JBICI Research Paper No. 15
Foreign Direct Investment and Development: Where Do We Stand?
JBIC Institute Japan Bank for International Cooperation
JBICI Research Paper No. 15 Japan Bank for International Cooperation (JBIC) Published in June 2002 © 2002 Japan Bank for International Cooperation All rights reserved. This Research Paper contains a selection of surveys and studies conducted at the JBIC Institute * for the internal use of JBIC and for the general public. The views expressed in the articles of this Research Paper are those of the authors and do not necessarily represent the official position of the Japan Bank for International Cooperation. No part of this Research Paper may be reproduced in any form without the express permission of the publisher. For further information please contact the Planning and Coordination Division of our Institute.
Japan Bank for International Cooperation (JBIC) was established in October 1999 as an organization that conducts Japan’s external economic policy and economic cooperation. JBIC is pursuing a more enhanced role by integrating the functions of two merged organizations: The Export-Import Bank of Japan (JEXIM) and the Overseas Economic Cooperation Fund, Japan (OECF). Upon the establishment of JBIC, the JBIC Institute (JBICI) was created as its research arm. Its research activities are geared toward improving the overall quality of JBIC’s operations through systematic analysis of various issues and policies related to JBIC’s activities. JBICI was established by merging the two former research institutes: the Research Institute for International Investment and Development (RIIID) of JEXIM and the Research Institute of Development Assistance (RIDA) of OECF.
Since 1993, the flow of foreign direct investment (FDI) rapidly expanded across the world. This growth was caused by an increase in mergers and acquisitions among OECD countries as well as by policy initiatives aimed at deepening regional integrity. As this trend gained force, policymakers in host countries started to ask a question: “What implications would these FDIs have on our country’s long-term economic growth? ” Whereas the traditional theory had it that “FDI could become the engine of growth for host countries through the transfer and diffusion of knowledge ”, there is now an increasing need to assess this claim by conducting empirical stud ies. This report is an elaborate analysis of the effects of direct investments on developing economies. The study was the result of collaboration between the Japan Bank for International Cooperation (JBIC) and the OECD Development Centre, led by Mr. Kiichiro Fukasaku. The Development Centre has wide knowledge and expertise of international economic and development issues. In particular, this report makes an empirical examination of the complex interdependence between FDI and economic growth in host countries. It also points to the significance of adopting more constructive, rules-based policies towards FDI. We hope this report will further the reflection on multilateral negotiations on trade and investment under WTO. It was presented to a workshop sponsored by JBIC in Tokyo in December 2001, and discussions at that workshop are reflected in the report. Finally, let me express my appreciation to many people from various institutions, at home and abroad, whose great support and cooperation has made this study possible. June, 2002 Koji Fujimoto Executive Director JBIC Institute
This study is the final product of a collaborative effort that began officially in August 2001 between the Japan Bank for International Cooperation (JBIC) in Tokyo and the Organisation for Economic Co-operation and Development (OECD) in Paris. The idea of undertaking a joint research project between the two organisations was first proposed by Mitsuhide Noguchi, then Director-General of the Research Institute for Development and Finance (hereafter referred to as the JBIC Institute), when he visited Paris in the summer of 2000. Subsequently, the idea received official support with endorsement from Koji Fujimoto, Executive Director of the JBIC Institute. The project was jointly managed by Kaoru Hayashi, Deputy Director-General of the JBIC Institute, and Kiichiro Fukasaku, Head of Research Division II at the OECD Development Centre. The research theme selected, “FDI and Development”, is both timely and important. The OECD Development Centre published a seminal work on this topic nearly 30 years ago (Reuber et al., 1973). Since then, many developing countries have significantly, and in some cases dramatically, altered their regulations and policies towards foreign direct investment. The topic therefore merits further investigation today, taking into account the development of new policy environments in host economies. It is also important that the research on this topic be conducted jointly. The JBIC Institute has accumulated a wealth of information on Japanese multinationals’ foreign affiliates located in Asia by conducting a survey every year since 1989. Owing to time constraints, however, the collaborative effort has been confined for the present to carrying out a literature review in several selected areas. Further research is thus warranted to cover other areas of interest for both home and host economies. The study was primarily drafted by Kiichiro Fukasaku, with substantial contributions from Federico Bonaglia (Chapters Ⅲ and Ⅶ), Andrea Goldstein (Chapter Ⅵ), Charles Oman (Chapter Ⅶ) and Ophélie Chevalier (Chapter Ⅶ), as well as from the OECD Directorate for Financial, Fiscal and Enterprise Affairs (Chapters Ⅱ and Ⅴ). Technical and secretarial support was provided by Ly Na Dollon, Morag Soranna and Mayrose Tucci and by Neil Beshers, external consultant. The authors are very grateful for useful discussions with and continuous encouragement from our colleagues, notably, Jorge Braga de Macedo, Ulrich Hiemenz, Helmut Reisen, David O’Connor, Koji Miyamoto, Maria Maher and Hans Christiansen. The first draft of the study was presented at a seminar in Tokyo on 4 December 2001, eliciting helpful comments and suggestions from Professor Shujiro Urata (Waseda University), Professor Shigeki Tejima ii
(Nishogakusha University) and other seminar participants. This final version has incorporated these invaluable contributions as much as possible within the framework of the project description agreed between the two organisations. Finally, the opinions expressed in this work are those of the authors and do not necessarily reflect those of the Organisation to which they belong.
Table of Contents Foreword ････････････････････････････････････････････････････････････ ⅰ Acknowledgements ････････････････････････････････････････････････････ ⅱ Table of Contents ･････････････････････････････････････････････････････ ⅳ List of Figures ･･･････････････････････････････････････････････････････ ⅵ List of Tables･････････････････････････････････････････････････････････ ⅵ List of Boxes ･････････････････････････････････････････････････････････ ⅵ List of Annex Tables･･･････････････････････････････････････････････････ ⅶ Abstract ･････････････････････････････････････････････････････････････ ⅷ Chapter Ⅰ
Trends in FDI: Global and Regional Perspectives ･･･････････････11 2.1 A Global Surge in FDI since 1993 ････････････････････････15 2.2 Regional Trends ･･･････････････････････････････････････19
The FDI-Growth Nexus ････････････････････････････････････27 3.1 Putting Theory to Work･････････････････････････････････28 3.2 FDI and Growth: Empirical Evidence ･････････････････････31 3.2.1 Threshold Externalities ････････････････････････････33 3.2.2 Local Financial Markets ･･･････････････････････････33
FDI-Trade Linkages ･･･････････････････････････････････････40 4.1 Standard Models ･･････････････････････････････････････41 4.2 The Substitution-Complementarity Hypothesis Revisited ････44 4.3 Aggregation, Causality and Endogeneity ･･････････････････45 4.4 Towards a Unified Approach ････････････････････････････48
FDI and Technology Transfer ･･･････････････････････････････56 5.1 Mechanisms of Technology Transfer ･･････････････････････58 5.1.1 Vertical Linkages with Suppliers and Buyers ･･････････58 5.1.2 Horizontal Linkages through Demonstration and Competition･･････････････････････････････････････60 5.1.3 Labour Turnover ･･････････････････････････････････62 5.1.4 International Technology Spillovers ･･････････････････63 5.2 Technology Transfer and Host-country Conditions ･･････････65
FDI, Privatisation and Corporate Governance･･････････････････71 6.1 Privatisation Trends in the 1990s ････････････････････････72 iv
6.2 Why Is Latin America More Advanced in Privatisation? ･･････76 6.3 Privatisation: Brazilian Style ････････････････････････････77 6.4 An Assessment of Privatisation Policy･････････････････････79 6.5 Implications for Corporate Governance････････････････････80 Chapter Ⅶ
Host-government Policies for Attracting FDI ･･･････････････････83 7.1 Incentive-based and Rules-based Measures ････････････････83 7.2 Some Observations from Case Studies ････････････････････87 7.3 The Cost of Investment Incentives ････････････････････････89 7.4 Towards Constructive Rules-based Policies ････････････････92
Summary and Conclusions ･････････････････････････････････110
List of Figures Figure 1.1. The Role of FDI in Development: Host-country Interests ･･･････････7 Figure 2.1. Trends in World Merchandise Exports and FDI Outflows･･････････12 Figure 2.2. World Merchandise Trade and FDI Flows, 1989-2000 ･････････････15 Figure 2.3. Trends in FDI Inflows, Cross-border M&As and Privatisation ･･････16 Figure 6.1. Global Trends in Privatisation Revenues ･･･････････････････････73 Figure 6.2. Privatisation Revenues by Major Non-OECD Region, 1990-99 ･･････74 Figure 6.3. Privatisation Revenues in Non-OECD Regions by Major Sector, 1990-99････････････････････････････････････････････････････74 Figure 6.4. Privatisation and FDI Inflows in Latin America･･････････････････75 Figure 6.5. Privatisation and FDI Inflows in East Asia･･････････････････････76
List of Tables Table 1.1. US Direct Investment Position in Selected Asian Economies, by Sector ････････････････････････････････････････････････････5 Table 2.1. Foreign Affiliates in Manufacturing Industry of Selected OECD Countries･･･････････････････････････････････････････････････14 Table 2.2. Cross-border M&As by Region/Economy of Seller･･････････････････17 Table 2.3. Cross-border M&As as a Percentage of FDI Inflows ･･･････････････19 Table 2.4. FDI Outflows by Home Region/Economy･････････････････････････20 Table 2.5. World’s 20 Largest Home Economies of FDI ･･････････････････････21 Table 2.6. FDI Inflows by Host Region/Economy ･･･････････････････････････22 Table 2.7. World’s 20 Largest Recipient Economies of FDI ･･･････････････････23 Table 2.8. Net FDI Exporters and Importers ･･････････････････････････････25 Table 7.1. Foreign Investment Regime in the Host Economy: Main Types of Regulatory and Incentive Measures ･････････････････････････････84 Table 7.2. Investment Incentives in the Automobile Industry ････････････････91
List of Boxes Box 2.1. International Investment Instruments････････････････････････････12 Box 2.2. Is China a Special Case?････････････････････････････････････････26 Box 3.1. Short-term and Long-term Impacts of FDI on Growth ･･･････････････30 Box 5.1. FDI as a Mode of Technology Transfer ････････････････････････････57 Box 6.1. The Power Crisis in California: What Went Wrong? ･････････････････72 Box 7.1. Policy Competition for FDI: A Race to the Bottom? ･･････････････････86 vi
List of Annex Tables Annex Table 3.1. FDI and Growth: Literature Survey ･･････････････････････37 Annex Table 4.1. FDI-Trade Linkages: Literature Survey ･･･････････････････50 Annex Table 5.1. FDI and Technology Transfer: Literature Survey ･･･････････68 Annex Table 7.1. FDI Policy Regimes in Latin America ･････････････････････96 Annex Table 7.2. FDI Policy Regimes in Asia･････････････････････････････102
Foreign direct investment (FDI) has been one of the defining characteristics of the world economy during the last two decades. Some developing economies have emerged as major recipients of FDI flows in recent years, while many others have attempted to attract such flows, often by offering fiscal and financial incentives to foreign investors. This work reviews and discusses recent empirical studies on key development issues related to FDI. The literature review focuses on five main areas of interest to host economies: the FDI-growth nexus; FDI-trade linkages; FDI and technology transfer; FDI, privatisation and corporate governance; and host-government policies for attracting FDI. Three of the conclusions reached in this study deserve special attention. First, a vast majority of existing empirical studies indicate that FDI does make a positive contribution to both income growth and factor productivity in host economies. FDI tends to “crowd in” domestic investment, as the creation of complementary activities outweighs the displacement of domestic competitors. Similarly, in the North-South context the relationship between FDI and trade is more one of complementarity than of substitution, owing to backward and forward linkages. Second, host countries will not be able to capture the full benefits associated with FDI until they reach a certain threshold level in terms of educational attainment, provision of infrastructure services, local technological capabilities and the development of local financial markets. The results of recent empirical studies based on microeconomic (firm- or plant-level) data indicate that the “spillover” effect of FDI on the productivity growth of local firms does not occur automatically, highlighting the complex nature of interactions between multinational enterprise (MNE) affiliates and local firms. Third, the role of FDI in development goes beyond the traditional areas of growth, trade and technology transfer to cover emerging areas of policy concern, such as mergers and acquisitions, privatisation, corporate governance and “policy competition”. Further research along these lines is certainly warranted to meet the future challenges of improved regulation and policy making. The study concludes by stressing that host-government policies should attach greater importance to the stability and predictability of the local business environment in which MNE affiliates operate. To this end, host countries should pursue both multilateral and regional approaches to adopting more constructive, rules-based policies concerning FDI.
Chapter Ⅰ Introduction
This study takes a fresh look at key development issues related to foreign direct investment (FDI) in developing and emerging economies in the light of country experience and policy-oriented work available in this area. It draws on existing empirical studies in the literature, including those conducted by the Development Centre and other parts of the OECD. Its main purpose is to advance the policy debate about the role of FDI in sustaining the long-term development of host economies. FDI has been one of the defining characteristics of the world economy over the past 20 years. At the aggregate level, this form of private capital flows has grown at an unprecedented pace for two decades. Over the 1998-2000 period the total amount of FDI flows into developing countries exceeded official aid flows by a factor of more than three. Along with international trade, FDI has been regarded as an engine for economic growth and for integrating developing countries into the world economy. The potential benefits of FDI can come through several channels. First, FDI is less volatile than other private flows and provides a stable source of financing to meet capital needs (Lipsey, 2000; Soto, 2000; Reisen and Soto, 2001). Second, FDI is an important — and probably the dominant — channel of international transfer of technology. Multinational enterprises (MNEs), the main drivers of FDI, are powerful and effective vehicles for disseminating technology from developed to developing countries and are often the only source of new and innovative technologies, which are usually not available in the arm’s-length market. Third, the technology disseminated through FDI generally comes as a “package” including the capital, skills and managerial know-how needed to exploit the technology appropriately. Finally, other potential benefits of FDI for host countries include increased competition in the products market, human capital development and an improvement in corporate governance standards and legal frameworks. Recent years have seen increased public concern that these benefits have yet to be demonstrated and that, where benefits exist, they may not be shared equitably in society1. The adjustment costs associated with FDI include higher
The impact of FDI on the development process has been discussed at length since the late 1960s. See inter alia Reuber et al. (1973) and Lall and Streeten (1977). The latter study provides an extensive review of the debate that emerged in the 1970s and offers one of the first empirical analyses on the effects of FDI on developing countries’ welfare. This question has attracted renewed interest among policymakers in developing countries as their economies have become increasingly open to foreign investment over the last decade.
short-term unemployment due to corporate restructuring and increased market concentration. Clearly, coherence between different sets of policies, institutional and regulatory conditions and the availability of skilled labour and infrastructure also affect a country’s ability to reap the full benefits of FDI. There are nonetheless divergent views on how public policy can help maximise the benefits (and minimise the costs) of FDI in order to meet a country’s development needs. This debate appears to have reached new heights at the beginning of the new millennium. A series of “globalisation backlash” demonstrations from Seattle to Genoa over the past two years expressed opposition to further liberalisation of international trade and investment. Slaughter (2000) attempts to explain the forces behind this phenomenon in the United States, the country whose economy is supposed to have benefited most from globalisation2. His empirical analysis highlights the importance of the “skills-preferences” cleavage among US workers and the breadth of anti-globalisation sentiments that it implies. In contrast, many developing and transition economies have been attracting foreign investors, often through direct incentives, to help modernise their antiquated telecommunication and other infrastructure and to kick-start new industries. Indeed, incentive-based competition for FDI has become a global phenomenon, involving governments at all levels (national and sub-national) in both OECD and non-OECD countries (Oman, 2000). Under these circumstances, it is important to inform the discussion by drawing lessons from country experiences and case studies available in the literature and to assist governments in identifying the conditions and policy requirements for maximising the benefits of FDI and minimising its risks and potential costs. The question is not merely academic, as this debate has important implications for concerted international efforts currently under way both at the United Nations (to achieve the “Millennium Development Goals”) and at the WTO (to launch a new round of trade negotiations)3. Analytical work on the role of FDI in development is also expected to contribute to the development and maintenance of open policies towards FDI. A crucial dimension of this work, therefore, is an
In 1999 the United States became the world’s largest net recipient of FDI (inflows minus outflows), registering a net inflow of $124 billion, compared with $38 billion in China, the second largest (UNCTAD, 2000, Annex Tables B.1 and B.2). 3 WTO Members successfully launched a new trade round at the Fourth Session of the Ministerial Conference held in Doha last November. The relationship between trade and investment is, however, one of several issues that remain politically so sensitive that “negotiations will take place after the Fifth Session of the Ministerial Conference on the basis of a decision to be taken, by explicit consensus, at that Session on modalities of negotiations” (WT/MIN(01)/DEC/W/1. Para. 20).
assessment of why some countries are better able to take advantage of the gains from liberalisation than others. 1.1 The Role of FDI in Development In the development context, FDI is often regarded as a politically sensitive and complicated issue 4. The main reason for the political sensitivity of FDI, as opposed to domestic private investment, is that the former is “not only owned and controlled by private groups in pursuit of private profits but also by private interests that are non-resident to boot” (Reuber et al., 1973, p. 16). The political and social concerns arising from this simple fact were probably the decisive factor in determining the attitudes and policies towards FDI that prevailed in developing countries from the 1970s through the mid-1980s (ibid., pp. 15-17). Since then, many developing countries have significantly, and in some cases dramatically, altered their attitudes and policies concerning FDI5. This alteration is closely linked to a fundamental shift in development policy from import substitution to export promotion, and more specifically in the design of government policies for attracting the type of FDI that is most desired. In this conjunction, it should be recalled that much of the FDI flowing into developing countries in the early years was attracted by the availability of natural resources. This traditional type of FDI has different implications for growth in the host country than do other types of FDI, such as export-oriented manufacturing FDI. This point was empirically tested by Hiemenz et al. (1991), whose regression analyses used a data sample covering 26 developing countries over the 1979-88 period. Their results indicate that natural resource-oriented FDI, in contrast to manufacturing FDI, tends to be undertaken independently of macroeconomic conditions and the degree of product market distortions in the host country. As exemplified by data on outward FDI by US firms, the sectoral composition of the FDI stock does vary substantially between resource-rich and resource-poor host economies (Table 1.1). In Indonesia, for instance, 90 per cent of the US FDI
A case in point is a critical review of Reuber et al. (1973) by Lall (1974), which demonstrates that academic views regarding the role of FDI in development were extremely divergent in the 1970s. While the degree of divergence appears to have narrowed considerably over the last 30 years, this subject remains politically sensitive, particularly for some developing countries. 5 A telling example in this respect is China’s reform experience since the late 1970s, and notably its gradual shift in FDI policy since the mid-1980s. For further discussion, see Fukasaku and Wall (1994) and Fukasaku et al. (1999). See also Lemoine (2000) and Démurger (2000) for a detailed analysis of the importance of FDI for China’s trade and growth.
stock was in petroleum in 1986, while in Chinese Taipei nearly three-quarters of this stock was in the manufacturing sector. It is interesting to note, however, that between 1986 and 1998 the services sector (including the telecommunication and power industries) considerably increased its relative share and became a main recipient of US FDI in several economies, notably Japan, Chinese Taipei, the Philippines and Singapore, whereas in China it was the manufacturing sector that expanded its relative share during the same period. Following the 1997-98 financial crisis, the relative importance of the petroleum sector increased markedly in resource-rich countries such as Indonesia and Thailand, and to a lesser extent in Malaysia6.
See Fan and Dickie (2000) and Thompson and Poon (2000) for further discussion on the importance of FDI for ASEAN economies.
Table 1.1. US Direct Investment Position in Selected Asian Economies, by Sector Host economy Sector
Japan 1994 1998
All sectors Petroleum Manufacturing Services Others
11,839 2,712 5,560 3,436 131
($ million) 37,027 41,423 6,124 4,396 15,844 11,428 14,159 25,311 900 288
55,606 (D) 15,173 34,753 (D)
183 (D) 43 (D) 7
($ million) 1,699 6,350 675 939 765 3,862 221 1,018 38 531
All sectors Petroleum Manufacturing Services Others
100 23 47 29 1
(percentage) 100 100 17 11 43 28 38 61 2 1
100 (D) 27 62 (D)
100 (D) 23 (D) 4
(percentage) 100 40 45 13 2
Host economy Sector
Hong Kong 1994 1998
100 15 61 16 8
Korea, Rep. 1994 1998
All sectors Petroleum Manufacturing Services Others
4,227 235 553 3,287 152
($ million) 11,986 17,548 552 597 1,902 2,597 9,015 11,366 517 2,988
23,308 202 3,283 16,396 3,427
897 8 341 526 22
($ million) 3,612 7,365 88 (D) 1,391 2,712 2,082 (D) 51 -24
All sectors Petroleum Manufacturing Services Others
100 6 13 78 4
(percentage) 100 100 5 3 16 15 75 65 4 17
100 1 14 70 15
100 1 38 59 2
(percentage) 100 2 39 58 1
100 (D) 37 (D) 0
Chinese Taipei 1994 1998
9,577 1,846 5,663 1,474 594
1,003 -11 745 250 19
($ million) 3,882 6,295 (D) 49 2,459 3,324 1,291 2,783 (D) 139
7,737 60 3,692 3,700 285
100 19 59 15 6
100 -1 74 25 2
(percentage) 100 100 (D) 1 63 53 33 44 (D) 2
100 1 48 48 4
9,432 (D) 3,954 3,563 (D)
100 (D) 42 38 (D)
Table 1.1. US Direct Investment Position in Selected Asian Economies, by Sector (continued) Host economy Sector
Indonesia 1994 1998
Malaysia 1994 1998
All sectors Petroleum Manufacturing Services Others
3,369 3,044 98 (D) (D)
($ million) 5,015 8,104 4,341 5,115 181 275 234 (D) 259 (D)
11,605 8,440 273 673 2,219
1,019 613 301 87 18
All sectors Petroleum Manufacturing Services Others
100 90 3 (D) (D)
(percentage) 100 100 87 63 4 3 5 (D) 5 (D)
100 73 2 6 19
100 60 30 9 2
Host economy Sector
Singapore 1994 1998
($ million) 2,382 5,629 396 1,048 1,582 3,679 351 914 53 -12 (percentage) 100 17 66 15 2
100 19 65 16 0
Thailand 1994 1998
All sectors Petroleum Manufacturing Services Others
2,481 455 1,456 543 27
($ million) 10,972 17,550 2,127 2,636 5,316 7,045 3,332 7,788 197 81
23,245 1,718 11,834 9,411 282
1,152 726 274 143 9
All sectors Petroleum Manufacturing Services Others
100 18 59 22 1
(percentage) 100 100 19 15 48 40 30 44 2 0
100 7 51 40 1
100 63 24 12 1
($ million) 3,762 5,209 1,185 1,209 1,341 2,313 842 1,236 394 451 (percentage) 100 31 36 22 10
100 23 44 24 9
Philippines 1994 1998
5,995 1,252 3,411 (D) (D)
1,274 123 558 428 165
($ million) 2,374 3,931 (D) 283 1,167 1,558 (D) 1,854 67 236
2,910 1 1,207 1,394 308
100 21 57 (D) (D)
100 10 44 34 13
(percentage) 100 100 (D) 7 49 40 (D) 47 3 6
100 0 41 48 11
7,124 2,666 2,767 1,459 232
100 37 39 20 3
Note : (D) Suppressed to avoid disclosure of data on individual companies. Sources: The figures for 1986 and 1994 were taken from Petri and Plummer (1998, Table 7.1); for other years, see US Department of Commerce, Bureau of Economic Analysis, International Investment Division.
This heterogeneity in the sectoral distribution of FDI, which reflects the differing motivations of investors, is what makes policy discussion on FDI complicated from the host country’s point of view. The present study focuses on the non-traditional type of FDI (in manufacturing and services) and its implications for sustained growth in developing countries. Figure 1.1 illustrates the role of FDI in development in its simplest form. This schematic diagram depicts the flow of capital, technology and managerial know-how from parent companies to their foreign affiliates and to local firms within and between industries. From the standpoint of host-country interests, the central question is the extent to which FDI contributes to long-term growth in developing countries through both direct effects (increased investment and trade) and indirect ones (technological spillovers to local firms, including privatised ones, and to local workers). Host-country policies and regulations are also important for both attracting FDI and maximising the benefits from it7. As this is basically an empirical question, much of this study is devoted to a review of recent empirical studies available in the literature. Figure 1.1. The Role of FDI in Development: Host-country Interests
A host country MNEs Parent companies
Foreign affiliates Capital Technology Know-how
Government policy & regulations
Vertical and horizontal linkages
Local firms and workers
The issue of home-country interests is beyond the scope of this study and will not be discussed here.
The rest of this study is organised as follows. ChapterⅡ presents a statistical overview of major trends in FDI flows from the global and regional perspectives. This chapter is intended to set the stage for subsequent discussions under the following five headings (Chapters Ⅲ to Ⅶ): •
the FDI-growth nexus;
• • •
FDI-trade linkages; FDI and technology transfer; FDI, privatisation and corporate governance; and
host-country policies for attracting FDI.
The study concludes by summarising major findings and policy lessons. The following paragraphs briefly describe the key questions addressed in individual chapters. 1.2 The FDI-Growth Nexus Does FDI contribute relatively more to growth than domestic investment? The answer to this question seems to be less controversial in theory than in practice. As noted above, FDI can be seen as an important vehicle for the transfer of capital, technology and knowledge to host countries, and hence as generating more growth opportunities than domestic investment8. In practice, however, opinions differ regarding its impact on growth. The growth-enhancing effect of FDI may depend in part on whether FDI “crowds out” or “crowds in” domestic investment. It may also depend crucially on the absorption capacity of a host country, as the efficient use of new technologies requires a minimum level of human capital on the recipient side. The literature on this FDI-growth nexus has grown rapidly over the past several years. Chapter Ⅲ reviews existing empirical studies on this topic and discusses implications for policy. 1.3 FDI-Trade Linkages Does FDI lead to trade, or does trade lead to FDI? Are FDI and trade substitutes or complements? These questions have been addressed repeatedly in the trade literature in different policy contexts, and a considerable amount of work has been dedicated to testing alternative hypotheses on the relationship between FDI and trade. On the one hand, the “complementarity hypothesis” is
This explains why China is so eager to attract FDI, even though the country has one of the highest saving rates in the world.
based on the premise that an MNE’s decision to invest abroad may be encouraged by significant bilateral trade between the home and host countries. There is also some evidence that FDI tends to enhance bilateral trade between Japan and Asian developing countries (Kawai and Urata, 1995; Urata, 2001). On the other hand, the “substitution hypothesis” affirms that FDI reduces bilateral trade by replacing home-country exports with local production in the host country. Chapter Ⅳ takes stock of recent empirical studies. 1.4 FDI and Technology Transfer Many developing countries view FDI as one of the most important means of acquiring new technologies and improving productive efficiency through technological spillovers. A central question here is the extent to which FDI contributes to the productivity growth of local firms. This is a key issue in the current debate on how to strengthen the trade capacity of poor countries. In the post-crisis context, it has also attracted renewed interest among Asian policymakers, with a view to enhancing local technological capabilities in the region. Chapter Ⅴ reviews and discusses the empirical literature on the relationship between FDI and technology transfer from the development perspective. 1.5 FDI, Privatisation and Corporate Governance Should privatisation be opened up to foreign investors? What are the implications of increased foreign participation in domestic business activities for corporate governance in host countries? This is a new — and politically sensitive — topic for host countries, notably Asian countries in the post-crisis context. Recent experiences in Latin America as well as in a number of OECD countries provide some useful lessons for them, since informed analysis and policy discussion on these questions remain limited in Asia (Fukasaku, 2001). Chapter Ⅵ attempts to synthesise the materials available and discusses some policy issues. 1.6 Host-government Policies for Attracting FDI Over the past decade, attracting FDI has become a leading item on national and regional policy agendas in both OECD and developing countries. Asia is no exception in this regard. Some argue, for instance, that the creation of the ASEAN Free Trade Area (AFTA) and ASEAN Investment Area (AIA) may be construed as a strategic response by the ASEAN countries to China’s ascendancy as the main destination of FDI flowing into the region during the 1990s. According to JBIC’s 9
latest survey of Japanese multinationals, 56.7 per cent of the 469 Japanese manufacturing firms surveyed regard China as a more attractive production location than ASEAN, while only 10.2 per cent of them hold the opposite view (JBIC, 2001, p. 13). Similarly, a recent proposal for forging strategic trade links between these two economies may be interpreted as a joint initiative to entice foreign investors from other regions. Many other regional trade and co-operation arrangements are currently in place or under consideration. Chapter Ⅶ addresses the question of policy competition for attracting FDI.
Chapter Ⅱ Trends in FDI: Global and Regional Perspectives
Foreign direct investment (FDI) has been one of the defining features of globalisation and has substantially reshaped the international business environment. A critical element of this phenomenon is the speed of change. Over the past 20 years, the dollar value of global FDI outflows has increased much faster than that of world merchandise exports, with a particularly strong surge recorded in 1986-90 and again in 1996-2000 (Figure 2.1). An underlying reason for this rapid expansion in FDI flows since the mid-1980s is the fundamental shift in policy regimes towards foreign investment at both the national and international levels. On the national front, a large number of countries, in OECD and nonOECD areas alike, have introduced substantial changes in their FDI regulations to create a more favourable environment for foreign investment. Chapter Ⅶ takes a closer look at such regulatory changes in selected host economies in Asia and Latin America. At the same time, efforts have continued at the international level to provide proper protection for foreign investment. First, bilateral investment treaties (BITs), which provide legal security to foreign investors and their investments, have become increasingly important instruments for protecting and promoting investment flows. Most existing BITs were concluded in the 1990s, in parallel with the rise in investment flows: during the three decades leading up to 1990, only 500 BITs had been signed, whereas by the end of the 1990s this number had almost quadrupled. It is also noteworthy that in 1999 the vast majority of BITs were concluded between developing countries (UNCTAD, 2000). Second, international standards for the treatment of foreign investors have been strengthened in terms of national treatment, transparency in incentives and disincentives, codes of business conduct and the avoidance of double taxation (see Box 2.1). These favourable policy developments give firms greater freedom in making strategic decisions on where to produce and how to serve different markets from a global perspective.
Figure 2.1. Trends in World Merchandise Exports and FDI Outflows (average annual growth rates) 35 30
25 20 15 10 5 0 1981-85
World merchandise exports
World FDI outflows
Sources: UNCTAD, World Investment Report (various years), and IMF, International Financial Statistics.
Box 2.1. International Investment Instruments Although there are no universal rules governing international investment, OECD Member countries are committed to providing non-discriminatory treatment to inward direct investment and related financial flows by virtue of the legally binding OECD Codes of Liberalisation. The 35 countries that adhere to the OECD Declaration on International Investment and Multinational Enterprises have also undertaken a political commitment to accord national treatment to established foreign direct investors, to promote voluntary standards of corporate responsibility by multinational enterprises, to encourage restraint in the use of investment incentives and to avoid the imposition of conflicting regulatory requirements on multinational enterprises a. These instruments have provided an effective framework for international co-operation and have served to underpin the liberalisation achieved in recent decades. a
All 30 OECD Member countries have adhered, as well as Argentina, Brazil, Chile, Estonia and Lithuania. The OECD encourages non-members to adhere to this Declaration, which includes the Guidelines for Multinational Enterprises.
Multinational enterprises ( MNEs) have long played an important role in many countries, both developed and developing, but their importance has increased still further with the recent surge in global FDI flows. Today, over 63 000 parent companies world-wide have established about 690 000 affiliates in countries other than their own, and this stock of inward FDI was valued at roughly $4 800 billion in 1999. Foreign affiliates of MNEs are estimated to have generated total gross product of more than $3 000 billion and total employment of over 40 million in host countries. While 90 per cent of all parent companies are
located in OECD countries, more than half of all foreign affiliates operate in nonOECD economies, providing a major source of industrial production and employment in a number of developing host countries9. Although the available data are fragmentary and incomplete, the relative importance of foreign affiliates in host economies can be seen in Table 2.1. This table presents information for 1996 (or the latest available year) on foreign affiliates operating in manufacturing industry in 16 OECD Member countries, in terms of the number of enterprises and employees as well as the value of production, exports and imports10. It is evident from this table that by the mid1990s affiliates of foreign-owned MNEs were already playing a significant part in industrial activity in most of these countries (with the notable exception of Japan). Foreign affiliates had a particularly strong presence in economies such as Canada, Hungary and Ireland, accounting for more than half of total industrial production. The table also suggests that foreign affiliates of MNEs operating in manufacturing industry are on average larger and more trade-oriented than their domestic counterparts.
These figures are taken from UNCTAD (2000, Tables I.1 and I.4). The original data are extracted from OECD (2000) CD-ROM. It should be noted that international comparison of FDI trends still poses a major challenge to researchers, because the official data provided by national authorities are not strictly comparable. See, for example, Feenstra (1998) for further discussion. 10
Table 2.1. Foreign Affiliates in Manufacturing Industry of Selected OECD Countries (1996 or latest available year) (a) (1) No. of enterprises Units % of national total Canada (5)
Czech Republic (6)
Finland (3), (7)
Germany (5) Hungary (8)
(2) No. of employees Units % of national total
(3) Production $ % of million national total
(4) Exports $ % of million national total (2) 75,105 49.4
n.a. (14) 149,847
(5) Imports $ % of million national total n.a.
29,606 (14) 742,663
n.a. (16) 12,716
n.a. (16) 10,737
157,141 (15) 179,226 (15) 20,583
38,587 (15) 114,456 (15) 45,872
106,410 (14) 423,590 (14) 86,469
5,994 (16) 11,174
1.5 (17) 21.5
10,342 (16) 18,081
4.0 (17) 27.7
11,514 (15) 34,294
40,348 (14) 134,372
United Kingdom (5)
United States (13)
19.2 (17) 11.9
229,537 (15) 552,000
33.2 (17) 14.9
n.a. (16) 140,886
n.a. (17) 22.5
n.a. (16) 268,673
n.a. (17) 33.8
Ireland (9) Italy (2), (10) Japan (5) Mexico (4), (5)
Notes : (1) "Manufacturing industry" is defined as ISIC Rev. 3, and national currrencies are converted into US dollars using period-average exchange rates. (2) 1995. (3) 1994. (4) 1993. (5) Refers to majority foreign-owned firms. (6) Refers to majority foreign-owned firms with 100 or more employees. (7) Based on the Annual Industrial Statistics, which are establishment-level data. (8) All firms with foreign ownership of over 10 per cent of the share capital. (9) Includes all local units of multi-location enterprises (with more than three employees) where 50 per cent or more of the share capital is held by non-Irish residents. (10) Refers to all Italian enterprises in which a foreign person owned or controlled a direct or indirect interest of 50 per cent or more at the end of the fiscal year. (11) Refers to majority foreign-owned establishments. (12) Refers to majority foreign-owned non-financial firms. (13) Refers to non-bank firms in which 10 per cent or more of the voting shares are foreign-owned. (14) On a full-time equivalent basis. (15) Turnover. (16) Refers to total merchandise trade. (17) Authors' estimation. n.a. not available. Source: Fukasaku and Kimura (2001, Table 1).
The combination of the prominence of MNEs in industrial activity and the growing importance of FDI as a vehicle for international technology transfer has recently drawn the attention of policymakers to the potential benefits — and costs — of FDI. This chapter presents an overview of major trends in FDI flows from the global and regional perspectives. It is intended to set the stage for the policy analyses and discussions in the subsequent chapters. 2.1 A Global Surge in FDI since 1993 The rapid expansion in global FDI flows since 1993 stands in sharp contrast to the annual movement of world merchandise trade (Figure 2.2). Over the 199299 period, outflows of FDI increased fivefold, from just over $200 billion to around $1 000 billion, while world exports rose by less than 50 per cent. In 2000, FDI outflows rose by a further 15 per cent to $1 150 billion. Figure 2.2. World Merchandise Trade and FDI Flows, 1989-2000 (value index, 1989=100) 700 600
500 400 300 200 100 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Year World merchandise trade
World FDI flows
Note: World merchandise trade is defined as the average of exports and imports, FDI flows as the average of outflows and inflows. Sources: see Figure 2.1.
A key feature of the current FDI boom is that an increasingly large proportion of FDI takes the form of cross-border mergers and acquisitions (M&As), including acquisitions of public enterprises (i.e. privatisation) in host economies
(Figure 2.3)11. It is no exaggeration to say that the current FDI boom has been largely fuelled by international M&A activity. The annual average value of crossborder M&As has jumped more than sixfold in recent years, from $132 billion in 1993-95 to over $810 billion in 1998-2000. During the latter sub-period, crossborder M&As accounted for 80 per cent of global FDI inflows. Figure 2.3. Trends in FDI Inflows, Cross-border M&As and Privatisation 1400 1200
1000 800 600 400 200 0 1990
Sources: UNCTAD, World Investment Report (various years) and OECD (2001e).
Cross-border M&A transactions are heavily concentrated in OECD countries, which accounted for over 90 per cent of the world total in 1998-2000 (Table 2.2). Within the OECD area, the share of European Union (EU) member states has increased steadily over the last ten years, from 35 per cent in 1988-90 to 46 per cent in 1998-2000. This is primarily due to industrial restructuring in the context of the European Single Market Programme (Dunning, 1998). Sleuwaegen (1998) also points out that service sectors, such as business services, distribution, and banking and finance, have gained in relative importance in the recent M&A waves in Europe, together with the engineering and chemicals industries.
The number of “real” mergers is so low (less than 3 per cent of total cross-border M&As in 1999) that, for the purposes of this study, “M&A” may be taken to mean acquisitions of 10 per cent or more of a firm’s voting shares. Acquisitions of less than 10 per cent of the voting stock are classified as portfolio investment and not considered here as FDI.
Table 2.2. Cross-border M&As by Region/Economy of Seller (three-year average, $ billion and percentages) Region/economy World total (value, $ billion)
Annual average 1993-95
% shares OECD regions (30)
Europe (23) EU (15)
North America (3)
1.8 1.8 0.0 0.7
4.6 4.3 0.4 1.7
1.8 1.6 0.2 0.9
Europe & Central Asia
Latin America & Caribbean Mercosur (4)
Middle East & North Africa
Asia/Pacific East Asia & Pacific China ASEAN (10) South Asia
Notes : 1. The "OECD regions" are OECD Asia/Pacific (4), OECD Europe (23) and North Anerica (3). 2. The "OECD Asia/Pacific" countries are Australia, Japan, Rep. of Korea and New Zealand. 3. "OECD Europe" comprises the EU (15), Czech Rep., Hungary, Iceland, Norway, Poland, Slovak Rep., Switzerland and Turkey. 4. The "EU (15)" consists of Austria, Belgium-Luxembourg, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden and the United Kingdom. 5. "North America" comprises Canada, the United States and Mexico. 6. "ASEAN (10)" comprises Brunei Darussalam, Cambodia, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand and Viet Nam. 7. The "Mercosur (4)" are Argentina, Brazil, Paraguay and Uruguay. The regional groupings as noted above are defined as if their member countries remain the same throughout the period. Sources: UNCTAD, World Investment Report (various years).
There is substantial inter-regional variation in the importance of crossborder M&A transactions relative to FDI inflows (Table 2.3). Note that these two sets of data are not directly comparable. Figures on FDI are based on balance-ofpayments statistics, while those on cross-border M&As are not. Conceptually, FDI is composed of cross-border M&As and greenfield investment, but in practice the available data rarely allow us to distinguish between these two types of international investment12. Particularly troublesome is the fact that the value of cross-border M&As includes funds raised on local and international markets, while such funds are not, by definition, part of FDI. This problem appears to be far more serious in the case of developed countries with advanced domestic capital markets than in the case of developing countries, where cross-border corporate sales are likely to be financed by FDI flowing from purchasing countries13. With this caveat in mind, it can still be argued that the most marked difference is that between OECD and non-OECD regions. Among the latter, Latin America and the Caribbean, and in particular Mercosur, have long embraced cross-border M&As as the principal mode of FDI inflows. In 1998-2000, M&As accounted for 59 per cent and 71 per cent respectively of total FDI flows into these two regions. This is due to the large-scale privatisation programmes, including privatisation of public utilities, which have been implemented in Argentina, Brazil and other countries of the region (see Chapter Ⅵ). In contrast, two regions in Asia and the Pacific, namely ASEAN and South Asia, have jumped onto this M&A bandwagon more recently, in the aftermath of the 1997-98 financial crisis. Among major developing-country recipients of FDI, China remains an exception in this respect.
The United States is a major exception. The statistical discrepancies are clearly marked in the case of the OECD Asia-Pacific region in 1998-2000 and North America in 1988-90 and 1998-2000, as these three cases are off the scale (more than 100 per cent). 13
Table 2.3. Cross-border M&As as a Percentage of FDI Inflows (percentages) Region/economy
Annual average 1993-95
OECD regions (30)
Europe (23) EU (15)
North America (3)
16 16 0 11
10 9 2 11
15 14 4 47
Europe & Central Asia
Latin America & Caribbean Mercosur (4)
Middle East & North Africa
Asia/Pacific East Asia & Pacific China ASEAN (10) South Asia
Notes and sources: see Table 2.2.
According to the latest UNCTAD forecast (September 2001), the current FDI boom appears to have come to an end in 2001. The dollar value of global FDI flows (based on recipient-country data) is likely to have declined by 40 per cent during the year, the “first drop since 1991 and the largest over the past three decades” (TAD/INF/PR21/ Rev. 1, 18 September 2001). This decline is largely due to a 49 per cent drop in FDI flows to developed countries as a result of the fall-off in cross-border M&A activity. For developing regions, in contrast, the decline in FDI inflows is likely to be modest: a drop of some 6 per cent is predicted. 2.2 Regional Trends OECD regions as a whole continued to be the primary source of FDI, accounting for more than 90 per cent of total outflows in 1998-2000, with some fluctuations during the past decade (Table 2.4). Among major OECD regions, the 15 member states of the European Union were collectively the largest supplier of FDI, followed by two North American OECD Members, Canada and the United States. In contrast, the share of Asia-Pacific OECD Member countries fell dramatically, from 22 per cent in 1988-90 to merely 3.5 per cent in 1998-2000,
largely owing to the diminishing role of Japan as a source country relative to its European and North American counterparts14. Table 2.4. FDI Outflows by Home Region/Economy ($ billion and percentages) Home region/economy
Annual average 1993-95
World total (value, $ billion)
% shares OECD regions (30)
Europe (23) EU (15)
North America (3)
4.0 4.0 0.4 0.6
11.5 11.5 0.9 2.4
4.8 4.8 0.2 0.5
Europe & Central Asia
Latin America & Caribbean Mercosur (4)
Middle East & North Africa
Non-OECD regions Asia/Pacific East Asia & Pacific China ASEAN (10) South Asia
Notes and sources: see Table 2.2.
Among individual home countries of FDI, the G7 countries and several other European countries maintained their dominant positions throughout the 1990s (Table 2.5). The very high concentration of FDI outflows hardly changed, as the ten largest home countries continued to supply more than 80 per cent of total outflows. Only three non-OECD economies — Hong Kong-China, Chinese Taipei and Chile — ranked among the 20 largest home economies of FDI in 1998-2000.
14 Over the past 15 years, the policy debate on Japanese FDI outflows has centred on the question of the “hollowing out” of manufacturing industries, as overseas production facilities expanded in response to the yen’s appreciation against the dollar after the Plaza Accord in 1985. After the 199798 crisis, however, Japanese FDI outflows to East Asia started to fall. For a detailed discussion of FDI issues from the Japanese perspective, see inter alia Tejima (2000a and 2000b) and Sazanami et al. (2001).
Table 2.5. World’s 20 Largest Home Economies of FDI (annual average, $ billion) Rank
1 2 3 4 5 6 7 8 9 10
United Kingdom United States France Germany Belgium-Luxembourg Netherland Spain Hong Kong, China Canada Switzerland
192.5 137.6 113.9 82.3 78.0 57.2 38.2 33.1 32.3 31.4
Total of above (% of grand total)
11 12 13 14 15 16 17 18 19 20
Sweden Japan Denmark Finland Italy Chinese Taipei Chile Portugal Korea, Rep. Ireland Total of above (% of grand total)
28.6 26.6 22.0 16.1 10.4 5.0 4.1 4.0 3.7 3.4
1 2 3 4 5 6 7 8 9 10
United States Japan France United Kingdom Germany Netherland Hong Kong, China Belgium-Luxembourg Italy Switzerland Total of above (% of grand total)
11 12 13 14 15 16 17 18 19 20
Canada Spain China Sweden Chinese Taipei Australia Denmark Singapore Austria Korea, Rep. Total of above (% of grand total)
1991-93 49.1 26.1 25.0 20.7 20.2 13.3 9.6 7.5 7.5 7.3 186.3 (85) 5.0 3.2 3.1 2.7 2.1 1.9 1.8 1.5 1.5 1.3 210.5 (96)
Sources: see Table 2.2.
In contrast, non-OECD regions as a whole had become major recipients of FDI by the mid-1990s. They received 34 per cent of total inflows in 1993-95, up 20 percentage points from the 1988-90 level (Table 2.6). East Asia and, to a lesser extent, the Latin America and Caribbean region had emerged as the most favourable places for foreign investment outside the OECD area. Subsequently, however, the share of non-OECD regions declined substantially as a result of the 1997-98 financial crisis, which had a severe impact on several East Asian economies. The effect of the crisis is evident in the sharp drop in the relative share of ASEAN member states and China as destinations of FDI flows in 1998-2000, in contrast to the steady rise in the relative share of Mercosur countries throughout the period concerned.
Table 2.6. FDI Inflows by Host Region/Economy ($ billion and percentages)
Annual average 1993-95
World total (value, $ billion)
Europe (23) EU (15)
North America (3)
8.6 8.3 1.8 4.7
23.2 22.5 12.0 7.6
9.7 9.4 4.1 1.5
Europe & Central Asia
Latin America & Caribbean Mercosur (4)
Middle East & North Africa
% shares OECD regions (30) Asia/Pacific (4)
Non-OECD regions Asia/Pacific East Asia & Pacific China ASEAN (10) South Asia
Notes and sources: see Table 2.2.
Where individual host economies are concerned, there have been some major changes in the preferences of foreign investors over the past decade (Table 2.7). Malaysia, Singapore and Thailand, three emerging economies of East Asia, have dropped out of the list of the 20 largest recipients of FDI, being replaced by Brazil and two small European OECD countries, Finland and Ireland. An equally noteworthy development is that both Japan and the Republic of Korea (hereafter referred to as Korea) have entered the ranking for the first time, becoming attractive locations for foreign investment in the post-Asian crisis era.
Table 2.7. World’s 20 Largest Recipient Economies of FDI (annual average, $ billion) Rank 1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20
Recipient economy 1998-2000 United States United Kingdom Germany Belgium-Luxembourg Netherland China France Canada Hong Kong, China Sweden
250.2 94.7 85.4 76.5 45.2 41.6 40.7 37.0 34.6 34.0
Total of above (% of grand total)
Brazil Spain Argentina Ireland Mexico Denmark Switzerland Korea, Rep. Finland Japan Total of above (% of grand total)
31.1 22.2 14.2 14.1 12.2 11.5 9.9 8.7 8.3 8.1
Rank Recipient economy 1 2 3 4 5 6 7 8 9 10
United States France United Kingdom China Spain Belgium-Luxembourg Netherland Mexico Singapore Malaysia Total of above (% of grand total)
11 12 13 14 15 16 17 18 19 20
Australia Canada Sweden Italy Argentina Bermuda Germany Hong Kong, China Thailand Portugal Total of above (% of grand total)
1991-93 27.7 17.8 15.5 14.3 11.8 10.5 7.5 5.3 4.8 4.7 120.0 (66) 4.6 4.0 3.4 3.3 3.1 2.8 2.3 2.1 2.0 2.0 149.5 (82)
Sources: see Table 2.2.
The unprecedented FDI boom enjoyed by Hong Kong-China over the last two years should also be mentioned. Inflows to the territory jumped from $24.6 billion in 1999 to $64.4 billion in 2000, surpassing mainland China’s record of $40.8 billion. At the same time, outflows from the territory surged from $19.3 billion to $63.0 billion. As a result, Hong Kong-China’s inflows and outflows in 2000 accounted for 52 per cent and 79 per cent respectively of total FDI flows into and out of the East Asia and Pacific region (excluding OECD Members of the region). While this may be due in part to the so-called “round-tripping” phenomenon linking the territory’s economy to other offshore financial centres, China’s imminent accession to the WTO and prominent cross-border M&A deals in the telecommunications sector have certainly boosted FDI flows to Hong KongChina (UNCTAD, 2001, pp. 24-25). Overall, the experience of the 1990s suggests that regional patterns of FDI flows can change dramatically over the medium term. Between 1991-93 and 19982000 there were two major developments which deserve special attention (Table 2.8). First, the United States, which on a net basis (outflows minus inflows) had been the world’s second-largest source country, after Japan, became the largest
net recipient of FDI. This about-face stands in sharp contrast to the continued dominance of five EU member states (France, Germany, Italy, the Netherlands and the United Kingdom) and Switzerland as net exporters of FDI, although some other EU countries (most notably, Belgium-Luxembourg and Spain) became new major suppliers of FDI on a net basis. The second important development over the last ten years is that China remains by far the largest developing-country recipient of FDI (see Box 2.2)15. Only a few emerging economies (other than China) succeeded in attracting a sizeable amount of FDI on a net basis, though there were some changes in the ranking of these economies over time. These observations suggest that the role of host-country characteristics must be taken into explicit account when examining the relationship between FDI and growth, which is the subject of the next chapter.
The “special case” of China has been discussed in detail in the literature (see inter alia Hill and Athukorala, 1998, pp. 27-30).
Table 2.8. Net FDI Exporters and Importers (annual average, $ billion) Economy 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 37 38 39 40
United Kingdom France Switzerland Japan Spain Netherland Denmark Finland Italy Indonesia Chinese Taipei Belgium-Luxembourg Portugal South Africa Turkey New Zealand Russian Federation Nigeria Hong Kong, China Hungary Chile Saudi Arabia Malaysia Austria Bermuda Germany Thailand Venezuela Singapore Canada Korea, Rep. Sweden Australia Poland Ireland Mexico Argentina Brazil China United States
97.8 73.2 21.5 18.5 16.0 12.1 10.5 7.8 3.5 2.6 2.3 1.5 1.2 0.5 -0.3 -0.3 -0.9 -0.9 -1.5 -1.6 -1.7 -1.7 -2.1 -2.5 -2.9 -3.1 -3.5 -3.6 -3.7 -4.7 -5.1 -5.4 -6.1 -7.7 -10.7 -10.8 -12.7 -28.8 -39.4 -112.6
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 37 38 39 40
Sources: see Table 2.2.
Japan United States Germany Hong Kong, China France Netherland Switzerland United Kingdom Italy Chinese Taipei Canada Austria Korea, Rep. Denmark South Africa Ireland Finland Russian Federation Chile Venezuela Saudi Arabia Nigeria New Zealand Sweden Turkey Poland Brazil Portugal Indonesia Hungary Thailand Australia Argentina Belgium-Luxembourg Bermuda Singapore Malaysia Mexico Spain China
1991-93 24.3 21.4 17.9 7.5 7.2 5.8 5.5 5.2 4.2 1.1 1.0 0.7 0.5 0.4 0.1 0.0 -0.2 -0.4 -0.4 -0.6 -0.6 -0.7 -0.7 -0.7 -0.7 -0.9 -0.9 -1.5 -1.6 -1.7 -1.8 -2.7 -2.9 -2.9 -3.0 -3.3 -3.9 -5.0 -8.6 -11.2
Box 2.2. Is China a Special Case? Lee and Houde (2000), following John Dunning’s eclectic approach to the determinants of decisions on FDI location, identify six main characteristics of countries that constitute advantages in attracting FDI, along with the types of FDI flows they might attract. The authors argue that China has all these advantages. • Market size and growth prospects. Factors like market size, prospects for market growth, the degree of development and per capita incomes of host countries are important determinants in the location decisions made by MNEs. Host countries with larger markets, faster economic growth and a higher degree of economic development will offer better opportunities for enterprises to exploit their ownership advantages and to realise economies of scale. FDI attracted by these advantages is called “marketoriented” FDI. • Natural and human resource endowments, including the cost and productivity of labour. Factor cost advantages and the availability of natural and human resource endowments are a driving force behind FDI. In particular, FDI oriented towards exports (either back to the home country or to third countries) seeks to exploit comparative advantages related to low labour costs or the abundance of natural resources. Recently, attention has shifted from natural endowments of resources and labour to acquired endowments of resources, such as intermediate goods and skilled labour. • Physical, financial and technological infrastructure. Differences in the quality of infrastructure, such as transportation and telecommunications, influence the decision on FDI location not only among candidate countries but also among different regions within a country. FDI is more likely to flow to areas that are easily accessible and consequently offer lower transportation costs. • Openness to international trade and access to international markets. China’s economic reform, open-door policy and other efforts to promote trade have attracted export-oriented FDI. The country’s geographic position, adjacent to the region’s industrial powerhouses Japan and Korea, is also a significant factor in attracting this type of FDI. • The regulatory and policy framework and policy coherence. General economic, political and social stability forms the background of a host country’s FDI policy. A transparent and well-functioning legal framework and business environment is of the first importance, since it lowers the (political) risk of doing business in an unfamiliar environment. Rules and regulations regarding the entry and operations of foreign firms, as well as standards of treatment of foreign firms, are particularly relevant in this respect. Chinese efforts to comply with international standards in the context of WTO accession negotiations have certainly had a positive impact. • Investment protection and promotion. Proper protection of investments is considered a minimum requirement for attracting FDI. There has been no case of expropriation of foreign investment in China since 1979, when the country opened up its economy to FDI. Like many other countries, China also offers investment promotion packages to attract FDI. Such incentive packages include tax and other financial incentives that affect net profit rates, which appear to be the primary concern of foreign investors, and may thus influence the decision on where to locate FDI.
Chapter Ⅲ The FDI-Growth Nexus
The conventional approach to examining the relationship between FDI and growth relies on a variant of the so-called “resource-gap” model as the main analytical tool. In this model, developing countries find themselves trapped in a low-growth path because their lack of financial resources prevents them from attaining optimal growth rates. The inflow of foreign capital can foster growth in the host economy by easing the shortages of capital, foreign exchange and skills. The growth process can become self-sustained if backward and forward linkages emerge from MNEs to the host economy and if FDI helps to raise the profitability of domestic investment. Additional benefits may accrue to the government through higher tax revenues. Some economists point out, however, that FDI may harm the host economy by “crowding out” domestic investment and suppressing local entrepreneurship. Concern has also been expressed over a possible deterioration in the balance of payments due to increased imports and profit repatriation, and over reduced tax revenues as a result of transfer-pricing practices, tax allowances and other financial incentives granted to foreign firms. Thus, although this conventional analytical model helps to assess the benefits of private capital flows, it does not settle the question of what role FDI plays in development (see inter alia Reuber et al., 1973, and Lall and Streeten, 1977). More recently, following the impressive surge in FDI flows into several developing and emerging economies, there has been renewed interest in empirical analysis of the FDI-growth nexus. Such interest has also been stimulated by new developments in growth theory. The “endogenous growth” model identifies knowledge accumulation as the driving force explaining the long-term growth of the economy (OECD, 2001a). FDI, which provides a channel for knowledge acquisition and dissemination, can therefore act as an engine of growth for the recipient economy. Furthermore, FDI tends to be less volatile than other capital flows, thereby exerting durable positive effects on growth (see, for example, Lipsey, 2000, and Reisen and Soto, 2001). Following this line of analytical work, this chapter aims to shed some light on the growth impact of FDI by presenting economic arguments, discussing methodological problems and reviewing empirical findings available in the growth literature.
3.1 Putting Theory to Work According to endogenous growth models, the impact of FDI on growth depends crucially upon the existence of production and knowledge externalities. In the standard neo-classical model, production is represented by a technology providing constant returns to scale, relating the level of output to input bundles. FDI enters this model as an additional production input. More precisely, FDI is treated as additional investment that increases the domestic capital stock. This is not, however, the only channel through which FDI can affect growth. Industrial organisation studies point to the peculiar nature of FDI, which is better described as a “combination of capital stock, know-how and technology” (de Mello, 1997). The central question in empirical analysis of the FDI-growth nexus is thus whether FDI significantly affects the growth rate of income or factor productivity. One can address this question by using various econometric techniques. In the early literature, for instance, many standard growth-accounting exercises were conducted to break down the growth rate of aggregate output into the respective contributions from the growth of capital and labour inputs and from technological change. These analyses, however, provide only a mechanical decomposition of output growth into its various sources, without explaining how these changes are affected by the “fundamentals” of the economy. Such limitations can be overcome by estimating growth equations based on neo-classical production theory (see Barro and Sala-i-Martin, 1995, Chapter 10). Some further comments are in order concerning the theoretical formulation of the FDI-growth nexus. First, one should analyse explicitly the extent to which FDI may substitute for domestic investment. This question has been addressed by including domestic investment directly in the growth equations (Borensztein et al., 1998) or by estimating investment equations that incorporate FDI (Agosin and Mayer, 2000; McMillan, 1999). Second, since the long-term effects of FDI on growth depend on technological and knowledge externalities that spill over to domestic firms in the host country, it is necessary to investigate whether these externalities indeed exist. Empirical studies addressing this question are more microeconomic in nature and will be discussed in Chapter Ⅴ; only a few attempts have been made to take such spillover effects explicitly into account in the macroeconomic context (Bende-Nabende et al., 2000). Finally, as discussed in Chapter Ⅱ, local conditions such as technological capabilities, human capital and the development of domestic financial markets are likely to play an important role in determining the location of FDI flows. It is thus necessary to examine
empirically whether any necessary pre-condition (or threshold) has to be met in the host economy for the FDI-driven growth to materialise16. In endogenous growth models, as was noted above, knowledge accumulation and diffusion play the key role. The existence of technological and knowledge externalities counterbalances the effects of diminishing returns to capital accumulation and keeps the economy on a sustained long-term growth path. FDI can make a substantial contribution to the increase in the host economy’s stock of knowledge, not only by introducing new capital goods and production processes (embodied technical change), but also by providing new managerial know-how and skills improvement that can spread to domestic firms (disembodied technical change). Skills may be upgraded through formal training or learning by doing within foreign affiliates. By enhancing the local stock of knowledge, FDI will therefore have both short-term and long-term impacts on the host economy and boost long-term growth (see Box 3.1 for a more technical discussion).
This argument is related to an extensive literature on threshold externalities (see Azariadis and Drazen, 1990), according to which the attainment of certain minimum critical thresholds in the host economy is required to trigger the development process.
Box 3.1. Short-term and Long-term Impacts of FDI on Growth Endogenous growth theory challenges the conventional view of economic growth by providing sound theoretical arguments for preventing the unbounded decline in the marginal productivity of capital. According to this theory, FDI can improve both short-term and long-term rates of growth in the host economy through knowledge and technology spillovers from the R&D and job training activities of MNEs. The growth impact of FDI can best be understood by considering a simple production function, specified in equation (1). Y is domestic output, A denotes technology available at a given time, L is the labour force and K H and KF are respectively the domestically owned and foreign-owned stock of capital.
Y = AF (η H K H , ηF K F ,θ L)
While A incorporates disembodied technical change, embodied technical change is captured by the factor-specific parameters ηH, ηF and θ. Note that, according to the standard assumptions of the model, each production input has a positive marginal productivity (F i > 0, i = K, L), but this productivity increases at a decreasing rate (Fii < 0). Equation (1) provides a simple description of the twofold impact of FDI on domestic production. In the short run, output expands, as the foreign-owned stock of capital increases (F F> 0). In the long run, FDI exerts an indirect effect on Y through the changes induced in the other inputs (ηH, ηF and θ) and in the technological parameter (A). Here the question of whether FDI is substituting for or complementing domestic investment is of crucial importance to the effect of FDI on Y through KH. If foreign investment helps to increase the profitability of domestic capital in the recipient country, then both KH and output increase (FH F > 0). If, on the other hand, FDI crowds out domestic investment, KH decreases and so does output (F HF < 0). The net effect on output, therefore, will depend on the magnitude of F HF < 0 and FF > 0. Moreover, technological improvement through FDI is measured by changes in the A and η parameters, while skills improvement is reflected by changes in θ. An increase in these parameters induces an unambiguous positive effect on output Y.
From the methodological point of view, however, the inclusion of FDI in neoclassical growth equations poses two major problems. One is the problem of reverse causality. GDP growth by itself or factors that affect GDP growth (such as well-functioning institutions) may influence FDI as well. If causality runs from growth to FDI, the use of ordinary least squares (OLS) estimation techniques would yield biased results. Instead of verifying whether FDI inflows foster GDP growth, the econometric analysis may have picked up how much the latter influences the former. The second major problem is that of “spurious correlation” caused by omitted variables in growth equations. FDI is likely to be significantly correlated with other explanatory variables that are also expected to affect growth. In this case, omitting some important variables from the right-hand side of the growth equation would result in biased estimation of the growth coefficient of FDI, since this coefficient is most likely to pick up the impact of these omitted variables. As a corollary, one needs to know how and to what extent FDI interacts with other
explanatory variables. Many factors that are expected to exert a positive impact on growth, such as domestic capital formation and international trade, may be stimulated by FDI as well17. In order to disentangle the full effects of FDI on growth, various spillover effects must therefore be specified and estimated in an appropriate manner18. 3.2 FDI and Growth: Empirical Evidence This section reviews and discusses the main findings of 15 empirical studies based on estimations of endogenous growth models. This literature review focuses on the four major questions that arise from the above discussion in the macroeconomic context: (1) Does FDI significantly affect the growth of income or productivity? (2) Does FDI “crowd out” or “crowd in” domestic investment? (3) Do technology and knowledge spillovers occur in the domestic economy? (4) Are there any necessary pre-conditions (e.g. human capital, technological or financial market development) that must be met if these positive effects are to materialise? The results of this review are presented in Annex Table 3.1, with an indication of which questions are addressed by individual studies. In this section, much of the discussion will address questions (1) and (4), which are the focal point of existing empirical studies in the growth literature. The vast majority of the studies reviewed here indicate that FDI does make a positive contribution to both income growth and factor productivity in host countries. Using panel data for 16 OECD countries and 17 (mostly Asian) nonOECD countries over the 1970-90 period, de Mello (1999) find a positive and significant impact of FDI on output growth in both country groups, once countryspecific characteristics are taken into account19. FDI tends to increase output growth through higher productivity in OECD countries (technological leaders) and through capital accumulation in non-OECD countries (technological laggards). In a similar vein, Xu (2000), using US survey data on manufacturing MNEs, finds strong evidence of the positive effect of FDI on total factor productivity (TFP)
The question of FDI-trade linkages will be taken up in the next chapter. To address these methodological problems, researchers have taken various approaches, such as the application of Granger causality tests and cointegration analysis to time-series data (e.g. de Mello, 1999, and UNCTAD, 2000), the use of instrumental variable (IV) techniques to identify the autonomous impact of FDI on growth (e.g. Carkovic and Levine, 2000; Reisen and Soto, 2001; and Lensink and Morrisey, 2001), and the construction and estimation of a full structural model based on three-stage least squares (3SLS) or full-information maximum-likelihood methods (BendeNabende et al., 2000). 19 Such characteristics are dealt with by introducing country-specific and group-specific dummies and applying a standard fixed-effect estimation technique. 18
growth in recipient countries, but the technology-transfer effect is found to be statistically significant only for developed countries. He argues that the absorption of MNEs’ technology may require a certain level of human capital accumulation on the recipient side and that many developing countries cannot meet such a threshold condition. Soto (2000) and Reisen and Soto (2001) investigate the growth impact of short- and long-term capital flows using a panel of 44 developing countries over the 1986-97 period. After controlling for potential reverse causality, their estimation results find a robust and positive correlation between FDI and portfolio equity flows on the one hand and GDP growth on the other. The superiority of equity over debt flows in stimulating growth is also established for those economies with underdeveloped banking systems. Since high volatility in capital flows may wreak havoc on the economic performance of a developing country, the apparent lower volatility of FDI relative to other kinds of capital flows is another possible growth-enhancing feature20. In contrast, Carkovic and Levine (2001), whose panel data cover 72 countries over the 1960-95 period, find no significant impact of FDI on growth. The exogenous (i.e. independent of GDP or TFP growth) component of FDI exerts no significant positive impact on output growth, nor is it strongly linked to productivity growth. The impact on capital accumulation is found to be statistically significant and positive, but this relationship is not robust to different specifications of the regressions with respect to other determinants of capital growth. Such discrepancies in estimation results may be explained at least in part by the choice of sample countries and periods. The panel data used by Reisen and Soto (2001) cover almost exclusively middle- and low-income countries over a shorter period when significant changes in capital flows have taken place.
Although researchers hold divergent views as to the growth impact of volatility in capital flows, one point on which most economists agree is that shocks from short-term capital flows are transmitted more quickly between countries than those arising from FDI and other long-term flows. As to the volatility of FDI in particular, Lensink and Morrisey (2001) argue that while such volatility is found to have a consistently negative effect on growth, this effect is actually due not to the volatility of FDI per se, but to unobserved variables whose growth-retarding effects are captured by FDI volatility. The swings in FDI might reflect political and economic uncertainty in the host country, a factor that is widely acknowledged to hamper economic growth.
3.2.1 Threshold Externalities The recent literature shows that developing countries need to have reached a certain threshold level of development in terms of education or infrastructure before they can capture the benefits associated with FDI ( Saggi, 2000). A useful survey by de Mello (1997) points to differences in the growth impact of FDI across countries: since recipient countries’ technological capabilities are likely to determine the scope for spillovers from foreign to domestic firms, the growth impact of FDI tends to be limited in technologically less advanced countries. Borensztein et al. (1998) address the technology-gap question by developing a growth model in which FDI contributes to technological progress through capital deepening, i.e. through the introduction of new varieties of capital goods. Recognising that such beneficial effects are likely to depend on the skills of the domestic labour force, they interact the FDI variable with a measure of human capital development (secondary school attainment). The authors find that FDI contributes to growth, though the magnitude of this effect depends on the stock of human capital available in the host economy. In particular, they argue that FDI raises growth only in those countries where the labour force has reached a minimum threshold of educational attainment. They also find that FDI tends to “crowd in” domestic investment, suggesting that the attraction of complementary activities outweighs the displacement of domestic competitors21. Similar results are obtained by Blomström et al. (1994). The authors find that the positive impact of FDI on growth, though robust to different sample specifications, vanishes when the sample is limited to lower-income developing countries. They argue that FDI is a source of growth only for a country already at a relatively high level of development and that low-income countries lack the capabilities needed to absorb the FDI-related technology transfer. This issue is reviewed and discussed in further detail in Chapter Ⅴ. 3.2.2 Local Financial Markets The development of domestic capital markets may be another requirement for realising the potential benefits of FDI in the host country. The impact of financial market development on growth has been widely studied theoretically
21 McMillan (1999) also argues that FDI can play a strong catalytic role for domestic investment in developing countries. According to Agosin and Mayer (2000), such a “crowding in” effect of FDI can be found for Asian countries but not necessarily for other developing regions.
(among others, Acemoglu and Zilibotti, 1997) and empirically (Beck et al., 2000). Imperfect and underdeveloped financial markets are likely to penalise domestic firms more than foreign affiliates of MNEs. Alfaro et al. (2001) develop a model in which FDI induces higher growth by increasing output directly in the MNE sector and indirectly, via spillovers, in the domestic sector. In this model, financial market constraints hinder the ability of domestic firms to invest and thus to benefit from the spillover effects of FDI. This model was tested empirically by introducing measures of both FDI inflows and financial market development, as well as an interactive term encompassing the two, in the augmented growth regression. The interactive term is found to have a positive and significant impact on GDP growth, while the coefficient of the FDI term is statistically significant but has a negative sign22. The authors interpret these results as showing that “there is a threshold level of the development of financial markets below which FDI will not have any beneficial effect on growth” (ibid., p. 12)23. Similarly, Hermes and Lensink (2000) argue that the development of the recipient economy’s financial system is an important pre-condition for any positive impact of FDI on economic growth. Many of the growth-enhancing effects of FDI work through the adoption of new technologies and skills, which, in turn, depend upon the availability of financial resources. The existence of well-developed financial systems that mobilise savings efficiently and screen investment projects is thus an important pre-condition for the FDI-growth nexus to materialise. The empirical research conducted to date supports this claim: only in those countries with a sufficiently developed financial system (as measured by the ratio of private sector bank loans to GDP) did FDI boost the growth of GDP per capita. Finally, some comments on threshold externalities may be in order from the standpoint of estimation techniques. First, recent developments in growth empirics using panel-data analysis have opened the door for more rigorous testing of the FDI-growth nexus (see Annex Table 3.1, “Estimation technique” column). Panel-data estimation makes it possible not only to exploit both cross-section and time-series variability of the data, but also to account for unobserved country heterogeneity by introducing country-specific effects. The application of these
This result appears to be robust with respect to the use of different measures of financial market development. 23 Since both the volume of FDI and the efficiency of financial markets are likely to be higher in faster-growing economies, reverse causality could yield a biased result on the interactive term. However, instrumental variable estimation of the cross-country growth regression does confirm previous findings, pointing to a positive and significant contribution of FDI to growth in countries where financial markets are sufficiently developed.
techniques to dynamic models, however, may violate the statistical assumptions that ensure unbiased estimation. The dynamic nature of the data can introduce correlation between the error term and the explanatory variables. This additional source of endogeneity generates the risk of reverse causality discussed above. Arellano and Bond (1991) and Arellano and Bover (1995) have developed an appropriate instrumentation technique for dynamic panel data (DPD) that makes it possible to control for the potential endogeneity of all explanatory variables, thereby yielding unbiased results. Second, the use of country-specific effects in panel-data estimation permits the introduction of a certain degree of heterogeneity among the countries under consideration, but this technique does not solve the heterogeneity problem. All other estimated coefficients are in fact assumed to be equal across countries. This assumption of homogeneity is not problematic if the causal link can be supposed to operate in more or less the same way in all countries. Evidence from empirical studies based on microeconomic data, however, points to firm-specificity in acquiring FDI-related spillovers (see Chapter Ⅴ for further discussion). Empirical results based on macroeconomic data, which are reviewed here, support this microeconomic evidence, pointing to the existence of pre-conditions or thresholds that must be attained if recipient countries are to benefit from FDI. The heterogeneity problem thus seems to be an important feature. Failing to acknowledge this problem in the empirical estimation leads to serious bias and inconsistency. The interpretation of the result can be misleading, since imposing homogeneity of coefficients implies that the causal relationship under investigation either occurs everywhere or occurs nowhere in the panel (Usha Nair and Weinhold, 2001). Two recent papers address this issue, applying heterogeneous panel estimation techniques (i.e. assuming that the slope coefficient can differ from one country to the next) to the FDI-growth relationship. De Mello (1999) finds empirical evidence of the hypothesis of cross-country heterogeneity in the FDIgrowth relationship by comparing the results of the fixed-effect estimator with those of the mean group estimator. Separate regressions are estimated for each country and the estimated coefficients are averaged for each group. While the assumption of homogeneity seems to be appropriate for OECD countries, it seems clear that heterogeneity prevails among non-OECD countries, where the aggregate parameter estimates differ from the average of individual country coefficients.
Usha Nair and Weinhold (2001) apply a somewhat different technique to a sample of 24 developing countries from 1971 to 1995. They begin by estimating a dynamic panel model under the assumption of homogeneity, finding that the FDI growth rate has a strong positive impact on GDP growth. They also use interactive terms to check whether either the level of human capital or trade openness affects this relationship, finding no statistically significant impact for the FDI-schooling term but a significant and negative impact for the FDIopenness term. This would mean that FDI has less impact on growth in more open economies. Heterogeneity is then introduced and the model re-estimated using a mixed (fixed and random) effect model. The results confirm the existence of a positive impact of FDI on growth. Furthermore, the mean coefficient estimated for the FDI-openness term is positive, though not statistically significant. These findings support the claim that there is substantial cross-country variation in the way FDI interacts directly and indirectly (i.e. through its impact on other growth determinants such as human capital or openness) with growth.
Annex Table 3.1. FDI and Growth: Literature Survey Authors (year)
Agosin and Mayer (2000)
UNCTAD. 32 developing countries over the 1970-96 period.
Alfaro, Chanda, Kalemil-Ozcan and Sayek (2001)
Net FDI inflows from IMF, IFS. Three samples (39-41 countries). Data averages over the 1981-97 period. WB data on FDI inflows as a percentage of GDP. Five ASEAN countries over the 1970-94 period.
FDI inflows from IMF. 78 developing countries over the 1960-85 period. Gross FDI outflows from OECD countries. 69 countries, two periods: 1970-79 and 1980-89.
Bende-Nabende, Ford and Slater (2000)
Blomström, Lipsey and Zejan (1994) Borensztein, de Gregorio and Lee (1998)
Questions addressed (2)
(1), (2), (4)
Three investment equations (one for each region) on pooled data using SUR (seemingly unrelated regression). Cross-country OLS (ordinary least squares) and IV (instrumental variables) regressions. System of equations estimated using 3SLS (three-stage least squares). A specific equation is estimated for each endogenous dependent variable in the growth regression (six channel equations). The model is estimated separately for each of the five countries. Granger causality.
In Asia, there has been substantial crowding in of investment, while crowding out has been the norm in Latin America. In Africa, FDI has increased overall investment one-to-one. The positive impacts of FDI on domestic investment are not assured. FDI contributes significantly to economic growth, but the positive effects do not materialise unless local financial markets are sufficiently developed.
Two-equation (one for each decade) system estimated using SUR and IV.
FDI and growth: FDI exerts a positive effect on growth only when a minimum level of human capital exists. FDI and domestic investment: the complementarity between foreign and domestic investment is not robust to different specifications.
FDI has a positive and significant coefficient in the growth equation for three out of five countries. The negative sign of FDI in Singapore and Thailand is attributed to the specific characteristics of capital formation in these countries. Authors claim that FDI boosts growth in countries with a fair balance of domestic private capital and FDI. Furthermore, FDI is positively associated with positive spillover effects that lead to human resource development, transfer of technology, expansion of trade and learning by doing. The spillover process is positively related to the level of economic development.
FDI Granger-causes economic growth.
Annex Table 3.1. FDI and Growth: Literature Survey (continued) Authors (year)
Questions addressed (1), (2), (4)
Carkovic and Levine (2001)
Gross FDI inflows from new WB database and IMF. Period: 196095.
Dynamic panel data estimator (GMM).
Net FDI inflows from IMF’s Balance of Payments Statistics. 16 OECD and 17 non-OECD countries over the 1970-90 period.
Lensink and Morrissey (2001)
WB data on FDI as a percentage of GDP. 67 least developed countries, average of 1970-95 data. WB data on FDI/GDP over the 1975-98 period in 115 countries.
Stationarity and cointegration analysis plus dynamic panel estimation (fixed-effect and mean group estimators). Cross-country OLS with stability tests.
The impact of the exogenous component of FDI on GDP growth is not significantly different from zero, nor is FDI strongly linked to productivity (TFP) growth. These results are robust after controlling for the level of human capital and financial development. The FDI-growth nexus is not robust in all countries. Where the positive relationship holds, it depends on country-specific factors. FDI enhances output growth through higher productivity in OECD countries, and through capital accumulation in nonOECD countries. The growth impact of FDI tends to be lower in technological leaders and higher in laggards. FDI enhance growth once a country has reached a given threshold of human capital and financial market development. For most developing countries (30 of 67, almost all countries in sub-Saharan Africa), this threshold has yet to be attained. FDI exerts a robust positive impact on growth. This result is not conditional on the level of human capital. Volatility of FDI has a negative impact on growth, but it probably captures the growthretarding effects of unobserved variables such as political uncertainty.
De Mello (1999)
Hermes and Lensink (2000)
IMF and UNCTAD. 1970-96.
WB data on net FDI inflows. 44 non-OECD countries over the 1986-97 period.
OLS and IV for crosssection using the 197598 average values. Fixed-effect panel using three ten-year periods. Dynamic panel data on investment equations. Dynamic panel data.
FDI is a strong catalyst for domestic investment in developing countries. Lagged FDI has a stronger effect on private domestic investment than does lagged private domestic investment itself. Different types of capital inflows have different impacts on growth. FDI and portfolio equity flows show a positive and significant correlation with growth; debt inflows show a negative correlation.
Annex Table 3.1. FDI and Growth: Literature Survey (continued) Authors (year)
UNCTAD data on FDI inflows. Five-year periods from 1970 to 1995 for more than 100 LDCs.
Usha Nair and Weinhold (2001)
Questions addressed (1)
WB data on net FDI inflows as percentage of GDP for 243 developing countries over the 1971-95 period.
Share of MNE affiliates’ value added in host-country GDP. 40 countries over the 1966-94 period. Data from the US Direct Investment Abroad Benchmark Survey. Inward FDI stock from WB and UNCTAD/TNC for 11 Latin American and East Asian countries. Period: 1970-95.
Non-dynamic fixed-effect panel, first-differenced instrumented panel and mixed (fixed and random) effect model (heterogeneous panel) Instrumental variables panel data estimation with country- and time-specific effects.
Results from analysis of time-series characteristics of the explanatory variables show that: (1) FDI is always positively related to contemporaneous growth in per capita income; correlation with past growth rates is not robust; and (2) FDI is not related to past investment, while it is correlated with past trade. Growth regressions including lagged FDI and investment and other controls over individual and pooled periods have poor explanatory power. Lagged FDI is found to exert a positive but not statistically significant impact on growth. It turns out to be significant only when interacted with the level of schooling. Standard fixed-effects estimation points to a significant and positive impact of FDI growth on GDP growth. Results from the dynamic model under the assumption of heterogeneity reinforce this claim and show how the indirect impact of FDI on growth works differently across countries.
FDI boosts total factor productivity growth. Strong evidence of technology diffusion from US affiliates to developed countries, but only weak evidence for developing countries.
FDI is found to promote growth in five out of 11 countries, four of which are Asian. The impact of FDI on growth is country-specific and tends to be positive where policies favouring free trade and education are adopted to encourage export-oriented FDI.
Chapter Ⅳ FDI-Trade Linkages
A long-standing debate among academic researchers in connection with the role of FDI in development concerns the linkages between FDI and trade. Does FDI lead to trade, or does trade lead to FDI? Are FDI and trade substitutes or complements? Given that global FDI flows have expanded much faster than world merchandise trade over the last three decades (see Chapter Ⅱ), these questions have attracted considerable attention in the economics profession. Much of the early literature of the 1970s and 1980s was directed to quantifying the homecountry effects of FDI. This focus was partly motivated by concerns that multinational enterprises (MNEs) were replacing trade and “exporting” jobs from home countries when they expanded overseas production24. Contrary to the prediction of the standard trade theory, however, most of the early studies based on US and Swedish data found a positive association between outward FDI and home-country exports (see below)25. Such findings subsequently led to numerous empirical studies, many of which are reviewed in Annex Table 4.1. Meanwhile, the environment for international business activities has changed dramatically. The combination of rapid advances in information and communication technologies and improvements in transportation service has made international transactions much easier, thereby stimulating outward FDI (mostly from developed countries) over the past decade. Since MNEs are significant players in world trade26, a major shift in the mode of international transactions is likely to affect the volume and pattern of international trade. At the same time, many developing countries’ attitudes towards FDI are more open and welcoming today than some ten years ago. This change reflects growing recognition on the part of recipient countries that under the right conditions FDI can play a critical role in the development process through international transfer of capital, knowledge and technology. Indeed, an increasing
A recent study by Brainard and Riker (1997), based on US firm-level data for 1983-92, finds that labour employed by parent firms in the United States is substituted only at the margin by labour employed by their foreign affiliates. Labour substitution is found to be far greater between foreign affiliates located in host countries at similar levels of development. 25 Owing to the limited availability of firm-level data, empirical analysis of FDI-trade linkages has been restricted to several OECD countries, notably Japan, Sweden and the United States. In addition, the hypothesis is typically tested by relating affiliate sales in host countries to homecountry exports to host countries. 26 See Fukasaku and Kimura (2001) for further discussion.
number of developing countries have embarked on both trade and investment liberalisation, albeit to varying degrees, either unilaterally or in the context of regional or multilateral initiatives. The experience of China and several Southeast Asian countries since the early 1990s is a case in point. Empirical analysis of FDItrade linkages has therefore attracted renewed interest in recent years among policymakers and academic researchers in both developed and developing countries. Against the backdrop of these developments, this chapter presents a survey of recent empirical studies that examine the relationship between FDI and trade. As a preliminary step, it briefly reviews what can be predicted from standard models of international trade and investment. 4.1 Standard Models In its simplest form, the standard trade theory based on the so-called Heckscher-Ohlin model offers a dramatic prediction about the relationship between international capital movements and trade flows. It is shown that these two flows are substitutes for each other (Mundell, 1957). In this model, crosscountry capital movements created by international factor-price differentials lead eventually to the elimination of international price differentials on factor markets as well as on the goods market, generating an outcome that is identical to that of a free-trade equilibrium with immobile factors between countries. In contrast, the standard theory of the MNE conventionally regards exports and FDI as alternative strategies for a profit-maximising firm. In a model presented by Caves (1996, Chapters 1 and 2), the MNE is defined as an enterprise that controls and manages production plants located in (at least two) different countries and maximises total revenues accruing from its intangible assets. Such a firm supplies a foreign market either through production by an affiliate (or by another firm in the host country under a licensing agreement) or through exports from the home country27. He calls this model as the “intangible-assets model” of the horizontal MNE — a multi-plant firm producing the same line of goods from plants located in different countries. The crux of this model is to explain the existence of such an MNE from three perspectives: (1) it must possess some kind of intangible assets; (2) there must be locational forces that justify the dispersion
Caves (1996, pp. 3-7) attributes MNEs’ apparent preference for direct investment over licensing to the problems of market failure associated with arm’s-length transactions in intangible assets.
of plants in different countries; and (3) there must be some transactional advantage to placing these plants under common administrative control28. This model was used extensively in the early literature to analyse what determines the firm’s choice between exporting and FDI. It is shown, for instance, that any change in government policy that favours local production in the host country (such as tariffs) discourages exports. Note, however, that this model is concerned with a single-product firm whose market share in the host country is treated as a fixed amount. Accordingly, foreign production simply replaces direct exports from the home country. The question of FDI-trade linkages has also attracted considerable attention in the international business literature, where the sequence and pattern of corporate internationalisation are analysed from both the historical and the industrial organisation viewpoints. A recent survey article by UNCTAD (1996, pp. 75-93) emphasises that the dominant characteristic of internationalisation is the precedence of exporting over outward FDI as a way of entering foreign markets. It has been argued that most firms, particularly those in manufacturing, tend to build up overseas activities step by step: they typically start by exporting; then set up representative offices; establish marketing, distribution and after-sales facilities; and finally build up local production facilities in some host countries29. This linear sequence of development over time may be best explained by the transactional approach to the MNE: the “successful firm runs out its successes in the domestic market before incurring the transaction costs of going abroad” (Caves, 1996, p. 12). Thus, the business literature again points to substitution between FDI and trade as the dominant pattern of internationalisation30.
In a similar vein, Dunning (1977) proposes an eclectic approach which highlights three key requirements for a firm to undertake direct investment: (1) ownership advantage, (2) location advantage and (3) internalisation advantage. There is a fairly large body of literature on the determinants of FDI; for further discussion, see the survey articles on this topic by Aggarwal (1980), Lizondo (1990) and Petri and Plummer (1998). 29 The same UNCTAD report points out that a similar step-by-step sequence can also be identified in many natural resource–based industries, except that imports (not exports) induced by home-country demand precede FDI. In the case of service industries, this sequence may be “truncated” because many service companies often go abroad through FDI to support the foreign operations of customer companies in their home countries. 30 Horizontal FDI in response to actual or threatened import protection in host countries (“tariffjumping FDI”) abounds in business history. In this case, however, the sequence of internationalisation runs from restrictive trade policy imposed by a host country to reduced exports from a home country and then to foreign production through FDI (see below).
In what follows, this chapter reviews 20 empirical studies conducted since the early 1980s that test statistically the relationship between FDI and exports31. As reported in Annex Table 4.1, some of these studies do find concrete examples of substitution between foreign sales (of final goods) and home-country exports. Perhaps most noteworthy of all is a detailed study conducted by Blonigen (1999) at the product level. Blonigen’s empirical results for 21 specific products show substitution of foreign affiliate production for home-country exports in most cases, and such substitutions are often large one-time shifts rather than gradual changes over time. As predicted by the standard theory of the MNE, trade policy can play an important role in the firm’s decision on whether to export or invest abroad. Belderbos and Sleuwaegen (1998) present an interesting case based on the experience of Japanese electronics firms in the late 1980s. They find that a substantial part of these firms’ FDI in Europe was induced by the EC’s antidumping rules and other trade measures targeting Japanese firms — an example of the so-called “tariff-jumping FDI” that substitutes for exports from the home country. At the same time, sub-contractor firms supplying parts and components to their parent firms in a vertical production (keiretsu) system are found to export more to Europe. Another important case study is that conducted by Gopinath et al. (1999) on the US food-processing industry. Following more closely the standard theory of the MNE, they develop a four-equation model in which foreign affiliate sales, exports, employment and the demand for FDI are jointly determined. Their regression results point to little substitution between foreign affiliate sales and exports: a 10 per cent rise in the price of exports leads to a 1.1 per cent drop in foreign affiliates’ sales and a 0.6 per cent increase in exports. Moreover, a rise in agricultural protection in foreign countries tends to reduce US exports and increase foreign sales by affiliates, though the net impact of protection is small. Finally, a curious finding reported in this study is that the price of exports has a very small but positive effect on demand for FDI. They argue that such a complementary relationship may be interpreted as suggesting that foreign sales activities of MNEs include additional marketing and other support services.
See Hufbauer et al. (1994) for a survey of earlier studies dating back to the late 1960s.
4.2 The Substitution-Complementarity Hypothesis Revisited One major conclusion arising from this survey is that the empirical relationship between FDI and trade is much more complex than what the standard models of international trade and investment would predict. The majority of the 20 studies reviewed in Annex Table 4.1 indicate a strong complementary relationship between FDI and trade, despite differences in the data sets and estimation techniques used. Such results are not necessarily surprising, however, as the trade literature also indicates various possibilities that the relationship between FDI and trade is more one of complementarity than of substitution, once several restrictive assumptions imposed on the standard models are relaxed (see e.g. Markusen, 1983, and Wong, 1986). Three cases deserve special attention. First, foreign production may have important demand-enhancing effects in host countries by creating local goodwill and customer loyalty (especially for brand names), facilitating marketing and distribution (at lower costs and more reliable delivery) and generating spillover effects on other export goods (for multi-product firms). This is what Brainard (1997) calls the “proximity advantage” of establishing local production in host countries. Such “horizontal FDI” is thus likely to increase overall demand, resulting in higher exports from the home country (demand complementarity). Second, if the production process is divided into upstream (parts and components) and downstream (assembly) stages, and only the latter stage is transferred abroad, then the newly established assembly plant’s demand for parts and components can be met by exports from home-country suppliers. This is what Lipsey and Weiss (1981, 1984) and other researchers describe as “vertical FDI”, whose aim is to exploit scale economies at different stages of production arising from vertically integrated production relationships. This type of FDI is also likely to increase exports (of parts and components) from the home country, partly offsetting the substitution of foreign production for exports of final products. A recent study by Head and Ries (2001), using micro data on Japanese manufacturing firms over the 1966-90 period, provides further evidence of the importance of vertical FDI. It confirms that enterprises having a higher degree of vertical integration show greater complementarity between manufacturing FDI and exports. Wholesaling FDI is also found to exert a statistically significant positive effect on exports from the home country, which supports the earlier findings reported by Yamawaki (1991). In contrast, separate regression analyses 44
conducted only for assembly firms in the automobile and electronics sectors found that in this case, the relationship between manufacturing FDI and exports is one of substitution. Third, a production affiliate established in a host country may serve as an export platform to third countries in a particular region, instead of simply meeting local demand in the host country. The trade impact of such “export-oriented FDI” is clearly different from that of “market-oriented FDI”, though empirical research remains very limited in this respect. A study by Svensson (1996), based on firmlevel data for 1974-90, finds that while the overall effect is trade-creating for Swedish MNEs operating in the European Union, affiliate exports tend to replace exports from parent firms to other European countries. On the other hand, a study by Kawai and Urata (1995) on Japanese manufacturing firms operating in East Asia finds that FDI tends to generate “reverse imports” to the home country, as is postulated in Vernon’s (1968) product-cycle model32. Hufbauer et al. (1994) report similar findings at the aggregate level where Japan is concerned. More generally, the export-promoting effect of production by affiliates, together with country-specific characteristics (such as skills and openness to trade and investment) and firm-specific characteristics (such as R&D), tends to increase both parent-company exports and affiliate production. For instance, Eaton and Tamura (1996) find that an increase in human capital has a significant effect on both exports and FDI from Japan and the United States, while the “distance” variable tends to discourage exports more than FDI. Similarly, Pfaffermayr (1996), working with panel data for seven Austrian industries, highlights the importance of R&D intensity as a common determinant of both exports and FDI, though the latter effect is found to be statistically insignificant. Overall, the export-promoting effect of FDI is likely to outweigh any tendency for affiliate production to replace parent-company exports (Lipsey, Ramstetter and Blomström, 2000). 4.3 Aggregation, Causality and Endogeneity A second major conclusion arising from this survey is that in addition to the paucity of firm-level data, which are currently available only for a few OECD countries, existing empirical studies face three methodological problems. One is the aggregation problem. As Blonigen (1999) argues, it is certainly possible to find
See also Urata (2001).
substitution between foreign sales and home-country exports for a given product33. This does not necessarily mean, however, that substitution dominates at the industry level — a reasonable level of aggregation for policy discussion on matters such as taxes and tariffs. If the aim of empirical research is to offer informed discussion about macroeconomic policy, a higher level of aggregation would be more appropriate34. The problem of aggregation should thus be placed in a specific policy context, subject to the availability of relevant data. Second, although the choice of estimation technique is often dictated by the availability of data, the use of cross-section analysis makes it difficult, if not impossible, to address the problem of causality between FDI and trade. Alternative approaches are to use time-series or panel-data analysis. For example, a pioneering study by Pfaffermayr (1994) applies Granger causality tests to quarterly data on Austria’s outward FDI and exports, finding a bi-directional causal link between the two. A more interesting result of this study is that a third variable (real GDP of OECD countries) is found to have a significant impact on both outward FDI and exports. The problem of causality has been further investigated by de Mello and Fukasaku (2000), who extend the time-series analysis to 16 selected Latin American and Pacific Asian countries. In their regression results, the theoretical prediction that imports precede inward FDI is supported by several country cases, but the evidence is far from conclusive. As is expected from some previous studies, they also find that net FDI flows have a negative impact on trade balances in Pacific Asian countries over the full sample period (1970-94) and in Latin America over the reduced sample period (1970-84). In a similar vein, Wang et al. (2001) apply Granger causality tests to the case of China, the largest recipient of FDI flows among non-OECD countries. In their study, causality seems to run from China’s imports to FDI inflows and then to exports back to the home countries. In a different context, Pain and Wakelin (1998) have also addressed the problem of causality by investigating whether the expansion of FDI has had any significant impact on the trade performance of either home or host countries. Based on macroeconomic data for 11 major OECD countries, their regression
His study is based on US import data at the 10-digit level of the Harmonised System. Developing host countries often express concern over a potential deterioration in the balance of payments due to increased imports from and royalty payments to parent companies located in developed countries (WTO, 2001). 34
results show that the trade effect of FDI varies both in sign and in magnitude among the OECD countries concerned. Their overall conclusion is that a small negative impact of outward FDI on home-country exports tends to be offset by a corresponding positive impact of inward FDI on host-country exports. A study by Goldberg and Klein (1997) indicates that exchange-rate movements are an important determinant of both trade and FDI flows between Japan and several Southeast Asian countries. For instance, a real appreciation of the Japanese yen against the US dollar tends to induce Japanese FDI flows into these countries, thereby contributing to the development of local export capacity, which supports the earlier findings of Kawai and Urata (1995). The results of the time-series and panel-data analyses described above lead us to a third methodological problem, which is the question of endogeneity. Since changes in FDI are not exogenous, it is difficult to disentangle the effect of FDI on exports from other factors that may have affected both FDI and exports simultaneously. Amiti and Wakelin (2000) put it simply: “[T]he problem is more than just one of endogeneity. There is also a conceptual issue: what does it mean economically for the partial derivative of one endogenous variable to be positively or negatively related to another? The question is how does a shock to an exogenous variable affect the two endogenous variables [FDI and exports]?” (p. 3). Conceptually, it is possible to distinguish between “substitutes” and “complements” in several ways. Most empirical studies reviewed so far (with the notable exception of Gopinath et al., 1999) define this relationship, explicitly or implicitly, in terms of volume. To test this hypothesis along the lines of the standard demand theory, Clausing (2000) has tried to introduce the “price of operating abroad” (e.g. tax rates and wage costs in host countries) as one of the key explanatory variables on the right-hand side of US export equation. In other words, FDI and exports are likely to be substitutes when the estimated cross-price elasticity of exports with respect to the cost of direct investment abroad is positive, and to be complements when this elasticity is negative. His estimation results show a complementary relationship. Similarly, Amiti and Wakelin (2000) have incorporated the “investment cost” index developed by the World Economic Forum (a qualitative index constructed on the basis of country business surveys) into a standard gravity-trade model of US exports. They find that estimated cross-price elasticities of exports with respect to
investment costs are negative but vary significantly among partner countries and over time35. 4.4 Towards a Unified Approach Until recently, economic analysis of the MNE (including the determinants of FDI) was regarded as a distinct area of research, based on partial equilibrium models and largely separate from the general equilibrium analysis of international trade. During the 1980s, the industrial organisation approach to international trade opened a new frontier in the trade literature (the so-called “new” trade theory) by incorporating imperfect competition and scale economies into general equilibrium models. Yet, in the words of Markusen (2000, p. v), the MNE was “generally missing, in spite of having precisely these characteristics”. A new approach to the theory of the MNE has been developed over the past several years, blurring the boundary between these two areas of research36. This approach is now called the “knowledge-capital model” of the MNE, after the seminal work of Markusen et al. (1996) and Carr et al. (2000). This model allows both horizontal and vertical FDI to arise endogenously, depending on country characteristics (such as market size, income level, skill differentials and distance) and the level of trade costs (such as transportation costs and tariffs). Following the “eclectic approach” developed by Dunning (1977), this model distinguishes between the knowledge-based service activity of the MNE as a source of ownership advantage (referred to as “headquarters services”) and the goodsproducing activity. This distinction leads to the possibility that these two activities can be geographically separated while remaining within a single firm. It is also assumed that headquarters services serve collectively as an input in goods production and are more skill-intensive than production. This model shows that horizontal FDI will substitute for exports, depending on the multi-plant scale economies realised relative to trade costs. In contrast, vertical FDI will complement exports, since the home country supplies headquarters services and/or intermediate products to the host country. Ultimately, it is country characteristics and trade costs (net of trade and other government policies) that determine which type of FDI will dominate. Thus, the knowledge-capital model of the MNE can provide a unified approach to
Note that their empirical study explicitly distinguishes the cost of direct investment from the cost of international trade (e.g. tariffs and non-tariff barriers). 36 Two articles by Markusen (1995 and 2000) provide an excellent overview of this topic.
international trade and production with testable hypotheses, and a third major conclusion that can be drawn from this survey is that empirical analysis of FDItrade linkages has entered a new era, as exemplified by the recent work of Amiti and Wakelin (2000) and Clausing (2000).
Annex Table 4.1. FDI-Trade Linkages: Literature Survey Authors (year) Lipsey and Weiss (1981)
Data/coverage 14 manufacturing industries based on BEA survey data on US MNEs for 1970.
Estimation technique Cross-section, OLS regression estimation of a gravity trade model for exports of United States and 13 other countries, including net sales or net local sales of US MNEs as an explanatory variable.
The same as for Lipsey and Weiss (1981).
Blomström, Lipsey and Kulchycky (1988)
7 manufacturing groups based on IUI (Stockholm) survey data on Swedish MNEs for 1970 and 1978. 30 manufacturing industries based on BEA survey data on US MNEs for 1982.
Cross-section, OLS regression estimation of parent company exports equations based on a gravity-type model, including net sales or net local sales of US manufacturing affiliates and sales of US non-manufacturing affiliates. Cross-section, OLS and 2SLS (two-stage least squares) regression estimation of a gravity-type model for Swedish exports, including net sales or net local sales of Swedish MNEs as an explanatory variable. Cross-section, OLS regression estimation of a gravity-type model for US exports, including net sales or net local sales of US majority- and minority-owned affiliates as explanatory variables.
Major findings Overseas production by US affiliates tends to increase US exports (complementarity effect). Its impact on exports from 13 foreign counties is generally negative. Estimated coefficients are higher for metals and machinery than other industries. Foreign production by a US firm does not on balance substitute for exports by that firm to the area where foreign production takes place. This complementarity effect is particularly strong for intermediate products. No evidence that host-country production substitutes for exports from Sweden. The higher initial level of foreign production (1970) leads to a larger increase in Swedish exports between 1970 and 1978, with only one exception (metal manufacturing). Predominance of positive relationships between affiliates’ net sales and US exports, with a few exceptions in which substitution effects are found.
Annex Table 4.1. FDI-Trade Linkages: Literature Survey (continued) Authors (year) Pfaffermayr (1994)
Data/coverage Quarterly data on Austria’s outward FDI flows, exports of goods and real OECD GDP during the 1969-91 period.
Hufbauer, Lakdawalla and Malani (1994)
Cross-section data on FDI (stocks and flows) and trade (exports and imports) for 1980, 1985 and 1990 for Germany, Japan and the United States. Pooled sample data on Japanese bilateral trade with and FDI outflows to 48 countries over the 1980-92 period for eight manufacturing industries.
Eaton and (1996)
Panel data on Japanese and US exports and outward FDI to 72 host countries over the 1985-90 period. Panel data on exports and outward FDI for seven Austrian industries over the 1980-94 period.
Estimation technique Time-series analysis of the causal relationship between FDI and exports based on Granger causality and cointegration tests. Regression estimation of standard gravity-type FDI and trade equations; in the latter, lagged FDI is included on the right-hand side of the equation. OLS regression estimation of standard gravity-type trade and FDI equations, with lagged FDI or trade variables, respectively, included on the right-hand side. Tobit regression estimation of export and FDI decisions based on a modified gravity model, including several characteristics of destination countries. Logit-transformation regression estimation of export and FDI equations, with factor intensities as common determinants.
Major findings Granger causality tests indicate a bi-directional causal link between FDI and exports. Real OECD GDP (exogenous) has a significant impact on both FDI and exports. A given amount of outward FDI from Japan generates about twice as many Japanese imports as exports. This is not the case in the United States and Germany (except for 1980). Two-way complementary interactions were found to exist between Japan’s FDI outflows and exports, except for some industries. FDI tends to increase imports more than exports. Increases in human capital have a significant effect on both exports and FDI from Japan and the United States. Distance tends to reduce exports more than FDI. A significant and stable complementary relationship was found to exist between exports and outward FDI with bi-directional causality. R&D intensity tends to affect both FDI and exports positively, though the coefficient is statistically significant only in the latter case.
Annex Table 4.1. FDI-Trade Linkages: Literature Survey (continued) Authors (year) Svensson (1996)
Data/coverage Pooled cross-section data of Swedish MNEs (at the firm level) over the 1974-90 period.
Estimation technique A variant of 2SLS regression estimation (the Tobit method) with limited dependent variables to take into account the possibility of simultaneity in the relationship between FDI and exports.
Panel data on trade and FDI flows between two home countries (Japan and the United States) and seven host countries (Argentina, Brazil, Chile, Indonesia, Malaysia, the Philippines and Thailand) over the 1978-93 (or 94) period.
Panel-data regression estimation of trade and FDI equations with fixed country effects based on a gravity-type model, including real bilateral exchange rates as explanatory variables.
Pain and Wakelin (1998)
Semi-annual panel data on manufactured exports and outward and inward FDI stocks for 11 OECD countries over the 1971-92 period.
Panel-data regression estimation of dynamic export demand equations.
Major findings Increased foreign production by affiliates tends to replace exports from their parent firms and to complement parent firms’ exports of intermediate products. The net effect of these can be negative, though quantitatively small. At the same time, exports from foreign affiliates tend to create a strong substitution effect in third countries (i.e. replacing parent company exports), though the overall effect is trade-creating for Swedish multinational operations in the EC. Appreciation of the yen-dollar real exchange rate tends significantly to raise Japanese FDI into Southeast Asia, but not into Latin America. US FDI tends to substitute for US trade with Southeast Asia, while Japanese FDI leads to more outward-oriented local production in the area. The trade effects of FDI vary in both sign and magnitude among the OECD countries under study. Overall, a small negative impact of outward FDI on home-country export performance is offset by a corresponding positive impact from inward FDI on host-country export performance.
Annex Table 4.1. FDI-Trade Linkages: Literature Survey (continued) Authors (year) Belderbos and Sleuwaegen (1998)
Data/coverage Firm-level data for Japanese manufacturing firms producing 35 electronics products.
21 products (10 automobile parts for 1978-91 and 11 final consumer products over the 1972-94 period). NBER trade data on US imports of Japanese products and firm-level data on Japanese local production in the United States. Panel data on US foreign sales, exports, affiliate employment and FDI flows with ten host countries over the 1982-94 period.
Gopinath, Pick and Vasavada (1999)
Amiti and Wakelin (2000)
Panel data on US bilateral trade flows (non-agricultural goods) with 35 partner countries over the 1986-94 period.
Estimation technique Logit model regression analysis of Japanese firms’ decision to undertake direct investment in the EC, using the latter’s trade policy variables. Logit model regression analysis of Japanese firms’ export intensity to Europe. Time series, seemingly unrelated regression estimation of US import demand functions for Japanese products.
Panel-data regression estimation of a fourequation system in which four variables are jointly determined.
Panel-data regression estimation of US bilateral trade flows with fixed country effects based on a gravity-type model, including two indices representing trade and investment costs as explanatory variables.
Major findings Tariff-jumping FDI by Japanese firms tends to substitute for their exports. Affiliated exporters tend to expand the supply of components to EC assembly plants operated by their core firms. Product-level data allows separate identification of both substitution (of local production for exports) and complementarity effects (from vertical production relationships between final products and intermediate inputs). Substitution of foreign production for exports often occurs in large one-time shifts. Foreign affiliate sales are found to substitute for exports at the aggregate food-processing level. Agricultural protection in foreign countries hurts US exports, while encouraging affiliate production. The price of exports has a very small but positive effect on FDI. Estimated cross-price elasticities of exports with respect to investment costs are negative but vary significantly across countries and over time. Overall, a fall in investment costs in a host country tends to increase US exports to that country.
Annex Table 4.1. FDI-Trade Linkages: Literature Survey (continued) Authors (year) Clausing (2000)
Data/coverage Two panel data sets constructed from BEA surveys for the 1977-94 period: one on the operations of US parent firms in 29 host countries and the other on the US operations of foreign affiliates of firms based in 29 home countries.
De Mello and Fukasaku (2000)
Annual data on total imports, manufactured exports and net FDI inflows for 16 selected Latin American and Pacific Asian countries over the 1970-94 period.
Estimation technique Panel-data regression estimation of US export and import equations with or without fixed country effects based on a gravitytype model, including net local sales as an explanatory variable. Separate regression estimation of US export equations, incorporating the “price” of operating abroad (e.g. taxes paid to host countries) among the explanatory variables. Time-series analysis of temporal causality between trade and FDI flows over both short and long terms, based on Granger causality and cointegration tests.
Lipsey, Ramstetter and Blomström (2000)
Ten manufacturing industries based on MITI survey data on Japanese MNEs for 1986, 1989 and 1992.
Cross-section, regression estimation of Japanese parent exports based on a gravity-type model.
Major findings Complementarity between net local sales and trade is far greater for intra-firm trade than inter-firm trade. As the price of operating abroad increases, exports fall, which suggests that foreign production and exports are complements.
The theoretical claim that imports precede inward FDI is supported by several country cases, but the evidence is far from conclusive. The prevailing impact of FDI on the trade balance is negative in Pacific Asia over the full sample period and in Latin America over the reduced sample period (1970-84). Increased production in a region by a parent firm’s affiliates is associated with greater exports to the region from the parent firm. The more the firm produces abroad, the higher Japanese parent company employment tends to be for a given level of parent production.
Annex Table 4.1. FDI-Trade Linkages: Literature Survey (continued) Authors (year) Head and Ries (2001)
Data/coverage Panel data for 932 Japanese manufacturing firms over the 1966-90 period. Explicit distinction between manufacturing and distribution affiliates set up by Japanese MNEs. Explicit distinction between major assemblers and their parts suppliers.
Panel data regression estimation of Japanese manufacturing firms’ export functions with or without fixed effects, including time-varying firm characteristics among the explanatory variables. Separate regressions estimated for ten automobile firms and nine electronics firms (assemblers in both cases) and for 96 parts suppliers to these two groups.
Wang et al. (2001)
Panel data on China’s exports, imports and inward FDI stocks (in real terms) with respect to 19 home countries/regions over the 1984-98 period.
Granger causality tests based on a standard VAR (vector autoregression) model.
Firms that are more vertically integrated show a greater degree of complementarity between manufacturing FDI and exports. The coefficients of wholesale FDI on exports are found to be positive and statistically significant. For these two groups of assemblers, manufacturing FDI and exports exhibit a substitution relationship. The manufacturing investment of assemblers has a positive and significant effect on exports by suppliers of automobile parts. Causality is found to run from China’s imports to FDI inflows and then to its exports back to the home countries. Two-way complementary relationships between China’s exports and imports.
Chapter Ⅴ FDI and Technology Transfer
Technology transfer through FDI is another topic of long-standing interest to academic researchers in the development field. Multinational enterprises (MNEs) are among the most important players in the world in terms of generating and controlling new technology37, and they derive much of their profits from putting the tangible and intangible resources available in different countries to the most productive use. As part of these global profit-making operations, FDI inherently involves the transfer of capital, technology and know-how from home to host economies (see Figure 1.1). This “packaging” of better technologies with the capital and skills needed to exploit them offers developing countries the opportunity to increase productivity and hence, in the long run, economic growth and development. While FDI is only one of several channels for international transfer of technology, many countries view it as the most important means of acquiring better technologies to upgrade their own production bases (see Box 5.1). Nonetheless, it is difficult to paint an unambiguous picture as to how MNEs transfer technology through FDI and how this technology contributes to productivity growth in the host economy38. This difficulty stems from the very nature of technology transfer through FDI. As discussed in Chapter Ⅲ, knowledge accumulation and dissemination are key to the long-term growth of the host economy. According to endogenous growth models, FDI can exert a positive impact on domestic output through its effect on the overall level of technology (disembodied technical change), through technological improvements in the capital stock and through skills improvements in the labour force. As such “spillover” effects are hard to quantify individually, most existing empirical studies have been directed to examining how significantly FDI affects total factor productivity (TFP) growth.
The generation of new technology is highly concentrated in a handful of advanced industrial countries and takes place mainly in large firms, typically MNEs. For example, G7 countries as a whole account for 90 per cent of world R&D expenditures, and the United States alone for 40 per cent. In the United States, just 50 leading firms accounted for nearly 50 per cent of industrial-based R&D expenditures in 1996 (UNCTAD, 1999, p. 199). 38 In this chapter, the term “technology” is used in a broad sense to signify the knowledge that is embodied in products, processes and practices, the latter including managerial skills and know-how.
Box 5.1. FDI as a Mode of Technology Transfer While FDI is only one of several means available for a firm to transfer technology outside its home country (e.g. exporting products that embody the technology or licensing its technology to an agent abroad), for developing countries it remains the most important means of acquiring new technology. Technology transfer through FDI offers benefits that other modes of transfer do not, for at least three reasons: −
Unlike trade in goods, where host countries must try to imitate and learn from reverse engineering, FDI involves the explicit transfer of technology. In addition to the technology itself, FDI brings needed complementary resources such as management experience and entrepreneurial abilities, which can be transferred through training programmes and learning by doing. A World Bank study using firmlevel survey data on Czech enterprises of different ownership shows that domestic firms receiving FDI or involved in joint ventures tend to provide training programmes and acquire new technologies more frequently than those with no foreign partners (Djankov and Hoekman, 1999).
Technologies used by foreign affiliates are not always available in the arm’s-length market, and even when they are available, some technologies may be more valuable or less costly when applied by MNEs that developed them, rather than by outsiders. This is especially the case when local employees need to develop specific skills in using the technology. MNEs also offer brand names and access to regional and global markets.
The entry of foreign affiliates in local markets provides an incentive for domestic firms to innovate in order to protect their market shares and profits. This “demonstration effect” alone is likely to lead to productivity increases in local firms (see the discussion below, under “Horizontal Linkages”).
The process of technology transfer to local firms may take longer than host governments would normally expect from foreign partners, particularly in technologically advanced sectors. A recent survey of 20 leading EU companies with investments in China, conducted by a group of business economists, finds that these companies show a measure of reluctance to transfer their core technologies and to base R&D activities there. One reason for this reluctance is the insufficient legal protection of intellectual property rights in the host country (Bennett et al., 2001). More generally, the characteristics of the host country and its domestic absorptive capacity are key determinants of international technology transfer (Radosevic, 1999).
A key question to be addressed in this chapter is how and the extent to which the productivity growth of local firms is influenced by the presence of MNE affiliates. Technology transfer is not instantaneous; it takes place over time, often years rather than months. As a result, empirical work on this topic would require detailed microeconomic data on the production performance of individual firms or plants of different ownership (local, foreign or joint venture), and the coverage would have to extend across industries and over several years in order to observe both intra-industry and inter-industry linkages and spillovers (Kugler 2000). It would indeed be a daunting task to meet such extreme data requirements. As no comprehensive analysis of this character has ever been made, one normally has to
look for circumstantial evidence of technology transfer through FDI on the basis of detailed case studies (Blomström and Kokko, 1996). Bearing in mind these methodological and data limitations, this chapter presents a review of recent empirical literature on the relationship between FDI and technology transfer in the development context. It begins by describing various mechanisms that may be at work when technology is transferred through FDI and goes on to discuss the main results of existing empirical studies based on microeconomic data (see also Annex Table 5.1). 5.1 Mechanisms of Technology Transfer As noted above, it is widely recognised that FDI provides an important channel for international transfer of technology. This does not, however, explain exactly how this transfer and the resulting spillovers take place. The literature suggests that FDI transfers technology to the host economy, either directly or indirectly, via four mechanisms: •
Vertical linkages: MNE affiliates may transfer technologies to local firms that
supply them with intermediate goods, or to buyers of their own products. Horizontal linkages: Local firms in the same industry or phase of the production process may adopt technologies through imitation, or may be forced
to upgrade their own technologies due to increased competition from MNE affiliates. Labour turnover: Workers trained or previously employed by MNE affiliates
may transfer their knowledge to other local firms when they switch employers or set up their own businesses. International technology spillovers: MNEs may enhance local technological capability through R&D activities in the host country or through intra-firm transfer of technology (i.e. from the parent firm to its foreign affiliates).
In the following sub-sections, we discuss each of these four channels in more detail. 5.1.1 Vertical Linkages with Suppliers and Buyers It has long been recognised that MNEs can benefit the host economy via the backward and forward linkages they generate. Backward linkages are relations with local suppliers, while forward linkages refer to relations with buyers — either consumers or firms that use intermediate and capital goods (including
machinery) produced by MNE affiliates. While the formation of inter-firm linkages does not inherently ensure that technology is actually transferred or spilt over to local firms, such linkages can be an important channel for technological spillovers, except in a self-contained “enclave” situation. Moreover, it is unlikely that MNEs will be able to appropriate the full value of these explicit and implicit transfers to their host-country business partners (Blomström et al., 1999). Some theoretical and empirical works have studied the factors that promote vertical linkages. It seems, first, that the larger the host market and the greater the technological capabilities of local suppliers, the more pronounced are the linkages. Second, according to the model developed by Rodríguez-Clare (1996), more linkages are created when the production process of the MNEs uses intermediate goods intensively; when the costs of communication between headquarters and the affiliate production plant are large; and when the home and host countries are not too different in terms of the variety of intermediate goods produced. Third, government policies can also promote linkage creation, e.g. through policies requiring a minimum of local content, although the efficiency and usefulness of such policy requirements have been debated in the literature39. Whether MNE affiliates in host countries actually form such vertical linkages largely depends, however, on affiliates’ decisions on how to source inputs (Chen, 1996). Although in some cases local content starts at a very low level, in general local vertical linkages are extended over time, which could be a consequence of technology transfer. Studies of the Asian electronics industry have generally shown that linkage creation was negligible at first, but had grown substantially five years later (Rasiah, 1994). When addressing the impact of linkage creation, it should be borne in mind that MNEs improve welfare only if they generate linkages beyond those already generated by the local firms they displace. MNEs can provide raw materials and intermediate goods, or assist local suppliers in purchasing these inputs, as well as helping prospective suppliers to set up production facilities. They can also provide training in management and organisation, and help suppliers to diversify by finding additional customers. Empirical evidence of such backward linkages is found in many studies in the early literature, including Lall’s (1980) study on Indian truck manufacturers,
See UNCTAD (2001, Chapter V) for further discussion.
Wanatabe (1983), UNCTC (1981) and Behrman and Wallender (1976). Many more recent examples of such linkages can be found in UNCTAD (2001, Chapter IV). Forward linkages, in contrast, are formed with local buyers. These may be either distributors, which can benefit from the marketing and other knowledge of MNEs, or — in the case of intermediate products — downstream firms which can use higher-quality and/or lower-priced intermediate goods in their own production processes. Downstream firms may also benefit from lower prices arising from increased competition in their supply market (Pack and Saggi, 1999), and consumers as well may benefit from lower-priced final products. 5.1.2 Horizontal Linkages through Demonstration and Competition The diffusion of technology through horizontal linkages (i.e. to competitors of the MNE affiliate in the host country) works through either demonstration or competition effects40. The term “demonstration effect” refers to the fact that exposure to the superior technology of the MNE may lead local firms to upgrade their own production methods (Saggi, 2000). When an MNE starts using a specific technology that has not previously been used in the host economy, its competitors may start imitating the technology. Often, the introduction of a new technology by an MNE reduces the (subjective) risk for local firms of using the same technology. Local firms may lack the capacity, financial resources or information required to acquire the necessary knowledge or to adapt the technology to local circumstances. However, when a certain technology used by MNE affiliates proves successful in the local environment, it may be adopted more widely by local firms. While FDI may expand the range of technologies available to local firms, it also usually increases competition in the local market. Moreover, demonstration and competition effects reinforce each other. The entry of an MNE affiliate increases competition, which is in itself an incentive to upgrade local technologies. This further stimulates competition, which in turn causes an even faster rate of adaptation of new technologies. Wang and Blomström (1992) also stress that the more competition the MNE affiliate faces from domestic firms, the more technology it has to bring in to retain its competitive advantage, and hence the larger will be the potential for spillovers.
Saggi (2000) argues that a strict definition of spillovers would count only the demonstration effect, because innovation in domestic industry induced by increased competition cannot be seen as pure externalities but as a “benefit enjoyed by the host country that works its way through the price mechanism” (ibid., p. 18). In practice, however, such a distinction is very difficult to make.
The effects of increased competition are usually seen as beneficial. Increased competition encourages both productive efficiency and more efficient allocation of resources. This may be especially true when MNEs enter industries where high entry barriers limit the degree of domestic competition (e.g. utilities). Case studies indicate, however, that substantial improvements in productivity are obtained not so much through better resource allocation as through a reduction in slack or Xinefficiency (WTO, 1998). This is the case when efficiency increases result from local firms enforcing stricter or more cost-conscious management and motivating employees to work harder, instead of imitating technology. In theory, competition generally improves efficiency and welfare, but entry by MNE affiliates does not necessarily increase competition. In fact, it may lead to increased concentration. Economies of scale are an important determinant of industrial structure, and when a foreign affiliate enters a relatively small national industry and increases the average firm size, this may initially improve resource allocation. There are concerns, however, that strong MNE affiliates may outcompete local firms, or at least force them to merge, and the consequent increase in industrial concentration can result in market power. The abuse of market power by the MNE (and possibly local firms) would then reduce allocative efficiency. Empirical evidence of demonstration and competition effects is difficult to obtain. Both effects are most likely to occur at the industry level (Saggi, 2000). One method of checking whether local efforts to adopt new technologies are encouraged by FDI is to relate R&D expenditures by industry to foreign presence, but these expenditures must be controlled for the effect of FDI on market structure, which is very difficult. Still, some general studies address the issue of horizontal linkages. Blomström et al. (1999), comparing foreign-owned and domestically owned firms, find that new technology is frequently introduced sooner by foreign-owned affiliates and that competition spurs quicker adoption of innovations by both types of firms. On the basis of plant-level data for Venezuela, Aitken and Harrison (1999) find a positive relationship between foreign equity participation and plant performance, which implies that foreign participation does benefit the plants that receive it. However, this effect is found to be robust only for small plants with fewer than 50 employees. For larger plants, foreign participation brings no significant improvement in productivity relative to domestic plants with no foreign participation. In general, productivity in domestic plants with no foreign participation declines as foreign investments in other plants increase. This may 61
result from a crowding-out effect: foreign competition may have forced domestic firms to lower output, thereby forgoing economies of scale. On balance, however, the results show that the effect of FDI on the productivity of the entire industry is weak but positive. Djankov and Hoekman (1999) also find that FDI has a positive impact on the TFP growth of recipient firms in the Czech Republic, but joint ventures and FDI appear to have a negative spillover effect on domestic firms that have no foreign partners. Such findings need not imply, however, that host countries have nothing to gain from FDI. Positive impacts such as improved resource allocation take time. When foreign firms bring in more efficient production methods, it is not surprising that some local firms suffer in the short run. 5.1.3 Labour Turnover Labour turnover is another channel through which technology may be transferred and disseminated in a host country. Workers employed by the MNE affiliate acquire knowledge of its superior technology and management practices. When these workers switch employers or start up their own businesses, they spread the technology. MNE affiliates usually try to avoid this kind of spillover by paying an “efficiency wage” with a premium to keep employees from switching jobs to domestic competitors (Globerman et al., 1994). If disclosing secrets to local managers would create unacceptable risks, e.g. due to managers’ tendency to take jobs with competitors, the MNE may consider using expatriate managers rather than local ones. The effects of labour turnover are difficult to establish. Several studies have been undertaken that may provide some insights, although they show quite different results. Katz (1987) finds that many managers of local firms in Latin America were trained in the MNE affiliates where they started their careers. In a study of 72 top and middle managers in Kenya, Gershenberg (1987) shows that MNEs offered more training to their managers than did local private firms, but he also finds evidence that only a small percentage (16 per cent) of job changes involved movement from multinationals to domestic firms. In the case of Mexico, Venezuela and the United States, Aitken et al. (1996) show that higher levels of FDI were associated with higher wages in all three countries. In the first two, they also find that multinationals paid higher wages than local firms but uncovered no evidence of wage increases by local firms.
In a detailed study of the development of the Korean electronics industry, Bloom (1992) finds that substantial technology transfer occurred in the 1970s when production managers switched to local firms. Pack and Saggi (1997) finds similar results for Chinese Taipei in the mid-1980s: among employees of MNE affiliates who changed jobs, almost 50 per cent of all engineers and 63 per cent of all skilled workers joined local firms41. UNCTAD (1999) examines the experience of Desh, a Bangladeshi garment firm which the Korean conglomerate Daewoo supplied with technology and credit. The study found that, over time, no fewer than 115 of the 130 initial workers left Desh to set up their own firms or to join newly established local garment firms. 5.1.4 International Technology Spillovers Multinational firms are among the world’s foremost creators of knowledge and technology. Of the top 25 R&D players world-wide, the first five positions are held by governments (in descending order, the United States, Japan, Germany, France and the United Kingdom), but the next 20 positions are dominated by MNEs (Van Tulder et al., 2001). Many of these firms concentrate their R&D activities in their home countries or other developed countries. Developing countries account for only an estimated 6 per cent of global R&D expenditures (Freeman and Hagedoorn, 1992). Moreover, such expenditures are highly concentrated even among developing countries: according to UNCTAD (1999) estimates using US firms as a proxy, the top four developing economies (Brazil, Chinese Taipei, Mexico and Singapore) account for 77 per cent of total R&D expenditures in developing countries. The rationale for this concentration lies in the need for efficient supervision and scale economies in the R&D process itself42. In addition, concentration of R&D offers a major advantage — from the firm’s perspective — in the form of “agglomeration economies”. This means that it is more efficient to cluster specific R&D expertise in a certain region, using local research institutions and other organisations to form an “innovation system”. This kind of locational advantage is fairly durable over time, so that MNEs tend to keep most of their R&D centralised at their headquarters (Globerman, 1979). Many developing countries do not offer the infrastructure and institutions needed for fruitful interaction between 41
See also Natarajan and Miang (1992) for the case of Singapore. The authors argue that the transfer of technology through the establishment of new local enterprises by those engineers and technicians who used to work for MNE affiliates has played an important role in building up a pool of indigenous supporting industries in Singapore. 42 See Caves (1996, Chapter 7) for further discussion.
academia, government and industry43. Another reason is the lack of protection of property rights, including intellectual assets, in host countries. In a recent study of EU-based companies regarding technology transfer to China, Bennett et al. (2001) highlight weak protection of intellectual property rights as a main obstacle to building or expanding the R&D base of these companies in the host country. Saggi (2000, p. 17), referring to UNCTAD data on trade in technology, points out that in 1995 over 80 per cent of global royalty payments — the explicit sale of technology — were made within the ambit of MNEs (i.e. from foreign subsidiaries to their parent firms). In addition, FDI may contribute directly to the generation of knowledge in developing countries through the internationalisation of MNEs’ R&D activities. Thus, a major policy concern for many developing countries is the extent to which technology is transferred from the parent firm to its foreign affiliates. Some argue that where R&D has been transferred to foreign subsidiaries, the primary purpose has been to carry out adaptive tasks, drawing on a few local resources in order to better serve the local market (Correa, 1999). MNEs are often blamed for failing to adopt technologies that are appropriate for the factor prices prevailing in developing countries. Nonetheless, where R&D is performed in developing countries, the expenditures have been found to generate significant efficiency gains in those countries, both within and across industries (Bernstein, 1989). Since foreign affiliates have access to the superior knowledge base provided by the parent firm, their R&D expenditures may perform better than local R&D expenditures. The question of intra-firm technology transfer has been addressed directly in two recent studies by Urata (1999) and Urata and Kawai (2000), on the basis of firm-level data on Japanese manufacturing affiliates operating in Asia. They conducted regression analyses to examine the determinants of intra-firm technology transfer. The dependent variables used for empirical analysis include both qualitative and quantitative indicators reflecting the extent and type of technologies transferred44. One quantitative indicator used for regression analysis
Rasiah (2001) argues, in reference to the development of small and medium-scale industries in Malaysia, that one reason why small machine tool firms in Penang perform better than those in the Kelang Valley is differences in the quality of government-business co-ordination. 44 Urata (1999) applied ten different kinds of qualitative technology indicators to the probit regression analysis. These indicators include operational technology, maintenance and inspection, process and quality control, design technology, development of new products and the introduction of new technology.
is the ratio of a foreign affiliate’s TFP level to that of its parent firm45. Their regression analyses show both expected and unexpected results. First, among the characteristics of host countries, the level of education is found to be statistically significant and positive in all cases. Second, reliance on parent firms in the forms of equity holdings, supply of personnel and purchase of capital goods is shown to promote intra-firm technology transfer. Third, experience in industrial activities also tends to have a positive effect on intra-firm technology transfer, but this is statistically significant only for Asian affiliates in textiles and electrical machinery. Fourth, in contrast, the technical capability of foreign affiliates, measured in terms of R&D expenditures and royalty payments, is found to be statistically insignificant in most cases and to have the wrong sign. Finally, technology transfer requirements imposed by host governments are shown to have an unexpected negative effect on intra-firm technology transfer in many cases. 5.2 Technology Transfer and Host-country Conditions As discussed above, FDI may disseminate technology in a host country in various direct and indirect ways. Although some studies have attempted to examine the specific effects of each of the modes of transfer discussed, it is hardly possible to disentangle the effects of the various channels when assessing how technology transfer through FDI affects the productivity growth of host-country firms. Researchers take the view that the technological gap which may exist between local firms and MNE affiliates is revealed in the observed difference in the level of TFP. The effect of technology spillovers then should be captured by changes in the level of TFP observed at the firm (or plant) level, after controlling for the impact of other variables that may influence the firm’s productivity performance. Following this methodology, most empirical and technical studies attempt to test various hypotheses regarding the beneficial impact of FDI as a mode of technology transfer to the host economy46. The main results of 15 recent studies based on microeconomic data are summarised in Annex Table 5.1. A number of studies indicate efficiency gains as a result of technological spillovers from MNE affiliates to local firms in the same industry. These include Blomström and Persson (1983), based on Mexico’s manufacturing census data for 1970; Blomström and Sjöholm (1999), using 45
It is assumed that the difference in technological level between an affiliate and its parent firm is expressed as the ratio of their TFP levels. The cross-section regression equations were estimated separately for the textiles, chemicals, general machinery and electrical machinery industries, as well as for those industries combined, with industry dummies (see Urata and Kawai, 2000). 46 As mentioned above, Urata (1999) also used several qualitative technology indicators.
Indonesia’s industrial survey data for 1991; and Haskel et al. (2001) on the basis of the United Kingdom’s annual census data for 1973-92. Other studies suggest, however, that the effects of FDI have not always been beneficial for local firms. Haddad and Harrison (1993) find no positive results for Morocco in the late 1980s. Aitken and Harrison (1999) find a positive correlation between foreign presence and TFP growth in Venezuela, but conclude that this may be misleading if MNEs are attracted by sectors that are more productive in the first place. Okamoto’s (1999) empirical study on the US auto parts industry also finds only a modest degree of technology transfer from Japanese assemblers to US independent parts suppliers, and the author argues that the improvement in productivity observed in the 1980s and early 1990s may be a result of increased competitive pressures rather than technology transfer per se47. The diverse experiences of these and other countries suggest that the positive spillover effects of FDI are not automatic, but may be affected by various host-industry and host-country characteristics. Several of these characteristics have been studied and tested. One of the most prominent is the “technology gap” between MNE affiliates and local firms in the host country. According to this argument, spillovers should be easier to identify empirically when the technological attributes of local firms match those of the MNE affiliates. Kokko (1994) and Kokko et al. (1996) provide evidence for this hypothesis and find that, for Mexico and Uruguay, spillovers are difficult to identify in industries where foreign affiliates have much higher productivity levels than local firms. Furthermore, spillovers can be more easily recognised when foreign firms are not “self-contained enclaves”. Kokko et al. (1996) argue that a high technology gap combined with low competition prevents spillovers to the host economy. Given the level of local firms’ capabilities, one of the most critical issues regarding the transfer of technology is whether these technologies are appropriate for local firms and can enable them to compete effectively in the global market. Many studies have suggested that this is not always the case, and that firms will in fact have to make a variety of investments to benefit from technology inflows. The capability of host-country firms to “absorb” foreign technology appears to be an important determinant of the size of the realised spillovers. As noted earlier, a defining characteristic of joint ventures and firms affiliated with foreign MNEs, as
The issues of reverse causality (observed higher productivity in the MNE sector is the cause instead of the consequence of FDI) and omitted variables (higher productivity is due to unobserved factors, independent of FDI), as well as the econometric techniques designed to redress them, are discussed at length in Chapters Ⅲ and Ⅳ.
opposed to local firms with no foreign partners, is the relative importance attached by the former groups of firms to the provision of training programmes and the acquisition of new technologies (Djankov and Hoekman, 1999). In addition to the relative difference in the technological capabilities of the MNE and the local firm, the absolute level of absorptive capacity is also important. Keller (1996), for example, states that access to foreign technologies alone is not enough to increase growth rates if the country’s stock of human capital remains unchanged. Some evidence for this argument is provided by Perez (1998), whose study is based on firm-level data on the Italian manufacturing sector for the years 1989-91. His analysis shows that foreign presence is likely to have a positive effect on the productivity growth of domestic firms in specialist and scale-intensive sectors (e.g. electrical components, precision machinery, base chemicals, metal products and motor vehicles), while no similar evidence was found for sciencebased sectors, such as the pharmaceutical, fine chemicals, electronics and information technology industries. Citing the latter examples, Perez argues that strong foreign presence does not necessarily help domestic firms to develop in technologically more advanced sectors. More generally, many developing countries do not meet the human capital threshold required for recipient countries to benefit from technology spillovers through FDI (see Chapter Ⅲ). A recent study by Kokko et al. (2001) finds similar evidence for Uruguay, pointing to the importance of past experience in industrialisation as a pre-condition for international transfer of technology. This result may be interpreted as an indication that local firms in the host economy lack absorptive capacity.
Annex Table 5.1. FDI and Technology Transfer: Literature Survey Authors (year) Blomström and Persson (1983)
Data/coverage Mexico’s 1970 Census of Manufactures at the four-digit level (215 industries), broken down by ownership (private and foreign); a 15 per cent cut-off point is used for foreign ownership. State-owned enterprises are excluded. Panel data on Moroccan industrial firms during the 1985-89 period.
Estimation technique Cross-section, OLS regression analysis of the determinants of labour productivity in domestically owned private firms, with the foreign share of employment in each industry as an explanatory variable.
Major findings A positive and statistically significant coefficient was found for the relationship between labour productivity and the foreign share variable.
Panel data regression analysis of the impact of foreign affiliates’ asset share on productivity growth.
Mexico — essentially the same data base as that used by Blomström and Persson (1983).
Mexico — see Blomström and Persson (1983) and Kokko (1994).
Cross-section, OLS regression analysis of the determinants of labour productivity in domestically owned private firms, with focus on different market characteristics. 3SLS estimation of a simple simultaneous model of labour productivity in locally owned and foreign-owned firms.
No significant relationship found between higher productivity growth in domestic firms and greater foreign presence in the sector (partly due to the distortionary effects of import protection). The interactive term of large productivity gaps and large foreign market shares tends to exert a negative impact on productivity in local private firms. Labour productivities of local and foreign firms are determined simultaneously because of competition when suspected “enclave” industries are excluded from the sample. The coefficient of foreign presence (foreign output share measured at the four-digit industry level) is found to be positive and statistically significant only in the subsample with small technology gaps.
Kokko, Tansini Zejan (1996)
A plant-level survey of the Uruguayan manufacturing sector in 1988, involving 159 private locally owned plants.
Cross-section regression analysis of the impact of foreign presence on labour productivity in private locally owned plants
Annex Table 5.1. FDI and Technology Transfer: Literature Survey (continued) Authors (year) Blomström and Sjöholm (1999)
Aitken and (1999)
Djankov and Hoekman (1999)
Data/coverage 1991 survey data on over 13 000 Indonesian establishments with more than 20 employees, with indications about domestic and foreign ownership (majority or minority). Firm-level data on the Italian manufacturing sector for the years 1989-91, covering over 4 000 firms, both domestic and foreign.
Estimation technique Cross-section regression analysis of the impact of foreign ownership on labour productivity in Indonesian manufacturing and technological spillovers to local firms.
Major findings Foreign ownership is found to be a significant determinant of labour productivity in manufacturing. Positive intra-industry spillovers from FDI are found, but the degree of foreign ownership (minority or majority) has little impact on them.
OLS regression analysis of the impact of foreign presence (measured by the share of total employment) on the dispersion of labour productivity.
1980 and 1991 industrial survey data for Indonesia, with a sample of over 2 800 establishments having more than 20 employees for both years. Venezuela’s annual survey of industrial plants (more than 4 000 plants) over the 1976-89 period (except for 1980).
OLS regression analysis of the impact of foreign presence (measured by the share of gross output) on the growth of value added and labour productivity in local firms at the national, provincial and district levels. Panel regression analysis of the impact of foreign ownership on TFP and output at both plant and sector levels.
Firm-level survey of Czech enterprises for 1992-96 with a full sample of over 500 firms, both domestic and foreign (including joint ventures).
Panel data regression analysis of the impact of foreign presence (measured by the share of assets) on TFP growth of firms with different types of ownership.
Foreign presence is found to have a statistically significant impact on productivity improvements of domestic firms in both specialist and scale-intensive sectors, but no similar evidence is found for sciencebased sectors.６ Intra-industry spillovers are found to be statistically significant at the national level. The hypothesis that geographical proximity increases spillovers is not supported by the results of regression analysis. Small plants (employing fewer than 50 workers) with higher foreign ownership tend to exhibit positive productivity gains. However, foreign ownership has a negative effect on the productivity of domestic firms in the same industry. FDI is found to have a positive impact on TFP growth of recipient firms. Joint ventures and FDI appear to have a negative spillover effect on firms that do not have foreign partnerships.
Annex Table 5.1. FDI and Technology Transfer: Literature Survey (continued) Authors (year) Okamoto (1999)
Data/coverage Plant-level production data (1992) on the US auto parts industry, including samples of Japanese and US independent firms, joint ventures and other manufacturers. 1990 and 1995 industrial survey data for Indonesia with a sample of over 16 000 establishments having more than 20 employees, both foreign and domestic.
Estimation technique Cross-section regression analysis of the impact of business ties to Japanese auto assemblers on TFP growth rates in US independent suppliers. Decomposition analysis of aggregate TFP growth rates between 1990 and 1995 into several factors, calculated at ISIC three-digit level.
Urata and Kawai (2000)
Firm-level data compiled from a MITI survey in 1993, including 266 parent firms and 744 overseas affiliates in four industries (textiles, chemicals, general machinery and electrical machinery).
Haskel, Pereira Slaughter (2001)
Kokko, Zejan Tansini (2001)
Plant-level panel data for all UK manufacturing covering 1973-92, based on the country’s Annual Census of Production. Firm-level data for the Uruguayan manufacturing sector in 1988, including 126 foreign-owned firms, using a 10 per cent cut-off point for foreign ownership.
Cross-section regression analysis of the determinants of intra-firm technology transfer, measured by the ratio of the TFP level of an overseas affiliate to that of its parent firm, with both firm-specific and hostcountry characteristics included as explanatory variables. Panel regression analysis of the impact of foreign presence (measured by the share of total employment) on TFP growth of manufacturing plants. Cross-section regression analysis of the determinants of labour productivity in domestically owned firms, with the foreign share of total output at the four-digit industry level as an explanatory variable. Probit regression analysis of the determinants of domestic exports at the firm level, with the foreign presence variable included on the right-hand side of the equation.
Okamoto and Sjöholm (1999)
Major findings Only a modest degree of technology transfer from Japanese assemblers to US independent suppliers between 1982 and 1992 was found. The contribution of foreign plants to aggregate TFP growth exceeds their output share. Foreign plants have a positive fixed effect, which may be due to their access to parent firms’ technology. The educational level and past experience in industrial activity in the host country tend to have a positive effect on intra-firm technology transfer. However, a technology transfer requirement imposed by the host country does not yield the expected outcome. Productivity spillovers in the same industry are found to be positive and statistically significant (but regional spillovers are not). Foreign firms established before 1973 tend to generate positive spillovers to labour productivity of local firms, while the impact of foreign firms established after 1973 is the opposite. The impact of foreign firms established after 1973 on exports of local firms is found to be positive and highly significant, while there is no sign of export spillovers from foreign firms established before 1973.
Chapter Ⅵ FDI, Privatisation and Corporate Governance
Cross-border mergers and acquisitions (M&As), as opposed to greenfield investment, have played a central role in the recent surge in global FDI flows (see Chapter Ⅱ). Privatisation of state-owned enterprises accounts for a large share of M&As in many developing and transition economies, notably in Latin America, Eastern Europe and the former USSR. Privatisation has been a core component of the structural reform measures these economies have undertaken to reduce budget deficits, consolidate public finances and rebalance government resources to their core functions. In the 1990s, divestiture of state enterprises reached public utility sectors, including power and telecommunication services that had previously been considered natural monopolies. Since the 1997-98 financial crisis, a number of Asian countries have begun to embrace privatisation and deregulation policies in order to revive their economies. However, privatisation programmes, particularly those involving public utilities, have raised public concern and even given rise to protests in some parts of the region. California’s power crisis in 2001, although deregulation per se was not its primary cause, has also drawn attention to the potential risks of reforming public utilities (see Box 6.1). This incident appears to have delayed plans to deregulate monopoly electricity companies in Korea, Malaysia and other countries in Asia48. Moreover, foreign participation in national privatisation programmes is a politically sensitive issue in many Asian countries. As foreign participation is nonetheless a key instrument for achieving privatisation goals and improving corporate governance in newly privatised enterprises, and since empirical analysis and informed policy discussion concerning the relationship between privatisation and FDI in Asia remain limited, recent experiences in Latin America (Brazil in particular) as well as in a number of OECD countries can provide useful lessons to Asian policymakers. This chapter attempts to synthesise the materials available and discusses some policy issues. It begins with a brief overview of privatisation trends in the 1990s, followed by discussion of recent experiences in Latin America, notably in Brazil. The last sections present an assessment of privatisation policy and its implications for corporate governance.
See International Herald Tribune, 8 February 2001.
Box 6.1. The Power Crisis in California: What Went Wrong? “[W]hile the deregulation was indeed flawed, the flaws did not cause the catastrophe” (Paul Krugman, The New York Times, 20 February 2001). In 1996, California became the first state in the United States to deregulate its electricity market. According to the State Assembly bill, the move was supposed to create a “market structure that provides competitive, low cost and reliable electric service”. Before deregulation, the state’s electric sector was a government-regulated monopoly. Under deregulation, investor-owned utility companies, such as Southern California Edison and Pacific Gas and Electric, sold most of their power-generating plants to other (unregulated) private companies and became buyers of wholesale electric power. They were also required to transfer operational control of transmission lines and power grids to a private non-profit organisation, called the Independent System Operator. As a consequence, they retained ownership and control only of the distribution systems supplying electricity to homes and businesses. These utility companies must now buy their power at prices that are set daily at auction by the California Power Exchange (PX), which is overseen by the Federal Energy Regulatory Commission. What went wrong, then? As Krugman noted above, the design of deregulation was flawed. First, while the wholesale price was set by the market, the retail price continued to be regulated. Deregulation prevented the utility companies from passing on price increases to their consumers until at least 31 March 2002. Second, under deregulation, utilities were not permitted to negotiate long-term contracts with power generators. The rigidity of retail prices thus made it harder for the state to cope with the power crisis that loomed during 2000. Wholesale prices in 1998-99 were estimated at an average of 16 per cent above marginal cost. In 2000, however, prices skyrocketed a further 500 per cent (McMillan, 2001, p. 8). Unable to recoup these higher costs from their customers, the utility companies rapidly ran out of money, with the two largest companies claiming that their combined losses exceeded $9 billion at the end of 2000. The primary reason for California’s power crisis, however, predated deregulation: robust economic growth considerably raised the state’s demand for power, while generation capacity actually declined by 2 per cent between 1990 and 1999 (ibid., p. 6). Deregulation simply helped to reveal this fundamental problem of demand-supply mismatch. In short, California’s power crisis makes it necessary for policymakers to take a hard look at where prices should be to reflect electricity’s true cost and to encourage energy conservation by end-users. It also offers an expensive lesson to regulators: the problem of damaging price hikes or supply interruptions in deregulated markets will have to be taken seriously.
6.1 Privatisation Trends in the 1990s Over the last two decades, privatisation has become part and parcel of structural reform programmes in many countries, both OECD and non-OECD. The initially timid progress recorded in some pioneering countries (United Kingdom, Chile, New Zealand) has been matched and sometimes surpassed by other countries that have joined the privatisation bandwagon, first in the developing world (Argentina, Mexico, Malaysia), then in Europe (France, Italy), and finally in
transition economies49. In the 1990s alone, close to $1 trillion worth of state-owned enterprises have been transferred to the private sector world-wide (OECD, 2001b, p. 43). In dollar terms, global privatisation proceeds grew by 18 per cent per annum between 1990 and 1999, with temporary drops recorded in 1992 (the European currency crisis) and again in 1998 (financial crises in East Asia and Russia). The wave of privatisation is driven by a host of factors, including the need to reduce budget deficits, attract FDI flows, improve corporate efficiency, and liberalise and deregulate power, telecommunication and other service markets. In Europe, following the signing of the Maastrich Treaty in 1992, privatisation activities accelerated in the second half of the 1990s as countries tried to reach the goals set for the creation of the European Monetary Union (EMU). In 2000, however, global privatisation proceeds fell to $100 billion, nearly 30 per cent below the 1999 level. Much of this drop was due to reduced privatisation activity in the OECD countries, partly reflecting the fact that few state assets still remain to be sold in competitive sectors (see Figure 6.1). Figure 6.1. Global Trends in Privatisation Revenues 180
Source: OECD (2001e).
In non-OECD countries, more than half of total cumulative privatisation 49 Possibly the first large-scale sale of a public enterprise was the Volkswagen issue in the early 1960s, but the dramatic fall of the share price after markets were hit by the Cuban missile crisis had a lasting negative impact on equity ownership culture in Germany (see Goldstein, 2001).
revenues for the 1990-99 period accrued to one region, Latin America, followed by Eastern Europe and Central Asia. These regions taken together accounted for nearly 80 per cent of total privatisation revenues in the 1990s, while the share of East Asia (excluding Korea) was only 14 per cent (Figure 6.2). Moreover, privatisation in non-OECD countries was heavily concentrated in infrastructure projects. Three sectors — infrastructure, financial services and other services — accounted for almost two-thirds of total privatisation revenues during this period (Figure 6.3). Figure 6.2. Privatisation Revenues by Major Non-OECD Region, 1990-99 9
Latin America East Asia
Europe & Cen. Asia Other regions
Source: World Bank (2000 and 2001).
Figure 6.3. Privatisation Revenues in Non-OECD Regions by Major Sector, 1990-99 4 12
Infrastracture Industry Primary sector Financial services Other services 16
Source: see Figure 6.2.
Privatisation has also served as a major channel for attracting FDI flows to Latin America. The aggregate value of privatisation revenues between 1990 and 1999 reached $146 billion, which is equivalent to 45 per cent of total FDI flows into the region during the same period (Figure 6.4). In contrast, privatisation in East Asia has been rather limited from a quantitative point of view, though policies in this domain have existed since the 1980s in Korea and several ASEAN countries50. Most FDI inflows to the region have taken the form of greenfield investment, as privatisation has attracted relatively little FDI (Figure 6.5). Such regional differences in the relationship between privatisation and FDI deserve further discussion from the political-economy point of view. Figure 6.4. Privatisation and FDI Inflows in Latin America 90 80 70
60 50 40 30 20 10 0 1989
Source: see Figures 2.3 and 6.2.
See Fukasaku (2001) for a detailed analysis of privatisation trends in East Asia.
Figure 6.5. Privatisation and FDI Inflows in East Asia 90
Source: see Figures 2.3 and 6.2.
6.2 Why is Latin America More Advanced in Privatisation? Privatisation is as much a political as an economic process, requiring the creation of supportive coalitions against strong vested interests. Privatisation is also a typical example of a “collective action” situation in which the actors likely to gain from it are widely dispersed. It is thus politically easier for beneficiaries of the status quo to organise opposition than it is for future beneficiaries to create a support group. Moreover, in developing countries with underdeveloped capital markets, the appeal of privatisation through the public offering of state-owned enterprises would be rather limited in terms of coalition building in the political process. Experiences in the 1990s suggest that the intrinsic allure of free-market economics has proven less important a factor for the success of privatisation programmes than political conjunctures (Armijo, 1998). The nature and gravity of economic crises have deeply influenced both the perceived need for radical political changes and the central themes of the political debate. The more severe the government’s fiscal deficits and the greater the contribution of state-owned enterprises to these deficits, the quicker the government resorts to divestiture to
solve its structural problems. Active participation by foreign investors can be seen as key to successful privatisation programmes, not only for the obvious reason that it enlarges the pool of potential buyers, but also because foreign investors may provide access to modern technologies and management methods and add pressure for “getting regulation right” in the newly privatised sectors. Another important issue is the choice of firms to be sold and the method of sale. It is vital that the economic goals to be met through divestiture are understood, in particular the choice between an immediate reduction of macroeconomic imbalances and a medium-term improvement in corporate efficiency. The least productive enterprises are hard to dispose of, which may make it necessary to concede extremely easy terms in addition to a low price, unless of course the government decides to eliminate such enterprises altogether. At the same time, countries in dire economic straits need quick “leading cases” to build their reputations in the eyes of foreign investors. Thus, governments have often chosen to start with the best managed and more profitable enterprises, such as those in the telecommunications industry, where state assets are expected to be more attractive to international investors. Unfortunately, such industries present technological and tariff complexities, requiring sophisticated tools for asset valuation and regulation that few developing countries possess. 6.3 Privatisation: Brazilian Style During the 1950s and 1960s, successive Brazilian governments used stateowned enterprises (SOEs) as an indirect form of public management to accelerate the country’s industrialisation and infrastructure development. The adverse economic situation — domestic and external — after the first oil shock made it necessary to re-orient this policy, and privatisation thus entered the agenda. For a variety of economic and political reasons, however, it was not until the early 1990s that the government made privatisation one of its top priorities for structural reform. Soon after taking office in March 1990, the Collor government sent to Congress Provisional Measure 155, later converted into Law 8031, which created the National Privatisation Programme (PND) and established most of the rules and procedures governing it today. The PND focused on selling large public enterprises in the industrial sectors (e.g. steel, petrochemicals and fertiliser) that had been considered strategic in the development model of the past. The sale that best exemplified this change was probably that of Usiminas, an integrated steel company, in October 1991. 77
In September 1992, President Collor was impeached and replaced by Vice President Itamar Franco (now governor of Minas Gerais). The Franco government sought to change the institutional framework in order to place more emphasis on fiscal considerations and to widen participation in the PND to include foreign investors. The fiscal impact of privatisation was negligible at that stage, however, simply because the PND was a small-scale programme in macroeconomic terms. The most pressing problem for the Cardoso government which took office in 1995 was — and still is — the weakness of public finances. President Cardoso stated that “[w]ith the Federal Treasury carrying, directly and indirectly, debts of more than $200 billion, it is doubtful that the government will be able to undertake large public investments over the medium term” (Cardoso, 1995, p. 10). Under these circumstances, the government sought to strengthen the institutional side of the privatisation programme by creating the cabinet-level National Council for Privatisation (CND), which is directly accountable to the President; to broaden the scope of privatisation by launching privatisation programmes at the level of the states and extending the PND to public utilities and mining; and to increase cash revenues and investment commitments in the sectors to be privatised. This policy initiative led to the enactment of the Concessions Law (Law 8987) in February 1995 and to various constitutional amendments later that year. By breaking public monopolies and granting concessions to the private sector, this law paved the way for privatisation in a number of sectors: telecommunications, electricity and gas, railways, highway construction, ports and airports, water supply and basic sanitation services. As a result, the Brazilian privatisation programme had generated revenues of close to $90 billion as of March 1999, one of the highest totals in the world. At the same time, the speed and scope of privatisation have raised some concern over the uneven development of regulatory regimes across sectors. Establishing a modern regulatory system (adequate, transparent and stable rules for private operation) remains a key challenge in Brazil. New regulations to promote market competition were put in place in the telecommunication sector before the sale of Telebrás, the former national monopoly, in July 1998. The changes have been impressive. A long-standing state monopoly regulated by a ministry has been converted into a competitive private industry, with the participation of both domestic and foreign investors. In the power sector, however, regulatory reform has been introduced only gradually. This reflects the complexity of this sector’s institutional and regulatory 78
environment: the federal government owned the power generation and transmission assets, while state governments owned the distribution companies as well as some vertically integrated utilities. The combination of ownership problems and debt in many state-owned utilities has prevented privatisation from being implemented as a one-shot, all-inclusive deal. A key challenge for this sector is to create a regulatory and market environment that leads private investors to make the long-term investments needed to meet the country’s growing demand for electricity51. 6.4 An Assessment of Privatisation Policy There is a general consensus that, in OECD countries, ownership changes have considerably improved firm-level performance in terms of productivity and profitability (see e.g. Meggison and Netter, 2001). Privatisation has also helped to increase the depth and liquidity of equity markets, although early observers of the British experience argued that these goals should be achieved through instruments other than the sale of state enterprises at a considerable discount (Jenkinson and Mayer, 1988). Developing economies have recorded significant increases in profitability, operating efficiency, capital spending, output and employment, and these increases are usually greater in countries with higher per capita income (Boubakri and Cosset, 1998)52. Overall, privatisation has increased consumer welfare, but the degree of economic success depends crucially on the post-privatisation market structure shaped by the regulation that is put in place. The substitution of private for public ownership has been accompanied by the emergence of new regulatory challenges for policymakers. First, in the natural monopolies that previously formed the core of the public enterprise sector (telecommunications, power, water, transport services), it is important to prevent the new private owners from simply pocketing monopoly rents. The policy issues and problems facing Brazil’s public utility sectors are widely shared by OECD Member countries. Indeed, sales of public utilities accounted for almost 70 per cent of total privatisation revenues realised by
For a full treatment of the Brazilian privatisation programme, see Goldstein (1999) and Fukasaku and Pineiro (1999). 52 In the case of transition economies, empirical tests of the relationship between enterprise performance and ownership generally refute the hypothesis that privatisation per se is associated with improved performance (e.g. Estrin and Rosevaer, 1999). The belief that mass privatisation would unleash entrepreneurial energies by providing powerful incentives for efficient restructuring has also proved naive. The chances of fostering entrepreneurship appear to be greater in a gradualist environment permitting negotiated solutions to restructuring as opposed to market-driven reforms (Spicer et al., 2000).
OECD countries from 1993 to 1998. As in Brazil, OECD governments have sought to achieve multiple objectives — economic, political and financial — in pursuing their privatisation programmes. Weaknesses in the regulatory framework have sometimes reduced the benefits of privatisation and deregulation, particularly in the electricity industry (Gonenc et al., 2001). Second, there is broad consensus that public authorities must devise sectoral policies that introduce and maintain competition, while at the same time establishing a sound regulatory framework in remaining monopolies. Public authorities must also ensure that transactions are transparent; convince investors that their investments are secure; and negotiate, monitor and enforce contracts with private suppliers of management and financing. In addition, they need to ensure that resources from privatisation sales are put to productive uses, and to manage the political and social tensions that inevitably arise as enterprise reforms are implemented, especially the critical issues of foreign ownership and labour layoffs. There is much less agreement, however, on how countries that need to consolidate their initial reforms should approach the next set of challenges (known as “second-generation” issues). In general, these issues are related to postprivatisation disputes and renegotiations between governments and the private sector and to the mechanisms necessary to promote competition and investment in the reformed industries53. 6.5
Implications for Corporate Governance Corporate governance is a burgeoning subject in the new institutional
economics concerned with the mechanisms whereby economic systems (in their broadest possible sense) cope with the information and incentive problems inherent in financing investments. In developing countries, corporate ownership is highly concentrated in the hands of either a small number of families or the state54. It is also difficult to monitor managers because well-connected firms, and a fortiori SOEs, do not face a credible bankruptcy threat. Privatisation and corporate governance are linked in two main ways. First, the sale of SOEs exposes them to take-over and bankruptcy threats, thereby easing 53
See OECD (2001d) for further discussion. See Claessens et al. (1999) for a detailed discussion of the structure of corporate control in East Asia. See also Oman (2001) for country case studies on corporate governance (Argentina, Brazil, Chile, China, India, Malaysia and South Africa). 54
the corporate governance problems proper of public ownership. Second, privatisation provides an opportunity to modify the distribution of ownership rights among different classes of investors, by extending public listing among large firms, increasing the number of small shareholders and reducing ownership concentration. In view of the goal of creating a “people’s capitalism”, it is thus important to reconcile two opposing imperatives: providing an appropriate mechanism for protecting small shareholders, while at the same time allowing management the flexibility required to pursue long-term corporate goals. The potential improvements in technical efficiency following transfer of control may be jeopardised if corporate control is not contestable: in this case, and especially if conduct regulation proves insufficient to open up protected markets, managers can exploit rents accruing from market position without having to worry about the threat of take-overs. Privatisation policies should take such factors into account, making it imperative to introduce reforms and redress perceived inefficiencies. While creating “people’s capitalism” has always ranked high among governments’ goals, tackling the issues that remain after ownership is transferred from public to private hands and when no (absolute) majority shareowner emerges has seldom been an overriding concern. Governments have great discretion in pricing the state-owned enterprises they sell, especially those being sold via public share offering, and they use this discretion to pursue political and economic ends. Most experiences, however, suggest that privatisation has only limited power to change the modes of governance prevailing in each country’s large private companies. In the United Kingdom, a country whose privatisation policies are often referred to as a benchmark, “control [of privatised companies] is not exerted in the forms of threats of take-over or bankruptcy; nor has it for the most part come from direct investor intervention” (Bishop et al., 1994, p. 11)55. The slow but steady decline in the number of small shareholders, after the steep rise in the immediate aftermath of privatisation, highlights the difficulty of sustaining people’s capitalism in the longer run. In Italy, privatisation was accompanied by a legislative effort aimed at providing non-controlling shareholders (both individual and collective investors) with more adequate safeguards and at creating conditions that allow them to monitor managers. Successive governments, however, were unsuccessful in broadening the number of large private business groups, even though enhancing the mobility of control to investors outside of the traditional core 55
Following three successive fatal accidents (19 September 1997, 5 October 1999 and 17 October 2000), Britain’s rail infrastructure operator, Railtrack, was declared bankrupt on 7 October 2001 and put into receivership. This incident suggests that financial deregulation has pushed banks into giving up their role as rescuers of sick but socially important public service companies.
of Italian capitalism was explicitly included among the authorities’ strategic goals. Those countries which have chosen the mass (voucher) privatisation route have done so largely out of necessity and face ongoing efficiency problems as a result. More successful experiences, such as those of the United Kingdom and Chile, indicate that mass sell-offs require the development of new institutional investors, such as pension funds, which may later play an active role in corporate governance56. In sum, although the promise of privatisation has frequently been oversold, not least by many international organisations, its ills have also been greatly exaggerated. When ownership transfer is accompanied by market liberalisation and proper implementation of the new regulatory system, consumers and endusers in OECD and non-OECD countries alike have benefited in terms of wider choice, higher quality and lower prices. Privatisation has also opened the door for foreign investors, notably in Latin America. Regulatory agencies still face substantial challenges, however, if they are to ensure that the maximum benefits are reaped from privatisation.
In Chile, for example, the take-over of the country’s dominant electricity utility, Enersis, one of the largest in emerging markets, was stalled for some months in 1998 as pension funds objected to lucrative additional terms that managers had negotiated for themselves on the basis of important agreements concerning the future strategic direction of Enersis — agreements about which they never informed the other shareholders.
Host-government Policies for Attracting FDI
In recent years, host-country governments throughout the world have played an active role in attracting FDI. Many see FDI as vital to reinvigorate their domestic economies and strengthen their ability to compete in global markets. A consequence of this trend has been intensified competition for FDI by both national and sub-national governments. Yet, while host-government policies may substantially influence the magnitude and quality of FDI inflows, the relative role of different policy measures remains poorly understood. Moreover, developing countries have expressed some concern in the context of the multilateral trading system under the aegis of the WTO that the ability of host governments to regulate the pattern of FDI inflows may have been significantly curtailed. It is thus important to have more informed discussion about the role of host-government policies for attracting FDI. This chapter aims to fill this gap by reviewing foreign investment regimes in selected host economies and discussing their implications for development policy. Before doing so, however, a brief discussion of the typology of host-government policies is in order. 7.1 Incentive-based and Rules-based Measures The foreign investment regime in a host economy comprises a wide variety of government measures aimed at regulating and attracting foreign investment. Table 7.1 lists the main types of regulatory and incentive measures applied by host governments. Conceptually, it is important to distinguish two broad categories of measures, “incentive-based” and “rules-based” measures, though in practice this distinction is not always self-evident57.
A case in point is the establishment of export-processing zones or special economic zones (e.g. in China). This requires host governments to set rules for firms (both foreign and domestic) operating in such zones, with legislation distinct from that governing the rest of the economy and often combined with certain incentive measures.
Table 7.1. Foreign Investment Regime in the Host Economy: Main Types of Regulatory and Incentive Measures Types of measures 1.Screening, admission and establishment
Examples Closure of certain sectors, industries or activities to FDI Minimum capital requirements Restrictions on modes of entry Admission to privatisation bidding procedures Establishment of special zones (e.g. EPZs) for FDI with legislation distinct from that governing the rest of the country 2.Fiscal incentives • Reduction in standard corporate income tax rate • Tax holidays • Reduction in social security contributions • Accelerated depreciation allowances • Duty exemptions and drawbacks • Export tax exemptions • Reduced taxes for expatriates 3.Financial incentives • Investment grants • Subsidised credits • Credit guarantees 4.Other incentives • Subsidised service fees (electricity, water, telecommunications, transportation, etc.) • Subsidised designated infrastructure (e.g. commercial buildings) • Preferential access to government contracts • Closure of the market to further entry or granting of monopoly rights • Protection from import competition 5.Performance • Local content (value added) requirements • Minimum export shares • Trade balancing • Technology transfer • Local equity participation • Employment targets • R&D requirements Sources: OECD (2001c, pp. 5-6), UNCTAD (1996, pp. 176-181), WTO (1998, Table 2). • • • • •
Incentive-based measures provide for fiscal, financial and other incentives to foreign firms. Common fiscal incentives include a reduction in the standard income tax rate for a particular category of foreign investors, tax holidays, accelerated depreciation allowances, duty exemptions and drawbacks, and export tax exemptions. Among the most important financial incentives are direct investment grants, subsidised credits and credit guarantees. In addition, “other incentives” are often used to provide foreign investors with privileged access to certain types of infrastructure services. Incentive-based measures may be either granted automatically or targeted to specific purposes, with conditions relating to certain pre-determined performance criteria. Rules-based measures are a much broader category of government measures that are applied to regulate directly or indirectly the scope and magnitude of foreign business activity in a host economy. Examples include rules on market entry and rights of establishment (including protection from market competition), protection of property rights (including intellectual property rights), the
establishment of various types of “zones” within an economy, participation in privatisation programmes, rules and procedures regarding dispute settlement, and domestic regulations on environmental protection and workers’ rights. Where environmental protection and workers’ rights are concerned, the use of government regulations and legal standards as a means of attracting FDI has generated strong concern in both home and host countries. It is often claimed that in competing to attract FDI, host governments may overtly or covertly relax the enforcement of such standards, thereby putting pressure on other governments to follow suit. Although there is little empirical evidence of the so-called “race to the bottom” in either environmental or labour standards (see Box 7.1), this debate has enhanced public awareness about host-country policies for attracting FDI58. The next section begins by reviewing available materials (mostly obtained through the Internet) with respect to foreign investment regimes in ten selected Asian and Latin American economies. This is followed by discussions on some controversial issues regarding the use of incentive measures to attract FDI inflows. Much of the analysis and discussion presented in this section draws heavily on the results of recent studies conducted at the OECD Development Centre and elsewhere59.
58 The case of export-processing zones( EPZs) has attracted particular attention in this debate, because the phenomenal growth of EPZs since the 1970s is seen by some as evidence of deliberately lowering of labour standards. A recent study by the ILO (1998) finds that adequate labour standards and a sound system of labour-management relations are indeed widely lacking in EPZs, but it also leaves little doubt about the dynamics of change that market forces are imposing on firms that operate in EPZs: “Today, globalisation places the emphasis on speed, efficiency and quality as well as cost … shifting the focus from cheap labour to productive labour. For countries to remain competitive, they must get this mix of cost and quality factors right by raising the capacity of their human resources, ensuring stable labour relations, and improving the working and living conditions of zone workers.” The study further notes that “countries which have established trade union presence in the zones do not appear to have suffered any loss of investment”, and that “none of the enterprises interviewed [for the ILO study] stated that a lack of worker organisation was an incentive to invest” (ibid., p. 12). 59 See inter alia Moran (1998), OECD (1998), Oman (2000) and UNCTAD (1999 and 2001).
Box 7.1. Policy Competition for FDI: A Race to the Bottom? The environmental and social dimensions of foreign investment have recently become a matter of intense controversy between certain home and host countries. The concern is that governments of OECD and developing countries alike, caught in a prisoner’s dilemma, might find it difficult to resist competitive pressures, causing a “race to the bottom” in environmental and labour standards. There is, however, little empirical evidence supporting this view. Environmental Standards Recent studies have found little evidence that US firms, or other OECD-based firms, have moved production to take advantage of lower environmental standards in another country 60. A main reason is that the costs of complying with anti-pollution laws have proved relatively modest on the whole, and any attractiveness that weak environmental rules may have is simply swamped, in most cases, by other factors. It is often more efficient to adhere to a single set of environmental practices world-wide than to scale back practices at a single location. The high local visibility of large multinational investors can make them particularly attractive targets for local enforcement officials. The memory of such events as the Bhopal disaster (India) in 1984 and the ensuing problems faced by Union Carbide have heightened many investors’ awareness of their potential environmental liabilities when they invest abroad. Moreover, competition for FDI in knowledge-intensive activities can create upward pressure on environmental standards. This reflects the growing desire for corporate managers to locate these activities in communities where their productivity levels and competitiveness benefit from high standards of environmental protection. The spillover benefits to firms of “clustering” and “agglomeration economies” tend in turn to attract other corporate investors, thus amplifying the attractiveness of communities that set and enforce relatively high standards of environmental protection. Core Labour Standards 61 Data gathered by the OECD on freedom-of-association rights (arguably the most important core labour standard, along with the right to bargain collectively) reveal a marked difference between OECD and non-OECD countries. However, these data do not support the hypothesis of a race to the bottom in core labour standards, because they show no significant deterioration of freedom-of-association rights in developing countries (or OECD countries) over the last 20 years. On the contrary, they show that the move to democracy in developing countries has been accompanied in several — notably in Latin America — by some improvement in the protection of workers’ right to associate. Furthermore, some recent studies show that since 1980, countries with high labour standards have maintained or increased their share of global FDI inflows (notwithstanding the importance of flows to China, a low-standards country). Similarly, low-standards countries have not increased their share of global exports. Two-thirds of 39 countries with low labour standards have seen their international competitiveness stagnate or decline, as measured by unit labour costs, while 14 of 18 high-standards countries have increased their international competitiveness 62.
See Oman (2000) for a detailed discussion. Core labour standards are defined as workers’ right to associate, i.e. to form independent unions of their choice, and to bargain collectively; the prohibition of forced labour and of exploitative child labour; and non-discrimination in employment. 62 See Raynauld and Vidal (1998). See also OECD (2000) for further discussion. 61
7.2 Some Observations from Case Studies Annex Tables 7.1 and 7.2 present a snapshot of the legal and policy framework currently applied to foreign investors in ten Asian and Latin American host economies: Argentina, Brazil, Chile, China, Chinese Taipei, Colombia, India, Indonesia, Malaysia and Mexico. The countries were selected on the basis of two criteria: the accessibility of legal materials and policy documents (including articles and policy analyses by consulting firms and newspapers) through the Internet, and the goal of presenting broadly comparable pictures of foreign investment regimes in the two regions. Each annex table lists key regulatory and incentive measures in three major areas of interest to foreign investors: (1) openness to FDI (measured in terms of screening procedures for entry and establishment of foreign firms, national treatment, existence and scope of negative lists and performance requirements), (2) fiscal and other incentives, and (3) protection of investments, notably protection of intellectual property rights and dispute resolution procedures. This review does not present an exhaustive list of regulatory and incentive measures maintained by the host government. Rather, it should be seen as illustrative, providing a broad-brushed picture of the regulatory environment facing foreign investors today. It should be emphasised at the outset that despite recent initiatives undertaken by many developing countries to liberalise their foreign investment regimes, there remain considerable differences across regions and countries. Several issues deserve particular attention. First, the creation of a “level playing field” for domestic and foreign firms is widely regarded as one of the pre-conditions for attracting FDI. In terms of screening procedures for entry and the right of establishment, as well as the granting of national treatment, it appears that Argentina, Chile and other Latin American countries have instituted a more liberal legal framework than their Asian counterparts. Although the latter have gradually streamlined registration procedures and introduced automatic approval for FDI, many countries in the region, such as China, India and Indonesia, have maintained considerable control over foreign investment at the entry phase and retained lengthy negative lists. In contrast, Chinese Taipei and, more recently, Malaysia offer a more liberal environment. Ceilings on foreign equity, negative lists and performance requirements such as local content requirements are being gradually phased out. Notwithstanding the considerable progress made in the aftermath of the 1997-98 financial crisis and the commitment to creating the ASEAN Investment Area (AIA), most Asian countries remain reluctant to liberalise their foreign investment 87
regimes fully. Second,
competitiveness has generated renewed interest in human resource development and skill formation by host governments in the context of FDI policy. In this respect, many of the countries under review regulate the hiring of foreign workers and impose training requirements on foreign investors. For instance, Malaysia, a country that has no legal minimum wage, has provided various incentive schemes to promote technical and vocational training. In particular, the government has established a Human Resource Development Fund to help finance the education and re-training of workers, although incentives are linked to certain performance requirements. Foreign companies are allowed to bring in the required personnel in areas where there is a shortage of trained Malaysians to do the job, but only for a certain period, and subject to the condition that Malaysians are trained to take over the posts eventually63. Export requirements are still in place in most special zones, but apply to domestic as well as to foreign investors. Third, protection of intellectual property rights (IPRs), which has a strong influence on the magnitude and quality of technology transfer, is another key characteristic of the foreign investment regime in the host economy. Foreign investors often request host governments to strengthen the legal framework for IPRs, and this issue is particularly sensitive for trade in electronic and pharmaceutical products, because these products are easily pirated. The scope and strength of IPR protection differ significantly among host economies. In India’s pharmaceutical industry, for example, patents apply to the manufacturing process but not to the products themselves. India will have to bring its patent laws regarding the pharmaceutical sector into conformity with the WTO’s TRIPS Agreement by 2005. Almost all of the countries under review have ratified major international agreements and conventions regarding protection of IPRs, and national legislation has been amended and improved accordingly. With respect to enforcement, however, there is still much room for improvement. Fourth, regulations on tax treatment at the national and international levels have become highly complex in many host economies, such as China. The rentseeking and even corrupt practices of local officials in the provision of various tax incentives have given rise to some concern (see, for example, Oman, 2000, pp. 5054, and the studies cited therein). Some mechanisms adopted to counter this
See US State Department, Bureau of Economic and Business Affairs, “Malaysia Country Commercial Guide 2000”, Chapter 7.
tendency are the requirement that enterprises appoint accounting firms (e.g. in India) to deal with tax matters and computerise records of imported inputs and export sales. In addition, the legal requirement to preserve documents related to transfer pricing will have a positive impact on the tax system on the whole. For instance, India’s Finance Bill for 2001 has made it mandatory for persons entering into international transactions to keep information and documents. In Mexico, documentation is required for all taxpayers, and compliance with the documentation requirement entitles the taxpayer to a 50 per cent reduction in penalties64. Finally, various forms of tax incentives have been widely used as a means of attracting FDI in almost all the host economies under review, notably those in Asia. In 1999, Chinese Taipei (a net capital exporter in East Asia) introduced a comprehensive tax incentive scheme to attract FDI65. Where financial incentives are concerned, however, there is little information available in the public domain. The cost of incentive measures will be discussed in more detail in the next section. 7.3
The Cost of Investment Incentives
Economists have long argued that governments’ use of discretionary, targeted fiscal and financial subsidies to attract investors is ineffective. An abundance of studies, using both investor surveys and econometric methods, support this view66. The studies find that the overwhelming majority of major investors base their decisions concerning investment location much more on a site’s economic and political “fundamentals” than on government subsidies. Among the key fundamentals are the size and perceived growth potential of the market that a site is well placed to serve, the likely long-term political and macroeconomic stability of the site’s location, an adequate supply of productive, trainable workers, and the availability of modern transportation and communications infrastructure. All of this research and advice from economists seems, however, to have little influence on the behaviour of the politicians and government officials whose job it is to attract corporate investors. These people keep launching new incentive programmes, often argue vehemently that they cannot do their jobs without such measures and seem largely to ignore the arguments of economists. Why?
“Transfer Pricing in Mexico”, PriceWaterhouseCoopers, www.pwcglobal.com/extweb/indissues. Asia Pacific Tax Notes, No. 10, July 1999, PriceWaterhouseCoopers, p. 18. 66 See, for example, Reuber et al. (1973), Lim (1983), UNCTC (1992), Shah (1992), Rolfe et al. (1993) and Pirnia (1996). 65
The reason is that studies showing that incentives are ineffective have largely failed to take adequate account of the fact that an investor’s choice of location, especially for a major new investment project (the kind governments most want to attract), usually involves a two-stage, or multi-stage, decision-making process. It is standard practice for an investor first to draw up a “short list” of sites that satisfy its “fundamentals test” — i.e. that meet or surpass the investor’s minimum criteria with respect to fundamentals, irrespective of potential host governments’ willingness to provide incentives — and then, in a second or later stage, to consider the availability of incentives in making the final site selection. Thus, although investors attach much greater overall importance to the fundamentals than to the availability of subsidies, discretionary incentives can make a difference — even a decisive difference — in an investor’s final choice of investment location. It is common, in fact, for investors to seek actively to play governments off against each another to bid up the value of incentives, once the competing sites have passed investors’ fundamentals tests and have been shortlisted as good potential sites. While economists rightly claim that investors attach much more importance to fundamentals than to incentives, the politicians and government officials charged with attracting FDI are often equally right when they argue that incentives can be effective, even decisive, and that it can be very difficult for them not to offer incentives if they want to attract a major investment project. In short, governments that wish to compete for such investment often face a prisoner’s dilemma67. In principle, one should look both at the relative costs — i.e. compare the costs of incentives to the benefits to be derived from the additional FDI attracted by incentives (FDI that otherwise would not have come to the country) — and at the determinants or components of the costs per se. For our purposes, however, it is largely sufficient to focus on the costs per se, though it is important to consider both direct and indirect costs. We can then address the question of how governments might best respond to the prisoner’s dilemma. One type of cost is the direct fiscal or financial cost of the incentives that governments offer to investors, including both current costs and the present value of commitments to provide subsidies in the future (e.g. tax holidays). Although conceptually easy to measure, even these direct costs are very difficult to gauge in 67
See also Graham (2000), especially pp. 63-67.
practice because, for both legitimate and questionable reasons, the governments that supply them and the investors that receive them are generally unwilling to disclose the amount of investment incentives68. Data gathered from unofficial sources leave little doubt, however, that the direct cost of financial and fiscal subsidies paid by governments (predominantly sub-national governments) to attract FDI in major automobile factories rose substantially during the 1980s and 1990s. In countries as diverse as Brazil, India, Germany, Portugal and the United States, the direct cost of investment incentives amounted to hundreds of thousands of dollars for each job the investment project was expected to create (see Table 7.2). Table 7.2. Investment Incentives in the Automobile Industry Date of package
Amount per job* (US dollars) 1980 United States Honda 4 000 early 1980s United States Nissan 17 000 1984 United States Mazda-Ford 14 000 mid-1980s United States GM Saturn 27 000 mid-1980s United States Mitsubishi-Chrysler 35 000 mid-1980s United States Toyota 50 000 mid-1980s United States Fuji-Isuzu 51 000 early 1990s United States Mercedes Benz 168 000 1992 Portugal Ford-Volkswagen 265 000 1995 Brazil Volkswagen 54 000 – 94 000 1996 Brazil Renault 133 000 1996 Brazil Mercedes Benz 340 000 1997 Germany Volkswagen 180 000 1997 India Ford 200 000 – 420 000 * Estimated value of fiscal and financial incentives supplied by national and sub-national governments to a particular investment project, divided by the number of jobs the project was expected directly to create. Source: Oman (2000, Table 2.1, p. 80).
Impressive as these orders of magnitude are, it is difficult to conclude — given reasonable assumptions about the likely long-term direct and indirect benefits accruing to the host economy from a major new investment project — that the net impact on the host economy is negative, unless the factory created by the investment shuts down within a few years (which happens more frequently than one might expect). The cost of incentives is nonetheless very large in some cases. It may certainly be concluded that, in such cases, the host economy would have derived significantly higher net benefits from FDI if it had been able to attract the 68
“Legitimate” reasons include investors’ unwillingness to disclose information that could significantly benefit competitors, and governments’ desire to avoid an upward ratchet effect on the value of incentives demanded by investors, which is likely if each successive potential investor knows how much the government has been willing to pay to previous investors. More questionable reasons relate to the danger of various forms of corrupt payments, as discussed further below.
investment with less costly incentives. The indirect costs of incentives are even more difficult to quantify, but they may in many cases be significantly greater than the direct costs. One type of indirect cost stems from the fact that incentives invariably discriminate against all the sectors and investment projects, actual and potential, that are not targeted by incentives. Incentives thus typically discriminate against smaller investors and against local investors. They also discriminate against sectors or activities in which the economy might develop — or might have developed — a comparative advantage that the people responsible for targeting incentives simply did not foresee or even consider. The largest indirect cost, however, stems from the fact that in setting up a system to compete for FDI using discretionary fiscal and financial incentives, governments tend, over time, to develop a system of governance that lacks transparency and, ultimately, accountability. For a developing economy, the longterm effect can be devastating. An argument can therefore be made for establishing a clear multilateral framework of rules to limit harmful aspects of incentive-based competition, while at the same time providing transparent, stable and predictable conditions for FDI. 7.4
Towards Constructive Rules-based Policies In the late 1960s, a number of European investor countries began to negotiate
bilateral investment treaties (BITs) with developing countries with a view to protecting their investment assets in these countries. According to UNCTAD estimates, over 1 900 BITs were in operation as of end 2000, along with around 2 100 double taxation treaties. In the past, most of these treaties were concluded between developed and developing countries, but recently the number of BITs between developing countries has been increasing substantially (UNCTAD, 2001, pp. 6-7). The principal role of BITs is to make binding provisions on expropriation, the transfer of payments and compensation for losses due to armed conflict or internal disorder, these benefits being accorded on a national treatment or MFN basis. In addition, BITs usually provide for the resolution of investor-state disputes by private institutions, such as the arbitration centres of the International Chamber of Commerce or the International Centre for the Settlement of Investment Disputes (ICSID) under the aegis of the World Bank. However, few BITs contain provisions on investment-restricting measures such as performance requirements. 92
The WTO agreement that deals most directly with investment matters is the Agreement on Trade-related Investment Measures (the TRIMs Agreement). As stated in the Preamble, the TRIMs Agreement aims to “promote the expansion and progressive liberalisation of world trade and to facilitate investment across international frontiers so as to increase the economic growth of all trading partners, particularly developing country Members, while ensuring free competition”. Despite this declared intent, this Agreement is restricted in scope, being substantially related only to those aspects of FDI that affect trade in goods. The TRIMs Agreement requires that the trade-distorting performance requirements imposed by host governments on foreign firms be notified within 90 days of the Agreement’s coming into force and eliminated in a phased manner69. The Annex to the Agreement also provides an illustrative list of TRIMs that are deemed inconsistent with the obligations of national treatment (Article III: 4) and of general elimination of quantitative restrictions (Article XI: 1) enshrined in the 1994 GATT. This list includes local content requirements, requirements limiting imports to the value of exports (trade-balancing requirements), foreign-exchange balancing requirements and export requirements. In other words, the TRIMs Agreement constrains WTO members from making investment incentives conditional on these performance requirements, though developing countries are allowed some flexibility in implementing this provision. As such performance requirements are often “packaged” with incentive measures, the TRIMs Agreement may indirectly limit the ability of host governments to use investment incentives as a means of attracting FDI inflows. A number of other trade and investment agreements concluded at the multilateral and regional levels may potentially affect the scope and magnitude of investment incentives used by host governments. These include the WTO Agreement on Subsidies and Countervailing Measures (ASCM), the General Agreement on Trade in Services (GATS), the state aid provisions of the European Union, the North American Free Trade Agreement and the APEC Non-binding Investment Principles. A recent paper issued by the OECD (2001c), however, states that “with the exception of the EU regime on state aids, none of the above agreements impose direct disciplines on the granting of investment incentives” (p. 3). This paper also notes that “the fact that few if any international disputes have to date challenged an investment incentive programme indicates both the difficulty of
A transition period of two years is allowed for developed countries, five years for developing countries and seven years for the least developed countries.
stating a successful claim under the present rules, and the political economy constraints that complicate the development of disciplines in the area” (p. 4). In short, owing to the prisoner’s dilemma inherent in incentive-based competition, the establishment of a multilateral framework of rules to discipline such measures would help to increase the collective welfare of host countries. Unfortunately, this multilateral approach has a very long way to go. Considering that investment issues are highly contentious under the WTO, a second-best option would be a regional approach to adopting more constructive, rules-based policies towards FDI. New regional integration agreements (RIAs) among national governments, and moves to deepen or strengthen existing ones, which have proliferated since the mid-1980s, have proven to be effective policy instruments for attracting FDI. NAFTA, Mercosur and the deepening of European integration are all cases in point. Their considerable power to attract FDI is not difficult to understand. First and foremost, they tend significantly to enlarge the market that can effectively be served by an investment in the region. In doing so, they greatly strengthen one of the key “fundamentals” to which investors attach great importance in choosing an investment location. RIAs can help to improve other fundamentals as well. In particular, governments often use them as vehicles to achieve a greater degree of internal market deregulation, or to pass regulatory reforms, especially when resistance from powerful domestic interest groups makes needed regulatory reform difficult to achieve at the national level. At the same time, they can help to ensure that a necessary process of deregulation does not degenerate into an unmanaged and destructive process of competitive deregulation. Second, RIAs can facilitate the co-operation among governments that may be needed to defend standards and regulations — e.g. on the environment and workers’ rights — because such rules can be difficult for governments individually to defend, or enforce, when faced with domestic pressures and the prisoner’s dilemma nature of policy competition. RIAs can also serve to harmonise and regulate governments’ use of fiscal and financial incentives to attract FDI, as well as contributing to greater macroeconomic and political stability in a region. Finally, no discussion of rules-based competition for FDI, especially among governments in developing countries, would be complete without mention of the tremendous importance that some investors — especially those seeking sites for 94
long-term investment in major production facilities — attach to the stability and predictability of the operating environment of their investment sites. This is why political and macroeconomic stability is included among the key “fundamentals” noted above. While a system based on negotiated incentives to attract investors may appeal to many investors, as well as to some government officials, in the long run most investors profit more from the stability, transparency and predictability of a rules-based approach to FDI policy.
Annex Table 7.1. FDI Policy Regimes in Latin America Policy Openness
Argentina Brazil Chile No registration or prior All investments must be Pro forma screening by government approval registered with the Central the government foreign required1. Bank of Brazil within 30 investment committee. days of entering the country. Registration is a pro forma requirement, but is needed for remittance of profits abroad, repatriation of capital and registration of the reinvestment of profits.
Colombia Investment screening has been largely eliminated, and the mechanisms that still exist are generally routine and nondiscriminatory. Some types of investments, however, are subject to special regimes, e.g. in the financial, hydrocarbon and mining sectors 2. Foreign investments must be registered with the Central Bank’s foreign exchange office within three months of the transaction date to ensure the right to repatriate profits and remittances and to access official foreign exchange.
Mexico The 1993 investment law, which is consistent with the investment chapter of NAFTA, eliminated the requirement of government approval for about 95 per cent of total activities. Authorisation by the National Commission for Foreign Investment is required when the investment is greater than $25 million or when a majority share in a regulated sector is involved. If the agency does not respond within 45 working days, the project is deemed to be approved.
Note:1) Exceptions to national treatment are maintained in the following sectors: real estate in border areas, air transport, shipbuilding, nuclear energy, uranium mining, and fishing (see US State Department, Country Commercial Guide, 2001). 2) In any event, the Colombian Economic and Social Policy Council (CONPES) may designate sectors of economic activity in which the government has discretion on whether to admit foreign capital participation. 96
Annex Table 7.1. FDI Policy Regimes in Latin America (continued) Policy Openness
Measure National treatment
Argentina The Constitution grants national treatment to foreign investors.
Brazil Chile Foreign capital The Constitution grants receives same legal national treatment to treatment as that foreign investors 4. applicable to national capital 3.
No areas of economic activity reserved exclusively to national 6 capital .
Foreign capital participation barred or subject to limitations in the following sectors: media, nuclear energy, health care, rural property ownership, highway freight transport, concession of domestic airlines, equity ownership in financial institutions and insurance companies 7.
Foreign capital participation is restricted in fishing, media, financial sector, and real estate in border areas 8.
Colombia According to the principle of equality governing foreign investments in Colombia, foreign investors are subject to the same treatment as domestic investment. Thus, no discriminatory conditions or treatment may be imposed 5. Foreign investment is prohibited in the defence and national security sector and in toxic waste disposal. Restrictions are imposed on foreign participation in television, ships’ brokers, the national airline and shipping companies 9.
Mexico National treatment is granted for most foreign investment.
There are 45 activities that are not yet liberalised. Activities (a) reserved to the state (e.g. petroleum, and electricity); (b) reserved to Mexican nationals (e.g. ground transportation, professional and technical services), and (c) subject to specific regulations (e.g. air transportation, financial sector, railways ) 10.
3) Art. 2 of Law 4.131/62, mentioned in http://alca-ftaa.iadb.org/eng/invest/bra~1.htm. Constitutional amendment approved in 1995 to eliminate the distinction between foreign and national capital. 4) The one-year residency requirement for capital repatriation — mostly applying to portfolio investment — was abolished in March 2000. The 30 per cent reserve requirement applying to all foreign capital entering Chile (the encaje) was abolished in September 1998. 5) See “Statute of foreign investment in Colombia” in http://www.coinvertir.com/01-legal/06_estatu/1_ing.htm#2. 6) Mentioned in http://alca-ftaa.iadb.org/eng/invest/arg~1.htm. Foreign firms may participate in privatisation and publicly financed R&D programmes. 7) Foreign firms may participate in the privatisation process, but any foreign capital invested under the privatisation plan must remain in Brazil for at least 6 years. See www.buybrazil.org/econ.html and www.brasilemb.org/trdae/guide.html. 8) Vessels fishing in the Chilean Exclusive Economic Zone must have majority Chilean ownership. Reciprocity instead of national treatment is applied to coastal shipping trade. Foreign investors may participate in the privatisation process and enjoy the same benefits as nationals.
Annex Table 7.1. FDI Policy Regimes in Latin America (continued) Policy
Measure Argentina Brazil Performance No specific performance Local requirements requirement for FDI. requirements 11.
Chile worker Local requirements 12.
Colombia worker Local worker requirement13. Local content requirement in the automotive industry.
Mexico Almost all performance requirements (export requirements, capital controls, local content, etc.) have been eliminated, except for the automotive industry and for companies applying for in-bond industry programmes. Local worker requirements exist14.
9) Generally, foreign investors may participate in privatisation of state-owned enterprises without restrictions. Colombia imposes the same investment restrictions on foreign investors as on national investors. 10) For a complete listing, see “Basic Guide for Foreign Investors”, National Bank of Foreign Trade (http://www.bancomext.com/Bancomext2000/publicasecciones/secciones/192/ingles.pdf). 11) In firms employing three or more persons, Brazilian nationals must constitute at least two-thirds of all employees and receive at least two-thirds of total payroll. 12) The labour law requires that at least 85 per cent of the employees of a particular employer must be Chilean nationals. Companies with no more than 25 employees are exempt from this rule. Local content requirements in the automobile assembly sector expired in 1998; a request for extension is pending. 13) Without exemption, at least 90 per cent of a company’s labour force and 80 per cent of management must be Colombian nationals. 14) There are no formal performance requirements. However, Article 29 of the Foreign Investment Law indicates the criteria the Commission will use for evaluating requests submitted to it for consideration: a) impact on employment and worker training; b) contribution to technology; c) compliance with the environmental provisions in the relevant ecological ordinances; and d) in general, contribution to increasing the competitiveness of the country’s productive plant. In deciding the outcome of a request, the Commission may not impose requirements that distort international trade. Article 7 of the Federal Labour Law stipulates that employers in every enterprise or establishment must ensure that at least 90 per cent of employees are Mexican. Workers in technical and professional fields must be Mexican, except where no Mexican nationals can be found for a particular specialisation, in which case employers may hire foreign workers in proportions not exceeding 10 per cent of the staff with that specialisation. The employer and foreign employees are obliged to train Mexican workers in the specialisation in question. Physicians serving enterprises must be Mexican. The provisions of this article are not applicable to directors, board members or general managers (FTAA-ALCA, Legislation for Foreign Investment Statutes in Countries in the Americas Comparative Study – Mexico).
Annex Table 7.1. FDI Policy Regimes in Latin America (continued) Policy Incentives
Measure Fiscal incentives
Argentina Argentinean and foreign firms face the same tax treatment (33% of net profits).
Brazil Domestic and foreign firms investing in less developed areas receive equal tax benefits.
Chile Investments over $50 million may qualify for tax concessions. Corporate tax exemptions are available to both domestic and foreign firms investing in less developed areas.
Existing incentives managed by the Ministry for the Economy apply to both domestic and foreign firms.
Same as above 15.
No special incentives for FDI.
Colombia Tax and foreign exchange benefits are available to domestic and foreign investors operating in special zones and producing for export. Reinvestment of corporate profits in Colombia for at least five years grants exemption from the 7% dividend income withholding tax or remittance tax. Profits are taxed at 35%. No special incentives for FDI.
Mexico Most direct tax incentives have been eliminated. The only significant federal tax incentive is accelerated depreciation allowances.
Special incentives exist for both domestic and foreign investments in priority development zones and priority industries.
Annex Table 7.1. FDI Policy Regimes in Latin America (continued) Policy Measure Investment IPR protection
Argentina Brazil Member of WIPO. WTO Member of WIPO and legislation on IPR signatory of the Bern ratified as law in 1995 16. Convention, Paris Convention and Washington Treaty. The New Industrial Property bill concerning patents and trademarks, which came into effect in May 1997, satisfies the TRIPS agreement17.
Chile Member of WIPO and signatory of the Bern and Paris Conventions. IPR law promulgated in 1991.
Colombia The common regulatory system for industrial property (trademarks and patents) stems from Decision 486 of the Andean Community Committee, implemented by Decree 2591 of 12 December 2000 and Regulatory Resolution 210 of 15 January 2001. Intellectual property protection (copyrights) is provided by means of Law 23 of 1982, Law 44 of 1993 and Decision 351 of the Andean Community Committee and regulatory decrees. Colombia is a member of WIPO and ratified the Paris Convention in 1996. Colombia remains on the Special 301 Watch List of the US Trade Representative for not providing effective protection of IPRs.
Mexico Mexico is a signatory to most major international IPR conventions, as well as NAFTA and WTO. IPR protection is provided by the Mexican Institute of Industrial Property Rights, established in 1991, and the 1994 Intellectual Property Law. Intellectual property is protected by a new copyright law passed in 1996.
16) Patent protection is available but inadequate, especially for pharmaceuticals. However, most of Argentina’s TRIPS obligations came into force on 1 January 2000. No specific law on trade secrets exists, although penalties for unauthorised revelation of secrets are applied to a limited degree under commercial law. Argentina has signed the WIPO Treaty on Integrated Circuits, but has no law dealing specifically with the protection of layout designs and semiconductors. 17) The law, however, includes some compulsory licensing and local worker requirements.
Annex Table 7.1. FDI Policy Regimes in Latin America (continued) Policy Investment protection
Measure Dispute settlement
Argentina Domestic or international arbitration. Member of ICSID and MIGA. Signatory of the New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards (1989).
Brazil Not a member of ICSID, nor signatory to the New York Convention, but member of MIGA. Bilateral Investment Treaties (BITs) allow for international arbitration, but foreign arbitration awards require confirmation by a court of the country where the judgement is rendered and the Brazilian Supreme Court.
Chile Member of ICSID and MIGA. Signatory to the New York Convention (1975). BITs allow for binding international arbitration.
Colombia Colombia is a signatory to the New York Convention (1979) and a member of ICSID and MIGA. National Law 315 permits the inclusion of an international binding arbitration clause in contracts between foreign investors and domestic partners.
Mexico Mexico is a signatory to the New York Convention (1971), but is not a member of ICSID or MIGA. Disputes can be addressed under either NAFTA or WTO.
Note to Annex Table 7.1: Most of the information contained in this table comes from the Country Commercial Guides prepared by the US Department of State, from the FTAA (http://www.alca-ftaa.org) and APEC (http://tyr.apecsec.org.sg/download/ieg) Internet sites, and from national investment promotion agencies, whose Internet sites can be recovered from http://www.ipanet.net.
Annex Table 7.2. FDI Policy Regimes in Asia Policy Openness
China Chinese authorities attempt to guide FDI towards “encouraged industries and priority regions”. All investments must pass through a sequential screening process involving various levels of government, depending on the size of the project 1. Each project is evaluated against official guidelines to determine whether it is beneficial to the development of the Chinese econom y2.
India Foreign investment is freely allowed in all sectors (including services) except in cases where sectoral ceilings exist 3. An “automatic route” for foreign direct investment is available. Investors are required to file relevant documents with the Reserve Bank of India within 30 days after the issue of shares to foreign investors. Proposals that do not fulfil the conditions for automatic approval will require the approval of the Foreign Investment Promotion Board, within the Ministry of Commerce & Industry, or of the Cabinet Committee on Foreign Investment (investment in excess of $171 million).
Indonesia To obtain investment approval, investors must submit an application form to the Chairman of the Investment Coordinating Board (BKPM) 4 . Foreign investments exceeding $100 million no longer require approval by the President of Indonesia, but can now be approved by the Chairman of BKPM. Starting in January 2000, some provinces can accept foreign investment applications directly. With the implementation of decentralisation, each province will eventually be able to accept applications. Foreign investment in Indonesia is no longer subject to a prescribed minimum amount of capital.
Malaysia Foreign investments must be approved by the Malaysian Industrial Development Authority (MIDA, for the manufacturing sector) or by other regulatory agencies (nonmanufacturing sector). Each project is evaluated against the guidelines set forth in the Industrial Master Plan and by the Foreign Investment Committee5. Foreign equity participation has been governed by the level of exports. A temporary relaxation of the strict foreign equity policy was in place during the 1998-2000 period, and the government has extended its liberalised treatment of foreign investment until 31 December 2003. Foreign equity holders will not be required to divest or dilute their equity holdings in projects approved under the above guidelines.
Chinese Taipei Application for investment approvals, acquisitions and mergers are screened by the Foreign Investment Commission within the Ministry of Economic Affairs. Screening is routine and non-discriminatory. Investment above $49 million and investment in industries on the negative list are screened by an interministerial commission.
Annex Table 7.2. FDI Policy Regimes in Asia (continued)
Note:1) Projects below $30 million may be approved by Special Economic Zones, open cities, provincial capitals and some other cities. Smaller cities can approve projects valued at $10 million. Projects above $30 million need approval by the Ministry of Foreign Trade and Economic Co-operation and the State Development and Planning Commission. Projects above $100 million must obtain State Council approval. After approval, foreign investors have 30 days to apply for registration. Business licences are issued by the industry and commerce administration authorities. The date of the licence is the date of establishment of the enterprise. The approval process sets a fixed period within which the investment must be made; if it is not made within that period, the licence can be revoked. The licence includes a period of operation for the enterprise. Extensions must be obtained 180 days before expiration. Separations and mergers are subject to approval and registration. See Law of the People’s Republic of China on Foreign Invested Enterprises, 12 April 1986, Art. 6, 7, 9, 10 and 20. 2) A new investment must increase foreign exchange, introduce new technology or improve skills. The investment cannot be approved if it uses technology that is already available in the People’s Republic or if the existing production capacity can already satisfy domestic demand. http://tyr.apecsec.org.sg/download/ieg/. 3) Ceilings on foreign equity ownership exist in several sectors (banking/insurance, civil aviation and airport infrastructure, telecommunications, petroleum, drugs and pharmaceuticals, trading). For a complete list, see the Indian Investment Centre website at http://iic.nic.in/vsiic/iic3_a.htm#Investment%20Policy. 4) The new investor must apply for an initial investment approval, which serves as a temporary operating licence. The process is completed in 10-20 working days. Once the foreign firm starts production, it must apply for a permanent business licence. The business licence is granted for a period of 30 years from the commencement of commercial operation and may be renewed. 5) Detailed guidelines for project evaluation exist. Among other criteria, project approval depends on export orientation, local equity participation, source of financing and the potential for technological diffusion. See http://www.state.gov/www/about_state/business/com_guides/2001/eap/malaysia_ccg2001.pdf.
Annex Table 7.2. FDI Policy Regimes in Asia (continued) Policy Openness
Measure National treatment
China National treatment is extended, on a reciprocal basis, to contracting parties of international treaties to which China is a signatory. In practice, however, national treatment is denied in almost all service and most industrial sectors. In the context of its accession to the WTO, China has committed to granting unconditional national treatment under the 1994 GATT.
India In the majority of cases foreign investment is not granted national treatment.
Indonesia National treatment is not granted to most foreign investment 6. Under the ASEAN Investment Area Framework Agreement (AIA), Indonesia is committed to extending national treatment gradually, first to ASEAN investors and then to all investors, subject to the member economy’s specific reservations and 7 exceptions .
Malaysia Foreign direct investors established in Malaysia are generally accorded national treatment in all but equity limits. Furthermore, criteria for project approval are applied in a nondiscriminatory manner, except in instances where local and foreign firms propose identical projects. Malaysia is a signatory to the AIA Framework Agreement.
Chinese Taipei Foreign firms are generally accorded national treatment.
6) A revised Foreign Capital Investment Law is being prepared and may come into effect as early as 2002. This law would provide an appropriate legal framework for investment based on equal treatment of investors; protection against expropriation, confiscation or requisition of investments and unilateral alteration or termination of contracts; freedom to repatriate foreign investment capital and net proceeds thereon; and access to impartial, quick and effective mechanisms for the resolution of commercial and other investment disputes. 7) The objective of AIA is to build a liberal, transparent and attractive ASEAN Investment Area in order to promote investment into ASEAN from both ASEAN and non-ASEAN sources. It is expected that under the AIA Agreement member countries will “immediately open” all industries and “immediately grant” national treatment. However, Article 7 of the Agreement also provides that each member country may, depending on its specific conditions, draw up a Temporary Exclusion List and Sensitive List, composed of any industries that the country cannot open or investment-related measures for which it cannot grant national treatment status to ASEAN investors. The salient features of the AIA are: extending national treatment to ASEAN investors by 2010 and all other investors by 2020; opening all industries to ASEAN investors by 2010 and all other investors by 2020; exclusion allowed through the Temporary Exclusion List (TEL) to be phased out by 2010 and the Sensitive List (SL) to be reviewed by 2003; granting ASEAN most-favoured-nation treatment for ASEAN investors.
Annex Table 7.2. FDI Policy Regimes in Asia (continued) Policy Openness
Measure China Negative list The Provisional Regulations of Guidance of Foreign Investment restrict FDI that is not beneficial to the development of the Chinese economy and prohibit FDI that jeopardises national security or the social and public interest. Foreign investment is not allowed in many telecommunication services, as well as the news media and television sectors. Access to some sectors is conditional on establishing a joint venture where the Chinese partner has the leading position (e.g. construction, aviation industry, automobile industry, foreign trade).
India Defence and strategic industries are closed to foreign investment, as well as agriculture, rail and postal services, housing and real estate.
Indonesia According to the 2000 negative list, there are some sectors where investment is prohibited or reserved to the government (including sectors that may involve public health and safety hazards), some sectors closed to foreign investment and some sectors where foreign investors must establish a joint venture in which at least 5% of the issued capital is held by domestic investors 8.
Malaysia With the exception of activities designated in a specific negative or exclusion list, all new projects in manufacturing, including investments for expansion and diversification, are exempted from both equity and export conditions. Project owners can hold 100% of the equity, and need not meet any export requirement to do s o 9. Foreign equity ownership is still limited in commercial banking (30% of equity), insurance companies (51%), telecommunications (61%) and shipping companies (70%), even though the limits have been loosened.
Chinese Taipei Foreign investment is prohibited in agriculture, forestry, fishing, pesticides, explosives, firearms, military equipment, postal services and savings institutions, and wireless broadcasting. Many sectors have been removed from the negative list, such as power generation, transmission and distribution, real estate development, transportation. Foreign ownership ceilings (usually 50%) still exist for sectors such as telecommunications, airlines, electrical power and transportation.
8) Sectors closed to foreign investment include radio, television, Internet and print media, the film industry (film making, film technical services, film export and import business, film distribution and theatre operation and/or viewing services), trading and transportation. Sectors open to joint ventures include the building and operation of ports; electricity production, transmission and distribution; shipping; processing and provision of clean water to the public; public railway system; nuclear power generation; medical services; telecommunications and air transport. For details see http://www.usembassyjakarta.org/econ/investment0800.html. 9) Exclusion List of Activities: paper packaging, plastic packaging (bottles, films, sheets and bags), plastic injection moulding components, metal stamping, metal fabrication and electroplating, wire harness, printing, steel service centre.
Annex Table 7.2. FDI Policy Regimes in Asia (continued) Policy Openness
Measure Performance requirements
China Export requirements: although not formally required, they are included in many contracts. Local content: localisation of production is strongly encouraged. Foreign investors are often asked gradually to increase the percentage of local content. Technology transfer: although not formally required, technology transfer is included in many contracts and encouraged by means of incentives. Local worker requirements: rules for hiring Chinese nationals depend on the type of establishment.
India Local worker requirements: No local worker requirements. Restrictions on employing foreign technicians and managers have been eliminated. Other requirements: Local content requirement, export obligations and foreign-exchange balancing apply to all foreign automobile manufacturing investment in India.
Indonesia In June 1994, the government dropped initial foreign equity requirements and sharply reduced divestiture requirements. Indonesian law provides for both 100% foreign-owned investment projects and joint ventures with a minimum Indonesian equity stake of 5%. Foreign shareholders are no longer required to reduce their holdings to a minority position at some time in the future. There is a requirement that within 15 years of establishment, a 100% foreign shareholder shall sell at least a nominal percentage to an Indonesian citizen or entity. Local content and export requirements: none. Local worker requirement: A company may hire foreigners only for specifically designated positions. Employers must have labour training programmes aimed at replacing foreign workers with Indonesian. Foreign investors must contribute to the training and development of the domestic labour force.
Malaysia Divestiture requirements in manufacturing have been relaxed during the last two years, but they still exist in some sectors, such as telecommunications (foreign equity must be reduced to 49% after five years). Joint venture and local worker requirement: the government often requires foreign investors to share ownership with local partners (30% of equity) and hire Malaysian workers, in proportions reflecting Malaysia’s ethnic composition. Foreign workers may be employed only in the construction, plantation, service (domestic servants, hotels, training personnel) and manufacturing sectors. In addition, certain “key posts” may be permanently filled by foreigners. To ensure that foreign labour is employed only when necessary, an annual levy on foreign workers is imposed.
Chinese Taipei Like domestic firms, foreign companies must locate in designated zones and are subject to restrictions on the number of foreign employees that can be hired. To obtain investment authorisation, it is not necessary to meet technology transfer requirements or employ a minimum number of local workers. Firms in EPZs are required to export all they produce, although they may sell on the domestic market after paying import duties. Local content requirements remain in place only in the automotive industry.
Annex Table 7.2. FDI Policy Regimes in Asia (continued) Policy Incentives
Measure Fiscal incentives
China China has long granted rebates and other tax benefits to foreign investors. The tax incentive system is complicated, however, mainly because of overlapping of central, provincial and local government regulations.
Many incentives are available to investors in Special Economic Zones, as well as in less developed regions. Foreign investors may have to negotiate incentives and benefits directly with the relevant authorities. Some incentives and benefits may not be conferred automatically.
India No income tax on profits derived from export of goods. Complete exemption from customs duty on industrial inputs and corporate tax holiday for five years for 100% export-oriented firms and firms in Export Processing Zones.
Indonesia Fiscal incentives are available to both domestic and foreign investors. Special investment incentives are granted on a case-bycase basis by BKPM. Special tax facilities are designated for investors entering designated pioneer industries 10.
Malaysia Fiscal incentives are available to both domestic and foreign investors and are usually linked to performance criteria, such as export targets, local content and technology transfer.
Chinese Taipei Accelerated depreciation allowances and tax credits are offered to firms introducing new technologies or locating in less developed areas. The five-year tax holiday for new investment was re-introduced in 1995.
Various incentives are available for exporting firms and firms located in the Multimedia Super Corridor (MSC). In particular, firms with MSC status face no restrictions on recruitment of expatriates, are exempt from all capital controls and can apply for government funding for R&D. Non-fiscal export incentives consist of an export credit refinancing facility, an infrastructure allowance and various investment allowances 11.
10) The tax holiday system and other facilities for new investors are under review and may be abolished under the IMF-supported reform programme. 11) See http://www.mida.gov.my/policy/chapter3.html.
Annex Table 7.2. FDI Policy Regimes in Asia (continued) Policy Investment protection
China China is a member of WIPO and a signatory to most conventions on industrial and intellectual property rights. Furthermore, China has committed to full compliance with the TRIPS agreement upon accession to the WTO. Nevertheless, implementation of IPR protection remains problematic.
India India is a member of WIPO and a signatory to most conventions on industrial and intellectual property rights (though not the Paris Convention). The new Trademark Bill passed in 1999 raised trademark protection to international standards. Enforcement of intellectual property rights is improving but still weak. Patent protection is weak, since no patent can be granted for a new product that can be used as a food or drug 12. As a signatory to the WTO TRIPS agreement, India is committed to introducing a comprehensive system of product patents no later than 2005.
Indonesia Indonesia became a member of WIPO and a signatory to the major international conventions dealing with IPR only in 1997. Until 2001, Indonesia did not have laws on protection of industrial designs, layout designs of integrated circuits, trade secrets and plant varieties 13. There has been some improvement, but effective enforcement of IPRs is still extremely difficult due to the ineffectiveness of Indonesia's judicial system and some legislative drawbacks 14. On 30 April 2001, the US Trade Representative downgraded Indonesia from Watch List to Priority Watch List under the Special 301 provision.
Malaysia Malaysia is a member of WIPO and a signatory to the Bern and Paris Conventions. Legislation concerning protection of IPRs is adequate, having been amended in 2000 to comply with the WTO TRIPS Agreement. Enforcement of copyright laws is improving, although piracy rates for software and music and video discs are still very high 15.
Chinese Taipei Chinese Taipei is not a member of WIPO. Copyright, patent and trademark laws have been amended recently to conform to TRIPS standards, but only the trademark law and certain provisions of the copyright law have been implemented. The amended part of the patent law will not go into effect until Chinese Taipei accedes to the WTO.
In April 2001, the US Taiwan is on the US Trade Representative Special 301 Watch List. decided to keep Malaysia on the Special 301 Priority Watch List for its failure to obtain substantial reductions in the production and export of pirated optical discs.
12) Processes for making such products can be patented, but the patent term is limited to five years after the patent has been granted. 13) Five international conventions signed in 1997: Paris Convention for the Protection of Industrial Property, Patent Co-operation Treaty and Regulations under the PCT, Trade Marks Law Treaty, Bern Convention for the Protection of Literary and Artistic Work, WIPO Copyright Treaty (Presidential Decrees Nos. 16-19 of 1997). Three new laws, namely the Law on Industrial Design, Law on Layout Design of Integrated Circuits, and Law on Trade Secrets, were enacted in December 2000. Furthermore, two revised laws were enacted in August 2001, the Patent Law and the Trademark Law. In the meantime, the revised draft of the Copyright Law is still under consideration by the Parliament. In January 2001, the Appeal Commissions on Patents and Trademarks were instituted (http://www3.itu.int/MISSIONS/Indonesia/State.htm).
Annex Table 7.2. FDI Policy Regimes in Asia (continued) Policy Investment protection
Measure Dispute settlement
China Member of ICSID and signatory to the New York Convention (1987). Member of MIGA. Chinese authorities, however, prefer arbitration through national agencies, such as the China International Economic and Trade Arbitration Commission.
India India is a signatory to the New York Convention (1960), but is not a member of ICSID. A new Arbitration and Conciliation Act was approved in 1996 on the basis of the UN CITRAL model law.
Indonesia Member of ICSID and MIGA and a signatory to the New York Convention (1982). However, no law requires Indonesian courts to enforce the judgements of nonIndonesian courts, and the court system does not provide effective recourse for resolving commercial disputes.
Malaysia Malaysia is a member of ICSID and MIGA and a signatory to the New York Convention (1986). The government has also set up the Kuala Lumpur Regional Centre for Arbitration under the auspices of the Asian-African Legal Consultative Committee to offer international arbitration, mediation and conciliation for trade disputes.
Chinese Taipei Chinese Taipei is not a member of ICSID or MIGA, nor a signatory to the New York Convention. Disputes are usually resolved according to domestic laws and regulations.
Note to Annex Table 7.2: Most of the information contained in this table comes from the Country Commercial Guides prepared by the US Department of State, from the ASEAN (http://www.aseansec.org) and APEC Internet sites (http://tyr.apecsec.org.sg/download/ieg), and from national Investment Promotion Agencies, whose Internet sites can be recovered from http://www.ipanet.net.
14) The amended Patent Law, for instance, states that products and production processes are patentable for a period of 20 years commencing from the filing of the patent application, but, at the same time, allows for compulsory licensing and limits patent protection to patents that are “implemented” in Indonesia. In practice, this means that patented products must be manufactured in Indonesia to receive patent protection. The industry estimates the levels of music and business software piracy at 87 per cent, motion picture piracy at 90 per cent and game software piracy at 99 per cent (http://www.usembassyjakarta.org/econ/301review.html).
15) Malaysia’s production capacity for CDs far exceeds local demand plus legitimate exports. The Optical Disc Act, which became effective on 15 September 2000, gives the government greater enforcement powers and allows for stiffer penalties (including jail sentences) for the production and export of pirated optical media. According to the industry, software piracy rates in Malaysia fell from 71 per cent in 1999 to 66 per cent in 2000 (http://usembassymalaysia.org.my/ccg-2002.html).
Chapter Ⅷ Summary and Conclusions
This study reviews and discusses recent empirical research on key development issues related to foreign direct investment (FDI). The literature review focuses on five main areas: the FDI-growth nexus; FDI-trade linkages; FDI and technology transfer; FDI, privatisation and corporate governance; and hostgovernment policies for attracting FDI. The study also presents a statistical overview of trends in FDI flows from the global and regional perspectives. The aim of the study is to advance the policy debate about the role FDI can play in sustaining economic and social development in host developing countries. The main issues and policy lessons may be summarised as follows. Much of the FDI flowing into developing countries in the 1970s through the mid-1980s was attracted by the availability of natural resources. This traditional form of FDI has different implications for growth and development in host economies than FDI in manufacturing and services. This study focuses on the latter type, which has been one of the defining characteristics of the world economy over the last 20 years. The period since the mid-1980s, and particularly since 1993, has seen rapid expansion of global FDI flows: outflows of FDI world-wide increased more than fivefold, from just over $200 billion in 1992 to $1 150 billion in 2000. Much of this surge in FDI flows is accounted for by cross-border mergers and acquisitions (M&As) among OECD countries, triggered by policy initiatives aimed at deepening regional integration, such as the implementation of the EU’s Single Market Programme and the creation of NAFTA. Among non-OECD countries, the region comprising Latin America and the Caribbean, particularly Mercosur, has also embraced cross-border M&As as the principal mode of FDI inflows in the context of large-scale privatisation programmes, including privatisation of public utilities. In contrast, ASEAN and South Asia began to jump on the “M&A bandwagon” more recently, in the aftermath of the 1997-98 financial crisis (see Table 2.3). The experience of the 1990s highlights two other important developments in the regional pattern of FDI. The first is the emergence of the United States as the world’s largest net recipient of FDI. At the beginning of the decade, the United States was the second largest source country of FDI on a net basis (outflows minus inflows), after Japan. Several European countries have continued to play a dominant role as net suppliers of FDI, but Japan’s share in the supply of FDI diminished considerably during this period. The second important development is 110
the rapid expansion of FDI inflows to several developing and emerging economies, among which China has been by far the largest recipient (see Chapter Ⅱ). The global surge in FDI over the last decade has reinforced the already important role played by multinational enterprises (MNEs) in host countries. As measured by several indicators, such as industrial production, employment and foreign trade, MNE affiliates have become significant players in many OECD countries, with the notable exception of Japan (see Table 2.1). At the same time, the increased presence of foreign affiliates has sparked renewed interest among policymakers as to what specific contribution FDI can make to the long-term development of the host country. The answer to this question seems to be less controversial in theory than in practice. Theory suggests that, by providing a channel for knowledge acquisition and dissemination, FDI can act as an engine of growth for the host economy; but the question has yet to be answered empirically. The literature on the development dimension of FDI is very large, and still growing. Where do we stand today? The vast majority of existing empirical studies based on macroeconomic data indicate that FDI does make a positive contribution to both income growth and factor productivity in host countries. FDI tends to “crowd in” domestic investment, as the creation of complementary activities outweighs the displacement of domestic competitors. Similarly, in the North-South context the relationship between FDI and trade is more one of complementarity than of substitution, owing to backward and forward linkages. Many studies, however, suggest that host countries cannot capture the full benefits associated with FDI until they reach a certain threshold in terms of educational attainment, provision of infrastructure services, local technological capabilities or the development of local financial markets. This “threshold externalities” argument highlights the complex nature of interactions between FDI and growth in the host country; these interactions work through various channels, which require further analysis based on firm-level or plant-level data (see Chapters Ⅲ and Ⅳ). The results of recent empirical studies using microeconomic data suggest that the “spillover” effect of FDI on the productivity growth of local firms does not occur automatically. The estimated magnitude of technological spillovers through FDI depends crucially on various host-country and firm-level characteristics, such as the relative and absolute absorption capacities of individual host countries and firms. Detailed analysis on the nature of technological spillovers is often hampered by the extreme data requirements of such studies. Nonetheless, one conclusion emerging from the results of several country case studies — on Italy, Mexico and 111
Uruguay, for example — is that the existence of large technological gaps in a less competitive market environment tends to reduce the probability of technological spillovers from MNE affiliates to local firms (see Chapter Ⅴ). As noted above, cross-border M&As (as opposed to greenfield investment) played a central role in the FDI boom of the 1990s. Privatisation of state-owned enterprises has provided a major channel for FDI inflows not only in Eastern Europe and the former USSR but also in Latin America. Since the 1997-98 financial crisis, a number of Asian countries have begun to implement ambitious privatisation programmes in the utilities sector, in order to consolidate public finances and revive their economies. For many governments in the region, however, privatisation and foreign participation in such programmes remain politically sensitive issues. An assessment of privatisation policy in both OECD and developing countries indicates that, on balance, privatisation has increased consumer welfare, but that the degree of success depends crucially on the post-privatisation market structure shaped by the new regulatory system that is established. In other words, weaknesses in the regulatory environment tend to reduce the benefits of privatisation. Active participation by foreign firms can also be seen as key to successful privatisation programmes, as in the case of Brazil. Nonetheless, regulatory agencies still face substantial future challenges (see Chapter Ⅵ). The above discussions all point to the importance of host-government policies for attracting FDI. Conceptually, host-government policies fall into two broad categories — incentive-based and rules-based measures — although this distinction is not always clear in practice. An important point concerning of hostgovernment policies is that the use of discretionary, targeted fiscal and financial subsidies to attract investors is ineffective as long as economic and political “fundamentals” are not satisfied. Moreover, these incentive-based measures are too costly (and even wasteful) for developing countries facing severe resource constraints. Clearly, there is a need for more constructive, rules-based measures that can help in creating the sound business environment that firms need to make long-term investments, including the development of local skills and technological capabilities. In this context, a comparative survey was conducted with respect to foreign investment regimes in ten host economies in Asia and Latin America. The results indicate that legal and policy frameworks for foreign investment appear to be more open in Latin America than in Asia, despite the considerable efforts made by Asian 112
countries to liberalise FDI policy following the 1997-98 financial crisis. There are also substantial differences among these Asian countries in terms of control over foreign investors at the entry phase and the use of negative lists. Moreover, protection of intellectual property rights (IPRs) differs significantly in both scope and magnitude across countries. This is considered a key feature of the foreign investment regime in the host country, as the IPR issue has an important influence on the magnitude and quality of technology transfer from home to host countries (see Chapter Ⅶ). The study concludes by stating that host-government polices should attach greater importance to the stability and predictability of the local business environment in which foreign firms operate. To this end, the establishment of a multilateral framework of rules to discipline incentive-based competition would help to increase the collective welfare of host countries. Unfortunately, this multilateral approach has a very long way to go. Considering that investment issues remain very contentious under the WTO, a second-best option for host governments would be a regional approach to adopting more constructive, rulesbased policies towards FDI. The commitment to creating the ASEAN Investment Area (AIA) is a case in point. The evolution of AIA and other regional approaches (Mercosur, FTAA, etc.) in the coming years will have important implications for future multilateral negotiations on trade and investment issues under the WTO.
Bibliography ACEMOGLU, D.
F. ZILIBOTTI (1997), “Was Prometheus Unbound by Chance?
Risk, Diversification and Growth”, Journal of Political Economy, Vol. 105, pp. 709-751. AGGARWAL, V. K. (1980), “Determinants of Foreign Direct Investment: A Survey”, Weltwirtschaftliches Archiv, Vol. 116, pp. 739-773. AGOSIN , M. AND R. MAYER (2000), “Foreign Investment in Developing Countries: Does It Crowd in Domestic Investment?”, UNCTAD Papers, No. 146. AITKEN , B., A. H ARRISON AND R.E. LIPSEY (1996), “Wages and Foreign Ownership: A Comparative Study of Mexico, Venezuela and the United States”, Journal of International Economics, Vol. 40, No. 3/4, pp. 345-371. AITKEN , B.J. AND A.E. H ARRISON (1999), “Do Domestic Firms Benefit from Direct Foreign Investment? Evidence from Venezuela”, American Economic Review, Vol. 89. ALFARO L., A. CHANDA , S. KALEMIL -OZCAN AND S. SAYEK (2001), “FDI and Economic Growth: The Role of Local Financial Markets”, University of Houston Working Papers. AMITI, M. AND K. WAKELIN (2000), “Investment Liberalisation and International Trade”, GEP Research Paper 2001/12, University of Nottingham. ARELLANO, M. AND O. BOVER (1995), “Another Look at the Instrumental-Variable Estimation of Error-Components Models”, Journal of Econometrics, 68, 2951. ARELLANO, M. AND S. BOND (1991), “Some Tests of Specification for Panel Data: Monte Carlo Evidence and Application to Employment Equations”, Review of Economic Studies, 58, 227-297. ARMIJO, L. (1998), “Balance Sheet or Ballot Box? Incentives to Privatize in Emerging Democracies”, in P. OXHORN AND P. STARR (eds), The Problematic Relationship between Economic and Political Liberalization, Lynne Rienner, Boulder, Colorado. AZARIADIS , C. AND A. DRAZEN (1990), “Threshold Externalities in Economic Development”, Quarterly Journal of Economics, Vol. 105, No. 2, pp. 501-526. BARRO, R. AND X. SALA-I-MARTIN (1995), Economic Growth, McGraw-Hill, New York. BECK, T., R. LEVINE AND N. LOYAZA (2000), “Financial Intermediation and Growth: Causality and Causes”, Journal of Monetary Economics (forthcoming). BEHRMAN, J. AND H. WALLENDER (1976), Transfer of Manufacturing Technology within Multinational Enterprises, Ballinger, Cambridge, Mass. BELDERBOS , R. AND L. SLEUWAEGEN (1998), “Tariff Jumping DFI and Export Substitution: Japanese Electronics Firms in Europe”, International Journal 114
of Industrial Organization, Vol. 16, pp. 601-638. BENDE-NABENDE, A., J.L. FORD AND J.R. SLATER (2000), “The Impact of FDI on the Economic Growth of the ASEAN-5 Economies 1970-94: A Comparative Dynamic Multiplier Analysis from a Small Model with Emphasis on Liberalisation”, in J.L. FORD (ed.), Finance, Governance, and Economic Performance in Pacific and South East Asia, Edward Elgar. BENNETT, D., X. LIU, D. PARKER , F. STEWARD AND K. VAIDYA (2001), “Technology Transfer to China: A Study of Strategy in 20 EU Industrial Companies”, International Journal of Technology Management, Vol. 21, pp. 151-182. BERNSTEIN , J.I. (1989), “The Structure of Canadian Inter-Industry R&D Spillovers, and the Rates of Return to R&D”, Journal of Industrial Economics, Vol. 37, pp. 315-328. BISHOP, M., J. KAY AND C. MAYER (eds) (1994), Privatization and Economic Performance, Oxford University Press, Oxford. BLOMSTRÖM , M. AND A. KOKKO (1996), “Multinational Corporations and Spillovers”, CEPR Discussion Paper No. 1365. Published: Journal of Economic Surveys, Vol 12, No 2, 1998. BLOMSTRÖM , M. AND F. SJÖHOLM (1998), “Technology Transfer and Spillovers: Does Local Participation with Multinationals Matter?”, NBER Working Paper No. 6816. Published: European Economic Review, Vol 43, 1999. BLOMSTRÖM , M. AND H. PERSSON (1983), “Foreign Investment and Spillover Efficiency in an Underdeveloped Economy: Evidence from the Mexican Manufacturing Industry”, World Development, Vol. 11. BLOMSTRÖM , M. AND WOLF (1994), “Multinational Corporations and Productivity Convergence in Mexico”, in W. BAUMOL, R. NELSON AND E. WOLFF (eds), Convergence of Productivity: Cross-National Studies and Historical Evidence, Oxford University Press, Oxford. BLOMSTRÖM , M., R. LIPSEY AND K. KULCHYCKY (1988), “US and Swedish Direct Investment and Exports”, in R. BALDWIN (ed.), Trade Policy Issues and Empirical Analysis, University of Chicago Press, Chicago, pp. 259-297. BLOMSTRÖM , M., R. LIPSEY AND M. Z EJAN (1994), “What Explains Developing Country Growth?”, NBER Working Paper, No. 4132. BLOMSTRÖM , M., S. GLOBERMAN AND A. KOKKO (1999), “The Determinants of Host Country Spillovers from Foreign Direct Investment: Review and Synthesis of the Literature”, SSE/EFI Working Paper Series in Economics and Finance, No. 239. BLONIGEN , B. (1999), “In Search of Substitution between Foreign Production and Exports”, NBER Working Paper No. 7154. Published: Journal of International Economics, Vol 53, February 2001, pp. 81-104. BLOOM , M. (1992), Technological Change in the Korean Electronics Industry, 115
Development Centre Studies, OECD Development Centre, Paris. BORENSZTEIN , E., J. DE GREGORIO AND J.-W. LEE (1998), “How Does Foreign Direct Investment Affect Economic Growth?”, Journal of International Economics, Vol. 45, pp. 115-135. BOUBAKRI, N. AND J.C. COSSET (1998), “The Financial and Operating Performance of Newly Privatised Firms: Evidence from Developing Countries”, Journal of Finance, Vol. 53, pp. 1081-1110. BRACONIER , H., K. EKHOLM AND K.H.M. KANARVIK (2001), “Does FDI Work as a Channel for R&D Spillovers? Evidence Based on Swedish Data”, Working Paper No. 553, The Research Institute of Industrial Economics, Stockholm. BRAINARD, S.L. (1997), “An Empirical Assessment of the Proximity-Concentration Tradeoff between Multinational Sales and Trade”, American Economic Review, Vol. 87, pp. 520-544. BRAINARD, S.L. AND D.A. RIKER (1997), “Are US Multinationals Exporting US Jobs?”, NBER Working Paper No. 5958. CARDOSO, F.H. (1995), “Public Service Concessions in Brazil”, Presidencia da Republica, Brasilia, September. CARKOVIC , M. AND R. LEVINE (2001), “Does Foreign Direct Investment Accelerate Economic Growth?”, University of Minnesota Working Paper. CARR, D.L., J.R. MARKUSEN AND K.E. MASKUS (2000), “Estimating the Knowledge-Capital Model of the Multinational Enterprise”, American Economic Review, Vol. 91, pp. 693-708. CAVES , R. (1996), Multinational Enterprise and Economic Analysis, 2nd ed., Cambridge University Press, Cambridge (originally published 1982). CHEN , E. (1996), “Transnational Corporations and Technology Transfer to Developing Countries”, in UNCTAD, Transnational Corporations and World Development, ITBP, London. CLAESSENS, S., S. DJANKOV AND L.H.P. LANG (1999), “Who Controls East Asian Corporations?”, Policy Research Paper No. 2054, World Bank, Washington, D.C. CLAUSING, K. (2000), “Does Multinational Activity Displace Trade?”, Economic Inquiry, Vol. 38, pp. 190-205. CORREA , C. (1999), “Technology Transfer in the WTO Agreement”, in UNCTAD, A Positive Agenda for Developing Countries, Geneva. DE MELLO, L., JR. AND K. FUKASAKU (2000), “Trade and Foreign Direct Investment in Latin America and Southeast Asia: Temporal Causality Analysis”, Journal of International Development, Vol. 12, pp. 903-924. DE MELLO, L.R., JR. (1997), “Foreign Direct Investment in Developing Countries and Growth: A Selective Survey”, Journal of Development Studies, Vol. 34, pp. 1-34. 116
DE MELLO, L.R., JR. (1999), “Foreign Direct Investment-Led Growth: Evidence from Time Series and Panel Data”, Oxford Economic Papers, Vol. 51, pp. 133-151. DE SOTO, H. (2000), The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else, Basic Books, New York. DÉMURGER , S. (2000), Economic Opening and Growth in China, Development Centre Studies, OECD Development Centre, Paris. DJANKOV , S. AND B. H OEKMAN (1999), “Foreign Investment and Productivity Growth in Czech Enterprises”, World Bank Economic Review, Vol. 14, pp. 49-64. DUNNING, J.H. (1977), “Trade, Location of Economic Activity and the MNE: A Search for an Eclectic Approach”, in Ohlin, B., P-O. Hesselborn and P.M. Wijkman (eds.), The international Allocation of Economic Activity: Proceedings of a Nobel Symposium, held at Stockholm, pp. 395-418, London: Macmillan. _____. (1998), “The European Internal Market Program and Inbound Foreign Direct Investment”, in J.H. D UNNING (ed.), Globalization, Trade and Foreign Direct Investment, Elsevier, Oxford, pp. 49-115. EATON , J. AND A. TAMURA (1996), “Japanese and US Exports and Investment as Conduits of Growth”, NBER Working Paper No. 5457. Published in T. Ito and A. Krueger, eds., Financial Deregulation and Integration in East Asia, NBER and University of Chicago Press. ESTRIN , S. AND A. ROSEVAER (1999), “Enterprise Performance and Ownership: The Case of Ukraine”, European Economic Review, Vol. 43, pp. 1125-1136. FAN, X. AND P.M. DICKIE (2000), “The Contribution of Foreign Direct Investment to Growth and Stability: A Post-crisis ASEAN-5 Review”, ASEAN Economic Bulletin, Vol. 17, pp. 312-323. FEENSTRA, R.C. (1999), “Facts and Fallacies about Foreign Direct Investment”. Published in Martin Feldstein, ed. International Capital Flows, University of Chicago Press and NBER, pages 331-350. FREEMAN, C. AND J. H AGEDOORN (1992), Globalisation of Technology, a Report for the FAST Program, EC FAST and MERIT, University of Maastricht, Brussels. FUKASAKU , K. (2001), “Privatisation and Corporate Governance in East Asia” (in Japanese), in S. URATA AND H. KOHAMA (eds), East Asia’s Sustainable Development, Keiso-Shobo, Tokyo. FUKASAKU , K. AND A.C. PINEIRO (eds) (1999), Privatisation in Brazil: The Case of Public Utilities, Ensaios BNDES 10, Brazilian Development Bank, Rio de Janerio. FUKASAKU , K.
D. WALL (1994), China’s Long March to an Open Economy, 117
Development Centre Studies, OECD Development Centre, Paris. FUKASAKU , K. AND F. KIMURA (2001), “Globalisation and Intra-firm Trade: Further Evidence”, May, mimeo [forthcoming in P. LLOYD et al. (eds) (2002), Frontiers of Intra-industry Trade Research, Macmillan]. FUKASAKU , K., Y. MA AND Q. YANG (1999), China’s Unfinished Open-Economy Reforms: Liberalisation of Services, Technical Paper No. 147, OECD Development Centre, Paris. GERSHENBERG, I. (1987), “The Training and Spread of Managerial Know-How: A Comparative Analysis of Multinational and Other Firms in Kenya”, World Development, Vol. 15, pp. 931-939. GLOBERMAN, S. (1979), “Foreign Direct Investment and ‘Spillover’ Efficiency Benefits in Canadian Manufacturing Industries”, Canadian Journal of Economics, Vol. 7, pp. 42-56. GLOBERMAN, S., J.C. RIES AND I. VERTINSKY (1994), “The Economic Performance of Foreign Affiliates in Canada”, Canadian Journal of Economics, Vol. 27, pp. 143-156. GOLDBERG, L.S. AND M.W. KLEIN (1997), “Foreign Direct Investment, Trade and Real Exchange Rate Linkage in Southeast Asia and Latin America”, NBER Working Paper No. 6344. GOLDSTEIN , A. (1999), “Brazilian Privatization in International Perspective: The Rocky Path from State Capitalism to Regulatory Capitalism”, Industrial and Corporate Change , Vol. 8, No. 4. _____. (2001), “Privatization of the Economy”, in N. SMELSER AND P. BALTES (eds), Intenational Encyclopedia of the Social and Behavioral Sciences, Elsevier Science. GONENC, R., M. MAHER AND G. NICOLETTI (2001), “The Implementation and the Effects of Regulatory Reform: Past Experience and Current Issues”, in OECD Economic Studies No. 32, OECD Economics Department, Paris. GOPINATH, M., D. PICK AND J. VASAVADA (1999), “The Economics of Foreign Direct Investment and Trade with an Application to the US Food Processing Industry”, American Journal of Agricultural Economics, Vol. 81, pp. 442-452. GRAHAM, E.M. (2000), Fighting the Wrong Enemy: Antiglobal Activists and Multinational Enterprises, Institute for International Economics, Washington, D.C. H ADDAD, M. AND A.E. HARRISON (1993), “Are There Positive Spillovers from Direct Foreign Investment? Evidence from Panel Data for Morocco”, Journal of Development Economics, Vol. 42, pp. 51-74. H ASKEL , J.E., S.C. PEREIRA AND M.J. SLAUGHTER (2001), “Does Inward Foreign Direct Investment Boost the Productivity of Domestic Firms?”, Paper presented at the International Technology Diffusion/Sloan/ATP Project 118
Reports, NBER Summer Institute 2001 (http://www.nber.org/~confer/2001/si2001/slaughter.pdf). H EAD, K. AND J. RIES (2001), “Overseas Investment and Firm Exports”, Review of International Economics, Vol. 9, pp. 108-122. H ERMES , N. AND R. LENSINK (2000), “Foreign Direct Investment, Financial Development and Economic Growth”, University of Groningen, SOM Theme E Working Papers No. 27 (http://www.ub.rug.nl/eldoc/som/e/00E27/). H IEMENZ, U. et al. (1991), The International Competitiveness of Developing Countries for Risk Capital (Chapter 8), The Kiel Institute for the World Economy, J.C.B. Moh, Tübingen. H ILL, H. AND P-C. ATHUKORALA (1998), “Foreign Investment in East Asia”, AsianPacific Economic Literature, Vol. 12, pp. 23-50. H UBERT, F. AND N. PAIN (2000), “Inward Investment and Technical Progress in the UK Manufacturing Sector”, OECD Economics Department, Working Paper No. 268, Paris. H UFBAUER , G., D. LAKDAWALLA AND A. MALANI (1994), “Determinants of Direct Foreign Investment and Its Connection to Trade”, UNCTAD Review, No. 39, pp. 39-51. ILO (1998), Labour and Social Issues Relating to Export Processing Zones, Geneva. ITO T. AND A.O. KRUEGER (eds.) (2000), The Role of Foreign Direct Investment in Economic Development, University of Chicago Press. JBIC (2001), “Results of the 13th Annual Questionnaire Survey on FDI in Fiscal Year 2001”, mimeo, 26 November 2001 (in Japanese). JENKINSON , T. AND C. MAYER (1988), “The Privatisation Process in France and the UK”, European Economic Review, 32. KATZ, J. (1987), Technology Generation in Latin American Manufacturing Industries, St. Martins Press, New York. KAWAI, M. AND S. URATA (1995), “Are Trade and Direct Investment Substitutes or Complements? An Empirical Analysis of Japanese Manufacturing Industries”, Institute of Social Science, University of Tokyo, Discussion Paper No. F-50, reprinted in H. LEE AND D.W. ROLAND-H OLST (eds) (1998), Economic Development and Co-operation in the Pacific Basin: Trade, Investment and Environmental Issues, Cambridge University Press, Cambridge. KELLER , W. (1996), “Absorptive Capacity: On the Creation or Acquisition of Technology in Development”, Journal of Development Economics, Vol. 49, pp. 199-227. KOKKO, A. (1994), “Technology, Market Characteristics and Spillovers”, Journal of Development Economics, Vol. 43, pp. 279-293. KOKKO, A. (1996), “Productivity Spillovers from Competition between Local Firms and Foreign Affiliates”, Journal of International Development, Vol. 8, pp. 119
517-530. KOKKO, A. AND M. BLOMSTRÖM (1995), “Policies to Encourage Inflows of Technology through Foreign Multinationals”, World Development, Vol. 23, pp. 549-468. KOKKO, A., M. ZEJAN AND R. TANSINI (2001), “Trade Regimes and Spillover Effects of FDI: Evidence from Uruguay”, Weltwirtschaftliches Archiv, Vol. 137, pp. 124-149. KOKKO, A., R. TANSINI AND M. Z EJAN (1996), “Local Technological Capability and Spillovers from FDI in the Uruguayan Manufacturing Sector”, Journal of Development Studies, Vol. 34, pp. 602-611. KUGLER , M. (2000), “The Diffusion of Externalities from Foreign Direct Investment: Theory ahead of Measurement”, Discussion Papers in Economics and Econometrics, Department of Economics, University of Southampton (http://www.soton.ac.uk/~econweb/dp/dp00/). LALL, S. (1974), “Less-Developed Countries and Private Foreign Direct Investment: A Review Article”, World Development, Vol. 2, pp. 43-48. _____. (1980), “Vertical Inter-Firm Linkages in LDCs: An Empirical Study”, Oxford Bulletin of Economics and Statistics, Vol. 42, pp. 203-226. LALL, S. AND P. STREETEN (1977), Foreign Investment, Transnationals and Developing Countries, Macmillan. LEE, H-L. AND M-F. H OUDE (2000), “Recent Trends and Main Characteristics of Foreign Direct Investment in China”, Financial Market Trends, No.77, pp.105-120. LEMOINE, F. (2000), “FDI and the Opening Up of China’s Economy”, working paper No. 11, Centre d’Etudes Prospectives et d’Informations Internationales, Paris. LENSINK, R. AND O. MORRISSEY (2001), “Foreign Direct Investment: Flows, Volatility and Growth in Developing Countries”, CREDIT, University of Nottingham, Research Paper 01-06. LIM , D. (1983), “Fiscal Incentives and Direct Foreign Investment in Less Developed Countries”, Journal of Development Studies, Vol. 19, pp. 207-212. LIPSEY, R.E. (2000), “The Role of Foreign Direct Investment in International Capital Flows”, NBER Working Paper No. 7094, Published: International Capital Flows, Martin Feldstein, ed., Chicago: University of Chicago Press, 1999, pp. 307-331. LIPSEY, R.E. AND M. WEISS (1981), “Foreign Production and Exports in Manufacturing Industries”, Reviw of Economics and Statistics, Vol. 63, pp. 488-494. _____. (1984), “Foreign Production and Exports of Individual Firms”, Review of Economics and Statistics, Vol. 66, pp. 304-308. 120
LIPSEY, R.E., E.D. RAMSTETTER AND M. BLOMSTRÖM (2000), “Outward FDI and Parent Exports and Employment: Japan, the United States, and Sweden”, NBER Working Paper No. 7623. Published: Global-Economy-Quarterly, Vol. 1, No. 4, pp.285-302. LIU, X,.C. WANG, AND Y. WEI (2001), “Causal Links between Foreign Direct Investment and Trade in China”, China Economic Review, Vol. 12, 190-202 LIZONDO, S. (1990), “Foreign Direct Investment”, IMF Working Paper No. 90/63. Published: Aliber,-Robert-Z.; Click,-Reid-W., eds. Readings in international business: A decision approach, Cambridge and London: MIT Press, 1993, pp. 85-113. MARKUSEN , J. (1983), “Factor Movements and Commodity Trade as Complements”, Journal of International Economics, Vol. 14, pp. 341-356. _____. (1995), “The Boundaries of Multinational Enterprises and the Theory of International Trade”, Journal of Economic Perspectives, Vol. 9, pp. 169-189. _____. (2000), “Foreign Direct Investment and Trade Policy”, Centre for International Economic Studies, University of Adelaide, Policy Discussion Paper No. 00-19. MARKUSEN , J., A. VENABLES , D. KONAN AND K. ZHANG (1996), “A Unified Treatment of Horizontal Direct Investment, Vertical Direct Investment and the Pattern of Trade in Goods and Services”, NBER Working Paper No. 5696. MCMILLAN, J. (2001), “Using Markets to Help Solve Public Problems”, mimeo, Graduate School of Business, Stanford University. MCMILLAN, M. (1999), “Foreign Direct Investment: Leader or Follower?”, Tufts University, Department of Economics Discussion Paper 99-01. MEGGISON , W. AND J. N ETTER (2001), “From State to Market: A Survey of Empirical Studies on Privatization”, Fondazione ENI Enrico Mattei, Working Paper No. 01-99. Published: Journal of Economic Literature, Vol. 39 No. 2, pp. 321-389. MORAN, T. (1998), Foreign Direct Investment and Development, Institute for International Economics, Washington, D.C. MUNDELL, R. (1957), “International Trade and Factor Mobility”, American Economic Review, Vol. 51, pp. 321-355. NATARAJAN, S. AND T.J. MIANG (1992), The Impact of MNC Investments in Malaysia, Singapore & Thailand, Singapore: Institute of Southeast Asian Studies. OECD (1998), Foreign Direct Investment and Economic Development: Lessons from Six Emerging Economies, OECD, Paris. _____. (2000), International Trade and Core Labour Standards, Paris. _____. (2001a), The New Economy: Beyond the Hype: Final Report on the OECD Growth Project, Paris (http://www1.oecd.org/subject/growth/products). 121
_____. (2001b), New Patterns of Industrial Globalisation: Cross-Border Mergers and Acquisitions and Strategic Alliances, Paris. _____. (2001c), “Regulation on Investment Incentives: The Impact of Trade Agreements” (TD/TC/WP(2001)38), 12 September, Paris. _____. (2001d), Restructuring Public Utilities for Competition, Paris. _____. (2001e), “Privatisation Trends”, in Financial Market Trends, No.79, pp.4365, Paris. OKAMOTO, Y. (1999), “Multinationals, Production Efficiency and Spillover Effects: The Case of the U.S. Auto Parts Industry” Weltwirtshaftliches Archiv, Vol. 135, pp. 241-260. OKAMOTO, Y. AND F. SJÖHOLM (1999), “FDI and the Dynamics of Productivity: Microeconomic Evidence”, Stockholm School of Economics, Working Paper Series in Economics and Finance No. 348. OMAN, C. (2000), Policy Competition for Foreign Direct Investment, Development Centre Studies, OECD Development Centre, Paris. _____. (2001), “Corporate Governance and National Development”, Technical Paper No. 180, OECD Development Centre, Paris. PACK, H. AND K. SAGGI (1997), “Inflows of Foreign Technology and Indigenous Technical Development”, Review of Development Economics, Vol. 1, pp. 8198. PACK, H. AND K. SAGGI (1999), “Exporting, Externalities and Technology Transfer”, World Bank, Policy Research Working Paper Series, No. 2065. PAIN , N. AND K. WAKELIN (1998), “Export Performance and the Role of Foreign Direct Investment”, The Manchester School, Vol. 66, Supplement, pp. 62-88. PEREZ, T. (1998), Multinational Enterprises and Technological Spillovers, Harwood Academic Publishers, Amsterdam. PETRI, P.A. AND M.G. PLUMMER (1998), “The Determinants of Foreign Direct Investment: A Survey with Applications to the United States”, in H. LEE AND D.W. ROLAND-H OLST (eds), Economic Development and Co-operation in the Pacific Basin (Chapter 7), Cambridge University Press, Cambridge. PFAFFERMAYR , M. (1994), “Foreign Direct Investment and Exports: A Time Series Approach”, Applied Economics, Vol. 26, pp. 337-351. PFAFFERMAYR , M. (1996), “Foreign Outward Direct Investment and Exports in Austrian Manufacturing: Substitutes or Complements?”, Weltwirtschaftliches Archiv, Vol. 132, No. 3, pp. 501-521. PIRNIA , N. (1996), “The Role of Investment Incentives in Influencing Investors’ Locational Choices: A Literature Survey”, Foreign Investment Advisory Service, International Finance Corporation (World Bank Group). RADOSEVIC , S. (1999), International Technology Transfer and Catch-up in Economic Development, Edward Elgar, Cheltenham, UK. 122
RASIAH, R. (1994), “Flexible Production Systems and Local Machine Tool Subcontracting: Electronics Component Multinationals in Malaysia”, Cambridge Journal of Economics, Vol. 18, pp. 279-98. _____. (2001), “Government-Business Coordination and Small Business Performance in the Machine Tools Sector in Malaysia”, WBI Working Papers, World Bank Institute, June. RAYNAULD, A. AND J.P. VIDAL (1998), Labour Standards and International Competitiveness: A Comparative Analysis of Developing and Industrialised Countries, Cheltenham: Edward Elgar. REISEN , H. AND M. SOTO (2001), “Which Types of Capital Inflows Foster Developing-Country Growth?”, International Finance, Vol. 4, No. 1. REUBER , G., H. CROOKELL, M. EMERSON AND G. GALLAIS -HAMONNO (1973), Private Foreign Investment in Development, Clarendon Press, Oxford. RODRÍGUEZ-CLARE, A. (1996), “Multinationals, Linkages and Development”, American Economic Review, Vol. 86, pp. 852-873. ROLFE, R.J., D. RICKS , M. POINTER AND M. MCCARTHY (1993), “Determinants of FDI Incentive Preference for Multinational Enterprises”, Journal of International Business Studies, Vol. 24, No. 2, pp. 335-55. SAGGI, K. (2000), “Trade, Foreign Direct Investment and International Technology Transfer: A Survey”, World Bank, Policy Research Working Paper No. 2349. SAZANAMI, Y., S. YOSHIMURA AND K. KIYOTA (2001), “Japanese Foreign Direct Investment Flows to East Asia and the Real Exchange Rate – Lessons from the Asian Financial Crisis”, Paper presented at the 3rd International Conference organised by Université de Le Havre and Inha University, Le Havre, 26-27 September. SHAH, A. (ed.) (1992), Fiscal Incentives for Investment in Developing Countries, World Bank, Washington, D.C. SLAUGHTER , M.J. (2000), Worker Perceptions and Pressures in a Global Economy, mimeo, June, forthcoming from Institute for International Economics, Washington, D.C. SLEUWAEGEN , L. (1998), “Foreign Direct Investment, Cross-Border Mergers and Competition Policy in the European Union”, in K. FUKASAKU , F. K IMURA AND S. URATA (eds), Asia and Europe, Sussex Academic Press, Chapter 8, pp. 180-201. SOTO, M. (2000), “Capital Flows and Growth in Developing Countries: Recent Empirical Evidence”, Technical Paper No. 160, OECD Development Centre, Paris. SPICER , A., G. MCDERMOTT AND B. KOGUT (2000), “Entrepreneurship and Privatization in Central Europe: The Tenuous Balance between Destruction and Creation”, Wharton School, Reginald H. Jones Center Working Paper, No. 123
99-04. Published: Academy of Management Review, Vol. 25, No. 3, pp. 630649 SVENSSON , R. (1996), “Effects of Overseas Production on Home Country Exports: Evidence Based on Swedish Multinationals”, Weltwirtschaftliches Archiv, Vol. 132, pp. 304-329. TEJIMA, S. (2000a), “The Effects of the Asian Crisis on Japan’s Manufacturing Foreign Direct Investment in Asia”, in V. BLECHINGER AND J. LEGEWIE (eds), Facing Asia – Japan’s Role in the Political and Economic Dynamism of Regional Co-operation, Deutsches Institut für Japanstudien, München, pp. 199-216. _____. (2000b), “Japanese FDI, the Implications of ‘Hollowing Out’ on the Technological Development of Host Countries”, International Business Review, Vol. 9, pp. 555-570. THOMPSON , E.R. AND J.P.H. POON (2000), “ASEAN after the Financial Crisis: Links between Foreign Direct Investment and Regulatory Change”, ASEAN Economic Bulletin, Vol. 17, pp. 1-14. UNCTAD (1996), World Investment Report — Investment, Trade and International Policy Arrangements, United Nations, New York and Geneva. _____. (1999), World Investment Report — Foreign Direct Investment and the Challenge of Development, United Nations, New York and Geneva. _____. (2000), World Investment Report 2000 — Cross-Border Mergers and Acquisitions and Development (Chapter XI), United Nations, New York and Geneva. _____. (2001), World Investment Report 2001 — Promoting Linkages, United Nations, New York and Geneva. UNCTC (1981), Transnational Corporations Linkages in Developing Countries: The Case of Backward Linkages via Subcontracting, New York. _____. (1992), The Determinants of Foreign Direct Investment: A Survey of Evidence, New York. URATA, S. (1999), “Intrafirm Technology Transfer by Japanese Multinationals in Asia”, in D.J. ENCARNATION (ed.), Japanese Multinationals in Asia: Regional Operations in Comparative Perspective , Oxford University Press, pp. 143162. _____. (2001), “Emergence of an FDI-Trade Nexus and Economic Growth in East Asia”, in World Bank (2001), Rethinking the East Asian Miracle, Chapter 12, Washington D.C. (forthcoming). URATA, AND H. KAWAI (2000), “Intra-Firm Technology Transfer by Japanese Manufacturing Firms in Asia”, in T. ITO AND A.O. KRUEGER (eds), The Role of Foreign Direct Investment in Economic Development, University of Chicago Press. 124
URATA, S. AND H. KAWAI (2001), “Overseas R&D Activities and Intra-firm Technology Transfer: The Case of Japanese Multinationals”, mimeo. USHA NAIR , R. AND D. WEINHOLD (2001), “Causality Tests for Cross-Country Panels: New Look at FDI and Economic Growth in Developing Countries”, Georgia Institute of Technology, Center for International Business Education and Research Working Papers No. 99/00-012. Published: Oxford-Bulletin-ofEconomics-and-Statistics, Vol. 63, No. 2, pages 153-71. VAN TULDER , R. D VAN DEN BERGHE AND A MULLER (2001), “Erasmus (S)coreboard of Core Companies: The World's Largest Firms and Internationalisation”, Rotterdam: Erasmus University/Rotterdam School of Management. VERNON , R. (1966), “International Investment and International Trade in the Product Life Cycle”, Quarterly Journal of Economics, Vol.80, pp.190-207. WANG, J.-Y. AND M. BLOMSTRÖM (1992), “Foreign Direct Investment and Technology Transfer”, European Economic Review, Vol. 36, pp. 137-155. WATANABE, S. (1983), “Technological Linkages through Subcontracting in Mexican Industries”, in WATANABE S. (ed.), Technology Marketing and Industrialisation: Linkages between Small and Large Enterprises, Macmillan, New Delhi. WONG, K. (1986), “Are International Trade and Factor Mobility Substitutes?”, Journal of International Economics, Vol. 21, pp. 25-43. WORLD BANK (2000), Global Development Finance 2000, Washington, DC. _____. (2001), Global Development Finance 2001, Washington, DC. WTO (1998), “The Effects of Foreign Direct Investment on Development: Technology and Other Know-How Transfers and Spillovers”, Note by the Secretariat No. WT/WGTI/W/65. _____. (2001), “Foreign Direct Investment and the Balance of Payments”, WT/WGTI/W/95. XU, B. (1999), “Capital Goods, Trade and R&D Spillovers in the OECD”, Canadian Journal of Economics, Vol. 32, pp. 1258-1274. _____. (2000), “Multinational Enterprises, Technology Diffusion, and Host Country Productivity Growth”, Journal of Development Economics, Vol. 62, pp. 477493. YAMAWAKI, H. (1991), “Exports and Foreign Distributional Activities: Evidence on Japanese Firms in the United States”, Review of Economics and Statistics, Vol. 73, pp. 294-300. ZHANG, K. (2001), “Does Foreign Direct Investment Promote Economic Growth? Evidence from East Asia and Latin America”, Contemporary Economic Policy, Vol. 19, No. 2, pp. 175-185.
JAPAN BANK FOR INTERNATIONAL COOPERATION (JBIC) http://www.jbic.go.jp/ 4-1, Ohtemachi 1-Chome Chiyoda-ku, Tokyo 100-8144, Japan Telephone +81-3-5218-9720 (JBIC Institute) Facsimile +81-3-5218-9846 Osaka Branch 13th Fl., Aqua Dojima East 4-4, Dojimahama 1-Chome Kita-ku, Osaka 530-0004, Japan Telephone +81-6-6346-4770 Facsimile +81-6-6346-4779 Overseas Network Representative Office in Beijing 3131, 31st Fl., China World Trade Center, No.1 Jian Guo Men Wai Ave., Beijing 100004, The People's Republic of China Tel. 86-10-6505-8989, 3825∼8, 1196, 1197 Fax.86-10-6505-3829,1198
Representative Office in Jakarta Summitmas II 7th Fl., Jl. Jenderal Sudirman, Kav. 61-62, Jakarta Selatan, Jakarta, Indonesia Tel. 62-21-522-0693 Fax.62-21-520-0975
Hong Kong Office Suite 3706, Level37, One Pacific Place, 88 Queensway, Hong Kong Tel. 852-2869-8505∼7 Fax.852-2869-8712
Representative Office in Kuala Lumpur 22nd Fl., UBN Tower, Letter Box No.59, Jalan P, Ramlee 50250, Kuala Lumpur, Malaysia Tel. 60-3-2072-3255,2201,2202 Fax.60-3-2072-2115
Representative Office in Bangkok 14th Fl., Nantawan Building, 161 Rajdamri Road, Bangkok 10330, Thailand Tel. 66-2-252-5050 Fax.66-2-252-5514,5515
Representative Office in Manila Office 31st / Floor, Citibank Tower, Valero St. corner Villar St., Makati, Metro Manila, Philippines Tel. 63-2-848-1828, 752-5682 Fax.63-2-848-1833∼35
Representative Office in Hanoi 6th Fl., 63 Ly Thai To Street, Hanoi, Viet Nam Tel. 84-4-8248934∼6 Fax.84-4-8248937
Representative Office in Singapore 9 Raffles Place, #53-01 Republic Plaza, Singapore 048619 Tel. 65-6557-2806 Fax. 65-6557-2807 Representative Office in Colombo Level 13, Development Holdings 42, Navam Mawatha, Colombo 2, Sri Lanka Tel. 94-1-300470, 300471, 300472 Fax.94-1-300473
Representative Office in Dhaka IDB Bhaban (5th Floor), E/8-A, Begum Rokeya Sharani, Sher-E-Bangla Nagar, Dhaka-1207, Bangladesh Tel.880-2-811-4081, 6700 (Direct) Fax.880-2-811-3336（Direct ） Representative Office in Islamabad House No. 62, Ataturk Avenue, G- 6/3, Islamabad, Pakistan Tel. 92-51-2820119 Fax.92-51-2822546 Representative Office in New Delhi 3rd Fl., DLF Centre, Sansad Marg, New Delhi 110001, India Tel. 91-11-371-4362, 4363, 7090, 6200 Fax.91-11-371-5066, 373-8389 Representative Office in Sydney Suite 2501, Level 25, Gateway 1 Macquarie Place, Sydney, N.S.W. 2000, Australia Tel. 61-2-9241-1388 Fax.61-2-9231-1053 Representative Office in Moscow 123610 Moscow, Krasnopresnenskaya Nab.12, World Trade Center, Office No. 905, Russian Federation Tel. 7-095-258-1832, 1835, 1836 Fax.7-095-258-1858
Representative Office in Frankfurt Taunustor 2, 60311 Frankfurt am Main, Germany Tel. 49-69-2385770 Fax.49-69-23857710 Representative Office in London 4th Fl., River Plate House, 7- 11 Finsbury Circus, London, EC2M 7EX, U.K. Tel. 44-20-7638-0175 Fax.44-20-7638-2401 Representative Office in Paris 21, Boulevard de la Madeleine 75038 Paris Cedex 01, France Tel. 33-1-4703-6190 Fax.33-1-4703-3236 Representative Office in Cairo Abu El Feda Bldg., 16th Fl., 3 Abu El Feda Street, Zamalek, Cairo, Egypt Tel. 20-2-738-3608∼9 Fax.20-2-738-3607 Representative Office in Nairobi 6th Fl., International House, Mama Ngina Street, P.O. Box 49526, Nairobi, Kenya Tel. 254-2-221420, 221637 Fax.254-2-221569
Representative Office in Los Angeles 601 South Figueroa Street, Suite 4590, Los Angeles, CA 90017, U.S.A Tel. 1-213-627-3500 Fax.1-213-627-3900
Representative Office in Lima Av.Central,643, San Isidro Lima 27, Peru Tel. 51-1-442-3031 Fax.51-1-440-9657
Representative Office in New York 520 Madison Avenue, 40th Floor, New York, NY 10152, U.S.A. Tel. 1-212-888-9500, 9501, 9502 Fax.1-212-888-9503
Representative Office in Mexico City Paseo de la Reforma 265 Piso-16, Col. Cuauhtemoc, Mexico, D.F. 06500, Mexico Tel. 52-55-5525-67-90 Fax.52-55-5525-34-73
Representative Office in Washington, D.C. 1909 K st., N.W., Suite 300, Washington, D.C. 20006, U.S.A. Tel. 1-202-785-5242 Fax.1-202-785-8484 Representative Office in Bogota Calle 100, 8A-55, Oficina 715, World Trade Center, Torre C, Santafé de Bogotá, D.C., Colombia Tel. 57-1-621 -1828, 1838, 1845, 1849, 1878, 1889 Fax.57-1-610-7293 Representative Office in Buenos Aires Ave. del Libertador No.498, Piso19, 1001 Capital Federal Buenos Aires, Argentina Tel. 54-11-4394-1379, 1803 Fax.54-11-4394-1763
Representative Office in Rio de Janeiro Praia de Botafogo, 228-801 B,Botafogo, CEP.22359-900, Rio de Janeiro, RJ,Brazil Tel. 55-21-2553-0817 Fax.55-21-2554-8798 Toronto Liaison Office P.O.Box493, 2 First Canadian Place, Suite3660, Toronto, Ontario, M5X 1E5, Canada Tel. 1-416-865-1700 Fax.1-416-865-0124
JBICI Research Paper 1 1.
Issues of Sustainable Economic Growth from the Perspective of the Four East Asian Countries, December 1999
Organizational Capacity of Executing Agencies in the Developing Countries: Case Studies on
Bangladesh, Thailand and Indonesia, December 1999 Urban Development and Housing Sector in Viet Nam, December 1999
Urban Public Transportation in Viet Nam: Improving Regulatory Framework, December 1999*
Current Situation of Rice Distribution System in Indonesia, December 19992 Energy Balance Simulations to 2010 for China and Japan, March 2000
Rural Enterprises Finance: A Case Study of the Bank for Agriculture and Agricultural
Cooperatives (BAAC) in Thailand, November 20003 8-1. Issues of Sustainable Development in Asian Countries: Focused on SMIs in Thailand, December 2000 8-2. Issues of Sustainable Development in Asian Countries: Focused on SMIs in Malaysia, December 2000 9.
Policy Issues and Institutional Reform in Road Sector in Developing Countries, February 20014
10. Public Expenditure Management in Developing Countries, March 2001 11. INDIA: Fiscal Reforms and Public Expenditure Management, August 2001 12. Cash Crop Distribution System in the Philippines - Issues and Measures to Address Them –, March 2002 13. MERCOSUR Experience In Regional Freight Transport Development, March 2002 14. Regional Cooperation for Infrastructure Development in Central and Eastern Europe, March 20025 15. Foreign Direct Investment and Development : Where Do We Stand?, June 2002
*No.1~No.14 were published as ‘JBIC Research Paper Series’. JBIC Institute, Japan Bank for International Cooperation 4-1, Ohtemachi 1-chome, Chiyoda-ku, Tokyo 100-8144, Japan Tel: 03-5218-9720, Fax: 03-5218-9846 (Planning and Coordination Division) Internet: http://www.jbic.go.jp/
Full texts can be downloaded from JBIC homepage. In Japanese only. English summary can be downloaded from JBIC homepage. 3 In Japanese only 4 In Japanese only 5 In Japanese only 2