of tariﬀ reduction within ASEAN for both parts and assembled cars, tariﬀ reduction within East Asia for parts, and tariﬀ reduction within East Asia for both parts and assembled cars. These simulation results are based purely on spatial economic factors: market access (market size), cost of inputs and market crowding. Only tariﬀ rate and distance are used to express transport costs. The diﬀerence of wage rates among countries is also considered. Some important conditions such as the level of infrastructure, human capital and non-tariﬀ barriers have been omitted. The result under these conditions is the unlikely attraction of Cambodia as an advantageous location for the Japanese automobile industry. In other words, the results of the simulations suggest that Cambodia becomes an advantageous location for the Japanese automobile industry if these omitted conditions in Cambodia stand comparison with those in other countries. Thailand is also attractive to Japanese MNEs in two of the scenarios, that of tariﬀ reduction within East Asia for auto parts, and tariﬀ reduction within East Asia for both auto parts and assembled cars. Thailand’s hope of becoming a manufacturing hub for small cars in addition to pickup trucks (Vaidya 2006) may be fulﬁlled in the coming era of free trade in the automobile industry beyond AFTA that will cover all East Asia. Concerning the evolution of Japanese automobile MNEs in East Asia, the second scenario, that of tariﬀ reduction within East Asia only on auto parts, will push Japanese MNEs further toward locating in other countries in East Asia because the operational proﬁts of both parts and car manufacturers in Japan decrease and the incentive for Japanese ﬁrms to have plants abroad increases. In the last scenario of tariﬀ reduction within East

Location choices of Japanese MNEs

271

Asia for both parts and assembled cars, the operational proﬁts of both types of ﬁrms increase in Japan and make the country one of the strongest in the automobile industry in East Asia. The simulations in this chapter using a spatial economic model are useful when considering diﬀerent countries’ location advantages, but there are some problems. First, the analysis was carried out only on the automobile industry within a partial equilibrium framework, and we focused only on the attractiveness of locations for car assembly ﬁrms and parts manufacturing ﬁrms. On this basis it is not possible to predict the conﬁguration of locations where the automobile industry in East Asia will ultimately become concentrated. Furthermore, we cannot apply the results of the simulations to other industries because the location pattern of the automobile industry in East Asia is clearly diﬀerent from that of other industries. These problems need to be addressed in future research.

NOTES 1. 2. 3.

4.

5. 6. 7. 8.

9. 10. 11. 12. 13. 14.

Source: Overseas Business Activities of Japanese Companies 2003. Source: RIETI FDI database. Pupphavesa and Pussarungsri (1994) showed that there are multiple channels and levels of technology transfer through subcontract arrangements between MNEs and local ﬁrms in Thailand. They noted that (1) direct technology transfer such as training and advice on plant layout is at a low or medium level, (2) spillover type of technology transfer such as product design and exposure to the managing or organizing of manufacturing activity is at a medium or high level, and (3) learning facilitation such as testing and diagnostic feedback on quality or advanced indications of future targets is at a high level. Some regulations, such as local-content requirements and export-performance requirements, are supposed to be eliminated in the agreement on trade-related investment measures negotiated in the Uruguay Round. Thus theoretical models on MNEs which assume no or little intervention from the host country become more realistic. The focus in this chapter has been narrowed to the ASEAN-4, consisting of Indonesia, Malaysia, the Philippines and Thailand, because not enough data are available on the other ASEAN countries. Source: RIETI FDI database. Source: Overseas Business Activities of Japanese Companies 2003. China is the main host country of Japanese MNEs in East Asia. But Japanese MNEs are not concentrated in China to the degree that Korean FDI is, as pointed out in Chapter 8. Of total South Korean FDI in East Asia from 1968 to September 2006, China received over 80 percent. Various host country determinants aﬀect where MNEs choose to locate, and these are covered in UNCTAD (1998). However, in this chapter we are interested in the economic determinants. Source: RIETI FDI database. Source: ibid. Source: ibid. Source: Electric Home Appliance Industry Handbook 2006, published by the Association for Electric Home Appliances. Source: Korea Export/Import Bank.

272 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

25.

Case studies of East Asian economic integration Source: Distance table for world shipping edited by the Japan Shipping Exchange. Source: Electric Home Appliance Industry Handbook 2006, published by the Association for Electric Home Appliances. Smaller values in the table also show that transport costs in the industry are higher. Source: Kaigai Shinsyutu Kigyou Soran 2003. Source: Overseas Business Activities of Japanese Companies 2003. Automobile production ﬁgures are larger than those in Figure 9.6 because companies producing automobiles designed in Japan were categorized as Japanese foreign aﬃliates in the China Automotive Industry Yearbook 2003. Source: ibid. Source: Thailand’s Automotive Industry, edited by The Brooker Group PLC. Source: ibid. Initially in the model labor was set as a variable cost. But the car assemblers’ share of expenditure for labor becomes negative when using the estimation presented in the Appendix because high-wage countries such as Japan and Korea mainly export automobiles. Thus, for the purposes of the model, automobile ﬁrm labor costs are regarded as ﬁxed. Because tariﬀ rates do not aﬀect wage rates under the assumptions, we can say that the wage rate is constant, meaning that the proﬁts of assembler ﬁrms and parts ﬁrms do not change through their ﬁxed costs. This assumption simpliﬁes the calculations.

REFERENCES Ando, Mitsuyo and Fukunari Kimura (2003), ‘The formation of international production and distribution networks’, in Takatoshi Ito and Andrew K. Rose (eds), International Trade in East Asia, Chicago, IL: University of Chicago Press, pp. 176–216. Belderbos, René and Martin Carree (2002), ‘The location of Japanese investments in China: agglomeration eﬀects, keiretsu, and ﬁrm heterogeneity’, Journal of the Japanese and International Economies, 16 (2), 194–211. Belderbos, René and Leo Sleuwaegen (1996), ‘Japanese ﬁrms and the decision to invest abroad: business groups and regional core networks’, The Review of Economics and Statistics, 78 (2), 214–20. Bosworth, Barry P. and Susan M. Reinhat (1999), ‘Capital ﬂows to developing economies: implications for saving and investment’, Brookings Papers on Economic Activity, 1, 143–80. Cheng, Shaoming and Roger R. Stough (2006), ‘Location decisions of Japanese new manufacturing plants in China: a discrete-choice analysis’, The Annals of Regional Science, 40, 369–87. Dai, Erbiao (1998), ‘The transﬁguration of regional economic structure in China and FDI’ (in Japanese), Working Paper Series Vol. 98–20. Fujii, Mitsuo (1979), ‘The advancement of textile industry and the evolution of cotton spinning ﬁrms in China’ (in Japanese), in Mitsuo Fujii, Toshikazu Nakase, Keiya Maruyama and Masataka Ikeda (eds), The Historical Evolution of Japanese Multinational Enterprise. Fourin (1998), Fourin Jidousya Chosa Geppou, No. 159. Kim, Ho Yeon (2003), ‘Impact of trade liberalization on the location of ﬁrms: NAFTA and the Automobile Industry’, The Annals of Regional Science, 37, 149–73.

Location choices of Japanese MNEs

273

Marwah, Kanta and Akbar Tavakoli (2004), ‘The eﬀect of foreign capital and imports on economic growth: further evidence from four Asian countries (1970–1998)’, Journal of Asian Economies, 15, 399–413. Puga, Diego and Dan Treﬂer (2005), ‘Wake up and smell the ginseng: international trade and the rise of incremental innovation in low-wage countries’, NBER Working Paper 11571. Pupphavesa, Wisarn and Bunluasak Pussarungsri (1994), ‘FDI in Thailand’, Proceedings of AT9 researchers’ meeting. Tomiura, Eiichi (2003), ‘Foreign direct investment into Asia and domestic R&D Intensity of Japanese manufacturers: ﬁrm-level relationship’, Proceedings of the International Conference on New Development in Asia Paciﬁc Region: Economic Analysis using Microdata. UNCTAD (1998), World Investment Report 1998: Trends and determinants: Geneva. Urata, Syujiro and Hiroki Kawai (2000), ‘The determinants of the location of foreign direct investment by Japanese small and medium-sized enterprises’, Small Business Economics, 15 (2), 79–83. Vaidya, Vivek (2006), ‘Thailand’s attractiveness as a small-car hub’, Bangkok Post, 3 July. Zhou, Dongsheng, Shaomin Li and David K. Tse (2000), ‘The impact of FDI on the productivity of domestic ﬁrms: the case of China’, CMC Working Paper CMC2000-008-001.

274

APPENDIX

Case studies of East Asian economic integration

THE DATA AND THE DERIVATION OF PARAMETERS FOR THE MODEL

Wages and regional incomes are represented respectively by per capita GDP and GDP in 2003, and taken from the IMF World Economic Outlook and the China Statistical Yearbook. The number of Japanese, foreign and local automobile assemblers is represented by the numbers of plants reported in Ward’s World Motor Vehicle Data 2003 for both Japanese and non-Japanese foreign aﬃliates. For the number of parts manufacturers, the ﬁgure used is that reported in Kaigai Shinsyutu Kigyo Souran 2003 for the number of Japanese foreign aﬃliates. For the number in Japan, the number of member companies in the Japan Auto Parts Industries Association is used. Transport costs are composed of ad valorem tariﬀ rate and freight. The tariﬀ rate data are from the UNCTAD trade analysis and information system. The simple average is used as of 2003 for 8708, 840731, 840732, 840733, 840734, 840790, 840820 in the HS code for parts manufacturers; and 870321, 870322, 870323, 870324, 870331, 870332, 870333, 870392, 870421, 870431 for car assemblers. When the data for 2003 have not been available, the average of the year before 2003 and after 2003 has been used. Following calculation, the lowest resulting tariﬀ rate has been chosen among tariﬀ measures. The freight transport costs have been calculated by multiplying the freight rate per unit per mile by the distance. Distances are represented by the distance between capital cities. For distances from landlocked regions in China, the distance used is that via the nearest coastal region in China. The freight rate per unit per mile is derived from export and import data between Japan and Taiwan in 2003, taking into consideration that export data are recorded in f.o.b. prices while import information is available on a c.i.f. basis. Therefore, for parts and for automobiles in equation (9.1) of the model are respectively 0.000048 and 0.000036. Japan and Taiwan have been chosen because the unit is the same between the countries. Trade data on Taiwan are derived from World Trade Atlas and those on other countries are derived from UN COMTRADE. Trade partners in the simulations are ﬁxed using the above trade data. The values of parameters for the degree of substitution are recursively estimated for Japan, China, the NIEs and the ASEAN-4 in 2003 for cars and auto parts, following the method adopted by Kim (2003). The grouping of cars and auto parts in the trade data is the same as that used for the tariﬀ data. The ﬁgure used for the number of auto-parts manufacturers is from the industrial statistics database produced by UNIDO in 3420 and 3430 of the SITC code in the closest year to 2003, Chinese Automobile

Location choices of Japanese MNEs

275

Industry Yearbook 2004, and the report on the 2001 census of industry, commerce and services in the Taiwan – Fuchien area. From the regressions, the elasticities of substitution for auto parts and automobiles were derived as 0.198075 and 0.764817 respectively.

10.

Economic opening and industrial agglomeration in China Zhao Chen, Yu Jin and Ming Lu

10.1

INTRODUCTION

Since the launching of economic reform in late 1978, China has been integrating into the world economy at a pace as remarkable as its economic growth. While the role of economic opening in growth and inequality in China has widely been studied (Kanbur and Zhang 2005; Wei and Wu 2001; Zhang and Zhang 2003; Wan et al. 2007), little has been published on how economic opening and industrial agglomeration are related in China. Industrial agglomeration is signiﬁcant for economic eﬃciency and regional development, and given the fact that regional disparity is now a great concern in China, a clearer understanding of the pattern and causes of industrial agglomeration would have great implications for policy-makers. This chapter sheds some new light on how Chinese industries have been gradually relocating and concentrating under the impact of economic forces and relevant policies, particularly economic opening. Fujita and Hu (2001) and Kim and Knaap (2001) investigated some potential factors related to China’s industrial agglomeration; however, neither of these studies did further econometric work. Wen (2004) employed the theory of new economic geography to investigate the determinants of industrial agglomeration in China in 1992. Unlike Wen’s work, this chapter uses provincial panel data to ﬁnd out and investigate what factors contributed to industrial agglomeration in China, keeping an eye not only on new economic geography factors but also those of physical geography and government policies, especially economic opening. Second, this study also relates to the heated debate on globalization and inequality. Diﬀerent theories and empirical analyses often reach contrary results about whether trade and economic liberalization lead to greater concentration of economic activity or greater disparity (Krugman 1993; Venables 1998; Sachs and Warner 1995). The experience in China and other transitional economies shows that globalization leads to increases in inequality. Wan et al. (2007) provide the evidence from China that 276

Economic opening and industrial agglomeration in China

277

economic opening results in the dispersal of international trade and FDI, and thus regional disparity. Since a higher regional share of industrial GDP always leads to a higher share of GDP (Wen 2004), the present study suggests that trade may increase inequality by contributing to industrial agglomeration, a point less emphasized in the existing literature. Finally, this study contributes to the literature through its depiction of industrial agglomeration in China and exploring its causes using provincial panel data from 1987 to 2001. More speciﬁcally, this work provides empirical evidence in support of the new economic geography theory. Since Krugman (1991), the framework of new economic geography has come into being where standard building blocks of mainstream economics are used to model the trade-oﬀ between dispersal and agglomeration. New economic geography models stress the importance of transportation costs, the HME and backward and forward linkages (Fujita 1988; Krugman 1991). They have been able to explain agglomeration in a theoretical framework that is tractable, has solid micro-foundations, and makes testable empirical predictions. However, only a small number of empirical studies build directly on the new economic geography framework. For example, Combes and Lafourcade (2000) used a data set from France and found that when allowance is made for intermediate inputs, transport costs have a signiﬁcant impact on specialization patterns in nearly all sectors. The remainder of this chapter is as follows. Section 10.2 presents some stylized facts of economic opening and industrial agglomeration in China since 1987. Section 10.3 presents a case study of industrial agglomeration in the Yangtse Delta and a framework for analyzing three groups of factors contributing to industrial agglomeration: economic policies, economic geography and new economic geography respectively. Section 10.4 provides an explanation of the econometric model and data used in this study, together with the empirical results. Section 10.5 concludes with a discussion of policy implications.

10.2

ECONOMIC OPENING AND INDUSTRIAL AGGLOMERATION IN CHINA1

Since China launched its economic reforms in late 1978, its economy has gradually opened up to the rest of the world with increasing foreign capital inﬂow and international trade. Meanwhile, China’s industries have adjusted from a political-oriented pattern to a market-oriented one, and a trend of industrial agglomeration has come into existence since the reforms.

278

10.2.1

Case studies of East Asian economic integration

Growing International Trade

China’s international trade before 1979 was under central government planning, which controlled more than 90 percent of trade by monopolizing the imports and exports of over 3000 kinds of commodities. However, with the advent of reforms, international trade was liberalized step by step until by 1991 almost all exports had been deregulated, with only 15 percent controlled by specially appointed trading companies. Imports were also deregulated. The proportion of plan-commanded imports in the total volume of imports was reduced from 40 percent in 1985 to 18.5 percent in 1991.2 By 1994 almost all planning on imports and exports had been abolished. In pre-reform China, tariﬀs were high and represented the only form of protection. When China initiated signiﬁcant trade reforms in 1992, tariﬀ rates remained high, averaging 44.05 percent. Since 1992 China has cut its tariﬀ rates substantially every year. The average tariﬀ rate fell to 17.1 percent in 1998 (Yin 1998, p. 126). On the other hand, non-tariﬀ barriers were introduced in the early 1980s, and subsequently an increasing number of goods were placed under licensed trading and quotas. In 1992 some 25 percent of imports and 15 percent of exports were managed under licenses. However, the scope of license and quota management has been narrowed down since 1992. By 1997 only 384 categories of imports, a mere 5 percent of the total, were managed under quotas and licenses (Yin 1998, p. 129). Trade liberalization has stimulated remarkable growth in both exports and imports. The trend of growth was maintained even during the Asian ﬁnancial crisis. In 1978 China ranked 32nd in the world in terms of international trade. That ranking improved to 15th in 1989, 10th in 1997 and 6th in 2001. The ratio of international trade to GDP also rose from 9.80 percent in 1978 to as high as 49.03 percent in 2002. In 2002 total trade exceeded US$600 billion, representing about 50 percent of China’s GDP.3 This places China as the ﬁfth-largest trader in the world. The same has also happened with exports. The ratio of exports to GDP rose from 4.62 percent in 1978 to 25.72 percent in 2002 (see Figure 10.1). Since the mid-1980s, exports of manufactured goods have accounted for an increasingly larger share, while the corresponding share for imports has declined, though at a slow rate. Clearly, China has been industrializing and has become a major exporter of manufactured commodities. 10.2.2

Increasing Cross-border Capital Flows

In 1979 three Special Economic Zones (SEZs) were set up in Guangdong for attracting FDI.4 However, not until 1984 did FDI start to pour in, when 14 Coastal Open Cities and ten Economic and Technological Development

Economic opening and industrial agglomeration in China

279

% 60 50 40 30 20 10 0 1978 1980 1985 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year Trade/GDP

Figure 10.1

Export/GDP

Ratio of trade (and exports) to GDP

Zones were opened. Since that time, a growing number of SEZs have been developed to attract FDI and technology transfer and to enhance exports. The second wave of FDI inﬂow occurred in 1992 when Deng Xiaoping made his well-known tour of South China. For many years China was the largest recipient of FDI among developing countries, and the second largest in the world from 1993, following the USA. In 2002 it became number one in the world, attracting US$52.743 billion in FDI. The ratio of FDI to GDP was more than 4 percent in 2002. FDI in China is closely related to imports and exports since many foreignowned ﬁrms are export-oriented and sometimes involved in importing from overseas markets. Meanwhile, a large amount of foreign loans has been utilized in various areas of development.5 Also, China has seen an impressive growth in capital outﬂows in recent years, owing to the rapid growth of domestic enterprises. China’s investment abroad nearly tripled, from US$2562.49 million in 1997 to US$6885.398 million in 2001. 10.2.3

Further Opening up after WTO Accession

Since becoming a member of the WTO, China has taken several steps to promote economic liberalization. The ﬁrst day of 2002 saw China cut its average tariﬀ rate to 12 percent from a level of 15.3 percent in 2001. In particular, the rate for manufacturing goods was reduced from 14.7 percent to

280

Case studies of East Asian economic integration

11.3 percent. At the same time, China abolished quotas and license arrangements for grains, wool, cotton, chemical fertilizers and a number of other products. In the future tariﬀ rates will be cut further, and non-tariﬀ barriers will be removed for most manufacturing goods. China has also moved to abolish or modify laws and regulations or issue new ones in accordance with the rules of the WTO. Small- and mediumsized enterprises and foreign-owned companies are now entitled to participate directly in international trade. Anti-dumping and anti-subsidy laws have been implemented since 1 January 2002. New laws and regulations concerning the service trade, covering such areas as legal services, telecommunications, ﬁnancial institutions, insurance, audio and video products and tourism, have been issued since China’s entry into the WTO. Also, measures have been taken to ensure compliance with WTO rules on intellectual property, foreign investment and information transmission. While China is fast integrating into the world economy, diﬀerent regions are participating in this integration to greater or lesser degrees due to signiﬁcant regional heterogeneity. Exports and overall trade increased faster in the coastal regions in general as reforms took place. This is shown in Figure 10.2, where the ratio of regional export/GDP (or trade/GDP) is compared to the national average in selected years, which is a proxy for economic opening in this study. In the next sub-section we investigate this regional disparity in more detail by looking at the evolution of industrial agglomeration in diﬀerent regions. 10.2.4

Regional Industrial Agglomeration during Reform

The following historical comparison of the shares that diﬀerent regions have in industrial GDP will give us a clearer picture of the extent of industrial agglomeration since the start of economic reform, and show how industries have been relocating from former industrial bases to new industrial cores. In 1987, when industry sector reform was launched, the coeﬃcient of variance in the regional share of industry was 0.026; this increased to 0.030 in 2001, suggesting that industries have become more geographically concentrated following economic reforms. However, the industrial–core regions today are not the same ones as when reforms began. The four parts of Figure 10.3 illustrate the regional shift in the share of industrial GDP in China in 1978, 1987, 1995 and 2001 respectively.6 A trend toward agglomeration can also be seen in the ﬁgures. As Figure 10.3(a) shows, at the start of reforms, Chinese industry was not as concentrated as it became by 2001, particularly in the following four respects. (1) the share of industrial GDP for some coastal provinces, such as Zhejiang and Fujian was very low; both

Economic opening and industrial agglomeration in China

281

of these provinces had less than 4 percent; (2) two western provinces, Gansu and Shaanxi, had more than 2 percent of industrial GDP, which put them in the same group as most eastern and central provinces at the time; (3) in 1978 Chinese industry was also highly concentrated geographically in three northeastern provinces, Heilongjiang, Jilin and Liaoning, especially in the eastern part of the Liaoning Peninsula; (4) the three autonomous municipal cities had a high share of industrial GDP relative to their territorial size. To a large extent these four results were the consequence of historical events and development strategies arising from political concerns. The heritage of industrial distribution before 1949 partly explains the higher level of industrial agglomeration in northeastern China, Shanghai, Shandong and Tianjing. The inland development policies adopted out of political concerns in the ﬁrst ﬁve-year plan also explain why some western provinces like Gansu and Shaanxi had a higher share of industrial GDP than some coastal provinces at the time. However, the location of China’s industries has changed substantially since economic reforms and the country’s open-door policies. This process can be seen from the four parts of Figure 10.3. Compared with Figure 10.3(a), Figure 10.3(d) shows that the share of industrial GDP has increased signiﬁcantly in the coastal provinces. This is particularly true for Guangdong, Jiangsu, Shandong and Zhejiang, where the shares increased to 11.20 percent, 10.11 percent, 9.69 percent and 7.35 percent, respectively, in 2001. Meanwhile, the share of industrial GDP in northeastern and western China and the three autonomous municipal cities7 has decreased except for Sichuan. Another indicator of this trend in agglomeration is the fact that in 1978 there were only 11 provinces with less than a 2 percent share of industrial GDP and two regions with more than 8 percent; yet in 2001 these ﬁgures became 16 and 3 respectively. It can also be seen from the four parts of Figure 10.3 that during the reforms, more and more industries have relocated into the Yangtse River Delta, Pearl River Delta and Bohai belt.8 Table 10.1 shows that the top four provinces with the highest share of industrial GDP are all within these three great regions. However, within them, Guangdong seems to be the most attractive location due to the impact of Hong Kong and its earlier open-door policies, while Liaoning has been replaced by Shandong as a leading industrial base in the Bohai belt. Meanwhile, the Yangtse Delta has maintained its role as the largest industrial base in China, leading in both light and heavy industries including textiles, garments, machinery, electronics, steel, automobiles and petrochemicals. Wen (2004) has ranked the cities with the ﬁrst and second highest shares of industrial GDP in 25 industrial sectors in 1995, roughly half of them are in the Yangtse Delta. In 2001, 30.23 percent of total industrial output in China was conducted in

8 7

Coastal 6

Inner

Average trade: 1.82 Average export: 1.71

Average trade: 0.42 Average export: 0.50

5 4 3

282

Trade

Export

2 1 0

1 2 3 6 9 10 11 13 15 19 20 21 4 5 7 8 12 14 16 17 18 23 24 25 27 28 29 30 31

1987 Notes: 1 Beijing, 2 Tianjin, 3 Hebei, 4 Shanxi, 5 Inner Mongolia, 6 Liaoning, 7 Jilin, 8 Heilongjiang, 9Shanghai, 10Jiangsu, 11 Zhejiang, 12 Anhui, 13 Fujian, 14 Jiangxi, 15 Shandong, 16 Henan, 17 Hubei, 18 Hunan; 19 Guangdong, 20 Guangxi, 21 Hainan, 23 Sichuan, 24 Guizhou, 25 Yunnan, 27 Shaanxi, 28 Gansu, 29 Qinghai, 30 Ningxia; 31 Xinjiang.

Figure 10.2(a)

Ratio of regional export/GDP (trade/GDP) to the national average, 1987

8 7

Inner Coastal

6

Average trade: 0.39 Average export: 0.47

Average trade: 1.87 Average export: 1.76

5 4

283

3

Trade

Export

2 1 0

1

2

3

6

9 10 11 13 15 19 20 21 4

5

7

8 12 14 16 17 18 23 24 25 27 28 29 30 31

1991 Notes: As for Table 10.2(a).

Figure 10.2(b)

Ratio of regional export/GDP (trade/GDP) to the national average, 1991

8 7 Inner

Coastal 6

Average trade: 0.39 Average export: 0.43

Average trade: 1.87 Average export: 1.80

5 4

284

3

Trade

Export

2 1 0 1

2

3

6

9 10 11 13 15 19 20 21 4

5

7 8 12 14 16 17 18 23 24 25 27 28 29 30 31 1997

Notes: As for Table 10.2(a).

Figure 10.2(c)

Ratio of regional export/GDP (trade/GDP) to the national average, 1997

8 7 Coastal 6

Inner

Average trade: 1.93 Average export: 1.89

Average trade: 0.34 Average export: 0.37

5 4 285

3 Trade

Export

2 1 0

1

2

3

6

9 10 11 13 15 19 20 21 4

5

7 8 12 14 16 17 18 23 24 25 27 28 29 30 31 2001

Notes: As for Table 10.2(a).

Figure 10.2(d)

Ratio of regional export/GDP (trade/GDP) to the national average, 2001

(a) 1978

(b) 1987

286

(c) 1995

(d) 2001

287 Sources: The Leading Body of National Industrial Census of the State Council (LBNICSC), The Data of the 1985 Industrial Census of the People’s Republic of China, vol. 4. China Statistical Publishing House, Beijing, 1986. The State Statistics Bureau of the PRC, China Statistical Yearbook of Industrial Economy, China Statistical Publishing House, Beijing, 1997, 1999–2002.

Figure 10.3

Regional shares of industrial GDP in 1978, 1987, 1995 and 2001

288

Case studies of East Asian economic integration

Table 10.1 The top four provinces with the highest shares of industrial GDP Year 1 2 3 4

1981

1986

1991

1996

2001

Shanghai Liaoning Jiangsu Shandong

Jiangsu Liaoning Shanghai Shandong

Jiangsu Guangdong Shandong Liaoning

Guangdong Jiangsu Shandong Zhejian

Guangdong Jiangsu Shandong Zhejian

the Yangtse Delta. In eﬀect, more industries were concentrated in fewer regions in 2001, which indicates the trend of industrial agglomeration in China as economic reform has progressed.

10.3

DETERMINANTS OF AGGLOMERATION IN CHINESE INDUSTRY

The previous section sketched out a rough picture of the relationship between economic opening and industrial agglomeration in China. This section will give a more detailed description of China’s industrial agglomeration. The Yangtse Delta, China’s most prominent industrial center, will serve as the case study for this chapter. Based on the data of the Yangtse Delta and observations from other provinces, this chapter will present a theoretical summary of the determinants of industrial agglomeration. The key question we shall answer is why economic opening has been important in reshaping China’s industrial map. 10.3.1 The Mechanism of Industrial Agglomeration: Case Study of China’s Yangtse Delta Shanghai was one of the two industrial centers in China before 1949, the other being northeast China. In the post-reform era, the change in industrial allocation and the growth of manufacturing in the Yangtse Delta, which includes Shanghai, Jiangsu and Zhejiang, oﬀers an attractive case study for understanding the mechanism of industrial agglomeration. Interestingly, the three regions have experienced quite diﬀerent patterns of industrial development. The trend of intra-regional industrial agglomeration in the Yangtse Delta The previous section showed how manufacturing industries have agglomerated in the Yangtse Delta. It will also be quite useful to look at the trend

Economic opening and industrial agglomeration in China

289

Table 10.2 Average industrial concentration in the Yangtse Delta, 1998–2002

Shanghai Jiangsu Zhejiang Gap between Jiangsu and Zhejiang

1998

2000

2002

0.317 0.414 0.269 0.145

0.290 0.412 0.299 0.113

0.261 0.410 0.328 0.082

Source: Fan (2004) and authors’ calculation.

of intra-regional industrial agglomeration within the Yangtse Delta before summarizing the mechanism of industrial agglomeration. Fan (2004) did an excellent job of showing the trend of agglomeration between 1998 and 2002. The ﬁrst index Fan used is an index of average concentration of manufacturing industries, and written as follows: vi

k(vki) , where vk k

i

Eki

i Eki

.

(10.1)

In this index, Eki represents the total value of industrial production in industry k in region i. vi stands for the average concentration of manufacturing industries in region i that ranges from 0 to 1. The closer the index is to 1, the higher the region’s share in all the industries. The results show that Jiangsu is still the largest province in manufacturing, but its relative share is slowly decreasing. Shanghai’s share in manufacturing is falling even faster than that of Jiangsu. In contrast, Zhejiang has maintained a growing pace of industrial development. The gap between Zhejiang and Jiangsu in terms of the industrial concentration index has narrowed from 0.145 in 1998 to 0.082 in 2002. Fan also presented the geographical concentration at the industry level using the following SP index: SPk c

vkivkjij, i

(10.2)

j

where i, j, k and vki have the same deﬁnition as in the ﬁrst index. ij is the straight distance between the capital cities of the provinces. c is a constant that is set at 0.5. Although we do not have large samples, the SP index can still show how geographical concentration changes. Compared with the geographical Gini index and Herﬁndal index, which measure industrial geographic concentration, but not incorporating geographic distance, the SP index can measure diﬀerent geographic concentration, given the same

290

Case studies of East Asian economic integration

geographical Gini index or Herﬁndal index. The SP index ranges from 0 to 1. The closer the index is to 0, the more the industry is geographically concentrated. Table 10.3 shows that among 28 two-digit industries, 21 have seen their SP index rise. The data in Table 10.3 show that most of the industries are experiencing relocation. Shanghai has shifted toward capital-intensive industries, while Zhejiang is developing labor-intensive sectors. This shows a very clear trend of intra-regional industrial location and specialization in the Yangtse Delta. Geography, history, policy and the role of economic opening In the development of the Yangtse Delta, Shanghai, Jiangsu and Zhejiang have experienced diﬀerent paths of industrial growth. Geography, history and government policy have played key roles in determining the location of industries. Shanghai has been China’s industrial center since before World War II, and at the early stage of economic reforms, Shanghai was the largest producer of many of the country’s goods. Shanghai is located in the middle part of China’s east coast and is the largest coastal city, where the Yangtse River enters the East China Sea. These historical and geographic factors have sustaining eﬀects on the industrial development of Shanghai, the socalled ‘dragon-head’ of the Yangtse Delta, on the Yangtse River drainage area, and even on the whole Chinese economy. Being neighboring provinces, Jiangsu and Zhejiang have shared the historical and geographic advantages of Shanghai. However, being diﬀerent from Shanghai, the two neighbors have their speciﬁc history-dependent industrial development. Jiangsu has been a large recipient of FDI. It became the largest recipient in 2003, surpassing that of Guangdong and accounting for 20 percent of China’s total FDI.9 In comparison, Zhejiang ranked only ﬁfth in terms of total FDI. Table 10.4 provides a comparison between Jiangsu and Zhejiang of several representative economic indices in 2003. An interesting question is: why should these two provinces, which have similar conditions of geography, climate and level of economic development, have performed so diﬀerently in attracting FDI? The answer is that history does matter. Tu (2005) found that the high growth of FDI in Jiangsu is related to its history of state-owned enterprise (SOE) development. Although people believe that state ownership leads to poor eﬃciency, the SOEs have a large stock of assets and skilled workers. Besides, it is worth noting that when setting up joint ventures or cooperative enterprises, it is easier for foreign investors to enter the regulated industries that have higher proﬁts and large-scale enterprises. China issued a law on joint ventures in July 1979. From 1979 to 1985, cooperative enterprises

Economic opening and industrial agglomeration in China

291

were the major form of FDI. In 1986 the proportion of joint ventures attracting FDI surpassed that of cooperative enterprises; then in 1999 wholly foreign-owned enterprises became the most popular form of FDI. The distribution of FDI also shows that foreign investors tend to invest in manufacturing industries where state-owned enterprises have a big market share (Zhang and Zheng 1999). Jiang (2002) showed that by the late 1990s, multinational companies accounted for two-thirds of the largest ten enterprises in microelectronics, mobile telecommunications equipment, cars, pharmaceuticals and engineering machinery. In the late 1990s, foreignowned enterprises held 90 percent of the market share in mobile phones and two-thirds in cars. SOEs that have low-waged but skilled workers, considerable capital stock and governmental resources are very attractive to foreign investors. At the same time, SOEs also want a lending hand from overseas investors to reform their corporate government structure. At the end of 1999, the fourth plenary meeting of the 15th Central Committee of the CCP decided that SOEs should exit some competitive markets and actively absorb FDI to promote restructuring. SOEs that undertook joint ventures with FDI could get favorable treatment regarding taxes, loans and resource provision. A comparison of Zhejiang and Jiangsu shows that the former clearly lagged far behind the latter in terms of both SOE industrial output and workers. At the start of reforms in 1978, Jiangsu’s SOEs produced 20.752 billion yuan while Zhejiang’s produced 8.103 billion. In 1993, the second year after Deng Xiaoping’s tour to south China, the gap enlarged to 165.104 billion versus 81.586 billion. Correspondingly, the gap in the number of SOE workers was 3.664 million in Jiangsu versus 1.831 million in Zhejiang in 1978, and 5.740 million versus 3.006 million in 1993.10 History has also played a signiﬁcant role in the formation of the specialized industrial clusters for which Zhejiang has become well known. (See Table 10.5 for details of some representative industrial clusters in Zhejiang.) A very distinctive feature of Zhejiang’s industrial clusters is that they have developed as a grass-roots economy. A typical industrial cluster consists of many small-sized private enterprises that specialize in related industries, or diﬀerent parts of a production chain. Enterprises share in scale economies of demand and supply, sales channels and information, and in their totality, they account for a large share of international trade. Historically many of these industrial clusters have a long tradition of manufacturing. Jin and Zhu (2002) gave three case studies of the history-dependent development of industrial clusters: shoe-making in Wenzhou, garments in Ningbo and hardware in Yongkang. Their history of related manufacturing can be dated back to ancient China. This long tradition is important to modern industrial development, because industry-speciﬁc tacit knowledge accumulates over a long period of time and diﬀuses locally.

Table 10.3 SP index and the province with the largest market share, 1998–2002 SP index

292

Food processing Food production Beverage production Tobacco processing Textile industry Garments and other fiber products Leather, furs, down and related products Timber processing, bamboo, cane palm fiber and straw products Furniture manufacturing Papermaking and paper products Printing and record medium reproduction Cultural, education and sports goods Petroleum refining and coking Raw chemical materials and chemical products Medical and pharmaceutical products Chemical fiber

Province with the largest market share

1998

2000

2002

1998

2000

2002

0.5757 0.6185 0.6088 0.5843 0.5646 0.6195 0.5566 0.6135

0.5494 0.6177 0.5992 0.5839 0.5602 0.6105 0.5306 0.6239

0.5539 0.6103 0.5951 0.6005 0.5504 0.5996 0.4656 0.6179

JS (0.5534) SH (0.3921) JS (0.3935) SH (0.4770) JS (0.5290) JS (0.4065) ZJ (0.5183) SH (0.4244)

JS (0.5821) SH (0.3982) ZJ (0.4118) SH (0.4513) JS (0.5226) JS (0.3959) ZJ (0.5636) JS (0.3776)

JS (0.5730) SH (0.4280) ZJ (0.4183) SH (0.4135) JS (0.4840) ZJ (0.4117) ZJ (0.6508) JS (0.4417)

0.6201 0.6041 0.6205 0.6249 0.6240 0.5900 0.6237 0.6179

0.6098 0.5940 0.6109 0.6226 0.6164 0.5793 0.6190 0.6243

0.5939 0.5815 0.5983 0.6189 0.5914 0.5780 0.6161 0.5482

SH (0.3702) ZJ (0.4081) JS (0.4045) JS (0.3886) JS (0.3978) JS (0.5419) JS (0.3732) JS (0.4652)

ZJ (0.3675) JS (0.4376) SH (0.4201) JS (0.3651) ZJ (0.3580) JS (0.5623) JS (0.3594) JS (0.4167)

ZJ (0.4205) JS (0.4460) SH (0.3871) JS (0.3614) SH (0.4440) JS (0.5607) JS (0.3952) JS (0.5299)

293

Rubber products Plastic products Non-metal mineral products Smelting and pressing of ferrous metals Smelting and pressing of nonferrous metals Metal products Ordinary machinery Special purpose equipment Transport equipment Electric equipment and machinery Electronic and telecommunications Instruments, meters, cultural and clerical machinery

0.6104 0.6133 0.5998 0.5607 0.603 0.6223 0.6124 0.5984 0.5899 0.6242 0.6144 0.6178

0.6229 0.6147 0.5888 0.5839 0.6023 0.6249 0.6096 0.6031 0.5945 0.6204 0.6055 0.6251

0.6197 0.6069 0.5954 0.5926 0.5973 0.6209 0.6125 0.6095 0.5957 0.6110 0.6068 0.6246

SH (0.4114) JS (0.4463) JS (0.4998) SH (0.5575) JS (0.4972) JS (0.4173) JS (0.4864) JS (0.5249) SH (4985) JS (0.4052) JS (0.4354) SH (0.4047)

JS (0.3827) JS (0.4112) JS (0.5227) SH (0.5030) JS (0.4915) JS (0.4120) JS (0.4838) JS (0.5071) SH (0.4800) JS (0.4125) JS (0.4675) JS (0.4074)

JS (0.4237) ZJ (0.3941) JS (0.5016) JS (0.4964) JS (0.4708) JS (0.4095) JS (0.4631) JS (0.4678) SH (0.4727) JS (0.4063) JS (0.4686) JS (0.4015)

Note: Diﬀusion in location has happened in the industries in italics. Market shares are in parentheses. SH Shanghai, JS Jiangsu, ZJ Zhejiang. Source: Fan (2004).

294

Case studies of East Asian economic integration

Table 10.4 Comparison of Jiangsu and Zhejiang regarding the value of foreign-owned enterprises* in 2003 Zhejiang (a) Per capita GDP (yuan) No. of enterprises Industrial output (100 million yuan) Industrial value-added (100 million yuan) Assets (100 million yuan) Fixed assets (100 million yuan) Total revenue (100 million yuan) Profit (million yuan)

20147 3675 2585.8

Jiangsu (b)

(a)/(b) 100

16809 4799 6007.42

119.9 76.58 43.04

630.09

1573.41

40.05

2399 828.67 2534.88 179.07

5167.13 2083.87 5926.59 314.96

46.43 39.77 42.77 56.85

Note: Taiwan-, Hong Kong- and Macao-owned enterprises are included. Sources: Jiangsu Statistical Yearbook 2004, Zhejiang Statistical Yearbook 2004, and authors’ calculation.

Low ﬂow of tacit knowledge leads to agglomeration of skilled labor, and an industrial atmosphere that helps diﬀuse the tacit knowledge is formed locally (Jin and Zhu 2002). Not only the long tradition of manufacturing, but also the recent development of industries and markets have played a role in the formation of industrial clusters. Jiashan County is located on a plain that has no forests, but it has one of the biggest timber markets in China. Haining does not have local leather resources, but its leather market is the country’s largest. Cangnan has the largest trading center for ginseng and antlers, although it makes no local products from these items. These are very illustrative cases showing that once a place develops an industry, the positive feedback eﬀects will strengthen its advantage and lead to a history-dependent industry development.11 Government policy is another driving force in shaping the industrial map of the Yangtse Delta. For the most part local governments have been doing well in developing the economy. Shanghai is a typical city where the government is quite strong in making development policy. During the past decade the Shanghai municipal government has put much eﬀort into infrastructure construction and environment protection for the purpose of building a better investment environment. The opening of the Pudong area has given the city a new wave of FDI and industry development. Now Shanghai is developing at a high rate by transforming its industrial structure to capital- and skill-intensive industries and moving out labor-

Economic opening and industrial agglomeration in China

295

intensive ones. Zhejiang province has a diﬀerent style of governing the economy. The government is always prepared to enhance the private economy with a laissez-faire policy, as well as providing public goods that are helpful to local enterprises. Jin et al. (2005) gave an example of how the local government in Datang Town has helped develop the local sock industry by providing market and technology information, planning regional industry location, and improving the local investment environment. However, local governments have also adopted protectionist policies aimed at local industrial growth, maintaining employment and ﬁscal revenue. There are many studies showing that China’s domestic market is seriously segmented, e.g. Young (2000) and Poncet (2003) to cite only two; and although there is more evidence showing that China is integrating its domestic market (Naughton 1999, Xu 2002; Bai et al. 2004; Chen et al. 2007), no one denies that market segmentation, local protectionist policies and the duplication of industries are still problems in China. Under the system of ﬁscal decentralization, protectionist policies and market segmentation might increase local ﬁscal revenue and employment in the short run. Chen et al. (2007) showed evidence that government expenditure, proxied by the expenditure-to-GDP ratio, and employment pressure, proxied by SOEs’ share of employment, increase market segmentation. Lu et al. (2004) argued that local government also has the incentive to protect local industries to gain the learning-by-doing eﬀects, so that they can raise their bargaining position in sharing the future beneﬁts of inter-regional specialization or catch up with the developed regions. For instance, on 12 March 2002, the Xinhua Daily Telegraph reported on the ‘Zhejiang Valley of Photoelectric Industries’ in Fuyang, Zhejiang. Then on 15 March, only three days later, the same newspaper carried a story about the plan for a ‘China Valley of Photoelectric Industries’ in Changchun, Jilin. The next year, the Wenhui Daily, in its 20 June 2003 edition, published an article about the problems of such duplication of industries that are occurring in various high-tech industries. In the Yangtse Delta, Shanghai is one of the provinces with the highest index of market segmentation (Chen et al. 2007). An example of the protectionist policy in Shanghai is that the local government only allows the locally produced car, Volkswagen, to serve as taxis. Nevertheless, market forces seem to be dominating the eﬀects of segmentation policies as studies show that China’s domestic market is being integrated (Naughton 1999; Xu 2002; Bai et al. 2004; Chen et al. 2007). In a study examining market integration in the Yantse Delta, Xu and Li (2005) show that market segmentation does have negative eﬀects on economic development. However, the negative eﬀects are declining, especially since 1997 when 14 cities coorganized the Coordination Committee of Cities in the Yangtse Delta.

Table 10.5 The distribution of major specialized industrial clusters in Zhejiang Main

Location

products

Ties Low-pressure electronics Hardware machines Lighters 296

Ball pens

Pens Leather goods Scutcheon printing Socks Buttons

Shengzhou Liushi Town, Leqing Yongkang Wenzhou downtown Fenshui Town, Tonglu Wenzhou downtown Haining Cangnan Datang Town, Zhuji Yongjia

No. of enterprises

1200 3800*

Quantity of production

0.25 bn pcs

6500

Value of production (billion yuan)

8 15.5 19

2601

1 bn pcs

365

2.1 bn pcs

0.55

1511

7.5 bn pcs

1.6

5688 2089*

25 mn pcs

12.088 7.5*

10 000 1208

6 bn pairs

No. of employees (1000)

50

80 66

200

252 943

2

9 2.26

Share in the national total (%)

10

33 62.3 200* 150 20

25 40 (badge) 20 (printing)

Plastic goods Moulds Valves Light textiles Locks Shoes

297

Synthetic leathers Eyeglasses Timber (veneer) Child garment Garment

Taizhou Taizhou Yuhuan Yongjia Shaoxing Wenzhou Wenzhou downtown Wenzhou Wenzhou Jiashan Zhili, Huzhou Wenzhou Ningbo

10081* 979 835*

0.8 mn tons* 39700 tons 60900 tons

7744 400 6000*

2 bn meters

12.1*5 1.18 3.5* 4.5 41.687 5 30

140.6* 12.5 32*

3 mn m3

4.5 4.16*

6600

0.1 bn pcs/sets

2.5

2000 1300

0.88 bn pcs

22.3 11.06

380

65 20 70

40 31*

80 41.3 15 10

0.4

Notes: * indicates data for 2000; others are for 2001. 1. From the industrial association. 2. The market share of electronic equipment. The market shares of other hardware products range from 50% to more than 80%. 3. The market share of lighters with metal shells. 4. Only enterprises with sales revenue greater than 5 million yuan included. 5. Sales revenue. 6. Low-pressure copper valves. Source: Jin et al. (2005).

51*6 30

164*

8 500* 248*

14.3

298

Case studies of East Asian economic integration

Have labor-intensive industries shifted to inner China due to labor costs? The trend of market integration enhances inter-provincial specialization and the adjustment of the industrial structure in the Yangtse Delta. The diﬀerentials in industrial structure among the three provinces in the delta have continued to increase in recent years (Fan 2004). Due to rising labor costs, Shanghai has been moving labor-intensive industries out to other provinces. In contrast, Zhejiang has seen very rapid development of its labor-intensive industries, thanks to local policies allowing free interregional migration that helps check labor costs. Interestingly, when local enterprises in Zhejiang mature, they move to Shanghai seeking more opportunities to develop. A more interesting question is: have the labor-intensive industries shifted to inner China due to labor costs? To answer this question, we select four representative cities in diﬀerent parts of China: Shanghai in east China and the Yangtse Delta, Wuhan in central China, Xi’an in northwest China and Chongqing in southwest China. Table 10.6 shows the change in these cities’ shares of 28 industries in the national total. Shanghai has been losing its share in most of the industries except technology-intensive industries such as electronics and telecommunications, and capital-intensive industries such as ordinary machinery. Western cities, especially Chongqing, have been expanding their shares in most of the labor-intensive industries such as food processing, timber processing, bamboo, palm ﬁber and straw products, plastic products and electric equipment and machinery. It seems that labor-intensive industries are moving towards inner China. This means that diﬀerentials in labor costs are a latent force in shaping the countrywide map of industrial location. However, within the country as a whole, industrial agglomeration towards the coastal areas, especially the three great industrial regions, is the dominant feature of industrial development in China. With freer mobility of labor and capital, the interprovincial disparity in industrial development could continue to rise due to further industrial agglomeration, while within each great industrial region such as the Yangtse Delta, the gap in industrial development might narrow because of intra-regional market integration and relocation of labor-intensive industries. 10.3.2 Economic Policies, Economic Geography and the New Economic Geography: The Role of Economic Opening in Reshaping China’s Industrial Map Diﬀerent factors can be put forward to explain the substantial changes in the geographical location of Chinese industries. According to traditional economic geography theory, geographical factors play the main role in

Economic opening and industrial agglomeration in China

299

determining industrial agglomeration. For instance, locations close to natural resources and major ports will probably become industrial centers with industrial agglomeration. Such cases in China are Liaoning and Shanxi, which used to be two important industrial centers because of their abundant natural resources. However, traditional economic geography theory cannot explain some important observations about industrial development. First, some regions with less geographical advantage may attract more industries. Second, those with similar geographical conditions might be substantially diﬀerent in industrial development. For instance, Zhejiang has no advantage in resource endowment compared with Fujian; however, its share of industrial GDP has continued to increase much faster than that of Fujian. Similarly, Guangdong and Guangxi are very alike geographically, but Guangdong has become an important industrial center since reforms began, while Guangxi has lagged far behind. Economic policies might be one of the explanations for such diﬀerences. When industrial reforms and open-door policies began in China, priority was given to some coastal regions with the establishment of four SEZs in Guangdong and Fujian in 1980, and 14 open coastal cities later on. Consequently, these regions attracted not only early foreign investment but also domestic investment, which led to a rapid increase in their shares of industrial GDP. Economic opening also contributes to industrial agglomeration indirectly. For instance, local infrastructure has been improved in order to attract more FDI, which consequently makes it more attractive for ﬁrms. What is more, economic opening always brings better management and technology to local ﬁrms through FDI, which also helps attract more ﬁrms. The extent of local government interference in the market might also have an impact on industrial concentration. The more local government interferes, the less attractive the region is to investors, and consequently the less industrial concentration there will be. We test these hypotheses in the next section. Some of the existing literature also indicates that economic policies do have an impact on regional development in China, although agglomeration has not been investigated in these studies (see Kanbur and Zhang 2005; Démurger et al. 2002). However, new economic geography theory focuses on a diﬀerent point. By assuming increasing returns, which borrows from the new international trade theory, the new theory of economic geography oﬀers an explanation diﬀerent from the traditional one. For two geographically similar regions, it implies that industrial concentration might begin by chance in one region, and because of the increasing returns, this region will continue to attract industries as long as the inter-regional transaction cost is not high enough to segment the two regions; the result will be industrial

Table 10.6

The change in the industrial share of selected cities, 2000–03 Absolute change (%)

300

Food processing Food production Beverage production Tobacco processing Textile industry Garments and other fiber products Leather, furs, down and related products Timber processing, bamboo, cane, palm fiber and straw products Furniture manufacturing Papermaking and paper products Printing and record medium reproduction Cultural, educational and sports goods Petroleum refining and coking Raw chemical materials and chemical products Medical and pharmaceutical products Chemical fiber

Rate of change (%)

SH

WH

XA

CQ

SH

WH

XA

CQ

0.311 0.493 0.045 1.357 1.103 1.706 0.279 5.186

0.161 0.164 0.152 0.041 0.089 0.183 0.024 0.099

0.062 0.012 0.063 0.001 0.109 0.050 0.109 0.011

0.102 0.071 0.086 0.386 0.059 0.002 0.095 0.073

13.77 6.04 1.21 22.94 22.08 16.16 6.42 40.75

21.88 13.60 10.39 2.04 19.16 16.64 23.44 21.44

20.80 2.02 10.24 20.66 31.90 45.01 60.07 57.75

20.59 10.36 9.06 18.37 12.50 2.24 53.74 45.53

0.149 0.800 1.505 1.570 3.809 1.220

0.135 0.064 0.393 0.309 0.183 0.159

0.573 0.064 0.535 0.005 0.179 0.034

0.582 0.022 0.166 0.026 0.012 0.074

1.83 17.39 13.24 11.48 83.38 15.96

13.26 6.38 25.83 84.21 11.75 26.75

81.6 34.62 24.12 36.02 99.28 11.51

69.95 4.59 16.61 64.17 20.26 5.70

1.171 14.846

0.906 0.002

0.218 0.204

0.291 1.060

15.87 30.05 78.28 100.00

9.40 92.76

15.06 99.35

301

Rubber products Plastic products Nonmetal mineral products Smelting and pressing of ferrous metals Smelting and pressing of nonferrous metals Metal products Ordinary machinery Special purpose equipment Transport equipment Electric equipment and machinery Electric and telecommunications Instruments, meters, cultural and clerical Machinery

0.720 0.770 0.531 4.488 0.508 2.021 1.445 0.922 0.018 1.534 1.877 1.033

0.105 0.075 0.087 0.942 0.025 0.037 0.149 0.577 0.521 0.412 0.402 0.285

0.054 0.017 0.056 0.025 0.153 0.209 0.004 0.257 0.020 0.033 0.393 0.610

0.034 0.010 0.106 0.296 0.250 0.162 0.347 1.259 0.150 0.134 0.095 0.363

8.67 9.13 13.75 35.90 11.36 16.52 12.66 11.79 0.12 16.15 17.87 8.43

44.19 23.00 17.68 27.39 5.97 7.37 19.99 50.55 22.18 50.96 36.19 25.18

Note: SH Shanghai; WH Wuhan; XA Xi’an; CQ Chongqing. Bold numbers indicate increasing shares. Source: Authors’ calculations based on the data from China statistical Yearbook.

73.17 23.12 22.27 61.79 111.34 74.89 0.75 22.91 1.45 4.40 50.59 46.32

7.18 3.12 7.48 25.07 17.66 43.25 25.20 646.59 2.58 26.77 43.16 15.95

302

Case studies of East Asian economic integration

agglomeration in the one region. In the framework of the new economic geography, the following factors are essential to industrial agglomeration: 1.

2.

3.

4.

The number of ﬁrms. Because of forward and backward linkage, the larger the number of ﬁrms in one region, the more the ﬁrms can beneﬁt from easy access to raw materials and local markets; therefore, ﬁrms tend to cluster geographically. The local stock of human capital. The more the stock of human capital, the less the cost needed to invest in R&D and the easier to gain from innovation. Market size. Firms tend to agglomerate where the regional market size is large. Other things being equal, the size of the local market will be larger, given the higher per capita income and the larger population size. Trade costs, which are one of the key parameters in the new economic geography theory. Lower trade costs, such as lower transaction or transportation costs, will encourage industrial concentration.

Economic policies and factors in new economic geography could interact in industrial agglomeration. Taking the opening policies as an example, the preferential implementation of such policies in some coastal cities could directly attract ﬁrms to those cities. What is more, economic opening might also help lower transaction costs and enlarge market size, thus helping to attract ﬁrms indirectly through the mechanism of increasing returns. Taking another example, Zhejiang as a pioneer of marketization has never stopped its privatization and deregulation. This has continuously encouraged the private sector and hence stimulated industrial agglomeration. In summary, economic policies are usually self-enforcing once implemented. Such kind of positive feedback eﬀects of policies via the variables with increasing returns in new economic geography can last for a long time. It is noteworthy that new economic geography theory leaves space for other factors, such as economic policies and geography, to play roles. As stated by Neary (2001), when trade costs are in a certain range, both agglomeration and diversiﬁcation are possible equilibria, so history and policy have a potential role in inﬂuencing which equilibrium prevails. Accordingly, in this chapter we test the roles of both policies and history in inﬂuencing agglomeration. Neither of these has been emphasized enough in recent studies, e.g. in Wen (2004). For example, the open-door policies implemented in the Pearl River Delta at the early stage of reforms might explain why the two geographically similar provinces of Guangdong and

Economic opening and industrial agglomeration in China

303

Guangxi have been totally diﬀerent in attracting industries. The government’s urban-biased policies result in better education and infrastructures in east China, which has a higher level of urbanization, and consequently, east China might attract more industries than elsewhere. A similar example of opening causing agglomeration is Mexico, which has seen industries concentrate in regions bordering the USA since the North American Free Trade Agreement (NAFTA) was implemented (Rodríguez-Pose and Sánchez-Reaza 2005). In the next section we formally build a model to test the impact of economic policies, economic geography and new economic geography factors on industrial agglomeration in China.

10.4

ECONOMIC OPENING, GEOGRAPHY AND INDUSTRIAL AGGLOMERATION: AN ECONOMETRIC STUDY

This section undertakes an econometric analysis of the eﬀects of various determinants of industrial agglomeration using the provincial panel data for China. We focus on the role of economic opening, which has been the most important factor in reshaping China’s industrial map. 10.4.1

Econometric Model and Data Explanation

The model speciﬁcation is as follows: Yit01X1i2X2i,ti3X3i,tiit

(10.3)

where Yit is the regional share of industrial GDP for region i in year t, which measures regional industrial agglomeration as in Wen (2004). Vectors X1i, X2i and X3i represent three types of factors aﬀecting industrial agglomeration in China: economic policies, economic geography and the new economic geography respectively. To minimize the simultaneity error, X2i and X3i are both made predetermined before realization of Yit by lagging one year. 0, 1, 2, 3 and it are constant, vectors of coeﬃcient and error term respectively. The provincial panel data we use in this chapter are compiled based on Comprehensive Statistical Data and Materials for 50 Years of New China, as well as various issues of China Statistical Yearbook, both published by the State Statistics Bureau of the PRC (SSB). See the Appendix for the details of data construction. Largely due to the incompleteness of some data, the modeling exercise is conﬁned to the period after 1987. Excluding

304

Case studies of East Asian economic integration

Taiwan, Hong Kong and Macao, there are 31 provinces or regions in China, including four autonomous municipal cities. Tibet and Chongqing are excluded due to the lack of complete data. Therefore, a total of 29 regions will be covered in this analysis. Other variables are grouped into three types as follows. Economic policies As argued above, economic polices in China have a geographical impact. We test in particular the impact of two types of policies on industrial agglomeration: economic opening and local government involvement in economic activities. Regional economic opening is represented by export (or trade), which is the ratio of regional exporting/GDP (or trade/GDP) to the national average. We expect that export or trade will encourage regional industrial agglomeration. The ratio of the regional share of government expenditure12 in GDP to the national average (gov) serves as a proxy of local government involvement in economic activities, which we expect to discourage regional industrial agglomeration, since government involvement in the planned economy has distorted resource allocation as described in Section 9.2 of the previous chapter.13 Although the variance of the above-mentioned policy proxies across regions and periods makes panel data ideal for testing the impact of policies, those proxies might be the result of some unobserved variables other than policies themselves. In order to minimize the estimation bias from such kind of endogeneity problem, export and trade, which are proxies of economic opening, are instrumented in our estimation. Two instrumental variables are used. One is the ratio of regional exporting/GDP (or trade/GDP) to the national average in 1978 (export78 or trade78), which represents the eﬀects of history. The other is the shorter distance from provincial capital to the port of Shanghai or Hong Kong, which catches the eﬀects of geography. Only two ports are included here because they are the two largest and most important ports in China. Shenzhen is not included because it actually borders Hong Kong. This instrumental variable is borrowed from Wei and Wu (2001). These two instrumental variables reﬂect the indirect impacts of history and geography on regional industrial agglomeration via economic opening. Economic geography We control economic geography with two dummy variables. The region dummy coast is used to distinguish coastal regions from the others. The city dummy is used to distinguish three autonomous municipal cities14 from other provinces. As in most of the studies on regional disparity in China, we use the dummy to distinguish coastal and inner regions rather than

Economic opening and industrial agglomeration in China

305

eastern, central and western regions. This is because in geography, history and policy preferences, there is a signiﬁcant diﬀerence between coastal and inner regions; however, the western and central regions are very similar. We expect the coastal regions to have a higher level of industrial agglomeration. Since the autonomous municipal cities serve as political and commercial centers compared with other regions, we expect that the city dummy relates negatively to the regional share of industrial GDP. New economic geography The new economic geography theory challenges neoclassical economics by stressing the importance of externality based on industrial backward and forward linkages, human capital accumulation (Henderson 1974) and the HME (Fujita 1988; Krugman 1991) in the trade-oﬀ between agglomeration and dispersal. In order to test these factors, which are all based on increasing returns, we include the following variables in the econometric model: (i) the regional share in manufacturing ﬁrms ( ﬁrm), which measures relative industrial externality; (ii) the ratio of the regional average years of schooling to the national average (edu), which serves as a proxy for regional comparative advantage in human capital. Because of the incompleteness of the data, we use the estimated regional average years of schooling in Wan et al. (2007); (iii) the ratio of the logarithm of regional per capita GDP to the national average (pergdp), which measures the relative capacity of the local market;15 (iv) urban development; here we use the ratio of the share of the non-agricultural population to the national average (urban) as a proxy, which we think better represents the regional infrastructure than does the number of cities in the region as used in Wen (2004); (v) transaction cost, which is represented by two variables in this study. One is the ratio of the share of regional transportation, postal services storage and telecommunication in GDP to the national average (com), which captures the development of information and communication services. The other is the regional share in the length of highways16 (road), which measures the condition of transportation. Since lower transaction cost helps attract ﬁrms, these two variables should be positively related to the regional share of industrial GDP. 10.4.2

Empirical Results

Table 10.7 shows the results of the three groups of determinants of the regional share in industrial GDP for the period 1987–2001. Five models have been run. The ﬁrst two include all the independent variables and have been estimated by random eﬀects and ﬁxed eﬀects respectively. According to the Hausman test, the random eﬀects assumption could be accepted.

306

Table 10.7

Case studies of East Asian economic integration

The determinants of industrial agglomeration

Dependent variables (1) RE

Regional share of industrial GDP (2) FE

(3) IVRE

(4) RE

(5) RE2

Economic policies export 0.624*** 0.659*** 1.179*** 0.623*** 0.202*** (trade) (0.076)1 (0.074) (0.452) (0.075) (0.062) gov 0.784*** 0.588*** 0.927*** 0.786*** 0.692*** (0.149) (0.143) (0.210) (0.148) (0.159) Economic geography coast 1.305*** 0.778 1.324*** 1.657*** (0.472) (0.698) (0.472) (0.480) city 5.516*** 5.543*** 5.446*** 5.623*** (0.873) (0.987) (0.853) (0.895) New economic geography firm 4.248*** 1.485 0.563 4.211* 7.665*** (2.443) (2.350) (3.667) (2.439) (2.575) pergdp 15.557*** 14.866*** 13.583*** 15.519*** 17.172*** (1.586) (1.499) (2.250) (1.581) (1.687) edu 0.340 0.938 0.735 0.048 (0.838) (0.838) (0.921) (0.892) urban 1.571*** 2.107*** 1.479*** 1.591*** 1.760*** (0.327) (0.333) (0.371) (0.324) (0.347) road 24.741*** 21.439*** 24.223*** 24.812*** 26.902*** (4.917) (4.921) (5.197) (4.905) (5.257) com 0.297*** 0.434*** 0.324* 0.299* 0.308* (0.159) (0.150) (0.167) (0.159) (0.171) Constant 15.127*** 15.758*** 13.525*** 14.783*** 16.646*** (1.680) (1.646) (2.228) (1.450) (1.781) R2 (within) 0.568 0.577 0.537 0.568 0.496 R2 (between) 0.604 0.237 0.547 0.606 0.623 Hausman test (FE versus 5.693 50.75 83.21 RE) Instrumental export78 export78 trade78 variables port port Hausman test P-value (IVRE 0.981 0.953 0.972 versus RE) Notes: 1. The values in parentheses are standard errors. *, ** and *** denote significance higher than 0.10, 0.05 and 0.01 levels, respectively. 2. Economic opening is measured by trade in regression (5) and by export in the other regressions. 3. According to Stata Reference 7, the estimation with random eﬀects can be accepted when the Hausman test is negative.

Economic opening and industrial agglomeration in China

307

Moreover, export is a proxy for economic opening in both models, and we use trade instead of export later to do a robust test in model 5. Model 3 is estimated by random eﬀects, where export is instrumented by export78 and port. However, no signiﬁcant endogeneity bias is found according to the Hausman test. In model 4, where the insigniﬁcant variable (edu) is not included, all other independent variables remain signiﬁcant and keep their coeﬃcients almost unchanged. Moreover, according to the Hausman test, the random eﬀects assumption is still acceptable, and there does not exist signiﬁcant endogeneity bias with export. In model 5, economic opening is measured by trade rather than export, where the random eﬀects model is again accepted without signiﬁcant endogeneity bias. Interestingly, when checking whether trade78 and port could serve as eﬃcient instrumental variables for economic opening (trade), we found that port is no longer signiﬁcantly related with trade. That is to say, exporting rather than trade that includes both exporting and importing is closely related with the distance to major ports. The models in Table 10.7 indicate that the result of estimation in model 1 is robust. According to model 1, industrial agglomeration may be well explained by economic geography, new economic geography and economic policies. That also implies that industrial development and agglomeration are a process aﬀected jointly by various forces, so a reliable empirical study should consider all these factors. The results of the estimation are as follows. Economic policies The two variables of economic policies have robust and signiﬁcant coeﬃcients in model 1. The positive coeﬃcient of economic opening, whether measured by export or trade, implies that an export-oriented development policy in developing economies may encourage industrial agglomeration. When using an instrumental variable, we found that industrial agglomeration is indirectly and signiﬁcantly related with history and geography through economic opening. That is to say, openness at the beginning of reform (export78 or trade78) as a historical factor has signiﬁcant eﬀect on economic liberalizing measured by export or trade respectively. Moreover, the distance to main ports, port, as a geographical factor has signiﬁcant eﬀects on economic opening when measured by export.17 However, port has no signiﬁcant eﬀects on economic opening when measured by trade. This is probably because exporting and export-oriented industries, rather than importing, are closely related with geographical factors; or because historical factors have very strong inﬂuences on trade, and geography plays a less important role in determining openness since the start of reforms. As expected, local government involvement in

308

Case studies of East Asian economic integration

economic activities has a signiﬁcantly negative inﬂuence on industrial agglomeration. Economic geography According to regression 1, the coastal regions tend to have a higher share of industrial GDP, which is probably because those regions are closer to international markets. Moreover, since the autonomous municipal cities have entered the post-industrialization era, they tend to have a lower level of industrial agglomeration. New economic geography Our study does conﬁrm the new economic geography theory for explaining the trend of China’s industrial agglomeration between 1987 and 2001. The positive eﬀects of per capita GDP and the number of ﬁrms in the regional industrial share are both highly signiﬁcant. Hence, per capita GDP as potentials of market size (the HME in Krugman 1991) and the number of ﬁrms as industrial externality are both determinants of industrial agglomeration. The eﬀect of regional human capital on industrial agglomeration is insigniﬁcant, though not negative in our regressions. This might be because industrialization in China is still at the initial stage, which is not strongly subject to human capital stock. Finally, improved urban development, better transportation conditions and intensive use of information all have positive and signiﬁcant eﬀect on industrial agglomeration.

10.5

CONCLUSIONS AND POLICY IMPLICATIONS

Since the initiation of economic reforms in the late 1970s and driven by market forces, China’s manufacturing industries have relocated from northern and eastern China and Shanghai into three industrial centers: the Yangtse River Delta, the Pearl River Delta and the Bohai belt. Economic opening has played a critical role in the reshaping of China’s industrial map. In the heated debate on globalization and inequality, our study shows that opening leads to industrial agglomeration and enlarges regional inequality. This is consistent with the ﬁndings in Wan et al. (2007). The trend of industrial agglomeration in China is a result of economic geography, new economic geography and economic policies. The eﬀects of these three factors, along with history and physical geography, can be seen in the post-reform relocation and agglomeration of Chinese industries, even in a relatively small region such as the Yangtse Delta. Among the relevant factors, economic opening is the most signiﬁcant force in reshaping the industrial map. Our econometric modeling of the determinants of

Economic opening and industrial agglomeration in China

309

industrial agglomeration in China indicates that: (1) economic opening, which is also related with geography and history, encourages industrial agglomeration; (2) large market size, eﬀects of forward and backward linkage, a high level of urbanization, better infrastructure and less involvement of local government tend to facilitate industrial concentration; (3) coastal regions have a geographical advantage in attracting ﬁrms. These ﬁndings not only support the new economic geography theory from the evidence within China, but also emphasize the important role that policy itself may play directly in industrial agglomeration. Since industrial concentration among regions causes regional disparity to a large extent, our study provides the government with some policy suggestions. For instance, by improving transportation and facilities for information and communication, local government might attract more ﬁrms and hence develop local industries. More importantly, by hastening integration into the world economy and deregulating, even China’s less developed regions might accelerate their industrial development and thus narrow regional disparity.

NOTES 1. 2. 3. 4. 5. 6.

7. 8. 9. 10. 11. 12.

Most of this section is based on Wan et al. (2007). Commodities under government control can be classiﬁed into two categories: plancommanded goods (both the value and volume of trade were strictly controlled) and plan-guided goods (only the value of trade was controlled). Unless indicated otherwise, data quoted in this section are all from the State Statistics Bureau of the PRC (SSB) (various years). See Table 3 in Démurger et al. (2002) for the timeline of policy initiatives. The stock market represents another avenue for attracting foreign capital. Taiwan and Tibet are blank because data are unavailable. Industrial GDP for Chongqing is merged into that of Sichuan in Figure 9.3(d), although the same has not been done in the following descriptions and econometric analysis since data for Chongqing before 1997 are not available. As mentioned in note 6, the data for Chongqing, the newly established autonomous municipal city, are merged with that for Sichuan in 2001. The Yangtse River Delta includes Jiangsu, Zhejiang and Shanghai. The Pearl River Delta includes Guangdong. The Bo Hai belt includes Beijing, Tianjing, Hebei, Liaoning and Shandong. Authors’ calculation based on SSB, China Statistical Yearbook 2004, China Statistical Publishing House. Data source: Comprehensive Statistical Data and Materials for 50 Years of New China (SSB 1999), Jiangsu Statistical Yearbook, 2003, 2004, and Zhejiang Statistical Yearbook, 2003, 2004. Jin et al. (2005) is a very comprehensive review of the development of industrial clusters in Zhejiang Province. In the economic growth literature, government consumption is represented by the share of government expenditure in GDP excluding education and defense expenditures. Here we use a similar deﬁnition, except that defense expenditure is not included in the local government expenditure statistics. Expenditures for culture, education, science and

310

13.

14. 15.

16. 17.

Case studies of East Asian economic integration healthcare are all public expenditures for human capital and are combined together in the statistics, so they are subtracted from the total expenditures. Privatization has also been a great issue during China’s reforms, which according to existing theories is not related to regional industrial agglomeration. However, in this study we have tested the impact of the privatization of SOEs on industrial agglomeration, which is represented by the ratio of the regional share of workers in non-stateowned enterprises to the national average. However, it is not signiﬁcant, and the regression result does not change when this variable is excluded. Therefore it has not been included in this study. Beijing, Tianjin and Shanghai. Regional population is another important factor that determines local market size according to the new economic geography theory. However, in our data set, local population is signiﬁcantly correlated with the other two explanatory variables, road and ﬁrm (the correlation coeﬃcients are 0.7024 and 0.7196 respectively). If we include regional population into the control variables in the basic regression equation shown in Table 9.2, ﬁrm will become insigniﬁcant because of multicollinearity. Since ﬁrm here is an irreplaceable explanatory variable, which we use to test industrial externality, and population, as a measure of market capacity, could to a large extent be replaced by pergdp to test the HME, we dropped population in the equation. The length of railways is not included here since the data vary unreasonably due to serious measurement errors. In the study of the history of city development in Japan, Davis and Weinstein (2002) found that economic agglomeration and city size have been determined basically by geographical factors (mainly location), and even great shocks in the short run (such as the change of city size due to the A-bomb blasts in World War II) have no signiﬁcant impact on city size in the long run. Similarly, in our study, geography might be a long-lasting factor that determines regional openness in the long run.

REFERENCES Bai, C., Y. Du, Z. Tao and S.Y. Tong (2004), ‘Local protectionism and regional specialization: evidence from China’s industries’, Journal of International Economics, 63 (2), 397–417. Chen, M., Q. Gui, M. Lu and Z. Chen (2007), ‘Economic opening and domestic market integration’, in Ross Garnaut and Ligang Song (eds), China: Linking Markets for Growth, Canberra: ANU E Press and Asia Paciﬁc Press, pp. 369–93. Combes, P. and M. Lafourcade (2000), ‘Transportation cost and regional employment inequalities: evidence from France, 1993’, mimeo, CERAS. Davis, D.R. and D.E. Weinstein (2002), ‘Bones, bombs, and break points: the geography of economic activity’, American Economic Review, 92, 1269–89. Démurger, S., J.D. Sachs, W.T. Woo, S. Bao, G. Chang and A. Mellinger (2002), ‘Geography, economic policy, and regional development in China’, Asian Economic Papers, 1, 146–205. Fan, J. (2004), ‘The integration of Yangtse Delta, regional specialization and spatial shift of manufacturing’ (in Chinese), Management World, No. 11. Fujita, M. and D. Hu (2001), ‘Regional disparity in China 1985–1994: the eﬀects of globalization and economic liberalization’, The Annals of Regional Science, 35, 3–37. Fujita, M. (1988), ‘A monopolistic competition model of spatial agglomeration: diﬀerentiated product approach’, Regional Science and Urban Economics, 18, 87–124.

Economic opening and industrial agglomeration in China

311

Henderson, J.V. (1974), ‘The sizes and types of cities’, American Economic Review, 64, 640–56. Jiang, X. (2002), ‘FDI, market structure and the competing behavior of foreigninvested enterprises’ (in Chinese), Economic Research Journal, 9, 31–8. Jin, X. and X. Zhu (2002), ‘The origin and evolution of specialized industrial clusters: a history and theoretical study’ (in Chinese), Economic Research Journal, 8, 74–82. Jin, X., X. Zhu and J. Ye (2005), ‘The development of private economy and specialized industrial clusters: the case of Zhejiang Province’ (in Chinese), manuscript, Economics School, Zhejiang University. Kanbur, R. and X. Zhang (2005), ‘Fifty years of regional inequality in China: a journey through central planning, reform and openness’, Review of Development Economics, 9, 87–106. Kim, T.J. and G. Knaap (2001), ‘The spatial dispersion of economic activities and development trends in China: 1952–1985’, The Annals of Regional Science, 35, 39–57. Krugman, P. (1991), ‘Increasing returns and economic geography’, Journal of Political Economy, 99, 483–99. Krugman, P. (1993), ‘The narrow and broad arguments for free trade’, American Economic Review, 83, 362–6. Lu, M., Z. Chen and J. Yan (2004), ‘Increasing returns, development strategy and regional economic segmentation’, Studies in Regional Development, 36 (1), 275–306. Naughton, B. (1999), ‘How much can regional Integration do to unify China’s markets?’, paper presented for the Conference for Research on Economic Development and Policy Research, Stanford University. Neary, P. (2001), ‘Of hype and hyperbolas: introducing the new economic geography’, Journal of Economic Literature, 39 (2), 536–61. Poncet, S. (2003), ‘Measuring Chinese domestic and international integration’, China Economic Review, 14 (1), 1–21. Rodríguez-Pose, A. and J. Sánchez-Reaza (2005), ‘Economic polarization through trade: trade liberalization and regional growth in Mexico’, in R. Kanbur, and A.J. Venables (eds), Spatial Inequality and Development, Oxford: Oxford University Press, pp. 237–60. Sachs, J. and A.M. Warner (1995), ‘Economic convergence and economic policies’, NBER Working Paper 5039. SSB (The State Statistics Bureau of the PRC) (1999), Comprehensive Statistical Data and Materials for 50 Years of New China, Beijing: China Statistical Publishing House. SSB (The State Statistics Bureau of the PRC), various years, China Statistical Yearbook, Beijing: China Statistical Publishing House. Tu, J. (2005), ‘A study on the diﬀerentials in FDI distribution in Zhejiang and Jiangsu’ (in Chinese), master’s thesis, Fudan University. Venables, A.J. (1998), ‘The assessment: trade and location’, Oxford Review of Economic Policy, 14 (2), 1–6. Wan, G.H., M. Lu and Z. Chen (2007), ‘Globalization and regional income inequality: evidence from within China’, Review of Income and Wealth, 53 (1), 35–59. Wei, S. and Y. Wu (2001), ‘Globalization and inequality: evidence from within China’, NBER Working Paper 8611. Wen, M. (2004), ‘Relocation and agglomeration of Chinese industry’, Journal of Development Economics, 73, 329–47.

312

Case studies of East Asian economic integration

Xu, X. (2002), ‘Have the Chinese provinces become integrated under reform?’, China Economic Review, 13, 116–33. Xu, X. and X. Li (2005), ‘Domestic market segmentation and regional coordinate development’ (in Chinese), Economic Research Journal, 57–67. Yin, X. (1998), The Procedure and Eﬀects of China’s Reform of International Trade (in Chinese), Shanxi: Shanxi Economic Publishing House. Young, A. (2000), ‘The razor’s edge: distortions and incremental reform in the People’s Republic of China’, Quarterly Journal of Economics, 115, 1091–135. Zhang, F. and J. Zheng (1999), ‘The eﬀects of multi-national companies on China’s economic structure and eﬃciency’ (in Chinese), Economic Research Journal, 1, 45–52. Zhang, X. and K.H. Zhang (2003), ‘How does globalization aﬀect regional inequality within a developing country? Evidence from China’, Journal of Development Studies, 39 (4), 47–67.

Economic opening and industrial agglomeration in China

313

DATA APPENDIX (1) Unless indicated otherwise, data for the period 1987–98 are all from SSB (1999). Data for years 1999–2002, unless indicated otherwise, are from the China Statistical Yearbook, 2000–2003 (SSB, various years). exporti GDPi

(2) export

1 29

exportiGDPi

i 1, 2, . . ., 29.

i

tradei GDPi

(3) trade

1 29

tradeiGDPi

i 1, 2, . . ., 29.

i

(4) gov

1 29

government expenditureGDP in region i i 1, 2, . . . , 29. government expenditureGDP in region i

i

Government expenditure does not include expenditures for culture, education, science and healthcare. (5) port min{distance from provincial capital to Shanghai, distance to Hongkong} Distances from provincial capital to Shanghai and Hongkong are from China electronical map for transportation & touring (Beijing: Tuling software technology company). (6) coast Tianjing, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, Hainan and Guangxi are coastal cities. Beijing is also treated as a coastal city in this chapter since it is very close to Tianjing port and has many preferential policies as the capital city. Others are noncoastal provinces, including Shanxi, Inner Mongolia, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, Hunan, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia and Xinjiang. (7) ﬁrm

number of firms in regioni

number of firms in regioni

i 1, 2, . . . , 29.

i

Data for the number of ﬁrms are from the China Statistical Yearbook for each year.

314

Case studies of East Asian economic integration

(8) edu 1 29

average years of schooling in region i i 1, 2, . . . , 29. average years of schooling in region i

i

The China Population Yearbook reports the regional population by education attainment from 1987. Unfortunately, such data were not published for 1989, 1991 and 1992, and data for 1987 and 1988 are incomplete as the illiterate population is not reported. Also, unlike data for other years, the 1994 data did not include the population below the age of 15. To estimate the data for these years, the average years of schooling (education) was computed using the data for the other years and then ﬁt to the model: ln(edu) f(·) , where edu is per capita years of schooling, f(·) is simply a linear function of time trend and regional dummies, and is the error term. This model is estimated by the GLS technique, allowing for heteroskedasticity in the panel data. The R2 of the estimated equation is 0.966. To denote the value predicted by ^, we have: ^ 2), ed^u exp[ln(ed^u)] exp(0.5 ^ 2 is the estiwhere ln(ed^u) denotes the predicted values of ln(edu) and mated variance of . Data for 1987–89, 1991, 1992 and 1994 are estimated using the above model. (9) pergdp

per capita GDPi

1 29

per capita GDPi

i 1, 2, . . . , 29.

i

Regional per capita GDP is the weighted average of urban and rural per capita GDP, with non-agricultural and agricultural population shares as weights. Both urban and rural GDP are deﬂated by regional urban and rural CPIs. For Shanghai, Beijing and Tianjin, urban and rural CPIs are regarded as the same. urbanization in region i i 1, 2, . . . , 29. urbanization in region i 29

(10) urban 1

i

Urbanization is deﬁned as the proportion of non-agricultural population in the total population of each region. Except for Hebei, Heilongjiang and Gansu, 1999–2001 data for the agricultural and non-agricultural population are from provincial statistical yearbooks. The total populations of

Economic opening and industrial agglomeration in China

315

Hebei, Heilongjiang and Gansu in 2000 are from the China Statistical Yearbook, 2001. For these three regions, the 1999 population ﬁgures are the averages of the years 1998 and 2000, and the 2001 ﬁgures are forecasts based on the ﬁgures in 2000 and the growth rate during 1999–2000.

(11) com

GDP of transportation, post, storage and telecommunication/GDP in region i 1 29

i GDP of transportation, post, storage

i 1, 2, . . . , 29.

and telecommunication/GDP in region i (12) road

length of highway in region i i 1, 2, . . . , 29. length of highway in region i

i

11.

The home market eﬀect in ASEAN countries Ikumo Isono

11.1

INTRODUCTION

This chapter looks at the HME in ASEAN countries and describes the characteristics of ASEAN industries. The following are the main two contributions of this chapter. First, we ﬁnd tendencies toward the HME in a broad number of manufacturing industries in the ASEAN-5 countries (Indonesia, Malaysia, the Philippines, Singapore and Thailand). Second, we ﬁnd a strong linkage between Malaysia and Singapore in machinery industries. The HME is an essential factor in spatial economics, which helps to determine whether economic activities in an industry will be agglomerated or dispersed. The concept of HME predicts that a large market will have a much larger share of industries and exports to other markets. In the world of spatial economics, which is characterized by increasing returns to scale and transport costs, a site with the largest demand for a product in a certain industry becomes the most suitable place to produce it. The product will be exported to other sites. By contrast, in the world of comparative advantage, characterized by constant or diminishing returns to scale, a site with a large demand will be an important import destination for the good. On this point, there is distinct diﬀerence between the worlds of new economic geography and comparative advantage. If an industry has a strong HME, large markets attract more ﬁrms and small markets lose them. Especially for small countries that want to prevent outﬂows of ﬁrms, it is valuable to examine which industries have a strong HME and which do not. This research may oﬀer useful suggestions because the ASEAN countries have small markets and hope to realize a larger market through the economic integration of ASEAN. The economic integration of the ASEAN countries is proceeding based on two main driving forces today. One is FTAs. The other is the rise of China. FTAs in East Asia have mainly been created by ASEAN countries. AFTA (ASEAN Free Trade Area) is made up of the ten members of 316

The home market effect in ASEAN countries

317

ASEAN, and today, many tariﬀs have been lowered to 0–5 percent in the ASEAN-6 (Brunei, Indonesia, Malaysia, the Philippines, Singapore and Thailand) countries using the Common Eﬀective Preferential Tariﬀ (CEPT) scheme. China, Korea and Japan have also signed FTA agreements with ASEAN countries. The early harvest program of the China–ASEAN FTA started on January 2004, and tariﬀ reductions for manufacturing products have been in place since July 2005. A Korea–ASEAN FTA was signed on May 2005 by Korea and the ASEAN countries except Thailand. Institutional economic integration, particularly in the form of FTAs, can be seen as a signiﬁcant factor that reduces transport costs (broadly deﬁned) among the integrated countries. Each ﬁrm makes location decisions for its factories based on its own circumstances. While the concentration of products and relocation of factories by ﬁrms may lead to substantial economic integration, economic integration also changes their circumstances and can induce further relocation. Some countries may hope to restrain the speed of economic integration to prevent the outﬂow of industries. China has a large market and hence is a signiﬁcant production site in Asia. Ishida (2006) examines competition between the ASEAN-5 countries and China, and ﬁnds that ASEAN lost export share and export growth rate in labor-intensive industries from 1995 to 2000 and in machinery industries from 2000 to 2002. Foreign direct investments (FDI) into ASEAN countries have decreased since the Asian economic crisis of 1997. In 1997, China received nearly the same amount of FDI as ASEAN-4 as a whole (Indonesia, Malaysia, the Philippines and Thailand), but in 2004 it received more than six times as much. China poses formidable challenges for ASEAN countries. Meanwhile, for ASEAN countries, the China–ASEAN FTA may bring a golden opportunity for achieving integration with huge markets. It is said that the aim of AFTA is to achieve horizontal regional specialization and to greatly boost FDI by using the advantage of economies of scale through economic integration. ASEAN aims to establish an ASEAN Economic Community (AEC) in 2020 to promote further economic integration in non-tariﬀ trade. Twelve priority sectors for integration are to be fast-tracked by 2007 for the original six member countries: automotive, wood-based products, rubberbased products, textiles, agro-based products, ﬁsheries, electronics, air travel, tourism, ICT, logistics and healthcare. The HME has been discussed in many papers. Krugman (1980), and Helpman and Krugman (1985) construct a two-country model to explain the HME. Because of the presence of transport costs, ﬁrms want to reduce such costs, and a large market becomes the production base for industries with increasing returns to scale. Krugman (1991) argues that lowering

318

Case studies of East Asian economic integration

transport costs accelerates the cumulative causation process. He surmises that small markets will ‘deindustrialize’ due to the outﬂow of ﬁrms. Davis (1998) ﬁnds that if we add the same transport costs for homogeneous goods as well as diﬀerentiated goods into Krugman’s (1980) model, the HME may disappear. Krugman and Venables (1995) ﬁnd that linkage between intermediate goods and ﬁnal good assembly ﬁrms also yields a strong HME.1 It is thought that North–North trade is suitable for a world of increasing returns and that North–South trade is good for a world of comparative advantage. This is because diﬀerences in factor prices, technology levels and factor endowments vary less among developed countries than between developed and developing countries. However, it is worth discussing the HME in ASEAN countries because manufacturing industries are developing rapidly, and diﬀerences in economic development are narrowing in the ASEAN-5. For example, Toyota Motor Thailand assembles pickup trucks, and Toyota Motor Manufacturing Indonesia assembles minivans. These products are exported reciprocally and to the world. In 2005, Thailand and Indonesia took ﬁrst and second place in the ranking of Toyota’s car production and sales in Asia outside of Japan. This suggests that economies of scale also play an important role in the ASEAN countries. In this study we use the method of Behrens et al. (2004) to calculate the HME. They constructed an index for measuring the HME and looked for the presence of the HME in a broad sample of industries in OECD countries. There are two noteworthy features in that paper. First, the index is generated from a model based on new economic geography. Second, a multi-country model is built by extending Krugman’s (1980) two-country model. While Davis and Weinstein (2003) use a two-country model, Behrens et al. (2004) stress the importance of third-country eﬀects and accessibility to other countries. Third-country eﬀects cannot occur in a two-country model.2 We use the 2000 Asian International Input–Output Table by IDE (2006) to calculate the HME index. It allows us to easily obtain data on production share, expenditure share and accessibility for 12 manufacturing industries, as shown in Table 11.1. We limit our discussion to the ASEAN-5 countries, because the 2000 Asian International Input–Output Table does not contain data for the CLMV (Cambodia, Laos, Myanmar and Vietnam) countries or for Brunei Darussalam. The remaining portion of this chapter is organized as follows. An overview of manufacturing industries in ASEAN is given in Section 11.2. Foreign dependence is discussed in Section 11.3. The accessibilityunadjusted HME index is described in Section 11.4. The accessibilityadjusted HME index and foreign dependence are discussed in Section 11.5.

The home market effect in ASEAN countries

Table 11.1

319

Classification of manufacturing industries

008 – Food, beverage and tobacco 009 – Textile, leather, and the products thereof 010 – Timber and wooden products 011 – Pulp, paper and printing 012 – Chemical products 013 – Petroleum and petro products 014 – Rubber products 015 – Non metallic mineral products 016 – Metal products 017 – Machinery 018 – Transport equipment 019 – Other manufacturing products Source: 2000 Asian International Input–Output Table.

In Section 11.6, we examine the linkage between Malaysia and Singapore. We oﬀer a conclusion in Section 11.7.

11.2

MANUFACTURING INDUSTRIES IN ASEAN COUNTRIES

In this section, we look at how manufacturing industries in ASEAN countries have developed. Figure 11.1 shows manufacturing products in 1990 prices. Indonesia, Thailand and Malaysia have grown rapidly since the mid-1980s. The Philippines and Singapore have grown more slowly.3 Figure 11.2 shows the contribution of manufacturing products to the gross domestic product of each of the ASEAN-5 countries. It shows that the growth of manufacturing products allowed Malaysia, Thailand and Indonesia to achieve industrialization and to catch up with Singapore and the Philippines in the 1990s. The share of the manufacturing sector in Indonesia increased from just 9 percent in 1975 to 25 percent in 2002. That in Malaysia became the highest among the ﬁve countries. As a result, the contributions of manufacturing in the ASEAN-5 became very similar.4 The exports of ASEAN-5 countries also have been growing, with the ranking of the export share of machinery being diﬀerent from that of manufacturing products. Figure 11.3 shows the share of the machinery industry’s exports in the total exports of each country.5 Exports of machinery increased more rapidly than those of other industries in the ASEAN-5

320

Case studies of East Asian economic integration 60 50

US$ billion

40 Indonesia Thailand Malaysia Singapore Philippines

30 20 10 0 1975

1980

1985

1990

1995

2000

Source: UN Common Database.

Figure 11.1 Manufacturing products in ASEAN-5 countries (US$ billion)

% 35 30 25 Indonesia Thailand Malaysia Singapore Philippines

20 15 10 5 0 1975

1980

1985

1990

1995

2000

Source: UN Common Database.

Figure 11.2

Contribution of manufacturing products to GDP (%)

321

The home market effect in ASEAN countries % 90 80 70 60

Singapore Malaysia Philippines Thailand Indonesia

50 40 30 20 10

19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03

0

Source: UN Common Database.

Figure 11.3

Export share of machinery industry (%)

countries, and the exports of Singapore and Malaysia are now highly dependent on machinery. There are still great diﬀerences in the export shares of machinery among the ASEAN-5 countries, although they have been narrowing. Singapore still has the highest share, even though it decreased from 71 percent in 1980 to 39 percent in 2002. That of Malaysia rose from 22 percent in 1979 to 28 percent in 2002. While Indonesia had the lowest share, it increased from 1 percent in 1989 to 5 percent in 2000.

11.3

FOREIGN DEPENDENCE AND TRADE FREENESS

It is said that industries in ASEAN countries are export-oriented and highly dependent on foreign markets. Here, we examine the degree of dependence on foreign markets and trade-freeness parameter for several industries. It is helpful to understand the relation between foreign dependence and third-country eﬀects in the new economic geography. Third-country eﬀects refer to the impact of diﬀerences in transport costs for other countries on the location attractiveness of two countries.

322

Case studies of East Asian economic integration

tAB

A

tAC

B

tBC

C

Figure 11.4

Third-country effect

Suppose there are three countries, A, B and C, as depicted in Figure 11.4. They have the same expenditure share. If trade costs are the same between each pair of countries and in both directions, the production share of the three countries should be identical.6 However, if trade between Countries A and C is more accessible than trade between Countries B and C, the production share of the three countries may diﬀer. Country A may receive a much larger share of industries due to the advantage of easier access to Country C, or Country A may lose share due to ﬁerce price competition or due to the relocation of ﬁrms to Country C. We now consider the impact of falling transport costs on the distribution of production. As transport costs decrease, ﬁrms in Country A can ship more goods to Country C, and ﬁrms in Country C also can ship more goods to Country A. As a result, the foreign dependence of Country A and Country C will rise. Lowering transport costs induces ﬁrms to relocate to Country A or Country C because the HME is enhanced. Country A may receive most of the share of the industry due to the advantage of easier access to Country C, or it may lose most ﬁrms due to ﬁerce price competition or due to the relocation of all ﬁrms to Country C. Figure 11.5 shows diﬀerent foreign tendencies among countries and industries. The ﬁgure depicts production share and foreign dependence in terms of purchase and distribution for the ASEAN-5 countries in 2000. Foreign dependence in purchase is the ratio of products purchased by the country from foreign countries. Foreign dependence in distribution is the

50 45 40 35 30 25 20 15 10 5 0

AI010

Distribution – Indonesia

Production share (%)

Production share (%)

Purchase – Indonesia

AI011

AI008

AI009 AI018 AI015

AI012 AI016 AI014 AI019 AI013 AI017

0

20

40 60 Foreign dependence (%)

80

50 45 40 35 30 25 20 15 10 5 0

AI009 AI012 AI016 AI014

AI015

AI019 AI013 AI017

0

100

20

40

AM016

AM017

AM014 AM015

AM008

AM019 AM013 AM012 AM011 AM018 AM009

0

20

40 60 Foreign dependence (%)

80

100

50 45 40 35 30 25 20 15 10 5 0

AM010 AM016

100

AM017

AM014

AM015 AM019 AM011 AM018

AM013 AM008

AM012 AM009

0

20

40 60 Foreign dependence (%)

Source: 2000 Asian International Input–Output Table.

Figure 11.5

80

Distribution – Malaysia

Production share (%)

Production share (%)

323

AM010

60

Foreign dependence (%)

Purchase – Malaysia 50 45 40 35 30 25 20 15 10 5 0

AI010

AI011 AI008 AI018

Production share and foreign dependence in purchase and distribution

80

100

50 45 40 35 30 25 20 15 10 5 0

Distribution – Philippines

Production share (%)

Production share (%)

Purchase – Philippines

AP008 AP019 AP015 AP010

0

20

AP016 AP012 AP009 AP018 AP011 AP014

324

40 60 Foreign dependence (%)

80

AP017 AP013

50 45 40 35 30 25 20 15 10 5 0

AP008 AP019 AP015 AP013 AP012 AP016 AP011AP014

0

100

20

AS013

AS017 AS012 AS016 AS019 AS011 AS018 AS015 AS014 AS009 AS010 AS008

0

20

40 60 Foreign dependence (%)

40 60 80 Foreign dependence (%)

100

Distribution – Singapore

80

100

Production share (%)

Production share (%)

Purchase – Singapore 50 45 40 35 30 25 20 15 10 5 0

AP017

AP009 AP018

AP010

50 45 40 35 30 25 20 15 10 5 0

AS017AS013 AS012 AS016 AS011

AS019

AS018 AS015 AS010

0

20

AS014 AS008

AS009

40 60 Foreign dependence (%)

80

100

Distribution – Thailand

50 45 40 35 30 25 20 15 10 5 0

AT009 AT014 AT019

AT015

AT018 AT012 AT011 AT016

AT008

AT017

AT010

0

20

40

60

Foreign dependence (%)

Figure 11.5 (continued)

AT013

80

100

Production share (%)

325

Production share (%)

Purchase – Thailand 50 45 40 35 30 25 20 15 10 5 0

AT009 AT014 AT019 AT015 AT018 AT012 AT013 AT008 AT011 AT016 AT017

AT010

0

20

40 60 Foreign dependence (%)

80

100

326

Case studies of East Asian economic integration

ratio of the country’s sales to foreign countries. The production share is the country’s share of the industry of the ASEAN-5. Some tendencies to foreign dependence can be seen in industries. ASEAN countries depend on foreign markets because they are highly dependent on machinery. In the machinery sector (017), all countries have a greater than 40 percent foreign dependence in both purchase and distribution. In other manufacturing products (019), all countries have a greater than 40 percent foreign dependence in purchase, and four countries have a ratio greater than 40 percent in distribution. The ﬁgures for other industries are not so high. We can discuss the destination share of sales of each ASEAN-5 country in the machinery sector using the 2000 Asian International Input–Output Table. It suggests that there is a high linkage between Singapore and Malaysia in machinery goods. Surprisingly, only 8.7 percent of Malaysia’s machinery goods were for domestic sales, a ﬁgure lower than for Singapore (9.2 percent), Japan (10.4 percent), and the USA (26.7 percent). Malaysia received 11.1 percent of Singapore’s exports. Other linkages in machinery are relatively weak. Indonesia, the Philippines and Thailand received only 0.3 percent, 0.8 percent and 2.5 percent of the exports of Malaysia. These same countries received 0.3 percent, 1.9 percent and 2.3 percent of the exports of Singapore. Other industries show diﬀerent tendencies. Food, beverage and tobacco (008) and non-metallic mineral products (015) are mainly produced for domestic consumption. Only Singapore has a greater than 40 percent foreign dependence in purchase, and all countries have ratios under 40 percent in distribution in these industries. Petro products are sold mainly in the domestic market. In the petroleum and petro products industry (013), all four countries but Malaysia have greater than 40 percent foreign dependence in purchase, but only Singapore has a ratio above 40 percent in distribution. A large portion of textiles is sold in foreign markets. In textile, leather and products thereof (009), all four countries but Thailand have greater than 40 percent foreign dependence in distribution. We also ﬁnd a relation between production shares and foreign dependence among countries. In terms of purchase, industries with a higher production share have lower foreign dependence in Indonesia, the Philippines and Thailand. This can be interpreted as meaning that local suppliers make the industry attractive. Industries in Singapore seem to show the opposite tendency. Industries with a higher production share also have higher foreign dependence. This is because Singapore is a small country. In terms of distribution, industries with a higher production share have higher foreign dependence in Malaysia and Singapore. It can be assumed

The home market effect in ASEAN countries

327

that goods in an industry with a higher production share will be exported. For Indonesia, all industries except for machinery (AI017), transport equipment (AI018) and food, beverage and tobacco (AI008) show a similar tendency. Foreign dependence is determined by accessibility to other countries and by other parameters. To examine diﬀerences in accessibility, we examine the trade-freeness parameters of each pair of ASEAN-5 countries for 12 manufacturing industries. We construct the trade-freeness parameter between country i and country j in industry s by

XsijXsji sij Xs Xs ii jj

1 2

(11.1)

where Xsij represents goods in industry s produced in country i and shipped to country j. We use this as a proxy for transport costs. Transport costs in the new economic geography include logistics costs, time costs, tariﬀs, non-tariﬀ barriers, and other costs. It is diﬃcult to obtain all types of costs and aggregate them. We deﬁne trade freeness in this way because it can be assumed that an increasing proportion of exports reﬂects falling transport costs (see Head and Mayer 2004).7 If goods can be traded easily, the trade-freeness parameter rises. If ﬁrms cannot trade due to prohibitive trade costs, the tradefreeness parameter is close to zero. Table 11.2 shows the trade-freeness parameters of ASEAN countries obtained by equation (11.1). It suggests that trade linkages in the machinery industry and those between Malaysia and Singapore are stronger than other linkages. The trade-freeness parameter for machinery (017) is high between Malaysia and Singapore, Malaysia and Thailand, and Malaysia and the Philippines. This appears to reﬂect the lower transport costs between these pairs of countries. Behrens et al. (2004) ﬁnd that the highest trade-freeness parameter between France and Germany is 0.346 for other manufactured products, and that the highest trade freeness parameter between the USA and Canada is 0.525 for transport equipment. We ﬁnd, thus, that the trade-freeness parameter for machinery between Malaysia and Singapore is higher than the highest parameter between the USA and Canada.8 We can discover tendencies in the trade-freeness parameters by taking a simple arithmetic average for industries and for countries. Machinery (017) has by far the highest value of trade freeness (0.189). This suggests that transport costs in machinery products are especially low. Petroleum and

Table 11.2 Trade-freeness parameters of each pair of ASEAN-5 countries 008 – Food, beverage and tobacco

Indonesia Malaysia Philippines Singapore

014 – Rubber products

Malaysia

Philippines

Singapore

Thailand

0.0060

0.0006 0.0027

0.0070 0.0386 0.0033

0.0068 0.0108 0.0015 0.0141

009 – Textile, leather, and the products thereof 328

Indonesia Malaysia Philippines Singapore

Philippines

Singapore

Thailand

0.0095

0.0015 0.0035

0.0189 0.1329 0.0059

0.0044 0.0633 0.0054 0.0243

015 – Non-metallic mineral products

Malaysia

Philippines

Singapore

Thailand

0.0134

0.0057 0.0058

0.0021 0.0701 0.0055

0.0051 0.0107 0.0061 0.0132

010 – Timber and wooden products

Indonesia Malaysia Philippines Singapore

Indonesia Malaysia Philippines Singapore

Malaysia

Indonesia Malaysia Philippines Singapore

Malaysia

Philippines

Singapore

Thailand

0.0064

0.0041 0.0068

0.0020 0.0676 0.0033

0.0031 0.0128 0.0021 0.0134

Malaysia

Philippines

Singapore

Thailand

0.0292

0.0071 0.0055

0.0456 0.0960 0.0108

0.0161 0.0420 0.0068 0.0542

016 – Metal products

Malaysia

Philippines

Singapore

Thailand

0.0057

0.0002 0.0145

0.0011 0.0304 0.0025

0.0024 0.0318 0.0029 0.0019

Indonesia Malaysia Philippines Singapore

011 – Pulp, paper and printing

Indonesia Malaysia Philippines Singapore

017 – Machinery

Malaysia

Philippines

Singapore

Thailand

0.0143

0.0046 0.0039

0.0027 0.0537 0.0082

0.0081 0.0122 0.0054 0.0224

012 – Chemical products

329

Indonesia Malaysia Philippines Singapore

Philippines

Singapore

Thailand

0.0775

0.0167 0.3026

0.0628 0.6933 0.0891

0.0319 0.3406 0.1164 0.1591

Malaysia

Philippines

Singapore

Thailand

0.0049

0.0089 0.0066

0.0054 0.0054 0.0039

0.0044 0.0065 0.0323 0.0016

018 – Transport equipment

Malaysia

Philippines

Singapore

Thailand

0.0586

0.0109 0.0115

0.0344 0.1196 0.0144

0.0337 0.0421 0.0103 0.0350

013 – Petroleum and petro products

Indonesia Malaysia Philippines Singapore

Indonesia Malaysia Philippines Singapore

Malaysia

Indonesia Malaysia Philippines Singapore

019 – Other manufacturing products

Malaysia

Philippines

Singapore

Thailand

0.0095

0.0015 0.0140

0.0189 0.2686 0.0466

0.0044 0.0061 0.0021 0.0463

Source: 2000 Asian International Input–Output Table.

Indonesia Malaysia Philippines Singapore

Malaysia

Philippines

Singapore

Thailand

0.0089

0.0040 0.0563

0.0029 0.2045 0.0188

0.0040 0.0229 0.0308 0.0469

330

Case studies of East Asian economic integration

petro products (013/0.041), other manufacturing products (019/0.040) and chemical products (012/0.037) are in the second-highest group. Transport equipment (018/0.008), food, beverage and tobacco (008/0.009), and timber and wooden products (010/0.009) are the lowest group. Transport costs in these industries seem to be very high. Trade freeness is highest between Malaysia and Singapore (0.1484); between Malaysia and Thailand (0.0502), Singapore and Thailand (0.0360), and Malaysia and the Philippines (0.0358) it is relatively high. The ratios between the Philippines and Singapore, the Philippines and Thailand, and Indonesia and the other four countries are all lower. This reﬂects geographical proximity and other aspects, including tariﬀs and other institutional barriers. There are large diﬀerences in foreign dependence and accessibility among the ASEAN-5 countries. Diﬀerences in foreign dependence between industries and countries have diﬀerent impacts on the distribution of economic activities.

11.4

HOME MARKET EFFECT

The strength of the HME determines whether distribution in an industry will change drastically or not as transport costs decrease. We use the model of Behrens et al. (2004) to calculate the HME. The three-country model in Figure 11.4 is useful as an explanation, although we actually consider a multi-country model. This model is constructed as follows. Consider a world with three countries A, B and C. We assume there are S 1 industries. There are ﬁrms, and each ﬁrm produces goods in one industry. There are S diﬀerentiated goods, made with technology with the same ﬁxed and marginal labor requirements. This means that ﬁrms produce goods with technology characterized by increasing returns to scale. We adopt industry S 1 as the numeraire. Each country has consumers who are exogenously given and internationally immobile. They have preferences for a variety of diﬀerentiated goods and maximize their own utility. Iceberg-type transport costs (tijs ) are incurred for shipping goods of each industry from one country to another.9 This means that tijs ( 1) units of goods melt, becoming 1 unit, when the goods are shipped from Country i (A, B or C) to Country j (A, B or C). Transport costs vary according to industry.

The home market effect in ASEAN countries

331

Let

sAA s sBA sCA

sAB sBB sCB

sAC sBC , sCC

(11.2)

where ijs (tijs )1 , ijs means the trade freeness from Country i to Country j in industry s, and s 1 is both the own- and cross-price elasticity of demand for industry s.10 This means that the trade-freeness parameter rises if transport costs decrease or products are more diﬀerentiated. This is because less diﬀerentiated goods do not sell as diﬀerentiated goods in other countries due to transport costs. We assume that ﬁrms can ship goods within the country at no cost. We also assume that transport costs from one country to another are the same in the opposite direction, and that goods for S1 can be freely traded. This ensures factor price equalization. Each ﬁrm can choose the country in which to produce the goods. For example, let us assume that a ﬁrm in industry s is located in Country A. The ﬁrm produces products to fulﬁll the demand of Countries A, B and C. Due to iceberg transport costs, the ﬁrm must produce more units for countries B and C. The ﬁrm earns revenues from the three markets after price competition. Each ﬁrm incurs ﬁxed costs and wages for marginal labor requirements.11 Firms maximize their proﬁts given their circumstances, and we can determine the equilibrium for delivered price in price competition given the location pattern of ﬁrms. We can also determine the conditions for non-positive proﬁts because we assume that ﬁrms can freely enter and exit the market. Each condition encompasses the parameters of all trade-freeness parameters (SAA, SAB, …, SCC ), all expenditure shares (A, B, C) and all shares of the number of ﬁrms ( SA, SB, SC ). Finally, we obtain the production shares S* S* ( S* A , B , C ) for each industry in spatial equilibrium. We then consider two other types of production share. One is production share determined only by the expenditure shares of the three countries. It is deﬁned as the production share in equilibrium when we assume that all trade costs from one country to another are the same (SAB SBC SCA). We call this the attractive factor. The other is the production share determined only by accessibility to other countries. It is deﬁned as the production share in equilibrium when we assume that all expenditure shares of countries are the same (A B C). We call this the accessibility factor. Then we can decompose the production shares in spatial equilibrium into the attractive factor and the accessibility factor using all s

332

Case studies of East Asian economic integration

trade-freeness parameters (SAA, SAB, …, SCC ) and all expenditure shares (A, B, C). Next, we discuss the attractive factor in a multi-country model. We assume that all trade costs from one country to another are equal. Solving the model of Behrens et al. (2004), the production shares in interior equilibrium are obtained as

N s ( si ) size i 1 c s i N1 , c

(11.3)

where Nc is the number of countries and s is the common trade-freeness parameter between one country and another, assuming that all trade costs from one country to another are equal. If no goods of industry s are traded, the common trade-freeness parameter is close to zero, and the production share of each country becomes identical to the expenditure share of the country. When the common trade-freeness parameter is positive, a country with a larger expenditure share than the average of the three countries can get a much larger production share, and a country with a smaller expenditure share than the average gets a much smaller production share due to Ncs/(1–s) 0. If transport costs decrease, the common trade-freeness parameter rises, the production share of the country with the smallest expenditure share may reach zero, and as result, we cannot obtain interior solutions. From now on, we only consider interior solutions. If (11.3) holds completely for all countries, i j ⇒

( si ) size ( sj ) size i j

also holds for all countries where ( si ) size is the production share determined by the attractive factor. This refers to the HME, because it means that a market with a larger market share will have a much larger share of industries. Behrens et al. (2005) use the sign tests as Zs

i

j

( si ) size ( sj ) size (i j ) 0 i j

(11.4)

for industry s. If (11.3) holds, (11.4) must be positive. This is a test of the existence of the HME. If (11.4) is signiﬁcantly positive, we can conjecture that the HME exists for the industry. In the present chapter, a signiﬁcance test cannot be done because we focus on just ﬁve countries. We calculate only (11.4) for an HME index.

The home market effect in ASEAN countries

333

We calculate an unadjusted HME index assuming that all trade costs from one country to another are equal. We obtain the unadjusted HME index (Zs)unadjusted using the real production share instead of the production share determined by the attractive factor on (11.3). We ﬁnd that the indices for food, beverage and tobacco, petroleum and petro products, and machinery are negative, and that the indices for other industries are positive. Rubber products, textile, leather and products thereof, transport equipment, and pulp, paper and printing have relatively high values. This suggests the existence of an HME in these industries. The reason why the indices for petroleum and petro products and machinery are negative seems to be that Singapore has the largest production share in these industries among the ASEAN-5, whereas it has the smallest expenditure share.

11.5

ACCESSIBILITY AND HOME MARKET EFFECT

In this section, we assume a world where transport costs between two countries are uneven. We construct a multi-country trade-freeness matrix with (11.2). Using this matrix and the real expenditure share, we can determine the spatial equilibrium for the model-based production share. This can be decomposed into the accessibility factor and the attractive factor. The calculated attractive factor is obtained by subtracting the accessibility factor from the real production share. We calculate the calculated attractive factor and obtain the adjusted HME index. The adjusted HME index is adjusted because we abstract the calculated attractive factor from the real production share by excluding the diﬀerences in the ease of access to other countries. Next, we ﬁnd the interior solutions of the production share in spatial equilibrium for all manufacturing industries except machinery. For the machinery industry, we obtain the share of Indonesia (29 percent), Malaysia (281 percent), the Philippines (71 percent), Singapore (194 percent) and Thailand (87 percent) in a ﬁctitious interior solution. Although it is unrealistic, we see that Singapore seems to be an attractive country for machinery production. Malaysia’s share turns out to be the lowest. But this does not mean Malaysia is the least attractive country for machinery. If we change the method of data construction for accessibility to be

334

Case studies of East Asian economic integration

XsijXsji ~ sij Xs Xs ii jj

2 3

instead of (11.1), we obtain ﬁctitious production shares for Indonesia (15 percent), Malaysia (436 percent), the Philippines (63 percent), Singapore (237 percent) and Thailand (50 percent). This implies that Malaysia is also an attractive country for machinery.12 The above reﬂects three facts. First, trade costs in machinery in ASEAN countries are lower than in other industries and are low enough to allow agglomeration. Second, Singapore and Malaysia may have a relatively large domestic expenditure or good accessibility to other countries. Third, there is a strong linkage between Malaysia and Singapore. Due to the small transport costs between the two countries, agglomeration in either Singapore or Malaysia occurs in this calculation. It is similar to Country A and Country C in Figure 11.4. Because transport costs between the two countries are suﬃciently low, Singapore may receive most of the share due to the advantage of its easier access to the Malaysian market, or may lose most ﬁrms due to the relocation of all ﬁrms to Malaysia. Next, we merge the data of Malaysia and Singapore to calculate the HME index for the machinery industry. We assume that trade in machinery between Malaysia and Singapore can be carried out costlessly. Accordingly, we can obtain a positive calculated production share determined by the attractive factor for machinery. Figure 11.6 depicts the unadjusted HME index and adjusted HME index. We can detect a tendency toward an HME in many manufacturing industries in the ASEAN-5 countries. The adjusted HME index diﬀers from the unadjusted HME index in that the indices for food, beverage and tobacco, and machinery remain negative. We cannot detect any tendency toward an HME in those industries. The index for petroleum and petro products becomes positive even though the values are very low. This suggests that distribution of production will not change drastically as transport costs decrease in industries that have low or negative HME indices. Figure 11.7 shows the ratio of the accessibility factor and unadjusted HME index. Machinery has the highest ratio for the accessibility factor (35.1 percent). This means that 35.1 percent of the production shares in the ASEAN-5 countries are determined by diﬀerences in accessibility among ASEAN countries, instead of each domestic market. If an industry has a strong HME, large markets will attract ﬁrms and smaller markets will lose them. The distribution of production in the industry will change dramatically as transport costs decrease. In this way, ASEAN industries with a low accessibility factor and a high adjusted HME index seem to face great challenges in the age of FTAs, because the

335

The home market effect in ASEAN countries 1.2 1 0.8 0.6 0.4

Unadjusted Adjusted

0.2 0 008 009 010 011 012 013 014 015 016 017 018 019 –0.2 –0.4 –0.6

Note: * The adjusted SHME index for machinery is calculated from the sum of data of Malaysia and Singapore.

Figure 11.6

Unadjusted and adjusted HME index in ASEAN countries

ASEAN markets are smaller than China, are near China, and will see falling transport costs. Textile, leather and the products thereof (009), pulp, paper and printing (011), and transport equipment (018) fall under this category. Transport equipment has the lowest value for accessibility factor and a relatively high adjusted HME index. This is partly because high tariﬀs on transport equipment products in 2000 diminished accessibility. At present, barriers in transport industries are falling as the AICO (ASEAN Industrial Cooperation Scheme) and CEPT has come into broad use. Exports of passenger vehicles from Thailand to Indonesia expanded from US$4.2 million in 2000 to US$416.2 million in 2005, those from Thailand to the Philippines increased from US$0.7 million in 2000 to US$243.7 million in 2005, and those from Thailand to Malaysia increased from US$0.2 million in 2000 to US$56.2 million in 2005. It would be beneﬁcial for ASEAN to expand its home market by promoting further economic integration. However, it will still face diﬃculties in adjusting to China. The auto industry in ASEAN would require a strong HME generated by linkages between intermediate goods and ﬁnal good assembly ﬁrms, as argued by Krugman and Venables (1995). In textile, leather, and products thereof (009), products are mainly sold out of ASEAN, as mentioned in Section 11.3. The HME index suggests that ASEAN countries would face a very diﬃcult situation in the textile industries.

336

Case studies of East Asian economic integration 1.2 014

1

Adjusted HME

0.8 0.6 0.4

009 018

011

010 015

016

0.2

019 013

0 –0.2 –0.4

012

5

10

15

20

25

30

35 017 40

008

–0.6 Accessibility factor (%) Note: * The ratio for machinery is calculated from the sum of data of Malaysia and Singapore.

Figure 11.7

Ratio of accessibility factor and unadjusted HME index

Industries with a high accessibility factor and a low adjusted HME index seem to be little aﬀected by FTAs. Petroleum and petro products (013), machinery (017), and other manufacturing products (019) belong to this group. Chapter 7 tests the HME and the reverse HME, based on Feenstra et al. (2001).13 It ﬁnds the HME in transport equipment (018) for ﬁnal goods, rubber products (014) for intermediate goods, and pulp, paper and printing (011) and chemical products (012) for both ﬁnal and intermediate goods. In this chapter, we do not consider the USA, Japan, and the EU and treat them as a part of the rest of the world. The ASEAN-5 countries export a large amount of machinery goods to the USA, Japan, and the EU. If we were to take these countries into account, the accessibility factor for machinery would likely increase.

11.6

LINKAGES BETWEEN JOHOR AND SINGAPORE

The conclusions of the previous sections suggest a linkage between Malaysia and Singapore. In this section, we focus on the linkage between Singapore and the state of Johor in Malaysia. Singapore is located south of

The home market effect in ASEAN countries

337

25000

US$ million

20000 Singapore Selangor Johor Pulau Pinang Melaka Negeri Sembilan Other states (average)

15000

10000

5000

0 1985

1990

1995

2000

2002

Source: UN Common Database and Department of Statistics, Malaysia.

Figure 11.8 Manufacturing products in Malaysia and Singapore (US$ million) Johor, with the two being geographically linked by two bridges; the Causeway and the Second Link. Figure 11.8 lists manufacturing products in Malaysia and Singapore. We ﬁnd an uneven distribution. Manufacturing products from Selangor, Johor and Penang account for 63 percent of Malaysian manufacturing products, while those of Singapore are higher than the total of those three states. Among the Malaysian states, the shares of Selangor, Johor and Penang increased from 1985 to 1995, while those of Melaka, Negeri Sembilan and many other states decreased during that period, but improved from 1995 to 2002. We can surmise that the three large states were the pioneers of rapid growth, and that industrial development then spread across other states. A similar development pattern is observed among the ASEAN-5 countries, as discussed in Section 11.2.14 The increase of manufacturing products and exports in the mid-1980s was mainly the result of an increase in multinational foreign aﬃliates in Malaysia. Indeed, the number of ﬁrms established in Johor also increased in the mid-1980s. Before 1980, some ﬁrms such as Kyushu Matsushita or Mitsumi had located in Johor. After 1985, Asahi industries, Hitachi Electronic Devices, Aiwa, Funai and Mitsubishi established presences there. Flextronics, which was incorporated in Singapore in May 1990,

338

Case studies of East Asian economic integration

located in Johor in 1991. Soode, which established as Soode Optik in Singapore in 1987, set up Soode Johor in 1994. In 1995 Seagate moved in. These ﬁrms sought cheaper labor, easy access to Singapore and easy access to the world market through the logistics hub of Singapore. There were reasons for the rapid increase of foreign aﬃliates in the mid1980s. First, Singapore experienced a recession in 1985 due to the economic slump in the USA. An economic committee was formed and issued a report entitled ‘The Singapore economy: new directions’. The report included many policy proposals, including: (1) changing from a production base to being a regional operational headquarters; (2) promoting service sectors, such as banking and ﬁnance, transport and communications, and international services; and (3) promoting oﬀshore activities. The concept of the growth triangle was announced in 1989 by Goh Chok Tong. It aimed to promote a relocation of labor-intensive industries to neighboring areas such as Johor in Malaysia and Riau in Indonesia. Second, Malaysia continued its deregulation of FDI and provided incentives to ﬁrms. Malaysia responded to Singapore with the Promotion of Investments Act 1986, which oﬀered incentives including pioneer status and an investment tax allowance. Third, the Plaza Accord of 1985 encouraged Japanese ﬁrms to relocate abroad. For these reasons, Singaporean and Japanese ﬁrms came to Johor, and the linkage between Singapore and Malaysia was tightened. This linkage has been maintained by reinvestments by foreign aﬃliates. In 1990, total capital investment in approved new projects was RM 26 809 million, while that in approved expansion or diversiﬁcation projects was just RM 1359 million. However, in 2005, in the sum in approved new projects decreased to RM 13 843 million, while that in approved expansion or diversiﬁcation projects grew to RM 17 213 million. Reinvestment has a particularly high share in electronics and electrical products. In 2005, investment in approved new projects was only RM 2652 million, while that in approved expansion or diversiﬁcation projects was RM 11 142 million. Malaysian Industrial Development Authority data also show that the number of approved new projects decreased after the Asian crisis, while that of approved expansion/diversiﬁcation projects increased. Malaysia is still highly dependent on electronics and electrical products, and Johor is the largest creator of jobs dealing with electronics and electrical products. This reﬂects the growing importance of Johor. In 2005, electronics and electrical products received 44.4 percent of total capital investments and 41.2 percent of potential employment through the implementation of projects. For potential employment on all projects approved from 2002 to 2005, Johor’s share was 35.8 percent, the highest of all states. We may note that Singaporean ﬁrms tend to locate in Johor while US

339

The home market effect in ASEAN countries

Table 11.3 Source of approved projects in E&E industries by state (2005)

Johor Kedah Kelantan Melaka Negeri Sembilan Pahang Perak Perlis Pulau Pinang Sabah Sarawak Selangor Terengganu Kuala Lumpur Labuan

None

others

USA

Japan

Singapore

Total

4 4 0 1 2 1 2 0 12 0 1 12 1 2 0

7 2 0 3 4 0 4 0 24 0 1 15 0 0 0

4 4 0 0 0 0 1 0 14 0 0 1 0 0 0

9 2 1 2 0 0 2 0 1 0 0 10 0 0 0

42 1 0 8 0 0 0 0 9 0 0 11 0 0 0

66 13 1 14 6 1 9 0 60 0 1 49 1 2 0

Source: MIDA.

ﬁrms tend to locate in Penang. Table 11.3 shows the number of approved projects by source and state in 2005. The share of Singaporean projects in Johor was 63.6 percent of all Singaporean projects and 59.2 percent of the total Johor projects. Japanese ﬁrms tend to locate in Johor and Selangor. This table shows that the main actors in Johor are Singaporean and Japanese ﬁrms. For example, the numbers of Flextronics projects approved in Johor, Penang, Malacca and Selangor are 5, 3, 3 and 1, respectively. The numbers of JCY HDD projects approved in the same three states are 10, 3 and 1, respectively. JCY HDD is a Malaysian-owned company, although its headquarters is in Singapore.

11.7

CONCLUSION

We ﬁnd that many HME indices are positive, although we cannot present a signiﬁcance test in this chapter. In the machinery industry, we do not ﬁnd the existence of an HME. We have discussed transport costs and HME indices. If transport costs decrease in an industry, the HME is enhanced and large markets will gain ﬁrms and small markets lose them. In an era of FTAs, industries in ASEAN countries, which have high transport costs and a high HME index, seem to

340

Case studies of East Asian economic integration

face major challenges. The industries in which this eﬀect is particularly strong are textile, pulp, paper and printing, and transport equipment. It seems that an integrated ASEAN market will be required to prevent the outﬂow of ﬁrms. In fact, AFTA has induced the restructuring of locations in ASEAN countries. It appears that industries with low transport costs and a low HME index will be unaﬀected by FTAs. The petroleum and petro products, machinery, and other manufacturing products are in this group. Also, considering third-country eﬀects, we ﬁnd a strong linkage between Malaysia and Singapore in the machinery industry. Because our model does not include land constraints, an agglomeration in the machinery industry may occur in Singapore. There are various limitations to this analysis. One is that we neglect diﬀerences in wages, endowments and technology levels. Behrens et al. (2004) assume a costlessly tradable good, and solve the model to ensure factor price equalization. There are large diﬀerences in wages even in the ASEAN-5 countries. Another limitation is that we ignore the rest of the world. This chapter cannot incorporate the rest of the world because trade costs there cannot be zero, and we cannot construct the data for these trade costs. Particularly in the machinery industry, ASEAN countries export a large portion of their products to the rest of the world, such as the USA and Japan. Incorporating the rest of the world would increase the ratio of the accessibility factor, and that would likely change the HME index.

NOTES 1. 2. 3. 4.

5. 6. 7.

8.

Chapters 8 and 9 in this book take a detailed look at this aspect. We discuss third-country eﬀects in Section 11.3. While Vietnam has been growing rapidly in the early 2000s, its manufacturing products amounted to less than 10 percent of those of the Philippines in 2002. It is regularly asserted that the catch-up of the late starters in Asia was brought about by their export-oriented development strategy, which was often accompanied by a ﬂyinggeese pattern of industrialization. It was frequently observed in East Asia that industries spread from industrialized countries to surrounding less industrialized countries as the former came to specialize in more advanced products. We use the SITC (Rev. 2) data. We assume that transport costs are not suﬃciently low to induce agglomeration in one country. Throughout this chapter we assume that ﬁxed and variable costs are the same across countries. This means that we do not address comparative advantage but address diﬀerence in market sizes and accessibility. This is partly justiﬁed by the fact that diﬀerences in exports are more apparent than diﬀerences in development, as mentioned in Sections 11.2 and 11.3. We also assume factor price equalization for (11.1). See Section 11.4. Behrens et al. (2005) used ISIC data while we use 2000 Asian International Input–Output Table.

The home market effect in ASEAN countries 9. 10. 11. 12. 13. 14.

341

See Chapters 2 and 3 for iceberg-type transport costs. s In this model ij corresponds to (11.1), as discussed in Section 11.3. See note 7. ˜ ijs is a monotonic transformation of (11.1) and represents decreasing transport costs. There are no diﬀerences between sij and sij in rank order of accessibility among each pair of two countries. It uses the IDE Input–Output Table for East Asian countries. These facts suggest that the industrialization process in Asia follows the ﬂying-geese pattern, as discussed in note 4.

REFERENCES Behrens, K., A.R. Lamorgese, G.I.P. Ottaviano and T. Tabuchi (2004), ‘Testing the “home market eﬀect” in a multi-country world’, CEPR Discussion Paper 4468. Davis, D. (1998), ‘The home market, trade, and industrial structure’, American Economic Review, 88, 1264–76. Davis, D. and D. Weinstein (2003), ‘Market access, economic geography and comparative advantage: an empirical test’, Journal of International Economics, 59, 1–23. Feenstra, R.E., J. Markusen and A. Rose (2001), ‘Using the gravity equation to diﬀerentiate among alternative theories of trade’, Canadian Journal of Economics, 34, 430–47. Head, K. and T. Mayer (2004), ‘The empirics of agglomeration and trade’, in V. Henderson and J.F. Thisse (eds), Handbook of Regional and Urban Economics, vol. 4, Amsterdam: North Holland, pp. 2609–69. Helpman, E. and P. Krugman (1985), Market Structure and Foreign Trade, Cambridge, MA: MIT Press. Institute of Developing Economies (2006), Asian International Input–Output Table 2000, IDE-JETRO. Ishida, M. (2006), ‘Competition and cooperation between China and ASEAN in international trade’, in Y. Onishi (ed.), New Developments in Economic Relations between China and ASEAN: Entering an era of mutual investment and FTAs, IDE Research Series No. 549 (in Japanese), pp. 33–74. Krugman, P. (1980), ‘Scale economies, product diﬀerentiation and the pattern of trade’, American Economic Review, 70, 950–59. Krugman, P. (1991), ‘Increasing returns and economic geography’, Journal of Political Economy, 99, 483–99. Krugman, P. and A. Venables (1995), ‘Globalization and the inequality of nations’, Quarterly Journal of Economics, 110, 857–80. Ministry of Trade and Industry (1986), The Singapore Economy: New Directions Report of the Economic Committee, Singapore.

Index accessibility and home market effect 333–6 accessibility factor 331 AFTA (ASEAN Free Trade Area) 109, 316–17 agglomeration 15–17 in Chinese industry 288–303 East Asia 27–35 econometric analysis of determinants 303–8 and industrial opening, China 276–309 and regional inequality 35–8 agglomeration force 5, 43, 58–70 agricultural goods, effects of import tax 64–5, 69–70 Akamatsu, K. 181 Amiti, M. 208 Anderson, J.E 183, 201 anti-monde modeling 154–5 Arita, T. 32 ASEAN (Association of South East Asian Nations) 108–10, 150–51 soft infrastructure development 120–22 ASEAN–China FTA 110, 112, 129, 317 ASEAN Free Trade Area (AFTA) 109, 316–17 Asian currency crisis 109–11 Asian integration compared with European integration 150–52, 160–63 motivation for 150 Asian Triangle 77–8 industrial clusters 79–87 and Japanese clusters 90–91 Association of South East Asian Nations, see ASEAN attractive factor 331–2 automobile industry 28–30, 34 flowchart model 97–8, 103

impact of tariff reductions on location 263–70 Japanese MNEs 250, 256, 261–71 automobile industry clusters 79–80, 91–2 India 84 Malaysia 85 Thailand 85–6 backward linkage 16 balance of power Asia 20–21, 149, 151 European Union 148–9 Baldwin, R.E. 60, 165–6, 207 Bayoumi, T. 155 Behrens, K. 318, 327 Belderbos, R. 252 Bergstrand, J.H. 183 biotechnology industry, flowchart model 103 Bohai Rim economy 81–2 border barriers 93–4 Brander, J. 181 Braunerhjelm, P. 160 Brülhart, M. 207 Cambodia, Japanese MNEs 270 Canon 88 capacity-building for industrial clusters 87–9 Cheang, C.T. 205 Chen, M. 295 Cheng, S. 254 China automobile industry 28 capital investment 26–7 determinants of agglomeration 288–308 die-casting industry 83 economic development and E. Asian economy 20, 109–11, 165

343

344

Index

economic opening and industrial agglomeration 276–309 FDI 162–3 Japanese MNEs 114, 253–5, 258–60, 261–2 Korean investment 219–27 regional inequality 36–7 semiconductor industry 33 share of trade 21–6 soft infrastructure 122 spontaneous development 114 China–ASEAN FTA 110, 112, 129, 165, 317 China plus one 114 Choi, Y.H. 233 circular causation of agglomeration 16–17 clusters 77–94 and Asian economic integration 92–4 Asian Triangle 79–87 formation 87–9 cluster-to-cluster linkages 90–92 Combes, P. 277 communication industry, Japanese MNEs 256 comparative advantage, theory of 4, 180–81 competition and MNEs 249–50 composition of intra-regional trade, East Asia 175–8 core–periphery structures 15, 36–7 country fixed effects, intra-regional trade 186–8, 189, 196–7 Court of Justice, European 146 cross-border capital flows, China 278–9; see also foreign direct investment cross-border infrastructure physical 120 soft 120–25 cumulative FDI, Japan 258 currency crisis, effect on E. Asian economy 109–11 Cyber Port, Hong Kong 82 Dai, E. 253 Davis, D. 318 de facto integration 3 Deardorff, A.V. 183, 184, 233

decision-taking ASEAN 151 European Union 149–50 demand conditions, flowchart approach to cluster policy 103 destination of exports 179–80 destination share of sales, ASEAN 326 determinants of intra-regional trade 182–98 die-casting industry, China 83 Dixit, A.K. 60 domestic regions, industrial disparity 66–70 domino effects, FTAs 164 double trade creation 154 Dowling, M. 205 Dunning, J.H. 205, 207, 209 eclectic theory of FDI 205 Economic Agreement for a New-Age Partnership 113 economic development, spontaneous 113–14 economic geography and location of Chinese industries 298–9, 304–5, 308 economic opening and industrial agglomeration, China 276–309 economic policies, and location of Chinese industries 299, 304, 307–8 economic reform and industrial agglomeration, China 280–81, 288 Eichengreen, B. 155 electronics industry 30–33, 34 energy issues and regional integration 127–9 environment issues and regional integration 127–9 equilibrium location of firms 62–5 effects of trade liberalization 63–5 EU (European Union) 145–50 EU-25, intra-regional trade 175–9 Europe economic integration 152–60 integration dynamics 163–4 political integration 144–50 European Coal and Steel Community (ECSC) 145 European Commission 146, 147–8 European Council 146

Index European Court of Justice 146 European integration 143–50, 152–60, 163–4 effect on growth 158–9 European Parliament 146 European Union (EU) 145–50 EU-25, intra-regional trade 175–9 exports 21 by destination 179–80 machinery industry 319, 321 factor prices and fragmentation 53–5 Fan, C.C. 28 Fan, J. 289 FDI, see foreign direct investment Feenstra, R.C. 158, 184, 201 fiscal transfers, European Union 149 fixed capital formation 26 flowchart approach to industrial cluster policy 87, 96–104 flying-geese model 19–20, 181–2, 204–5 footloose-capital model 60–65 foreign affiliates, Malaysia 337–8 foreign dependence and trade freeness 321–30 foreign direct investment (FDI) Asia 161 China 278–9 East Asia 25–6 European Union 157–8 theories 204–8 forward linkage 16 fragmentation 4, 52–8 and declining transport costs 18 and factor prices 53–5 and tariff imposition 56–8 and trade costs 92–4 welfare effects 55 free trade agreements (FTAs) 112–13, 129, 316–17 effect on Korean FDI 236 free trade zone, Hong Kong 82–3 Freund, C. 153 FTAs, see free trade agreements Fujita, M. 15, 20, 36, 92, 223–4, 233, 276 GDP per capita, East Asian countries 35–6 GDP share, East Asian countries 21

345

GMS Program, Asian Development Bank 89 Gokan, T. 92 government policy and industrial agglomeration, China 294–5 gravity equation 183–4 gravity model, anti-monde modeling 155 Greater Mekong Subregion 77–8 Green, M.B. 233 Griffith, R. 156 Hanink, D.M. 209 Hanson, G. 158, 207 harmonious society 114 Head, K. 233 Heckscher–Ohlin (H–O) theory 181 Helpman, E. 183, 317 heterogeneity 16 high-tech industrial development zones 83–4 high-tech industries, Vietnam 86–7 home market effect (HME) 5, 59–65, 181, 316–40 and accessibility 333–6 calculation of 330–33 testing for 187, 189, 196 Hong Kong, IT industry 82–3 Hu, D. 36, 223–4, 276 Hyundai Automotive 231–2 import tax 47–8 effects on agricultural goods 64–5, 69–70 and fragmentation 56–8 reduction, effects of 49–50 imports, East Asian countries 21 income convergence 35–6 EU 158–9 India automobile industry clusters 84 Software Technology Park 84–5 industrial agglomeration, see agglomeration industrial clusters, see clusters inequality, regional 5, 35–8 infrastructure development 120–25 innovation capacity 34–5 institutions for integration, Europe 145–7

346

Index

integration–disparity nexus 5 integration dynamics 163–6 Asia 164–6 Europe 163–4 international trade theories 180–82 intra-industry trade 161 intra-regional trade 21–7, 173–200 composition 175–8 determinants of 182–98 by economic function 177, 179 structure of 174–80 volume 174–5 Investment Encouragement Law, Thailand 85–6 Ishida, M. 317 IT industry flowchart model 103 Japanese MNEs location 256 Jacquemin, A. 154 Japan automobile industry 30 as core economy 19–20 and environmental issues 127 industrial clusters and Asian Triangle 90–91 investment in China 114 logistics action plan 138 MNEs, see Japanese MNEs regional strategy 111 semiconductor production 32 soft infrastructure development 125, 137 Japan–China Joint Venture Dalian Industrial Zone 82 Japanese MNEs 249–71 automobile industry 90–92, 250, 256, 261–70 China 253–5, 258–60 determinants of location 251–5 distribution 255–8 impact of tariff reduction 263–70 Jee, M. 224 Jiang, X. 291 Jiangsu 254, 290 Jin, X. 291, 295 Johnston, R.B. 96 Johor, Malaysia 336–9 Jones, L.P. 227

Kawai, H. 258 Kim, H. 233 Kim, H.Y. 233 Kim, T.J. 276 Knaap, G. 276 Kojima, K. 181 Korea FDI 210–24 in China 90, 219–24 MNEs development of 209–24 location choices 227–32 Pusan Port 90 SMEs, location choices 232–5 Korea ASEAN FTA 317 Krugman, P. 181, 207, 317–18 Kuchiki, A. 82, 89 labor costs, effect on industrial agglomeration, China 298 Lafourcade, M. 277 Lall, S. 210, 213 Lamberte, M.B. 150 Lee, H. 233 LG Electronics 230 Li, X. 295 logistical efficiency and regional integration 121–2 Lu, M. 295 Malaysia automobile industry 85 foreign affiliates 337–8 links with Singapore 336–9 manufactured goods, composition of trade 176–8 manufacturing industries, ASEAN countries 319–21 manufacturing networks, MNEs 114; see also clusters market integration, EU 155–7 Markusen, A. 96 Markusen, J.R. 203, 205, 206, 208 Marwah, K. 249 Mason, M. 203 McCann, P. 32 McLaren, J. 153 Meyer, S.P. 233 middle classes, role in regional integration 113, 125–6

Index Midelfart-Knarvik, K.-H. 160 MNEs, see multinational enterprises monopolistic competition model 60 multinational enterprises (MNEs) benefits to host countries 249 Japanese, see Japanese MNEs Korean, see Korean MNEs Malaysia 337–8 share of trade 24–6 value chain managements 89–90, 91–2, 105 multinational foreign affiliates, Malaysia 337–8 NAFTA, intra-regional trade 175–80 Narula, R. 209 nationalism as impediment to integration 117–19 Navaretti, B. 206 Neary, P. 302 NEFTA (Northeast Asian FTA) 165 neoclassical framework 14 new economic geography 207 and location of Chinese industries 302, 305, 308 see also spatial economics new trade theory 206 Nishikimi, K. 60 non-tariff barriers, effects of reduction 50–52 North American Free Trade Agreement, see NAFTA Northeast Asian FTA (NEFTA) 165 Okita, S. 181–2 Ottaviano, G.I.P. 208 Overman, H.G. 160 patent applications 34–5 Pearl River Delta economy 80 Pohang Steel 232 political economy and integration Asia 150–52 Europe 144–50 Porter, M.E. 96 PPP (purchasing power parity) 35–6 product life-cycle theory 204 production share and foreign dependence 322–7 protectionism, China 295

347

Proton City Scheme 85 Puga, D. 208, 252 purchasing power parity (PPP) 35–6 Pusan Port, Korea 90 reciprocal dumping model 181 Redding, S. 201 regional diversity as impediment to regional integration 119 regional inequalities 5, 66–70 and agglomeration 35–8 regional integration, East Asia 19–27 development of 107–17 impediments to 117–19 promotion of 119–29 regional trade agreements (RTAs) 1; see also AFTA; European Union; NAFTA; NEFTA reverse home market effects 189, 196 Ricardo, D. 180 Robert-Nicoud, F. 208 Sakong, I. 227 Samsung Electronics 231 Samuelson, P. 181 Sapir, A. 154, 155 Scott, A.J. 28 sectoral specialization and European integration 153–4 semiconductor production 32–3 Shanghai 84, 290 Japanese MNEs 254 Shanghai International Automobile City 81 Singapore, links with Johor, Malaysia 336–9 Single Market Initiative (SMI) 156 Sleuwaegen, L. 252 soft infrastructure 120–25 software industry, Vietnam 87 Software Technology Park of India (STPI) 84–5 Soloaga, I. 155 South Korea, mobile phone industry 31–2 spatial economics 5, 14–19, 207–8 specialization force of economic integration 43, 44–58 spontaneous development 113–14 state aids to industry, EU 159–60

348 state-owned enterprises and agglomeration, China 290–91 Stern, R.M. 233 Stiglitz, J.E. 60 Stough, R.R. 254 straw effect 65 structure of intra-regional trade, East Asia 174–80 Sundararajan, V. 96 Taiwan, semiconductor production 32–3 tariffs and fragmentation 56–8 and Japanese automobile industry location 263–70 see also import tax Tavakoli, A. 249 textile industry, Japanese MNEs location 256 Thailand automobile industry 28–30, 85–6, 262–3 dual track policy 136 Japanese MNEs 262–3 regional inequalities 37 theory of comparative advantage 4, 180–81 third-country effects 321–2 Tomiura, E. 252 Toyota 91–2, 262–3, 318 Toyota City 15 trade, theories of 180–82 trade costs 92–4; see also non-tariff barriers; tariffs; transport costs trade creation and European integration 154–5 trade diversion and European integration 154–5 trade freeness and foreign dependence 321–30 trade growth Asia 160–63 China 278 Europe 152–3 trade liberalization China 278 and domestic regional disparity 69–70 effects on equilibrium 63–5

Index trade share, East Asian countries 21 transport costs 16, 17–19, 45–7 effects on domestic regional disparity 69 effects on production distribution 322 effects of reduction 50–52 transport networks and industrial clusters 88–9 Trefler, D. 252 Truman, E.M. 154 Tsuji, M. 104 Tu, J. 290 United States semiconductor firms in East Asia 32–3 strategic choices, East Asia and Western Europe 107–8 Urata, S. 209, 258 value chain management 89–90, 105 automobile industry 91–2 Van der Bulcke, D. 219, 224 van Wincoop, E. 201 Venables, A.J. 201, 203, 206, 208, 318 Vernon, R. 204 vertical specialization, Asia 161–2 Vietnam electric and electronics industry cluster 88–9 high-tech industries 86–7 volume of intra-regional trade, East Asia 174–5 wage rates differentials, EU 158–9 effect of fragmentation 54–5 effect of import tax 47–8 Wan, G.H. 276 welfare effects of fragmentation 55 of import tax 48 Wen, M. 276, 281 Wincoop, E. van 201 Winters, L.A. 155 worker mobility and agglomeration 16, 17 WTO accession, China 279–80 Wuxi–Singapore Industrial Zone 81

Index Xu, X. 295 Yang, P.S. 223, 231 Yangtze River Delta economy 81 industrial agglomeration 288–98

Zhang, H.-Y. 219, 224 Zhejiang 290–91 Zhou, D. 250 Zhu, X. 291

349

Economic Integration in East Asia: Perspectives from Spatial and Neoclassical Economics

Regional Integration in East Asia: From the Viewpoint of Spatial Economics

Economic Integration in Asia and India

Economic Integration in Asia and India

Growth and Productivity in East Asia (National Bureau of Economic Research-East Asia Seminar on Economics)

Growth and Productivity in East Asia (National Bureau of Economic Research-East Asia Seminar on Economics)

Spatial Disparities in Human Development: Perspectives from Asia

East Asia's Economic Integration: Progress and Benefit

Economics: Marxian versus Neoclassical

Prospects for Monetary Cooperation and Integration in East Asia

Towards Monetary and Financial Integration in East Asia

The Politics of Economic Regionalism: Explaining Regional Economic Integration in East Asia (International Political Economy)

Situational Analysis Beyond Neoclassical Economics

East Asia's De Facto Economic Integration

Fiscal Policy and Management in East Asia (National Bureau of Economic Research-East Asia Seminar on Economics)

Economic Liberalization and Integration in East Asia: A Post-Crisis Paradigm

Politeness in East Asia

Politeness in East Asia

Inequality and Economic Integration

Agriculture and Economic Development in East Asia: From Growth to Protectionism in Japan, Korea and Taiwan

Economic Development, Integration, and Morality in Asia and the Americas (Research in Economic Anthropology)

Regionalism and Global Economic Integration: Europe, Asia and the Americas

Regionalism and Global Economic Integration: Europe, Asia and the Americas

East Asian Visions: Perspectives on Economic Development

Pan-Asian Integration: Linking East and South Asia

Liberation from Liberalization: Gender and Globalization in South East Asia

Regionalisation, Growth, and Economic Integration (Contributions to Economics)

Multinationals and East Asian integration

Informal Politics in East Asia

Deregulation and Interdependence in the Asia-Pacific Region (National Bureau of Economic Research-East Asia Seminar on Economics)

Insolvency Law in East Asia

Economic Integration in East Asia: Perspectives from Spatial and Neoclassical Economics

Regional Integration in East Asia: From the Viewpoint of Spatial Economics

Economic Integration in Asia and India

Economic Integration in Asia and India

Growth and Productivity in East Asia (National Bureau of Economic Research-East Asia Seminar on Economics)

Growth and Productivity in East Asia (National Bureau of Economic Research-East Asia Seminar on Economics)

Spatial Disparities in Human Development: Perspectives from Asia

East Asia's Economic Integration: Progress and Benefit

Economics: Marxian versus Neoclassical

Prospects for Monetary Cooperation and Integration in East Asia

Towards Monetary and Financial Integration in East Asia

The Politics of Economic Regionalism: Explaining Regional Economic Integration in East Asia (International Political Economy)

Situational Analysis Beyond Neoclassical Economics

East Asia's De Facto Economic Integration

Fiscal Policy and Management in East Asia (National Bureau of Economic Research-East Asia Seminar on Economics)

Economic Liberalization and Integration in East Asia: A Post-Crisis Paradigm

Politeness in East Asia

Politeness in East Asia

Inequality and Economic Integration

Agriculture and Economic Development in East Asia: From Growth to Protectionism in Japan, Korea and Taiwan

Economic Development, Integration, and Morality in Asia and the Americas (Research in Economic Anthropology)

Regionalism and Global Economic Integration: Europe, Asia and the Americas

Regionalism and Global Economic Integration: Europe, Asia and the Americas

East Asian Visions: Perspectives on Economic Development

Pan-Asian Integration: Linking East and South Asia

Liberation from Liberalization: Gender and Globalization in South East Asia

Regionalisation, Growth, and Economic Integration (Contributions to Economics)

Multinationals and East Asian integration

Informal Politics in East Asia

Deregulation and Interdependence in the Asia-Pacific Region (National Bureau of Economic Research-East Asia Seminar on Economics)

Insolvency Law in East Asia

Recommend Documents