46448 Competitiveness, FDI and Technological Activity in East Asia Edited by Sanjaya LaB Professor of Development Economics at the International Development Centre, Queen Elizabeth House and a Fellow of Green College, Oxford University and Shujiro Urata Professor of Economics at Waseda University, Japan IN ASSOCIATION WITH THE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENTITHE WORLD BANK Edward Elgar Cheltenham, UK · Northampton, MA, USA © The International Bank for Reconstruction and Developmentffhe World Bank,2003 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any lorm or by any means, electronic. mechanical or photocopying. recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited Glensanda House Montpellier Parade Cheltenham Glos GL50 IUA UK Edward Elgar Publishing. Inc. 136 West Street Suite 202 Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Cataloguing in Publication Data Competitiveness, FDI and technological activity in East Asia / edited by Sanjaya Lall and Shujiro Urata. p.cm. "In association with thelnternational Bank for Reconstruction and Developmentffhe World Bank." Includes index. 1. Research. Industrial-Economic aspects-East Asia. 2. Technological innovations-Economic aspects-East Asia. 3. Investments, Foreign-East Asia. 4. Competition-East Asia. T. Lall. Sanjaya. II. Urata, Shujiro, 1950 HC460.5.Z9R432003 338'.064 ' 095-dc21 2003048851 ISBN I 84376 1149 Printed and bound in Great Britain by MPG Books Ltd, Bodmin. Cornwall Contents List offigures vii List of tables ix List of contributors xiii Preface and acknowledgements xvii Introduction and overview Sanjaya La!! and Shujiro Urata 2 Foreign direct investment, technology development and competitiveness: issues and evidence 12 Sanjaya Lall 3 Competitiveness and technology: an international comparison 57 Hiraki Kawai and Shujiro Urata 4 Building technological capabilities with or without inward direct investment: the case of Japan 83 Akira Goto and Hiroyuki Odagiri 5 Overseas R&D activities and intra-firm technology transfer: the case of Japanese multinationals 103 Shujiro Urata and Hiroki Ka>\'ai 6 The dynamics of technology development: lessons from the Korea experience 143 Linsu Kim 7 Technology acquisition and development in Taiwan 168 Bee- Yan Aw 8 From using to creating technology: the evolution of Singapore's national innovation system and the changing role of public policy 191 Poh Kam Wong 9 In search of balance: technological development in China 239 Yang Yao 10 Can the Philippines ever catch up? 268 Joy V Abrenica and Gwendolyn R. Tecson v vi Competitiveness, FD! and technological activity in East Asia 11 Industrial technology transition in Malaysia 305 Rajah Rasiah 12 Foreign direct investment, technology and competitiveness in Thailand 334 Peter Brimble 13 Technology development in Indonesia 375 Yumiko Okamoto and Fredrik SjOholm Index 397 Figures 2.1 Total and high-technology manufacturing: annual growth rates, 1985-97 14 2.2 Shares of MT and HT products in manufactured exports (1989) for 13 largest exporters 30 2.3 R&D financed by productive enterprises, 1995 42 2.4 RCA in high-technology exports and main industrial strategies, 1995 45 4.1 Number of new technical agreements and payments for technology imports, 1952-84 90 5.1 Japanese foreign direct investment by region, 1981-98 105 5.2 Japanese foreign direct investment by industry, 1981-98 105 6.1 Strategy for technology 165 7.1 International spillovers 177 8.1 Analytical framework for national innovation systems 196 8.2 Stylized profile of technological capabilities of Singapore over the four phases of NIS development 207 8.3 Emerging institutional framework for S&T policy in Singapore as of 2001 227 8.4 R&D funding priorities according to the National Science and Technology Plan, 1996-2000 229 9.1 Comparison of the distribution of R&D personnel in the USA, South Korea and China, 1996 250 9.2 Comparison of the distribution of R&D expenditure in the USA, ED, Japan and China, 1996 251 10.1 Distribution of Philippine exports, 1980-99 271 10.2 Composition of Philippine manufactured exports, by factor intensity 271 10.3 RCA of Philippine manufactured exports, 1965-96 272 lOA Composition of Philippine manufactured exports, by technology intensity 272 10.5 Composition of high-technology manufactured exports 273 12.1 FDI, technology and competitiveness, framework for analysis 335 12.2 Key thresholds in the current structure of intra-firm technological capabilities 366 vii viii Competitiveness, FDI and technological activity in East Asia 13,1 Total and manufactured exports in Indonesia. 1980~98 376 13.2 Inward FDI and manufactured exports in Indonesia, 1980~98 385 13.3 Machinery imports in Indonesia, 1981~97 388 Tables 2.1 Technological classification of exports 24 2.2 Growth rates and market shares of exports 25 2.3 Regional shares of developing countries' manufactured exports 26 2.4 Distribution of manufactured exports over technological categories, 1985 and 1998 29 2.5 RCAs for leading 13 developing country exporters by technological category, 1985 and 1998 32 2.6 Tertiary level enrolments and enrolments in technical subjects, 1995 38 2.7 Technical tertiary enrolments by country, 1995 39 2.8 R&D propensities and manpower in major country groups 41 2.9 FDI as percentage of gross domestic investment 43 2A.I Leading I3 exporters of manufactures in 1998, values of exports in 1985 and 1998 52 2A.2 Leading developing country exporters in 1998, distribution by technological category and growth rates, I 985~98 54 3.1 Level and growth of per capita GDP, 1970~97 59 3.2 Level and growth of total factor productivity, 1970--97 61 3.3 Educational attainment 64 3.4 Researchers and R&D 66 3.5 FDI inflows 69 3.6 Imports of machinery as a share of total imports 70 3.7 Trade (exports + imports) 71 3.8 Royalty payments 72 3.9 The determinants of growth rates of per capita GDP 74 3.10 The determinants of TFP growth 78 4.1 Indicators of R&D in selected countries, 1995 85 4.2 County share of world exports (by technology) 85 5.1 Overseas affiliates of Japanese firms 108 5.2 Overseas R&D units of Japanese firms Il2 5.3 Overseas R&D by Japanese firms 114 5.4 Researchers at overseas affiliates by Japanese firms 116 5.5 Functions of affiliates' research centres 118 5.6 The determinants of establishment of overseas R&D units 122 ix x Competitiveness, FDI and technological activity in East Asia 5.7 The determinants of R&D expenditure at overseas affiliates of Japanese firms 124 5.8 The determinants of the number of researchers at overseas affiliates of Japanese firms 126 5.9 Training programmes at overseas affiliates 130 5.10 Technology transfer at overseas affiliates of Japanese firms 132 5.11 The determinants of training programmes at overseas affiliates of Japanese firms 134 5.12 The determinants of the extent of technology transfer achieved 138 6.1 Korea's exports by technology intensity 144 6.2 Foreign technology transfer to Korea, 1962~99 147 6.3 Basic parameters and performance between Hyundai and Daewoo, 1982 149 6.4 Research and development expenditures, 1965~98 154 6.5 Patent applications and patents granted 156 7.1 Foreign direct investment 170 7.2 R&D expenditure indicators 174 7.3 Firms' views about the most effective ways in which government could promote a satisfactory technology level 175 7.4 Mean characteristics of Taiwanese manufacturing industries, 1986 180 7.5 Coefficient estimates of TFP growth regression by firm size 183 7.6 Coefficient estimates of TFP growth regression for non- exporters and non-FDI firms 184 8.1 Aggregate economic growth performance, 1960~2000 192 8.2 Singapore GDP distribution by sectors, 1960~2000 193 8.3 Comparative R&D indicators, Singapore and selected OECDI Asian NTEs 194 8.4 Average output of technical manpower from tertiary education institutions in Singapore, 1970~99 201 8.5 Growth of R&D in Singapore, 1978~99 202 8.6 Number of organizations performing R&D, 1978~99 203 8.7 R&D expenditure by sectors, 1978-99 204 8.8 Growth of Singapore's venture capital industry, 1985~2000 205 8.9 R&D output indicators for Singapore, 1993~99 206 8.10 Singapore's ranking in technological capability-related indicators 209 8.11 Distribution of private sector R&D expenditure by industry, 1993-99 213 8.12 Foreign companies' share of industry R&D expenditure, 1993~99 215 Tables Xl 8.13 Foreign companies' share of industry R&D expenditure, 1999 215 8.14 Deepening of Singapore's R&D system 221 9.1 TFP in China, 1953~95 244 9.2 China's export composition, 1985-98 247 9.3 The ten most exported products in China, 1985~99 248 9.4 Distribution of R&D personnel in Chinese research institutions 252 9.5 Distribution of R&D expenditure in Chinese research institutions 253 9.6 Sectoral distribution of institute R&D personnel in the Chinese manufacturing sector 253 9.7 Institute R&D expenditure in the Chinese manufacturing sector 254 9.8 Fund distribution of the '863' and Torch projects in 1997 255 9.9 Commercial applications of '863' projects, 198~96 255 9.10 Key Technology Project in 1997 256 9.11 The distribution of the Star Project in 1997 257 9.12 Regional distribution of technology exporters and selected importers, 1993 260 9.13 FDI concentration in the Chinese manufacturing sector, 1985 and 1995 263 10.1 Exports of technology-intensive manufactures, selected Asian economies 270 10.2 Measures of human capital formation 274 10.3 Comparative S&T performance of selected Asian countries 276 lOA Entry of MNCs in the electronics industry, 1994~96 277 10.5 Net foreign investment inflows in selected Asian economies 279 10.6 TFP growth rates in selected Asian economies 280 10.7 Inflows of foreign direct investment in the manufacturing sector 281 10.8 Selected ratios in the economics industry, 1974-94 282 10.9 Work done by others as a percentage of total output, 1994 283 11.1 Sectoral employment, output and export, selected sectors, 1970-98 308 11.2 Manufacturing export structure, Malaysia, 1968~97 309 11.3 Malaysia's share of world electrical and electronic exports, 1995 310 11.4 Manufacturing trade ratios, Malaysia, 1968-97 311 11.5 Foreign fixed assets ownership, Malaysian manufacturing, 1970~98 313 11.6 Educational enrolment, selected economies, 1970~92 316 xii Competitiveness, FDI and technological activity ill East Asia 11.7 Public education and technical orientation, selected economies 317 11.8 High-technology exports and patent applications filed 318 11.9 R&D in selected industries, Malaysia, 1992 319 11.10 Technology transfer agreements by industry, Malaysia, 1975-2000 321 ILl I Scientists and engineers/population 322 11.12 Research and development statistics 323 12.1 Thailand's key macroeconomic indicators, 1995-2000 337 12.2 Changes in Thailand's distribution of GDP by sector as compared to the NIEs 340 12.3 Distribution of manufactured exports by technological categories 341 12.4 Inward and outward FDI flows in Thailand 345 12.5 Net flows of foreign direct investment in Thailand by sector 346 12.6 Net flows of foreign direct investment in Thailand, by region/country 350 12.7 Foreign investor interest in Thailand: BOI approvals 352 12.8 Rankings of selected technology indicators, 2000 360 12.9 Computer and internet indicators 362 12.10 R&D comparisons in Asia, 1999 363 13.1 Average annual growth rates of manufactured exports 1990-96 and sectors' shares of total manufacturing export 378 13.2 Distribution of manufactured exports by technological categories, 1995 380 13.3 R&D expenditures as percentage of GNP in Indonesia and other selected Asian countries 380 13.4 R&D intensity by industry, 1995 382 13.5 Foreign presence in the Indonesian manufacturing sector 384 13.6 Sources of technological capability in seven Indonesian textile firms 387 13.7 Human resource development (HRD) in the Indonesian manufacturing sector, 1995 391 Contributors Dr. Joy V. Abrenica Society for the Advancement of Technology Management in the Philippines Room 249, School of Economics University of the Philippines Diliman Diliman. Quezon City 1101 Philippines Tel: 63-2-435-6313/16 Fax: 63-2-435-6313; or 63-2-921-3359 Email: satmp@skyinet.net Dr. Peter Brimble President The Brooker Group Limited 2 Floor, Zone D, Room 201/2 Queen Sirikit National Convention Center, 60 New Rachadapisek Road Klongtoey, Bangkok 10110, Thailand Tel: 66-2-229-3111 Fax: 66-2-229-3127 Email: peterb@Loxl.Loxinfo.co.th Professor Akira Goto Research Center for Advanced Economic Engineering (AEE) University of Tokyo Email: agoto@aee.u-tokyo.ac.jp Professor Hiroki Kawai John F. Kennedy School of Government Harvard University Belfer 315, 79 JFK Street Cambridge, MA 02138, USA Tel: 1-617-495-1397 Fax: 1-617-495-1635 Email: hk@econ.keio.ac.jp xiii xiv Competitiveness, FDI and technological activity in East Asia Professor Linsu Kim Professor of Management College of Business Administration Korea University, Seoul, Korea 136-70 I Tel: 82-2-3290-1919 Fax: 82-2-921-7246 Email: linsukim@unitel.co.kr Professor Sanjaya Lall Queen Elizabeth House University of Oxford 21 St. Giles Oxford OXl 3LA, UK Tel: 44-1865-559614 Fax: 44-1865-559614 Email: sanjaya.lall@economics.ox.ac.uk Professor Hiroyuki Odagiri Graduate School of Economics Hitotsubashi University Kunitachi Tokyo 186-8601 Japan Tel: 81-42-580-8813 Fax: 81-42-580-8882 Email: odagiri@econ.hit-u.ac.jp Professor Yumiko Okamoto Graduate School of International Development Nagoya University Furo-cho, Chikusa-ku, Nagoya Japan 464-8601 Tel: 81-52-789-5096 Fax: 81-52-789-4951 Email: plokamotoy@M.Gsid.nagoya-u.ac.jp Contributors xv Professor Rajah Rasiah Professor and Senior Research Fellow Institute for New Technologies United Nations University 6211-TC Maastricht Netherlands Tel: 31-43-3506330 Fax: 31-43-3506399 Email: Rasiah@intech.unu.edu Professor Bee-Yan Aw Department of Economics 501 Kern Graduate Building Pennsylvania State University University Park, PA 16802, USA Tel: 1-814-863-1996 Fax: 1-814-863-4775 Email: byr@psu.edu Professor Fredrik Sjoholm Email: fredrik.sjoholm@hhs.se Professor Gwendolyn R. Tecson School of Economics University of the Philippines Diliman, Quezon City 110 I Philippines Tel: 63-2-920-5481 Fax: 63-2-920-5462 Email: gtecson@pacific.net.ph Professor Shujiro Urata School of Social Sciences Waseda University 1-6-1 Nishiwaseda, Shinjuku-ku Tokyo, 169-8050,Japan Tel: 81-35-286-1414 Fax: 81-33-208-1032 Email: surata@mn.waseda.ac.jp xvi Competitiveness, FDI and technological activity in East Asia Professor Poh Kam Wong Director Center for Management of Innovation & Technopreneurship (CMIT) Associate Professor Department of Business Policy Business School National University of Singapore Singapore 117591 Tel: 65-874-6323 Fax: 65-775-3955 Email: tbawpk@nus.edu.sg Professor Yang Yao Associate Professor China Center for Economic Research Beijing University Beijing 100871 China Tel: 86-\ 0-6275-3 \03 Fax: 86-10-6275-1474 Email: yyao@ccer.pku.edu.cn Preface and acknowledgements This study on competitiveness, FDI and technological activity in East Asia is the outcome of a research project organized by the World Bank Institute under the Brain Trust Program which is financed by the government of Japan through its Policy and Human Resources Development Trust Fund. The principal objective of the Brain Trust Program is to conduct studies on the Japanese and East Asian development management experience and to disseminate the lessons of this experience to developing and transition countries. The current study is one of a series of such projects undertaken over the past decade. The objective of this study is to examine the degree to which foreign direct investment (FDr) and technological activity have contributed to export competitiveness and economic growth in East Asia. The links between export competitiveness and its main contributory factors, namely FDI and domestic technological effort which include R&D, learning-by doing, adaptation and copying have not yet been fully explored. The ways in which these links are forged differ among countries. Some coun tries have placed less emphasis on FDI and the presence of transnational companies (TNCs), relying instead on building domestic technological capacity through R&D efforts, adaptation and so on. Some others have depended largely on TNC presence for their technology development and upgrading. These differences in the strategies adopted by countries in their technol ogy development pose two important questions. They are: (i) what are the most effective ways in which technology transfer could take place through FDI? and (ii) how to adopt alternative ways of technology development in lieu of FDI? The first alternative - where technology transfer and market informa tion are imported through FDI has several forms. These forms include inward FDI, or externalized methods such as licensing, subcontracting, original equipment manufacturing (OEM) arrangements and so on. Inward FDI transfers take place between TNCs and their overseas affiliates and are called intra-jirm technology transfers. The externalized methods come under technology spillover where technology transfer takes place between TNCs' overseas affiliates and local firms. In the case of Japanese multinational corporations (MNCs), intra-firm technology xvii xviii Competitil'eness, FDI and technological actil'ity in East Asia transfer was quite common, while in the case of countries like Indonesia and Malaysia, spillover was the predominant form of technology trans fer from TNCs. The second alternative is to place more emphasis on domestic technolog ical effort. This involves building R&D capabilities, adaptation of new tech nology, copying and so on. Korea, for instance, took the path of building domestic R&D capabilities instead of relying on FDI for its technological growth. Its total R&D budget as a percentage of GOP has been quite high. Some other countries have taken an informal approach to technology trans fer by relying mostly on incremental improvements, learning-by-doing, adaptation and copying. There is, however, little correlation between suc cessful technology development and the existence of a strong domestic tech nological base. A number of countries have been able to undertake highly sophisticated export activity - such as exporting electronic equipment - by specializing in the final assembly of products by foreign affiliates even when they lacked a strong domestic technological base. Some others have encour aged affiliates to undertake advanced processes and design the products locally and even launch advanced R&D. A third category of countries has embarked on the path of developing local capabilities and networks which allow them to keep up with fast-moving technologies without having to rely on FDI. The alternatives chosen by each country depend on a number of factors. These include, inter alia, the prevailing policy environment, domestic tech nological capacity, development objectives and the institutional frame work. While these factors are unique to each country, it is possible to distill some general lessons and policy guidelines by looking at some of the country experiences in East Asia. For this reason, a number of countries have been identified for case studies. They include Japan, Korea, Taiwan, Singapore, Malaysia, Thailand, the Philippines and China. Technological progress is achieved through a continuous upgrading of technology, information and skills. The process becomes more complex in an environment where both export competition and technical change take place simultaneously and at very high levels. Depending on how this process of technological progress is managed, each economy develops a distinctive pattern of exports over time. These export patterns differ in terms of the product, market and technological specialization. The significant differences in the levels of technology transferred, upgrading of local content over time, and R&D undertaken by affiliates are influenced by the prevailing export patterns and vice versa. Similarly, the main agents responsible for technology transfer are also, to a large extent, influenced by these patterns. For instance, in Singapore, China, Malaysia, Thailand and Indonesia, TNCs have been the main agents of technology transfer Preface and acknowledgements xix whereas in Korea, Taiwan and Hong Kong, domestic firms have played the central role in this regard. In spite of the different strategies used, there is a rapid convergence in policy objectives as a result of the intensifying competitive pressures and changing 'rules of the game' on trade and investment. Countries which have labour-intensive TNCs are beginning to find that it is difficult to sustain their export growth without deepening local content, skill upgrad ing and increasing local R&D activity. The others with strong indigenous technological bases are also realizing that as technological innovation becomes more expensive and specialized it is imperative to form alliances with technology leaders or rely more on FDT. The most common response by countries to these emerging trends is to encourage labour-intensive processes to relocate to cheaper areas and thereby to undertake outward FDI of their own. A study such as this cannot be completed without incurring a heavy debt of gratitude. All those who participated in the two workshops in Bali, Indonesia and Sapporo, Japan have made immense contributions to this study at various stages. We owe special gratitude to Mr Farrukh Iqbal who was the manager of the Brain Trust Program at the time when this study was conceptualized. He helped organize the study and provided valuable guidance in the two workshops as well as in subsequent discussions. We are grateful to Ms Vasumathi Rollakanty for providing assistance with the organization of the workshops and for attending to all other logistical matters in the management of this project. We would be remiss if we failed to acknowledge the support of Ms Valentina Kalk of the External Affairs Office of the World Bank and the publisher, Edward Elgar Publishing Ltd, UK, without whose encouragement and support, it would not have been possible to publish this study in such a short time. Needless to say, all of those who are associated with this study owe special gratitude to the two editors, Professor Sanjaya Lan of Oxford University and Professor Shujiro Urata of Waseda University, whose relentless effort, good judgement, vast experience and knowledge helped shape this study from its inception. TSUTOMU SHIBATA (Program Manager and WBI Representative for Japan) K. MIGARA O. DE SILVA (Team Leader) Brain Trust Program World Bank Institute (WBI) 1. Introduction and overview Sanjaya Lall and Shujiro Urata East Asia is by practically any measure the most competitive and dynamic industrial region in the developing world. This is universally acknowledged but it is not fully understood, despite the industry that has grown up around the analysis of East Asian competitiveness. It is widely known that the suc cessful countries used very different strategies to build their industrial competitiveness, but the ingredients of these strategies still need further analysis. In particular, the means that the 'Tiger' economies used to access and absorb new foreign technologies over time - this process is the very life blood of industrial success - have useful lessons for other countries. As the intensity of international competition mounts and as developing countries open their economies more widely to global market forces, all policymak ers and analysts need to know how the most dynamic countries 'did it'. This book deals with one aspect of competitiveness strategies in East Asia: the interaction between foreign direct investment (FDI) and local technological activity in building export competitiveness. It highlights different strategic approaches at the national level and looks at capability development within industrial enterprises. ISSUES These differences in the strategies adopted by countries to promote tech nology development pose important questions. How, for instance, are foreign technologies most efficiently transferred from industrial to develop ing countries? Does the mode of technology transfer matter for subsequent technology development? What is the need for local technological effort in developing countries that are 'latecomers' to the industrial scene and can draw upon the vast cornucopia of technological knowledge in the devel oped world? Are TNCs essential for export competitiveness in technology intensive industries today? And so on. Not all these questions can be addressed in this book but many are. Some of the points that emerge in the analysis of country experiences are as follows. 1 2 Competitiveness. FDI and technological activity in East Asia · The transfer of technology and market information from foreign companies can take several forms. These include 'internalized' modes like inward FOI (where transfers take place within the company), or 'externalized' modes like licensing, subcontracting, original equip ment manufacturing arrangements (where transfers take place across companies). While all countries use both methods to access foreign technologies, the balance between them varies greatly. · Having accessed foreign technology, the developing country has to learn to use it effectively, adapt it to local conditions and improve it over time to keep up with competition and the international technol ogy frontier. Simply importing new technology does not mean that it can be used competitively: there is always a need for local effort to create the capabilities to use new knowledge. · The mode of technology transfer affects the need for, and the extent of, domestic effort. · Reliance on internalized technology transfer reduces the need for independent local effort (though some learning is always involved). It also often provides access to state-of-the-art technologies along with established brand names and entry into global markets. FOI is thus a very effective way to transfer and operationalize new technologies for export competitiveness. · Externalized technology transfer entails greater effort and risk on the part of the recipient country, and may not allow access to the most val uable new technologies (which are closely held by the innovators). Exploiting externalized technologies in world markets is also more difficult because of the need to build export marketing capabilities and channels. Where externalized transfers are not managed effectively, with the considerable local investment in capability development, they can result in technological inefficiencies, lags and subsequent uncom petitiveness. · Yet there may be sound reasons for encouraging the development of local technological capabilities. In most medium to large sized econ omies, local enterprises account for the bulk of industrial activity, and their competitiveness depends on externalized technology trans fers. Externalized transfers involve greater effort - and yield greater learning benefits. With industrial diversification and deepening, countries have to undertake more advanced technological functions, such as design, development and research. Internalized modes of technology transfer may not lead to the same pace of upgrading as externalized modes: there are good economic reasons for multina tional companies to keep advanced technological effort centralized in a few industrialized countries. Introduction and overview 3 · Technology development strategies have a large element of cumula tiveness and path-dependence: once launched, they tend to persist along particular trajectories for long periods. At the same time, there are forces making for greater convergence. The international 'rules of the game' are changing, constraining the use of certain government interventions (such as protection, subsidies, local content rules and exclusion of FDI) in promoting technology development. The exter nal environment is also changing, making externalized modes more costly and risky. At the same time, countries relying heavily on inter nalized modes are seeing the need for greater domestic capability development to sustain export competitiveness. The East Asian countries have chosen different combinations of internal ized and externalized modes of technology transfer, and are converging towards a more uniform set of policies at different rates. Korea, for instance, has taken the path of building domestic R&D capabilities instead of relying on FDI for its technological growth. Taiwan is rather similar, but has been slightly more open to FDI and less interventionist in building domestic industrial champions. Both are now more open to inward FDI while becoming major direct investors on their own. The Philippines and Thailand have chosen to rely heavily on the import of internalized technology and have invested relatively little in domestic R&D effort. This has not prevented them from entering highly sophisti cated and dynamic areas of export activity, but it is seen as a constraint on sustaining competitiveness in the longer term. Singapore has relied on FDI but has made strong efforts to induce foreign investors to deepen into local design and development. China combines the Korean and Singaporean strategies. Its total R&D budget as a percentage of GDP has been quite high. And so on. The choice of strategy has, of course, depended on many country-specific factors. These include, inter alia, the prevailing policy environment (in turn reflecting history, culture and political economy), domestic technological and human capital, development objectives and the institutional frame work. As noted, however, there is considerable policy convergence between Asian economies. Countries that have built competitiveness by specializing in labour-intensive segment of MNC activities are finding that they can sustain export growth only by deepening local content, skills and technolog ical activity. Those with strong indigenous technological bases are realizing that innovation is becoming expensive and specialized, and that production is being increasingly organized in internalized MNC systems spanning several countries. It thus becomes necessary to form alliances with technol ogy leaders and participate more fully in global production systems. 4 Competitiveness. FDI and technological activity in East Asia This is a highly simplified picture of a complex and dynamic scene, with many variations within and across East Asian countries. The chapters that follow adopt different approaches to the analysis of technology strategy and competitiveness. We have not sought to impose a uniform format or approach on the studies, encouraging authors to explore aspects of tech nology development and FDI from their own perspectives. CHAPTERS IN THIS BOOK Lall (Chapter 2) sets the scene for the country studies by analysing the general relations between technology transfer, FDI and local technology development. He describes the nature of technological learning and effort in developing countries, and the possibility of a divergence between the private interests of the multinational company and the social interests of the host economy in terms of long-term technology development. He goes on to trace recent patterns of export growth and competitiveness in manu factured products, noting the rising share of technology-intensive products and the growing competitiveness of East Asia. The advantage of East Asia over other developing regions is greatest in technology-intensive products, and the most dynamic exporters (including China) have rapidly moved up the technology ladder as they have expanded exports. The burgeoning competitiveness of East Asian economies draws upon quite distinct strategies. Lall distinguishes between four broad approaches. The first is 'autonomous' (the building of domestic technological capabil ities, restricting FDI entry and investing heavily in domestic skills and capa bilities), as in Korea and Taiwan. The second is 'strategic FDI dependent' (relying heavily on FDI but using industrial policy extensively to induce it to deepen into advanced activities and functions), as in Singapore. The third is 'passive FDI dependent' (also heavily FDI reliant but without the use of industrial policy to deepcn thc technological structure). The fourth is 'lSI restructuring' (inducing domestic market-oriented activities to restructure for export markets), as in China. Each of these policies has been successful in its own way in boosting export competitiveness, though each faces different strategic challenges. Lall shows with some comparative data the varying implications or conse quences of the strategies in terms of skills, R&D and FDI. Cluster analy sis is used to supplement the descriptive statistics and illustrate graphically the differences in FDI dependent and autonomous strategies. Kawai and Urata (Chapter 3) examine the performance of East Asian economies in terms of per capita GDP and total factor productivity (TFP) growth for 1970~97. They also investigate statistically the determinants of Introduction and overview 5 per capita GDP and TFP growth for 137 countries with a focus on domes tic technological capability and inflows of foreign technologies. They find that East Asian economies as a group perform better than other developing countries (despite some variations among the individual Asian economies). Improvements in domestic technological capability and greater inflows of foreign technology both contribute to increases in per capita GDP and TFP in East Asia. High educational attainment plays an important role in improving domestic technological capability while capital goods imports and FDI inflows are the main sources of foreign technolo gies. In some cases, foreign technologies are used most effectively when combined with high domestic technological capability. Though East Asian economies in general perform better than do other developing economies, there are wide variations among them. Many are facing difficult challenges in upgrading technological levels because of increased competition and rapid technological progress. Kawai and Urata argue for a business-friendly environment for foreign and domestic firms by pursuing open policies, improving hard and soft infrastructure, and main taining stable macroeconomic performance. Goto and Odagiri (Chapter 4) analyse how Japan acquired foreign tech nologies in the manufacturing sector during the period from 1945 to the early 1970s. The most striking feature of Japan's strategy was the limited use of FDI and the reliance on technology licensing and purchase of capital goods for the acquisition of foreign technologies. The reasons for its competitive success lie in accumulated technological capability, a rapidly growing market, intense competition from domestic rivals and potential foreign entrants and government policies. There was a 'virtuous circle' between the import of foreign technology and domestic R&D, reSUlting in the development of technological capability. The measures used by the Japanese government included the protection of domestic industries and promotion of R&D. Goto and Odagiri draw two policy implications from Japan's experience. The first is that the acquisition of foreign technology requires strong tech nological capabilities and enormous efforts on the part of the acquiring country. The government played two important roles in technology acqui sition. First, it nurtured technological absorptive capabilities by promoting general education and R&D spending. Second, it opened the economy to trade and FDI gradually and with careful preparation. It announced liber alization in advance and in a policy environment in which Japanese firms could raise technological levels in order to compete effectively with antici pated competition from foreign countries. The second implication concerns FDI policy. While accepting the poten tial benefits of FDI for technology transfer, they identify some potential 6 Competitiveness, FD! and technological activity in East Asia drawbacks. such as the domination of domestic industries by foreign firms. Goto and Odagiri argue that restrictive FDI policy contributed to the development of local firms by providing high profits and so allowing them to invest in capacity and capabilities. However, they warn that such restric tions have to be short term: domestic firms must face a potential entry threat. In the globalized economic environment and taking account of the wide technological gap between developed and developing economies, Goto and Odagiri question the advisability of restrictive FDI policy for developing economies today. Urata and Kawai (Chapter 5) deal with overseas R&D and intra-firm technology transfer by Japanese MNCs. They distinguish between two types of MNC technology transfers: intra-firm (from the parent to overseas affiliates) and spillovers (from affiliates to local firms). Both can contribute to upgrading technological capability in host economies. Urata and Kawai find that Japanese affiliates transfer technology to affiliates and also under take R&D in host developing economies (mainly to support production). The determinants of affiliate R&D include strong dependence on local sources for intermediate inputs and long operational history in a given location. The determinants of successful intra-firm technology transfer are the availability of educated personnel and R&D resources, the size and development of the local manufacturing sector and accumulated Japanese FDI. Urata and Kawai draw the implications of their analysis for developing host economies. First, it is important to build strong supplier industries to promote R&D by foreign firms. Second, developing economies should improve the quality of human capital, on the shop-floor, in engineering and in research activity, in order successfully to assimilate technology from foreign firms. Finally, developing economies should provide the economic environment, legal system. infrastructure and so on that attracts FDI. Kim (Chapter 6) argues that rapid technological learning by Korean firms was the most important factor in its phenomenal industrial and export growth. Korea relied heavily on the inflow of new technology via licences, capital goods and original equipment manufacture (OEM) con tracts rather than FDL Domestic efforts and capabilities were essential in assimilating foreign technology in these externalized forms, since foreign firms were reluctant to transfer key technologies and capabilities. Kim warns that FDI reduces the efforts on the part of domestic firms and increases their dependence on foreign firms. In Korea domestic technolog ical capability was nurtured by both supply and demand factors. The main supply-side factors were the availability of strong human resources, active R&D by private firms and the creation of large firms, supplemented by credit provided on favourable terms. The main demand-side factor was the lntroduction and overview 7 very competitive environment facing Korean firms, engendered by the strong emphasis on exports which offset the protection offered on the domestic market in terms of providing incentives for capability building. Kim derives four policy implications for developing countries. The first is the importance of skill building. The second is the role of strong export orientation in creating an intensely competitive environment. The third is the use of the 'brain drain' as an input into domestic capability building: high calibre nationals abroad can be important sources of technical knowl edge and skills. The fourth is the role of government research institutes (GRIs) in providing a supply of capable researchers and engineers. The Korean experience is, however, difficult to emulate, in part because of the new international environment. For example, tightened intellectual property right protection has made it difficult for developing economies to 'reverse engineer' foreign products, one of the main methods of acquiring foreign technologies by Korean firms. Restrictions on FDI are more diffi cult, and the protection of infant industries is now impossible. Bee-Van Aw (Chapter 7) analyses econometrically the effect of different modes of technology acquisition on total factor productivity growth in Taiwan, using firm-level data for the electronics industry in 1986 and 1991. She distinguishes between the following modes: exporting, FDI, R&D, employee training, subcontracting and technical cooperation. The results suggest that exporting, R&D and FDI had positive effects on TFP. In terms of technology spillovers (the acquisition of technology through informal interaction between firms) FDI and exports have a significant impact in the same geographical region and in the same industry. This points to the importance of contacts with foreign firms through exports and FDI, par ticularly in the electronics industry, where the technology frontier shifts rapidly. Aw also finds that R&D and training improve technological ability, allowing local firms to assimilate technology more effectively. In Taiwan, the role of the government and industry associations in linking foreign investors with local firms, and in establishing an efficient system of sub contracting, was vital to technology transfer and absorption. All this required human capital, and policies to improve education and R&D man power were also critical. The impact of technological activity and spillovers greatly contributed to the competitiveness of the small and medium-sized enterprises which formed the bedrock of Taiwan's economic development. Wong (Chapter 8) traces Singapore's rapid economic growth and struc tural transformation since independence in 1965. This remarkable eco nomic performance was, according to him, made possible only by the upgrading of technological capability. Singapore underwent four distinc tive phases in technology development. The first was the industrial take-off 8 Competitiveness, FDI and technological activity in East Asia phase, with high dependence on technology transfer by foreign firms. The second phase involved deepening local process technological capabilities, both within MNC affiliates and in supporting local firms. In the third phase foreign and local enterprises and research institutions expanded applied R&D activity. The fourth phase is that of high-tech entrepreneurship and basic R&D. This evolution has moved Singapore from being a technology user to becoming a technology creator. The technology development of Singapore is to a large extent attributable to effective government policies. These include sound macroeconomic man agement, open trade and FDI policies (including effective promotion and targeting of MNCs), active science and technology policy and strong emphasis on education and training. One noteworthy feature of technology development in Singapore has been the willingness of MNCs to invest in R&D, partly in response to the strong incentives and enabling conditions offered by the government. Wong also notes the effectiveness of government measures in creating technological linkages between MNCs and local firms. Wong draws several implications from this experience for other economies. First, it is important to combine a long-term commitment to upgrade tech nological capability with the institutional capability for flexible and rapid policy change in response to changes in the global environment and the changing needs of the national innovation system. Second. a proper sequenc ing should be followed in building and deepening technological capabilities: operational and adaptive capabilities should be built before investing in basic research. Third, success in technology development depends not only on effective science and technology policies but also on a variety of other factors, including an open trade and investment environment, sound macroeconomic performance, political and social stability, a corruption-free public service sector and good infrastructure. Fourth, there is no universal set of technol ogy development policies that can be applied to all the countries. One country cannot copy the system that worked in another without modifying it to suit its own conditions. This modification itself needs significant government capabilities. Yao (Chapter 9) describes China's remarkable industrial growth since the start of liberalization in the late 1970s. He attributes it partly to improve ments in technological levels, revealed by rapid increase in total factor pro ductivity and by a shift in the export structure from resource-based products to labour-intensive and simpler technology-intensive products. He describes several characteristics of the technology system in China, such as the dominance of central and local government and public research institutes in R&D activity and the important (even excessive) role of basic research. The acquisition of foreign technologies has shifted from licensing to FDI, particularly in the 1990s. Introduction and oveniew 9 Yao assesses the effectiveness of the government, private firms and foreign firms in Chinese technology by descriptive and econometric analy sis. He finds several interesting patterns. First, collective, private and foreign firms (excluding those from Hong Kong, Macao and Taiwan) are more efficient than state-owned enterprises. Second, foreign firms seem to be reluctant to transfer state-of-the-art technology, but they have beneficial spillovers by training personnel and stimulating competition. Third, R&D by public research institutes has no significant impact on firms' technolog icallevels. Yao draws policy implications for China and other developing econo mies. First, he argues for a shift in emphasis in technology development from the public to the private sector and for close ties between basic research in public laboratories and the private sector. Second, he notes the importance of technology development in labour-intensive industries, in which countries like China have a strong comparative advantage. This entails a shift away from the current emphasis on high-tech. Third, he remarks on the importance of integrating foreign firms into the domestic economy. Abrenica and Tecson (Chapter 10) provide a provocative analysis of technology development in the Philippines. They contrast its very high-tech export specialization (the Philippines has one of the most technology intensive export structures in the world) with its weak technology base. They describe the various channels of technology transfer FDI was the most important and go on to explain the sharp divergence between the export structure and the technology base. They note that the high-tech export structure is due solely to the predominance of the electronics indus try, and within it to semiconductors. In this activity, the Philippines is con fined to the labour-intensive end of assembly and testing, with very low local content and little upgrading over time. The apparently high endow ment of human capital is also slightly misleading once the quality of edu cation is taken into account. While MNCs have transferred considerable production technology to the Philippines, they have not invested much in local R&D. Neither have local industrial firms. The bulk of R&D takes place within government laboratories, most of it de-linked from the productive sector. Technology strategy has not been coherent, and its implementation has been weak. The main science and technology body has a low status in the government. Trade liberalization has not stimulated a change in the technology culture in industry, and without this the country is exposed to considerable risk by its narrow specialization and low local content in export activity. Rasiah (Chapter 11) deals with technology development in Malaysia, which was, to an even greater extent than Thailand, dependent on foreign 10 Competitiveness, FDI and technological activity in East Asia investors to drive industrial and export growth. As with Thailand, local tech nology development lagged; Rasiah notes the contrast between Malaysia's experience and those of Japan, Korea and Taiwan, where technology devel opment and industrial competitiveness were largely driven by domestic effort. However, the government did play an important role in attracting FDI and providing good infrastructure and a conducive business environment for export-oriented industrialization. FDI has provided considerable benefit to the Malaysian economy, going beyond its direct contribution to exports, employment and manufacturing production. It has raised local content over time, and so helped the growth of supplier industries. However, rising costs and the emergence of cheaper sites in China and Indonesia place Malaysia under increasing pressure to upgrade its production structure and capabilities to higher value activities. Rasiah finds that Malaysia has not been very successful in meeting this challenge, largely because of its failure to make the requisite institutional changes in such areas as tertiary education and technological activity. While the public sector has actively supported R&D activities, the private sector has not responded adequately. In addition, there is insufficient co ordination between such technology support activities as finance, training, technical extension and R&D. Rasiah emphasizes the need for a compre hensive policy addressing all these issues. Brimble, in his analysis of Thailand in Chapter 12, draws a rather differ ent picture of technology development in another dynamic and competi tive economy. Thailand's rapid economic growth was based on the growth of labour-intensive, low-tech industries, with foreign investors playing a very important role. However, the development and deepening of local technological capabilities lagged. Indeed, Thailand lags behind many other Asian economies in terms of technological development; this is borne out by interuational comparison of such things as the quality of technology, management and labour. It is also shown by low R&D spending, the small number of scientists and engineers and the small number of students spe cializing in science and technology. Brimble attributes weak technology development in Thailand to a lack of awareness in the public and private sectors of the economic importance of science and technology. This is attributable to a tradition of passive reli ance on off-the-shelf technology import, with little subsequent effort to absorb and improve upon it. Thai entrepreneurs tend to display a short-run commercial orientation, which deters them from investing in the long-term development of technological capability. Yet the phase based on low wages and skills is coming to an end, and Thailand's future growth and competi tiveness depend on its ability to develop significant new technological capa bilities. This calls for substantial changes in its science and technology Introduction and overview II policies. Such changes extend well beyond the narrow scope of S&T, and cover manpower development, technological upgrading of small and medium-sized enterprises and the development of support industries. Brimble stresses the need to allocate more resources to training, R&D and other innovation activities with a particular emphasis on applied tech nology. He also points to the importance of the creation and use of link ages - between the government and the private sector, between foreign firms and public intervention and between foreign and local firms - to improve technological capabilities. All this also needs greater awareness on the part of the Thai government, business and the population at large of the impor tance of technology development and the role of FDI in development. Okamoto and Sjoholm (Chapter 13) treat the case of Indonesia, which made enormous economic strides betore it was struck by the financial crisis in 1997. Despite rapid industrialization, however, Indonesia still lags behind other East Asian countries in technological development. Analysing the results of other studies of Indonesia's experience, Okamoto and Sjoholm draw four lessons for technology development. First, in the early stage of industrial development agents like foreign buyers, trading companies and experts are important sources of technology; thus, it is important to have an open trade and FDI regime. Second, FDI can con tribute to development not only by introducing new technology but also by generating employment, expanding production and exports. However, the host economy has to build local skills and R&D capabilities if it is to benefit fully from MNC presence. Third, the government can play an effective role in acquiring, upgrading and disseminating technology, especially in an environment where private industrial linkages are deficient. Specifically, government can provide better education and training, and improve the technology infrastructure (standards, quality and testing). Fourth, technol ogy policies targeting high-technology industries (like aircraft in Indonesia) can be wasteful, unless backed by the prior development of skills and a technological, managerial and institutional infrastructure. 14 Competitiveness, FDI and technological activity in East Asia 35 o All production 30 o Allexports 25 · High-tech production 20 · High-tOCh exports 15 10 5 o~~~~--~~~~~~~~~ _sL----------------------------------------------------- USA Japan Germany France UK China Korea Taiwan Singapore Hong 68 Kong countries Figure 2.1 Total and high-technology manufacturing: annual growth rates, 1985-97 faster than those that do not. Figure 2.1 shows the growing significance of high-tech activities in production and trade (based on NSF, 2000). The 68 countries in the total sample apparently account for 95 per cent of global productive activity. Note two features of the table: exports in general grow faster than production, and high-tech production and exports grow much faster than overall production and exports. The speed, spread and nature of technical change pose difficult chal lenges for achieving competitiveness. Unlike a few decades ago, policy lib eralization makes it difficult to insulate tradable industrial activities from global technical pressures. In addition, the traditional buffers provided by high transport and information costs are diminishing in significance as new information technologies force all countries into the same competitive arena. Some analysts note that the nature of competition itself is changing in response to new technologies. What Best (1990) calls the 'new competi tion' entails flexible response, networking, customization and new forms of inter-firm organization (in clusters) rather than c1assie eost/price competi tion dominated by hierarehicallyorganized firms. Small enterprises and net works are establishing competitive niches in many aetivities, including those with high rates of innovation. At the same time, however, with growing trade and investment flows, the role of multinational corporations (MNCs) is rising in importance. 2. TECHNOLOGY AND CAPABILITY BUILDING ANALYTICAL FRAMEWORK Technological effort is vital to developing countries, even though they do not innovate. They import new technology, equipment, patents and so on from more advanced countries, but they have to learn to use these inputs Issues and evidence 15 effectively. Using new technologies is not an automatic or simple process. I t entails the conscious building of 'technological capabilities', a mixture of information, skills, interactions and routines that firms need to handle the tacit elements of technology. Received theory assumes that technology mastery and diffusion in developing countries are relatively easy, knowl edge is not tacit, and the markets involved relatively efficient. Thus, devel oping countries simply import and apply existing technologies, picking them in line with their factor prices. Once selected, technologies can be used effectively from the start (apart from minor learning-by-doing). In this setting, free international trade and investment flows maximize the inflow of beneficial new technology,3 This approach is over-simplified. The international technology market is far from perfect. 4 Once imported, using technology efficiently is not easy, costless or automatic. Micro-level research on developing countries, based on the evolutionary theories of Nelson and Winter (1982), shows how complex and demanding the task can be. Technology is not sold in 'embod ied' forms. Its tacit elements need effort and time to master. Its efficient use cannot therefore be assumed for poor countries that expose themselves to more world markets and technologies. Technological mastery entails build ing costly new capabilities; it takes time and investment and is uncertain (LaB, 1992, 1993; Westphal, 2002), Some important features of the capability building process are described in Box 2.1. The learning curve is not known in advance. Learning is tech nology and firm specific, and often occurs in an uncertain environment where the skills, information, networks and credit needed are not available. Many enterprises do not even know how to go about learning, and have to 'learn to learn'. They interact intensively with other agents, with extensive spillovers. Once launched, the process is difficult to change. The learning process is, in other words, rife with externalities, agglomeration, path dependence and cumulative effects. 5 Technology development can thus face market failures (Stiglitz, 1996). In sum, learning to use new technologies (new, that is, to a particular user or location) needs investment and conscious effort. Much of the effort lies within the firm, but a significant part lies outside, in other firms, factor markets and support institutions. While the capability building process is essential in both developed and developing countries, it tends to be more difficult in the latter, with weak enterprises, networks, markets and institu tions. Furthermore, mastering new technology is not a once-for-all task. Most developing countries start with comparatively simple, labour-inten sive technologies where skill needs are low, learning is short and relatively less risky and there is little need for inter-firm or inter-industry coordina tion. Once mastery is achieved, continued development (with rising wages) 16 Competitiveness, FDl and technological activity in East Asia BOX 2.1 TEN FEATURES OF TECHNOLOGICAL LEARNING IN DEVELOPING COUNTRIES 1. Technological learning is a real and significant process. It is conscious and purposive rather than automatic or passive. Firms using a given technology for similar periods need not be equally proficient: each would travel on a different learn ing curve according to the intensity and efficacy of its capa bility building efforts. 2. Firms do not have full information on technical alternatives. They function with imperfect, variable and rather hazy knowledge of technologies they are using. 3. Firms may not know how to build up the necessary capabil ities - learning itself often has to be learned. The learning process faces risk, uncertainty and cost. For a technological latecomer, the fact that others have already undergone the learning process is both a benefit and a cost. It is a benefit in that they can borrow from the others' experience (to the extent that this is accessible). It is a cost in that they are rel atively inefficient during the process (and so have to bear a loss if they compete on open markets). 4. Firms cope with uncertainty not by maximizing a well-defined function but by developing organizational and managerial satis ficing routines (Nelson and Winter, 1982). These are adapted as firms collect new information, learn from experience and imitate other firms. Learning is path-dependent and cumulative. 5. The learning process is highly technology specific, since technologies differ in their learning requirements. Some technologies are more embodied in eqUipment while others have greater tacit elements. Process technologies (like chemicals) are more embodied than engineering technolo gies (machinery or automobiles), and demand different (often less) effort. Capabilities built up in one activity are not easily transferable to another. 6. Different technologies have different spillover effects and potential for further technological advance. Specialization in technologies with more technological potential and spill overs has greater dynamic benefits than specialization in technologies with limited potential. 7. Capability building occurs at all levels - shop-floor, process Issues and evidence 17 or product engineering, quality management, maintenance, procurement, inventory control, outbound logistics and rela tions with other firms and institutions. Innovation in the sense of formal R&D is at one end of the spectrum of tech nological activity; it does not exhaust it. However, R&D becomes important as more complex technologies are used; some R&D is needed just for efficient absorption. 8. Technological development can take place to different depths. The attainment of a minimum level of operational capability (know-how) is essential to all activity. This may not lead to deeper capabilities, an understanding of the princi ples of the technology (know-why): this requires a discrete strategy to invest in deepening. The deeper the levels of technological capabilities aimed at, the higher the cost, risk and duration involved. The development of know-why allows firms to select better the technologies they need, lower the costs of buying those technologies, realize more value by adding their own knowledge, and to develop autonomous innovative capabilities. 9. Technological learning is rife with externalities and interlink ages. It is driven by links with suppliers of inputs or capital goods, competitors, customers, consultants, and technology suppliers. There are also important interactions with firms in unrelated industries, technology institutes, extension ser vices, universities, associations and training institutions. Where information flows are particularly dense, clusters emerge with collective learning for the group as a whole. 10. Technological interactions occur within a country and with other countries. Imported technology is generally the most important initial input into learning in developing countries. Since technologies change constantly, moreover, access to foreign sources of innovation is vital to continued technolog ical progress. Technology import is not, however, a substi tute for indigenous capability development - the efficacy with which imported technologies are used depends on local efforts to deepen the absorptive base. Similarly, not all modes of technology import are equally conducive to indig enous learning. Some come highly packaged with comple mentary factors, and so stimulate less learning. Source: Lall (2000b). 18 Competitil'eness. PDf and technological activity in East Asia involves the upgrading and deepening of technologies. Otherwise, countries that establish a competitive niche in a low-teehnology activity may stagnate at the bottom of the technology ladder. To sustain competitive growth, they must move into more advanced technologies and technological functions within activities. At each stage, learning needs new knowledge, skills, insti tutions and policies. This has always been true, but the new technological paradigm means that the challenges are greater. Continuous access to new technologies is essential to sustaining compet itiveness. While internalized modes necessarily involve MNCs, external ized ones may also involve MNCs selling technologies on contract (MNCs are the largest sellers of licensed teehnology). However, there are also other sources of technology: national enterprises without overseas investments, consultants, capital goods producers, research institutions or govern ments. The sale can take a variety of forms: minority joint ventures, fran chising, turnkey projects, sale of equipment, licences, technical assistance, subcontracting or original equipment manufacturing arrangements. Internalized transfers bring a package of supporting inputs to ensure their efficient deployment. Externalized transfers may involve additional inputs by the technology seller, but generally tend to call for greater learning effort by the recipient. 3. ROLE OF FDI IN TECHNOLOGY TRANSFER AND LEARNING The MNCs that dominate global FDI flows are also the main source of innovation: innovation is often the main competitive factor that allows them to become (and remain) multinational. Despite the recent growth of small technology start-ups, concentration in R&D remains high. For instance, in 19972 per cent (representing the largest employers) of manu facturing companies undertaking R&D in the USA accounted for nearly 80 per cent of industrial R&D spending (calculated from table 2-8, NSF, 2000). Such concentration is even higher in small OECD industrial coun tries (UNCTAD, 1999). It does not seem to have declined over time. As the major innovators, MNCs are the main sources of international technology transfer. Their role is naturally larger in high-technology activ ities where they possess the strongest advantages. Before considering trans fers to developing countries, let us highlight features of recent FDI (Box 2.2). In general, internalized technology flows are a very efficient means of transferring a package of capital, skills, information and brand names to developing countries. For many new technologies, internalized transfers are Issues and evidence 19 BOX 2.2 SALIENT FEATURES OF RECENT FDI · FDI flows are growing faster than other economic aggregates like national gross fixed capital formation, world trade and GDP. International production (by MNCs and affiliates) is steadily increasing its share in global production. · MNCs increasingly dominate world trade: around two-thirds of visible trade is handled by MNCs, and the share is growing par ticularly in activities with significant scale economies in produc tion, marketing or innovation. · Of the visible trade handled by MNCs, between 30 and 40 per cent is within MNC systems, between affiliates and parents or among affiliates. Such internalized trade contains the most dynamic exports today, moving within integrated international production systems, where MNCs locate differ ent functions or stages of production to different countries. Affiliates participating in such systems produce at massive scales and use the latest technologies, skills and manage rial techniques. Examples of complex integrated systems in which developing countries are important are automobiles (mainly in Mexico, Brazil and Argentina) and electronics (Malaysia, Singapore, the Philippines and Mexico). The glo balization of the value chain is likely to spread across many other industries, and linking local production chains to global chains will become a major source of growth, technology transfer and skill development. · Some MNCs are locating non-production functions like accounting, engineering, R&D or marketing to affiliates - these are high value activities that feed into manufacturing competi tiveness and local capabilities. This is what UNCTAD has termed 'deep integration' in international production, in contrast to earlier 'shallow integration' where stand-alone affiliates repli cated many functions and related to other affiliates or parents via trade. However, the transfer of functions such as R&D lags that of production, particularly in developing countries. Over 90 per cent of overseas R&D by US MNCs is in other industrial countries. MNCs from smaller countries are more international in terms of relocating R&D overseas, but MNCs from econo mies like the UK are also conducting very substantial amounts of R&D overseas. However, much of such R&D remains 20 Competitiveness, FDI and rechnological activity in East Asia confined to other industrial countries. For deep integration to occur, host countries have to be able to provide not just cheap labour but the whole array of modern skills, infrastructure, insti tutions, efficient business practices and supplier networks that MNCs need to be fully competitive in world markets. Very few developing countries are able to meet these needs. · Large companies with transnational operations increasingly dominate the process of innovation: the creation of new tech nologies and organizational methods that lies at the core of competitiveness in all but the Simplest activities. Most such companies originate in mature industrial countries. About 90 per cent of world R&D expenditure is in the OECD. Within this group, seven countries (led by the USA) account for 90 per cent, the USA alone for 40 per cent. Access to new technolo gies thus involves getting knowledge from technological leaders in these countries. Many are increasingly unwilling to part with their most valuable technologies without a substantial equity stake. Thus, FDI becomes the most important often the only - way of obtaining leading edge technologies. · Mf\JCs are often central to exports by local firms of technology intensive products. Many such products are difficult to export independently because of the need for expensive branding, distribution and after-sales servicing. Thus, 60-70 per cent of consumer electronics made by Korea and Taiwan is sold to MNCs on an OEM (original equipment manufacture) basis. The significance of OEM for Korea is shown by the following statis tics. In 1985, over 40 per cent of Korean exports were in the form of OEM. In 1989, around 50-60 per cent of VCR and TV exports and about 80 per cent of PC exports by Korea were under OEM. In 1990, 70-80 per cent of total Korean electron ics exports were under OEM. MNCs are also active in exports of low-technology products where factors like scale econo mies, branding, distribution and design are important. · Mf\JCs can help restructure and upgrade competitive capabil ities in import-substituting activities. Where the facilities are already foreign owned, MNCs are often better able to respond to liberalization than local firms by investing in new technolo gies and skills. They can also help local suppliers to upgrade, or attract investment by their suppliers overseas. This has been commonly found in Latin America. Where local firms own the Issues and evidence 21 facilities, MNCs help them to upgrade through mergers and acquisitions (M&As). While cross-border M&As are often regarded with suspicion or resentment, they can salvage exist ing facilities that would not survive in a liberalized environment. In fact, with globalization and liberalization, international M&As now constitutes the bulk of FDI flows, accounting for over 80 per cent of FDI in developed countries and around one-third in developing ones (UNCTAD, 2000). · FDI in services is rising rapidly as formerly homebound provid ers (as in utilities) globalize activities and take advantage of lib eralization and privatization in their industries. The entry of service MNCs can provide rapid improvements in the produc tivity and efficiency to host economies, not only in their indus tries but also to their customers (many of which are important exporters). the only possible mode of transfer, since innovators are unwilling to part with them to unrelated parties. Even where technologies are available at arm's length, internalization may be the most efficient way of transferring the tacit knowledge involved because of the commitment of the transferrer and its capability to support learning. If the technology is changing rapidly, internalization provides the most direct access to improvements. If the activity is export oriented, internalized transfers offer the additional advan tages of international marketing skills and networks, established brand names or, of increasing relevance, access to integrated production struc tures spanning several countries. However, internalized technology transfer may also have costs (Pack and Saggi, 1997). Profits are realized by the MNC on the package as a whole rather thanjust the innovation component. If the host country already pos sesses other elements of the package, it is cheaper to buy the technology separately (countries like Korea and Taiwan did this because their enter prises had the necessary capabilities to master the technology). In general, the more standardized and diffused the technology, and the more capable the buyer, the more economical will externalized modes be. However, there is a more subtle reason: the existence of learning benefits, deepening and externalities may tilt the choice in favour of externalization even for rela tively complex and difficult technologies. In such activities, reliance on foreign investment can shorten the learning period but reduce the other benefits of technology transfer and capability building. 22 Competitiveness, FD! and technological activity in East Asia A useful way to analyse this is to divide technological capabilities into four levels, At the bottom are the simplest (operational) ones, needed for running a technology efficiently: these involve basic manufacturing skills as well as some more demanding troubleshooting, quality control, mainte nance and procurement skills, At the intermediate level are duplicative skills, which include the investment capabilities needed to expand capacity and to purchase and integrate foreign technologies. Next come adaptive skills, where imported technologies are adapted and improved, and design skills for more complex engineering learned. Finally come innovative skills, based on formal R&D, that are needed to keep pace with technological frontiers or to generate new technologies. The advantage of internalized forms lies in the long-term commitment of the foreign partner to the project and its ability to provide the elements needed to operationalize new technologies. At the lowest level, therefore, foreign investment is a very efficient way of transferring technology. Since all technologies need adaptation and improvement, foreign affiliates, with their base of high-level management and technical skills, tend to be in the forefront of such activity in developing countries. In addition, MNCs have the experience of other affiliates in the developing world to draw on, and can shift knowledge and personnel across countries to help with the upgrading of local capabilities. As capability development progresses to the top level, where local inno vative efforts become viable, there can be a conflict of interest between the host country and the foreign investor. Internalized technology transfer and local capability development can, in other words, become competitive rather than complementary. There are good reasons for international inves tors to keep innovative work centralized at home or in a few developed countries: thesc include ease of coordination, skill availability, proximity to main markets, and more advanced science and technology infrastructures. At the same time, it is important for countries at a certain stage of indus trial development to deepen their capabilities and move into innovation. MNCs tend to transfer the results of R&D rather than the process itself, whereas the sustained technological growth of developing countries calls for increasing local innovation. There is clear scope for a clash between the social interests of the host economy and the private interests of MNCs. At this stage, there is a case for restricting reliance on internalized forms to promote local R&D capabilities based on externalized forms, or for inter vening in the FDr process to induce MNCs to transfer more advanced tech nological functions. Issues and evidence 23 4. PATTERNS OF EXPORT COMPETITIVENESS IN EAST ASIA The pattern of manufactured exports reflects the underlying structure of and competitive advantages in manufacturing. A sophisticated export structure, with growing world market shares, generally suggests that the exporting economy has built the production and innovative capabilities to use advanced technologies and, over time, keeps abreast of new technolo gies. This is not, however, always the case. The rise of integrated produc tion systems means that developing countries can enter sophisticated export activities at the simple labour-intensive end: FDI can introduce a wedge between indigenous capabilities and export structures. East Asia has examples of both kinds of competitive bases. Some countries have built strong domestic capabilities in technology-intensive activities, generally by restricting or manipulating FDI to boost local innovation. Others have enjoyed rapid export growth and technological upgrading by plugging into high-tech TNC production systems. This section shows how both strategies have led to dynamic competitiveness in East Asia. Both strategies have advantages and drawbacks. The domestic capability building strategy has many long-term advantages, but it is costly, risky and prolonged. And over time, as innovation accelerates and the structure of international production changes, it has to move towards greater reliance on MNCs. The FDI dependent strategy allows countries to raise exports and technological structures rapidly, but it faces the problem of sustainabil ity. MNCs can dynamize competitiveness in countries in which they locate integrated systems, but such activity may not allow the exporting country to maintain a competitive edge as wages rise, technologies grow more demanding or there is greater need for local suppliers. Thus, domestic capa bility building becomes necessary to retain the base set up by export oriented MNCs. While the two strategies move towards some convergence, however, there is considerable path-dependence once countries get launched along partic ular trajectories. This is clearly the case in East Asia, as the country studies in this book show. But let us paint a broad comparative picture here by ana lysing the export data. Technological Classification The analysis separates primary from manufactured exports and groups manufactures into four technology levels: resource-based (RB), low technology (LT), medium-technology (MT) and high-technology (HT) (Lall, 2000a). All such classifications suffer from data and definition 24 Competitiveness. FDI and technological activity in East Asia Table 2. 1 Technological classification of exports Primary products Fresh fruit, meat, rice, cocoa, tea, coffee, wood, coal, crude petroleum, gas Manufactured products Resource-based manufactures Agro/forest-based products Prepared meats/fruits, beverages, wood products, vegetable oils Other resource-based products Ore concentrates, petroleum/rubber products, cement, cut gems, glass Low-techn%gy manufactures Textile/fashion cluster Textile fabrics, clothing, headgear, footwear, leather manufactures, travel goods Other low-technology Pottery, simple metal parts/structures, furniture, jewellery, toys, plastic products Medium-technology manufactures Automotive products Passenger vehicles and parts, commercial vehicles, motorcycles and parts Medium-technology process Synthetic fibres. chemicals and paints, fertilizers, plastics, iron, pipes/tubes Medium-technology engineering Engines, motors, industrial machinery, pumps, switchgear, ships. watches High-technology manufactures Electronics and electrical Office/data processing/telecommunications equipment, transistors. turbines. generating equipment Other high-technology Pharmaceuticals, aerospace, opticall measuring instruments, cameras Other transactions Electricity, cinema film, printed matter, 'special' transactions. gold, art, coins, pets problems. It is not possible to capture an important aspect of technical change: upgrading within given product categories. 6 The categories are highly aggregated and can conceal differences within each category.? They do not show the processes involved in making a given product: the spread of integrated production systems mean that these differ substantially between countries. 8 Nor do they show local physical and technological content in exports: one country may be assembling imported components and another manufacturing the product from scratch. Nevertheless, some disadvantages can be overcome indirectly. For instance, we can use qualita tive information on local content and processes to allow for differences in technology content in export activity in different locations. lssues and evidence 25 Table 2.2 Growth rates and market shares of exports (% per annum) All products Primary All mfg. RB LT MT HT Growth 198598 World 8.6 3.4 9.7 7.0 9.7 9.3 13.1 Developed 8.4 4.4 8.8 7.0 8.5 8.5 11.3 Developing 8.7 1.3 12.5 6.0 11.7 14.3 21.4 Growth \985-90 World 13.1 5.6 \4.9 \1.4 16.3 \5.1 17.4 Developed 14.0 7.9 14.7 12.7 15.4 14.7 16.2 Developing 9.1 1.3 15.4 4.9 18.4 19.3 26.7 Growth 1990-95 World 8.2 4.4 8.9 7.4 8.3 7.8 13.2 Developed 7.0 4.9 7.2 6.2 6.2 6.6 10.1 Developing 12.0 2.7 15.3 10.3 11.3 16.6 25.4 Growth 1995-98 World 2.1 -1.9 2.7 -0.4 1.8 2.5 6.1 Developed 1.7 -2.3 2.2 -0.9 1.4 2.0 5.6 Developing 2.8 -1.1 3.5 0.9 2.2 3.1 7.0 Shares of products in world exports, 1985 and 1998 1985 100 21.7 73.8 21.1 13.7 30.2 12.4 1998 100 11.5 84.2 14.5 15.8 32.8 21.1 Shares of developing countries in world exports, 1985 and 1998 1985 24.3 52.1 16.4 26.3 26.7 8.3 10.7 1998 25.0 39.7 23.3 23.7 34.5 15.3 27.0 Global and Regional Patterns We start with the major categories for four sub-periods between 1985 and 1998.9 Table 2.2 gives growth rates and market shares for exports by the world, developed and developing countries. 10 Table 2.3 shows manufactured exports by main developing regions. I 1 SSA and LAC are shown with and without their major exporters South Africa and Mexico, the 'outliers' in their regions. 12 The figures show large, and generally increasing, disparities in export performance across the developing world, with enormous concentration of competitive capabilities in East Asia. Table 2.3 shows that with nearly 70 per cent of total manufactured exports by developing countries, East Asia is not just the dominant player, its share is rising over time. The largest loss of share is by LAC2 (excluding Table 2.3 Regional shares of developing countries' manufactured exports of developing world total) Year East Asia South Asia MENA LACI (incl. LAC2 (exc. Mexico SSAI (inc!. SSA2 (exc. Mexico) Mexico) S Africa) S Africa) All 1985 56.9 4.5 12.9 23.1 16.9 6.2 n/a 2.6 manufactures 1998 69.0 3.8 6.0 19.3 8.9 lOA 1.8 0.8 RB 1985 34.6 3.8 23.8 32.9 30.7 2.2 n/a 4.9 1998 47.5 4.7 15.0 28.0 24.0 4.0 4.8 104 ~ 0- Agro-based 1985 55.1 2.2 4.5 32.0 3004 1.6 n1a 6.2 1998 55.1 1.7 4.9 33.1 28.3 4.6 5.3 204 Other RB 1985 25.6 4.5 32.3 33.3 30.8 2.5 n/a 4.3 1998 41.4 7.2 23.1 23.1 20.6 2.5 404 0.6 LT 1985 71.7 8.3 7.3 11.9 10.2 1.7 n/a 1.8 1998 70.2 8.5 7.2 12.6 5A 7.2 L5 0.2 Textile cluster 1985 69.9 11.6 8.1 9.5 8.5 1.0 n/a 0.9 1998 67.3 12.1 9.1 lOA 4.9 5.5 l.l 0.8 Other LT 1985 75.2 1.7 5.7 16.6 13.5 3.1 n1a 0.8 1998 74.9 2.9 4.2 16.0 6.3 9.7 2.0 0.3 MT 1985 63.4 2.0 7.1 25.8 17.5 8.3 nla 1.8 1998 63.8 1.8 4.4 28.1 10.2 17.9 1.9 0.2 Auto 1985 40.6 2.7 5.9 50.3 32.9 17.4 nla 0.4 1998 39.8 1.4 2.9 54.2 16.9 37.3 1.7 0.1 Process 1985 53.4 2.3 13.8 28.2 25.2 3.0 nla 2.3 1998 65.6 3.3 8.4 19.9 13.0 6.9 2.8 0.5 Engineering 1985 73.0 1.7 3.5 20.1 10.4 9.7 nla 1.7 1998 72.5 1.1 2.6 22.4 5.8 16.6 1.3 0.1 N "l HT 1985 81.0 1.1 1.8 14.8 6.6 8.2 nla 1.3 1998 85.5 0.6 0.7 12.9 2.1 10.8 0.4 0.0 Electronic 1985 84.7 0.5 0.7 14.0 5.1 8.9 nla 0.1 1998 87.2 0.3 0.6 11.8 1.2 10.6 0.2 0.1 Other HT 1985 60.3 4.5 8.2 19.2 15.2 4.0 n/a 7.8 1998 66.9 4.2 1.9 25.0 12.2 12.8 0.0 0.3 Source: Lall (2000a). 28 Competitiveness, FDI and technological activity in East Asia Mexico, which shows a healthy increase), It is followed by MENA, which retains second position but after a hefty fall; its performance is strongly influenced by Turkey. South Asia, despite its substantial industrial sector, suffers a diminution of its small share, the continuing legacy of decades of import-substitution, Sub-Saharan Africa starts from a marginal position and deteriorates further. The share of SSA2 falls to below 1 per cent (only 0,2 per cent of the world total); even including South Africa only brings the figure to below 2 per cent (0.4 per cent of the world total), Table 2.4 shows the technology composition of manufactured exports by region. The world and the developed countries shift from RB to HT prod ucts. In the developing world, there is a more marked shift away for 'simple' (RB and LT) to 'complex' (MT and FIT) products, but with a massive increase in HT. East Asia has the most high-tech export structure (more than developed countries) and the largest upgrading; its reliance on LT products falls over time, particularly in the textile group. LAC has a complex export structure but mainly because of MT products (particularly autos). The Mexican presence is very significant, driven in turn by its maquiladora export activities. Both MENA and SSA2 reduce their heavy dependence on RB, but remain specialized in simple manufactured prod ucts, especially in the fashion cluster (led by Turkey and Morocco in MENA and Mauritius in SSA2). While MENA raises its share of HT and MT. SSA2 does the reverse. South Asia shows a similar trend to MENA, but with a much heavier reliance on LT products. Thus, East Asia emerges as the most competitive, dynamic and technologically advanced region in the developing world. Country-level Performance in East Asia Manufactured exports by developing countries are concentrated not just at the regional but also at the country level. In 1998, for instance, the leading five exporters accounted for 60 per cent, and the leading ten for over 80 per ccnt, of the developing world total. Interestingly, concentration tends to be greater the higher the level of technology, ranging from about 60 per cent for RB to 96 per cent for the top ten HT exporters in 1998. This suggests, as expected, that capability needs (and entry barriers) rise with technolog ical sophistication. Concentration levels also tend to rise over time, indicat ing growing capability needs. Annex Tables 2A.l and 2A.2 show values and breakdowns of manufactured exports by technological category for 13 main developing country exporters. All the countries in this study are in this group. Shares of MT and HT products in their 1998 manufactured exports are shown in Figure 2.2. In MT. the highest share is in Mexico, a result of its MNC dominated Table 2.4 Distribution of manufactured exports over technological catef.wries. 1985 and 1998 1985 RB RBI RB2 LT LTl LT2 MT MTl MT2 MT3 HT HTl HT2 World 23.7 9.8 13.8 18.6 8.3 10.3 40.9 12.2 9.7 19.0 16.8 10.9 5.9 Developed 21.0 9.6 11.5 16.1 5.8 10.3 44.7 14.3 10.3 20.1 18.2 11.4 6.8 Developing 38.0 11.6 26.5 30.4 20.1 10.3 20.6 2.1 6.7 11.9 11.0 9.3 1.7 East Asia 23.1 11.2 11.9 38.3 24.7 13.6 23.0 1.5 6.3 15.2 15.6 13.8 1.8 South Asia 32.3 5.6 26.7 55.8 51.8 4.0 9.2 1.3 3.4 4.5 2.8 1.1 1.7 MENA 70.1 4.0 66.1 17.1 12.6 4.5 11.3 1.0 7.1 3.2 1.6 0.5 1.1 LAC 1 54.2 16.1 38.2 15.7 8.3 7.4 23.1 4.5 8.2 10.4 7.0 5.6 1.4 LAC2 61.1 18.4 42.7 16.2 8.9 7.3 18.9 3.6 8.8 6.5 3.8 2.5 1.3 SSA2 70.7 27.3 43.3 10.1 7.0 3.0 13.8 0.3 5.9 7.6 5.5 0.5 5.0 '" '0 1998 RB RBI RB2 LT LTI LT2 MT MTl MT2 MT3 HT HTl HT2 World 17.3 8.9 8.4 18.8 8.2 10.6 38.9 11.5 8.8 18.6 25.1 18.2 6.9 Developed 16.8 9.1 7.7 15.5 5.2 10.2 43.2 13.6 9.1 20.5 24.5 16.0 8.5 Developing 17.6 7.8 9.8 27.8 17.1 10.8 25.5 5.2 7.6 12.8 29.1 26.6 2.5 East Asia 12.1 6.2 5.9 28.3 16.6 11.7 23.6 3.0 7.2 13.4 36.0 33.6 2.4 South Asia 21.7 3.4 18.3 61.6 53.6 8.0 12.1 1.9 6.5 3.7 4.6 1.9 2.8 MENA 44.3 6.4 37.9 33.7 26.1 7.6 18.8 2.5 10.6 5.7 3.3 2.5 0.8 LAC I 25.4 13.3 12.1 18.1 9.1 8.9 37.1 14.5 7.8 14.8 19.4 16.2 3.2 LAC2 47.2 24.5 22.6 16.8 9.3 7.5 29.1 9.8 11.0 8.3 6.9 3.5 3.4 SSAI 46.0 22.4 23.6 22.6 10.5 12.0 25.8 4.9 11.6 9.3 5.7 3.0 2.7 SSA2 51.3 38.6 12.7 35.0 27.4 7.6 11.5 0.6 7.7 3.2 2.2 0.7 1.5 30 Competitiveness, FDJ and technological activity in East Asia 80 70 60 50 (%) 40 30 20 10 0 co 0 ~ >. '" c ~ C u .~ "0 'N '" . '" ..::; d C '5. § '" .s 0'" " "; ~ ~ ~ '" ~ " ~ bJ.l C i:i'l " :;; ..c ""' :.a c:.. "0 .5 bJ.l c 0 :I: ""' Figure 2.2 Shares of MT and HT products in manufactured exports ( 1998) fiir 13 largest exporters automotive and engineering exports, followed by Korea (dominated by the chaebol) and Brazil (mainly MNCs). The highest shares of HT are in the Philippines, Singapore and Malaysia, all dominated by MNC export oriented operations. Of these countries, local content is highest in Singapore, followed by Malaysia. The Philippines remains at the stage of final assembly and testing operations. However, Korea and Taiwan have the greatest technological inputs and competence among the large HT exporters. Their domestic firms undertake much of the manufacturing and also provide high levels of physical, design, R&D and engineering inputs (see Hobday, 1995 and Lall, 1996). Because of product differentiation and branding, a large proportion of sophisticated exports have to be sold by MNCs under their own brands (called OEM, or original equipment manufacture). Hong Kong has a weak technology struc ture, and is the only Asian Tiger to suffer massive de-industrialization in recent years. This may in part be traced to its lack of targeted industrial and technological policies: Singapore, with higher wages and a strong service sector, has a growing industrial sector because it has been able to target and promotc higher value-added activities. Let us now consider revealed comparative advantage (RCA) by teelmol ogy categories for the leading exporters in 198598 (Table 2.5).13 There are several points of note. China. In 1985 no manufactured export category for China had an RCA of over I (at the more disaggregated level, its fashion cluster products had an advantage). Its main advantage lay in primary products (with an Issues and evidence 31 RCA of 1.7, not shown here). In 1998, China showed a revealed compara tive advantage only in LT products. Its RCA for HT products also increased strongly over the period, but not sufficiently to give an RCA of over unity. Its weakest RCA was in medium-technology products, while the lowest increase was in RB. M~ature Tigers. Korea started the period with a strong revealed com parative advantage in LT and a slight one in MT. By 1998, while it retained RCAs of over unity in these two groups, it had also developed a strong com parative advantage in HT products offset by a declining RCA in LT (and primary products). Taiwan started in 1985 with a stronger RCA in LT and had a slight advantage in HT. In 1998, it retained this pattern, but like Korea had suffered a large erosion of LT advantages while enjoying an increase in HT. Singapore had a strong comparative advantage in RB (petroleum refining and petrochemicals) and a somewhat lower one in HT products. By 1998, its only area of comparative advantage was HT, the only country in the group to show an increasing RCA in this set of products and declining ones in all others. Hong Kong remained overwhelmingly special ized in LT over the period, far more so than much lower-wage economies in the region. However, it also showed a slight positive comparative advan tage in HT products (mainly electronics-based light consumer goods like watches and toys) in both periods. In general, it showed the weakest and least dynamic of industrial structures among the Tigers, accompanying an anaemic export performance. 'New Tigers'. The Philippines started in 1985 with an RB-dominated export structure (with a slight comparative advantage in textile products). It then underwent a dramatic transformation, losing in RB and gaining in HT. However, this was driven almost entirely by semiconductor assembly and belies an otherwise weak competitive performance. 14 Malaysia also started with a strong advantage in RB but it already had an advantage in HT. Over the period, its RB suffered a decline but retained a (slight) advan tage, while its HT greatly increased its RCA. Thailand had advantages in RB and LT in 1985, and by 1998 had greatly increased its competitive edge in HT while losing it slightly in RB and LT. Indonesia started, like China, with no RCAs of over unity in any manufactured category (that is, its advantages lay in primary products). Over time, Indonesia acquired advan tages in RB and LT products. Its pattern of resource-based and low technology competitiveness development contrasts with that of most other countries in the region. Table 2.5 RCAs for leading 13 developing country exporters by technological category, 1985 and 1998 1985 RB RBI Ril2 LT LTI LT2 MT MTI MT2 MT3 HT HTI HT2 Total Total Total Total China 0.525 0.364 0.639 0.751 1.417 0.219 0.096 0.012 0.322 0.034 0.099 0.019 0.248 Korea 0.471 0.462 0.477 2.890 4.270 1.786 1.183 0.249 1.398 1.673 0.986 1.347 0.318 Taiwan 0.539 0.760 0.381 3.657 4.599 2.904 0.663 0.314 0.764 0.836 1.237 1.794 0.205 Mexico 0,462 0.253 0.610 0.366 0.322 0.401 0.547 0.390 0.267 0.790 0.693 0.987 0.148 Singapore 2.070 0.921 2.888 0.523 0.548 0.502 0.645 0.071 0.654 1.009 1.644 2.127 0.750 Malaysia 1.695 3.521 0.396 0.322 0.489 0.189 0.208 0.015 0.236 0.319 1.195 1.719 0.224 "" '" Thailand 1.124 2.009 0.495 1.336 2.279 0.582 0.378 0.022 0.574 0.506 0.197 0.260 0.080 Brazil 1.728 1.914 1.597 1.067 1.305 0.877 0.679 0.558 1.137 0.523 0.272 0.280 0.257 Philippines 1.695 3.142 0.666 0.930 1.258 0.667 0.157 0.051 0.499 0.052 0.467 0.693 0.048 Indonesia 0.893 1.252 0.637 0.234 0.445 0.064 0.044 0.001 0.168 0.009 0.050 0.054 0.042 India 1.610 0.392 2.476 2.289 4.653 0.397 0.231 0.148 0.269 0.266 0.227 0.135 0.398 Hong Kong 0.174 0.231 0.133 4.415 6.953 2.383 0.609 0.001 0.166 1.225 1.146 1.616 0.275 Turkey 0.908 1.021 0.828 2.814 4.402 1.543 0.567 0.144 1.325 0.452 0.093 0.101 0.079 All developing 1.0812 0.791 1.288 1.0982 1.633 0.670 0.3398 0.115 0.465 0.420 0.4387 0.571 0.193 countries 1998 RB RBI RB2 LT LTI LT2 MT MTi MT2 MT3 HT HTI HT2 Total Total Total Total China 0.619 0.520 0.723 2.883 4.151 1.896 0.563 0.105 0.773 0.745 0.862 1.016 0.455 Korea 0.671 0.479 0.874 1.210 1.493 0.991 1.073 0.889 1.525 0.971 1.288 1.619 0.418 Taiwan 0.362 0.334 0.391 1.839 1.870 1.814 0.804 00400 1.175 0.877 1.654 2.203 0.208 Mexico 0.409 0.426 0.390 1.070 1.152 1.006 1.187 1.696 0.604 1.151 1.260 1.561 0.469 Singapore 0.912 0.422 1.431 0.417 0.293 0.514 0.537 0.081 0.623 0.778 2.683 3.545 0.414 1.032 1.741 0.282 0.625 0.651 0.606 0.557 0.064 0.571 0.855 2.217 2.876 0.482 .... .... Thailand 1.080 1.342 0.803 1.302 1.782 0.928 0.508 0.191 0.596 0.663 1.339 1.729 0.311 Brazil 2.074 2.436 1.690 0.730 0.893 0.603 0.858 0.967 1.201 0.629 0.295 0.189 0.577 Philippines 0.471 0.702 0.227 0.873 1.463 0.414 0.317 0.134 0.158 0.505 3.042 4.133 0.169 Indonesia 1.470 1.957 0.954 1.148 1.630 0.773 0.311 0.066 0.729 0.264 0.252 0.318 0.079 India 1.623 0.343 2.979 2.409 4.363 0.887 0.348 0.231 0.741 0.234 0.244 0.140 0.519 Hong Kong 0.291 0.360 0.219 3.352 6.491 0.908 0.378 0.000 0.393 0.605· 1.158 1.306 0.766 Turkey 0.854 1.096 0.598 3.114 5.522 1.239 0.554 0.293 0.997 0.503 0.253 0.281 0.180 All developing 0.949 0.816 1.091 1.384 1.937 0.953 0.613 00421 0.801 0.641 1.083 1.365 0.339 countries Note: Calculated for all exports, including primary products (not shown here). 34 Competitiveness, FD! and technological activity in East Asia 5. MAIN DETERMINANTS OF COMPETITIVENESS Strategic Issues What were the strategies pursued by these countries to build industrial competitiveness? While a part of export growth was certainly based on the better exploitation of natural resources and unskilled or semi-skilled labour, the most dynamic performers have relied on the creation of new advantages in complex products. This has been based, in turn, on new domestic skills and technologies and/or on the attraction of MNC produc tion chains. The exact combination of the two has differed by country: some have relied on attracting FDI, others have built stronger domestic capabilities by intervening in factor and product markets. To reiterate the strategic issues: the development of competitiveness in all activities at all technological levels requires building capabilities in manage ment, production, engineering, design, distribution, marketing and so on. The realization of existing advantages (say, in resource-based or unskilled labour-intensive activities) tends to involve less effort, risk and externalities than the development of advantages in complex activities. Sustained export growth needs moving from easy to complex products and processes within activities, and across activities from easy to complex technologies. The choice between modes of technology transfer depends on the existing base of skills and the scale, marketing and technological demands of exporting. It also depends upon the ability of governments and institutions to help local enterprises build capabilities and tap externalities. MNCs and local firms face different factor markets and have to overcome different market failures in learning. MNCs have advantages over local firms in using new technologies. They have mastered and used the technologies elsewhere; they may have created the technology in the first place. They have large reserves of skill, technical support, experience and finance to design and implement learning. They have access to major export markets, established marketing channels and well-known brand names. They can transfer particular components or pro cesses from a production chain to a developing country and integrate it into an international system. This is much more difficult for local firms, not just because they may not have the experience or technological competence they inevitably face higher transaction and coordination costs in integrat ing into MNC corporate systems. While MNCs can be a highly effective means of transferring technolo gies and building production capabilities, they may be less effective in deep ening and broadening them. MNCs initially transfer equipment and technologies suited to existing skills and capabilities. They do invest in Issues and evidence 35 upgrading local skills, technological capabilities and supply chains but only to the extent that it is profitable in eommercial terms (to implement produc tion technologies). They will go beyond this, but only if the skill base is growing, local suppliers improving their capabilities, technology institu tions can provide more advanced services, and so on. This needs aetive government policies. Moreover, a policy to induce MNCs to enter more advanced activities by offering such inducements as specialized infrastruc ture and skills can accelerate the upgrading process. With a completely passive policy, MNC exports can remain at low, technologically stagnant, levels. Thus, an MNC-dependent export strategy needs a proactive element for dynamic competitiveness. More important, depending on FDI is not a substitute for strengthening domestic capabilities. There are many activities that MNCs do not enter, including many localized ones that tend to be populated by SMEs. They also need efficient local suppliers if they are to go beyond the assembly of imported components. Capturing the spillover benefits of foreign presence needs capable local firms. More important, a strong base of national enter prises can lead to broader, deeper and more flexible capabilities, since the technology development process within foreign affiliates may be curtailed as compared to local firms. The very fact that an affiliate can draw upon its parent company for technical information, skills, teehnological advances and so on means that it needs to invest less in its own capabilities. This applies particularly to functions like advanced engineering, design or R&D, which MNCs tend to centralize in industrial countries. As they mature industrially, it is imperative for developing countries to undertake these functions locally to support their future comparative advantage. This is why some countries choose to promote technology development in indige nous firms. Different countries make different strategic choices in these respects. Taking our sample of leading developing country exporters, we may distin guish between four: 1. Autonomous, based on the development of capabilities in domestic firms, starting in simple activities and deepening over time. This strat egy used extensive industrial poliey, reaching into trade, finance, edu cation, training, technology and industrial structure. It involved selective restrietions on FDI, and actively encouraged teehnology imports in other forms. All these interventions were carried out in a strongly export-oriented setting, with favours granted in return for good export performance. The prime examples are Korea and Taiwan. 2. Strategic FDI-dependent, driven by FDI and exports within inte grated production networks. There was a strong effort to upgrade 36 Competitiveness, FDI and technological activity in East Asia MNC activity according to strategic priorities, directing investments into higher value-added activities and inducing existing affiliates to upgrade their technologies and functions. This strategy involved extensive interventions in factor markets (skill creation, institution building, infrastructure development and supplier support), encour aging R&D and technology institutions, and in attracting, targeting and guiding investments. The best example is Singapore. 3. Passive FDI-dependent, also driven by FDI but relying largely on market forces to upgrade the structure. The main tools were a welcom ing FDI regime, strong incentives for exports, good export infrastruc ture and cheap, trainable labour. Skill upgrading and domestic technological activity were relatively neglected (though some countries had a relatively good base), and the domestic industrial sector tended to develop in isolation from the export sector. Malaysia, Thailand and the Philippines are good examples, along with the Special Economic Zones of China (and the maquilas of Mexico). 4. lSI restructuring, with exports growing from established import substituting industries where competitive (or nearly competitive) capa bilities had developed. The main policy tool was trade liberalization or strong export incentives (some, as in Latin America, within regional trade agreements). This led to considerable upgrading, restructuring and expansion of these industries along with their supplier networks. In some countries the main agents were domestic enterprises and in others they were MNCs. The main difference from the 'autonomous' strategy was the lack of clear and coordinated industrial policy to develop export competitiveness, with haphazard (and often weak) support for skills, technology, institutions and infrastructure. China and India are exam ples within Asia, the large Latin American economies elsewhere; ele ments of this strategy are also present in many other economies. These strategies are not exclusive. Countries often combine them and vary the combinations over time. Nevertheless, this typology is useful as an analytical tool. Let us now consider the main supply-side determinants of capability building (skills, technological activity and FDI attraction). Skills In the traditional setting, industrial development required improving the quantity and quality of primary schooling and basic technical education, and encouraging all forms of in-firm training. The data do not permit a rig orous comparison of skill formation across countries, particularly for enterprise training. What is possible is to compare educational enrolments. Issues and evidence 37 The most common comparisons are for the three general levels: primary, secondary and tertiary. Given the focus in technology, however, we concen trate on tertiary enrolments in technical subjects: science, mathematics and computing and engineering. Note, however, that national rankings are fairly similar even if other measures of skill creation are used (Lall, 1999), so the exact definition of the measure is not that important. Table 2.6 shows enrolments for the mid-1990s for East Asia and its comparator regions and countries, with some advanced industrial countries shown for reference. Enrolment data are not, as noted, the ideal measure of skills. They ignore on-the-job learning, other forms of training and quality differences in the education provided. Nevertheless, they are the only comparable data avail able and they do show the national base for skill acquisition. In percentage terms the Asian NIEs enrol over 33 times as many of their population in technical subjects than does SSA (including South Africa). The ratio is twice that of industrial countries, nearly five times that of Latin America and the new NIEs, and over ten times that of South Asia and China. The leading three countries in terms of total numbers of technical enrolments - China (18 per cent), India (16 per cent) and Korea (11 per cent) account for 44 per cent of the developing world's technical enrolments, the top ten for 76 per cent and the top 20 for 93 per cent. In terms of the intensity of technical skill creation (enrolments as a per centage of the population), however, the picture is quite interesting, partic ularly at the country level (Table 2.7). The world leader is Korea (1.65 per cent), followed by Finland (1.33 per cent). Taiwan, the next developing country, ranks fifth 0.07 per cent). Singapore comes in much later, in thirty-eighth position, below the Philippines and Hong Kong; however, this is misleading in that the polytechnics provide a great deal of technical edu cation in Singapore and students also study abroad, which is not captured in the UNESCO data. The new NIEs, apart from the Philippines, are well behind: Indonesia (54), Thailand (70) and Malaysia (75). China and India are even further (82 and 78 respectively). These figures have to be treated with care. The connection between techni cal enrolments and technological competence is not direct. The quality of the training and the ability of industry to exploit available skills in R&D or other technical effort matter a great deal. The accumulated stock of trained man power and, more importantly. its base of experience are extremely important. Technological Activity Technological activity in developing countries consists less of 'innovation' than of engineering and technical work for learning, adaptation and improvement. Given its diffuse nature, however, it is difficult to measure. Table 2.6 Tertiary level enrolments and enrolments in technical subiect,\~ 1995 Tertiary level enrolment Technical enrolments: numbers and 'Yr, of population Total no. 'X, pop. Natural science Maths, Engineering All technical students computing subjects numbers (,X,) numbers numbers ('X,) numbers - countries 35345800 0.82 2046566 0.05 780930 0.02 4194433 0.\0 7021929 0.16 Sub-Saharan Africa 1542700 0.28 111500 0.02 39330 0.01 69830 0.01 220660 0.04 MENA 4571900 1.26 209065 0.06 114200 0.03 489302 0.14 812567 0.22 Latin America 7677800 1.64 212901 0.05 188800 0.04 1002701 0.21 1404402 0.30 '-N '::c Asia 21553400 0.72 1513100 0.05 438600 0.01 2632600 0.09 4584300 0.15 4 mature Tigers 3031400 4.00 195200 0.26 34200 0.05 786100 1.04 1015500 1.34 4 new Tigers 5547900 1.61 83600 0.02 280700 0.08 591000 0.17 955300 0.28 S. Asia 6545800 0.54 996200 0.08 7800 0.00 272600 0.02 1276600 0.10 China 5826600 0.60 167700 0.02 99400 0.01 971000 0.10 1238100 0.13 Others 601700 0.46 70400 0.05 16500 0.01 11900 0.01 98800 0.08 Transition economies 2025800 1.95 55500 0.05 30600 0.03 354700 0.34 440800 0.42 Developed economies 33774800 4.06 1509334 0.18 1053913 0.13 3191172 0.38 5754419 0.69 12297400 3.17 876734 0.23 448113 0.12 1363772 0.35 2688619 0.69 N America 16430800 5.54 543600 0.18 577900 0.19 904600 0.31 2026100 0.68 3917700 0.49 805800 0.10 805800 0.10 Australia. Ne\1' Zealand 1128900 5.27 89000 0.42 27900 0.13 117000 0.55 233900 1.09 Source: Calculated from UNESCO (1997) and national sources. Blank spaces indicate data was not available. Issues and evidence 39 Table 2.7 Technical tertiary enrolments by country (% population), 1995 \ Korea 1.65 38 Bolivia 0.34 2 Finland 1.33 39 Costa Rica 0.34 3 Australia 1.17 40 Turkey 0.33 4 Taiwan 1.06 41 Ecuador 0.29 5 Spain 0.97 42 Vruguay 0.29 6 Ireland 0.90 43 Venezuela 0.29 7 Austria 0.78 44 El Salvador 0.26 8 Germany 0.77 45 Morocco 0.25 9 UK 0.75 46 Tunisia 0.24 10 Chile 0.73 47 Indonesia 0.23 11 Portugal 0.73 48 Nicaragua 0.22 12 Sweden 0.73 49 Honduras 0.20 13 Greece 0.72 50 Thailand 0.19 14 Canada 0.69 51 Brazil 0.18 15 Israel 0.68 52 S. Africa 0.17 16 N. Zealand 0.68 53 Hungary 0.16 17 USA 0.68 54 Malaysia 0.13 18 Norway 0.67 55 Egypt 0.12 19 Italy 0.64 56 India 0.12 20 Japan 0.64 57 Jamaica 0.11 21 France 0.61 58 Paraguay 0.11 22 Denmark 0.60 59 China 0.10 23 Panama 0.59 60 Zimbabwe 0.09 24 Netherlands 0.56 61 Bangladesh 0.08 25 Philippines 0.55 62 Nepal 0.08 26 Colombia 0.5\ 63 Sri Lanka 0.08 27 Switzerland 0.5\ 64 Cameroon 0.06 28 H. Kong 0.49 65 Madagascar 0.06 29 Romania 0.49 66 Pakistan 0.05 30 Argentina 0.47 67 Senegal 0.05 31 Singapore 0.47 68 Mauritius 0.04 32 Peru 0.46 69 Congo 0.03 33 Mexico 0.44 70 Kenya 0.02 34 Belgium 0.43 71 CAR 0.01 35 Jordan 0.42 72 Ethiopia 0.01 36 Algeria 0.41 73 Malawi 0.01 37 Poland 0.39 Source: Calculated from UNESCO (1997). 40 Competitiveness, FDI and technological activity in East Asia What we can measure is formal R&D. This is still useful, since R&D becomes an important input into competitiveness in countries at interme diate levels of industrialization. It is necessary to monitor technological developments overseas and select those relevant to local needs. This lowers the cost of technology transfer and captures more spillovers from the oper ation of TNCs. A growing R&D base permits better and faster technology diffusion within the economy and facilitates greater use of local resources. It makes it feasible and attractive for TNCs to locate their own design and development work there. Most importantly, it permits the industrial sector greater flexibility and diversification, and allows it greater autonomy. Table 2.8 shows R&D scientists and engineers and expenditures in the developing world. The patterns again reflect the technological depth of exports analysed earlier. Productive enterprise-financed R&D as a share of GNP . . . perhaps the best indicator of technologically useful R&D in the mature NIEs is nearly 400 times higher than in Sub-Saharan Africa, and around 10 times higher than in the new NIEs and Latin America. Asia as a whole accounts for 86 per cent of R&D scientists and engineers in the developing world, Sub-Saharan Africa for 0.3 per cent, and Latin America for 10 per cent. The proportion of enterprise-financed R&D in total R&D spending is highest in the mature NIEs, followed by the new NIEs, and lowest in Sub-Saharan Africa. Latin America and South Asia are similar, with below 10 per cent of national R&D financed by productive enterprises. The regional averages conceal variations at the national level. Figure 2.3 shows productive enterprise-financed R&D as a percentage of GNP for selected countries. Korea is again one of the leaders; its figure is the highest, not only in the developing world, but also, apart from Japan, in the world as a whole. 15 Taiwan comes next in the developing world, with a lower ratio than the UK but more than the Netherlands or Italy. Singapore comes next, though much lower in the world scale. While its high dependence on FDI has not held back the growth of private sector R&D (much of it in foreign affiliates), this has needed a strong government push, and the innovation base remains narrow. Hong Kong does not publish R&D data, but reports suggest that total R&D is only 0.1 per cent of GNP (NSF. 2000, table 2-14) and enterprise-financed R&D is a very small proportion of this. The other three Tigers are clearly a class apart in the developing world. Of the new Tigers, Malaysia leads while Thailand comes in last, surpris ingly lower even than Indonesia. 16 This reveals an important weakness in Thai competitiveness, the shallowness of its high-technology export activ ity (Lall, 2001). Malaysia has succeeded in raising R&D in MNCs (espe cially in electronics), adopting some of the same strategies as Singapore, but it has a long way to go before it can match the latter in technological competence. Table 2.8 R&D propensities and manpower in major country groups (latest year available) Countries and regions (a) Scientistslengineers Total Seetor of Source of R&D by in R&D R&D performance financing ('Yo ('%of distribution) GNP) Per mill. Numbers CX,of Productive Prod. Govt. Prod. Prod. pop. GNP) sector education enterprise sector Industrialized economies (b) 1102 2704205 1.94 53.7 22.9 53.5 38.0 1.037 1.043 Developing economies (c) 514 1034333 0.39 13.7 22.2 10.5 55.0 0.041 0.054 Sub-Saharan Africa (exc. S. Africa) 83 3193 0.28 0.0 38.7 0.6 60.9 0.002 0.000 North Africa 423 29675 0.40 nla nla nla nla nla nla Latin America & Caribbean 339 107508 0.45 18.2 23.4 9.0 78.0 0.041 0.082 Asia (excluding Japan) 783 893957 0.72 32.1 25.8 33.9 57.9 0.244 0.231 .... "" Mature NIEs (d) 2121 189212 1.50 50.1 36.6 51.2 45.8 0.768 0.751 New NIEs (e) 121 18492 0.20 27.7 15.0 38.7 46.5 0.077 0.055 S. Asia (f) 125 145919 0.85 13.3 10.5 7.7 91.8 0.065 0.113 Middle East 296 50528 0.47 9.7 45.9 11.0 51.0 0.051 0.045 China 350 422700 0.50 31.9 13.7 nla nla nla 0.160 transition economies (g) 1857 946162 0.77 35.7 21.4 37.3 47.8 0.288 0.275 (79-84 countries) 1304 4684700 0.92 36.6 24.7 34.5 53.2 0.318 0.337 Notes: (a) Only including countries with data, and with over 1 million inhabitants in 1995. (b) USA, Canada. West Europe, Japan. Australia and New Zealand. (c) Including Middle East oil states, (d) Hong Kong, Korea, Singapore, Taiwan (e) Indonesia, Malaysia, Thailand, Philippines. (I) India, Pakistan, Bangladesh, Nepal. (g) Including Russian federation. Source: Calculated from UNESCO (1997). Regional propensities for R&D spending arc simple averages. 42 Competitiveness, FDI and technological activity in East Asia 2.5 2.0 1.5 1.0 0.5 O.O-'-"".......----.."'-"'''-'''''..I:llL''''-''=....... -''''1.j''''''-'''''-''''~.wl.J''''_''''_'''''_''''iI.J!i!lUl!lLllI''_''''Lllm.__..'''__~_ _ Figure 2.3 R&D financed by productive enterprises, 1995 (% GNP) MNCs account for substantial portions of technological effort in Singapore, Malaysia, Brazil and Mexico. Interestingly, the latter two coun tries attract the most US MNC R&D in the developing world (UNCTAD, 1999), but are poor performers in overall terms. In Korea and Taiwan, R&D by local firms takes precedence, driven by strategies to restrict FDI inflows and reverse the passive reliance on foreign technologies that marks most developing countries. FDI As noted, trade and competitiveness are increasingly related to MNC activ ity. MNCs now account for large shares of world trade and their shares are higher in technologically advanced and differentiated products. A very large part of MNC trade is now intra~firm. In the USA, for instance, exports by MNCs to their majority-owned affiliates in 1996 comprised 48 per cent of parent company exports, up from 41 per cent in ] 977. Half of exports by foreign MNCs in the USA (accounting for 20 per cent of total US exports) were also intra-firm; similar trends are likely in other capital exporting countries. The propensity to engage in intra- as compared to inter-firm trade is higher in technologically complex and novel products. Within the web of intra-firm trade relationships is the emerging network of international production systems, with different stages of production and services located in different countries in accordance with relative costs and strategic considerations. Entry by developing countries into a large (and dynamic) segment of industrial activity and trade thus increasingly requires direct MNC (that is, equity-based) participation. Arm's length and joint venture relations with MNCs help, but they only provide partial entry into this segment, and only for countries that have strong local technological and innovative capabil Issues and evidence 43 Table 2.9 FD! as percentage of gross domestic investment (averages) 1980-85 1994-97 Singapore 18.72 27.81 China 0.87 13.24 Malaysia 1l.18 12.47 Hong Kong 6.90 9.93 Philippines 0.37 8.20 Indonesia 1.00 6.60 Thailand 2.41 4.18 Taiwan 1.50 3.05 Korea, Rep. 0.40 l.ll Cambodia 0.00 27.41 Laos 0.00 23.14 Vietnam 0.00 22.88 Pakistan 1.39 6.27 Sri Lanka 3.02 5.54 India 0.14 2.46 Bangladesh 0.00 0.49 Chile 6.23 23.14 Mexico 2.41 12.50 Argentina 2.98 11.39 Brazil 4.19 5.97 Ireland 4.14 15.05 UK 6.53 12.34 France 2.02 8.32 USA 2.74 5.81 Germany 0.60 1.53 Japan 0.09 0.09 Sources: World Bank, (2000), UNCTAD (2002). ities. Even for such countries, the needs of scale, specialization, and access to new technologies and international marketing makes it imperative to belong to MNC systems rather than stay outside them. This is why world technological leaders encourage interpenetration by each other's MNCs, strategic alliances and cross-border M&As. However, few developing countries participate in these emerging MNC systems. While FDI in developing countries is rising rapidly (from an average of US$29 billion in 1986-91 to US$208 billion in 1999), flows are highly con centrated. The top 10 developing countries account for nearly 80 per cent, and the top 25 for 95 per cent, of the total. Table 2.9 shows FDI inflows as a percentage of gross domestic investment in East Asia (including some 44 Competitiveness, FDI and technological activity in East Asia formerly socialist economies) and comparators in Latin America, South Asia and the OECD, averaged for 1980-85 and 1994-97, Singapore has been, and remains, the country most reliant on internal ized technology transfers by MNCs in the region. Malaysia has tradition ally been the second, but in recent years has been overtaken by China (which in absolute terms is the largest recipient of FDI in the developing world). The Philippines has greatly raised its attractiveness to foreign inves tors, as has Indonesia. Thailand remains a relatively modest player by regional standards. Taiwan, and to a greater extent Korea, have been rela tively restrictive on FDI inflows (and have been net exporters of FDI for some time). However, Korea has become far more receptive to inward investments after the financial crisis. In fact, it saw a dramatic upsurge of inward FDI in 1998-99, accounted for by a burst of cross-border M&A activity: inflows thus rose from US$3.1 billion in 1997 to US$10.3 billion in 1999 (UNCTAD, 2000). Thailand has seen a similar spurt, largely into the financial sector but also into manufacturing industries. Indonesia, by contrast, has suffered negative inflows as a result of the political instability following the crisis. The relationship between reliance on FDI (internalized technology inflows) and domestic R&D effort varies greatly within East Asia. The 'autonomous' countries, Korea and Taiwan, built up strong innovative capabilities by restricting internalized technology inflows and using a battery of promotional measures to encourage technological development and R&D by local enterprises (Lall, 1996). Of the FDI-dependent coun tries, only Singapore had a coherent and consistent policy for encouraging local R&D, encompassing both foreign and local firms and using tools and incentives to get MNCs to upgrade affiliate technological capabilities. The other economies (China apart) did not have technology promotion policies for MNCs (though Malaysia had a mild and largely ineffective approach along Singaporean lines); for domestic firms they had some general policies but without much 'bite'. As a result, their capabilities were largely confined to production technologies. This pattern suggests that countries with ambitious technological objec tives need to mount coherent and targeted strategies to induce firms to move up the skill and innovation ladder. Building a strong and diverse inno vation base in domestic enterprises seems to need selective policies to restrict FDI. Raising R&D in foreign affiliates needs a different set of pol icies, but it is clearly feasible if the host country is able to provide world class skills, infrastructure and business environment. Issues and evidence 45 2.5 Capability building Bubble size indicates average value of strategy a a RCA in higlHech exports 2 l~ Korea ,0 1.5 (UlW<1n FDI-dependence RCA 147 <:>(l \ strategy ~ "GECD Chma ~ RCA 086 Hong Kong ~ ~_ Malay,.. 0.5 Devclopmg countncs MexIco RCA: 0.12 ('0\ I PhliJppmes 0 +---r-----~-- ~ \....~d Smgapore ThaJland RCA 161 -0.5 I -2 0 2 4 6 8 10 12 14 16 18 FDI Figure 2.4 RCA in high-technology exports and main industrial strategies. 1995 6. CLUSTER ANALYSIS OF COMPETITIVENESS STRATEGIES We can illustrate the different strategies for Asian and other countries (73 in total) for 1995. The dependent variable is competitiveness, measured by RCAs in different technological categories. The independent variables are FDI (as a percentage of gross domestic investment) and technological effort (R&D by productive enterprises as a percentage of GOP). The technique used is K-means cluster analysis. l ? Cluster analysis groups' observations on the basis of selected characteristics, help in this case to identify groups of countries with similar export patterns and strategies. Strategies in turn are indicated by different degrees of reliance on FDI and domestic R&D. Figure 2.4 is a three-dimensional representation of RCAs and these two determinants of competitive performance. It shows clearly how Korea and Taiwan stand out from the other high performing Asian economies in terms of their higher reliance on domestic R&D rather than R&D. It also shows that FDI-dependent countries in the region have higher RCAs, being able to enter MNC international production systems. Clearly the growth of these production systems has transformed compet itive patterns in some countries very quickly, allowing them to bypass the slow and arduous process of building domestic capabilities. It does not, however, ensure a sustainable base of capabilities in the longer term, as their wages rise and technologies become more complex and skill intensive. In strategic terms, the data show clearly the differences between what we have termed 'autonomous' and 'FDI-dependent' approaches to building competitiveness. The implications are that countries that wish to pursue autonomous strategies need to mount enormous efforts to compensate for 46 Competitiveness, FDf and technological activity in East Asia the lack of internalized technology transfers. Those that are unable to do this would be wise to depend on MNCs and direct their efforts towards inte gration into their international production systems. However, most coun tries have not been able to do either the aim of these countries should be to discover what they need to do to follow one pattern or the other. Note, however, that the strategic choices illustrated by these figures relate to the past. The options for the future are much narrower, as the 'rules of the game' change (towards a smaller role for government) and are more rigor ously enforced. 7. CONCLUSIONS The figures on export competitiveness in East Asia are impressive. The region is the best placed in the developing world to sustain export compet itiveness. It is a dominant player in developing world trade. It is well posi tioned in fast growing products and shows great flexibility in adapting to changing patterns of technology and demand. Its specialization in high technology products points to considerable development of skill and tech nological capabilities. These generalizations conceal important national differences. Different Asian exporters base their competitiveness on different agents and factors. At one extreme, Korea and Taiwan have a strong domestic base of enter prises, skills, innovation and institutions. At the other, some very success ful exporters have weak domestic enterprises, a shallow technology base, an export sector isolated from local industry, and an inadequate skill creation system. This is not to say that the former set necessarily has better long-term competitive prospects than the latter. Each group faces its own risks and challenges. For instance, those with autonomous strategies expose them selves to greater risk the more they approach technological frontiers and the more rapid the pace of technical change. Those highly dependent on MNC networks to provide technology do not face the same technological risk. However, looking to the future, we can argue that the requirements of continued competitiveness are likely to be very different from the past. Passive FDI-dependent countries with low levels of skill and R&D may find themselves unable to attract the most advanced or dynamic technolo gies if their skill and supplier base does not rise to the level needed. Globalization does not, in other words, reduce the role of local capabil ities and innovative activity beyond the short term where an assembly base is established. On the contrary, it raises it because technical efficiency in each location becomes the final determinant of success. As rising wages and Issues and evidence 47 technical change force countries into more complex activities, they have to furnish more advanced capabilities. Skill development, industrial special ization, enterprise learning and institutional change are needed to create cumulative and self-reinforcing processes to promote further learning, regardless of how much countries rely on MNCs. Autonomous strategies as demonstrated by Korea and Taiwan entail a great deal of industrial policy and accompanying interventions in factor markets and institutions. They lead to a massive development and deepen ing of indigenous skills and technological capabilities, with the national ability to keep abreast of new technologies and for domestic enterprises to become significant global players in their own right. However, such strate gies are increasingly difficult and risky on economic grounds - the sheer pace of technical change and the growth of international production systems raise the costs of being left on the outside. They are also increas ingly constricted by the new rules of the game being laid down by interna tional agencies and developed countries. FDI-dependent strategies comprise two sub-strategies, targeted and passive. Targeted strategies - as in Singapore - also entail considerable industrial policy, but the intensity of government interventions is lower than with autonomous strategies. The sources of technical change remain largely outside, in the hands of MNCs; there is less need to intervene to promote learning in infant industries for this reason. However, industrial policy is needed to ensure the development of the relevant skills, capabil ities and institutions required to ensure that TNCs keep transferring new technologies and higher value functions. Passive strategies involve less industrial policy in export-oriented activities to start with (though there may be intervention in domestic-oriented activity). However, they need to evolve into more targeted strategies if countries are not to lose their com petitive positions and momentum. National technology strategies are now starting to converge. Autonomous countries are becoming more integrated into MNC systems (and have many capable MNCs of their own). FDI-dependent countries are trying to strengthen capabilities in domestic firms and build up the institutional struc ture for innovation. Those using passive FDI strategies are moving towards more targeted strategies. These changes are driven both by new technologies and globalization as well as by new rules of the game, and are likely to persist into the foreseeable future. This does not, however, mean that countries will converge technologically. There will remain significant differences in techno logical and competitive performance, even among the successful exporters of East Asia, because of differences in endowments (size, location, resources and so on) and in inherited structures of technological learning. National systems of technology development have elements of path dependence and 48 Competitiveness, FDI and technological activity in East Asia stability (Lall, 2000a) and can change only as the institutional, technological and human capital base evolves necessarily a slow process. Inherited struc tures also influence how flexibly and dynamically countries respond to new competitive challenges: this feedback process can let leaders maintain their advantage for very long periods. FDI can help change national technologi cal systems, but the real driver of change lies within each economy. Government policies and institutional structures playa vital role here, and this role remains even as its form and content evolves. NOTES I. I am grateful to Manuel Albaladejo for doing the cluster analysis reported at the end of the paper, and to various members of the World Bank Institute study, in particular Bee Van Aw and Rajah Rasiah. for helpful comments. I would like to acknowledge my debt to UNCTAD for classifying the export data at my request. The usual disclaimers apply. 2. See UNIDO (2002) for data on the values and concentration of global and national R&D spending. 3. Despite their emphasis on human capital and technology, endogcnous growth models also assume that in developing countries openness to trade and investment (both condu cive to technology flows) is both necessary and sujficienl. 4. The international technology market is fragmented and ill defined, and searching for the optimal technology deal can be costly and difficult. It is not easy to define the technol ogy 'product' or its price. The transfer can take many different forms (that is, the product is not well specified). Much depends on how much technical and other information the seller includes (or the buyer asks for) and how it transmits this information and modifies it over time. The seller knows more about the 'product' than the buyer does (otherwise it would have nothing to sell): the buyer thus operates under an information asymmetry, largely absent in transactions in physical products. Even with full information, the two parties can have different valuations of the technology depending on their market posi tions, expectations and technological eapabilities. Since technological inlormation is constantly changing, the valuation also depends on which vintage is being transferred and how its future evolution is foreseen. For these reasons. the price and terms of tech nology transfer are subject to bargaining and the accompanying uncertainty and non transparency. 5. These are explored in 'new economic geography'; see, for instance, Krugman (1995) and Venables (1996). 6. One may compare unit values of exports over time or across countries to get a rough indication of technical change and quality, but this is only possible for a few (relatively homogeneous) products and countries. When the objective is to build up a broad picture of export patterns by technology levels, this procedure cannot be applied. 7. Ideally, the data should distinguish between levels of technology at a fairly disaggregated level. but this is not possible. We use the UN Standard International Trade Classification (SITC) data at the 3-digit level (Revision 2). This level can put together products of different levels of technological complexity in the same category. For instance, telecom munications apparatus includes advanced telephone technology as well as simple tele phone receivers. The export data do not distinguish quality differences within categories, such as fashion clothing from mass-produced items. 8. For instance, semiconductor exports can be based on high-tech processes in the USA and simple assembly and testing in the Philippines: both appear as 'high technology'. 9. The export data are all in terms of current US dollars and do not show volume changes. Since the main purpose of the exercise is to compare regions and countries across tech Issues and evidence 49 nological categories, this does not matter very much, since a general price deflator would apply to all equally. However, it does mean that relative price changes between product categories cannot be taken into account. 10. Transition economies in Eastern Europe and Central Asia are excluded because of the very patchy nature of the data available over the period. Developed economies are defined to include [srael, South Europe but not Turkey (which is included in the Middle East group). Developing countries are defined to include the South Africa, the mature Asian Tigers, China and Asian transition economies (like Vietnam), and all Latin American countries (including Mexico). Data for 1998 have several missing values for developing countries like Bangladesh, Sri Lanka and many African countries, none of which are major exporters in the developing world. Data for 1980 could not be used because they had missing values for major Latin American exporters. II. 'East Asia' includes all countries in Asia east of Myanmar, including Myanmar and Vietnam (but not Laos or Cambodia) and China, and excludes Japan and Central Asian transition countries. 'South Asia' comprises India, Pakistan, Bangladesh, Sri Lanka, Maldives, Nepal and Bhutan. 'MENA' (Middle East and North Africa) includes Afghanistan and Turkey as well as all Arab countries (Sudan is counted under SSA). 'SSA' (Sub-Saharan Africa) includes South Africa (SSA I) unless specified (SSA2). 'LAC' (Latin America and the Caribbean) includes Mexico (LACl) and excludes it (LAC2) when specified. 12. Mexico is treated as an outlier because of its proximity to the USA and the unusual nature of its trading relations. Mexico has long been a base for export-oriented assem bly by US firms in its border maquiladoras, which were allowed to import duty-free inputs and sell the finished product to the USA with tariffs levied only on the value added. The formation of NAFTA in the mid-1990s gave offshore assembly a new fillip and brought Mexico into a position to challenge Asia. NAFTA allowed significant new privileges like allowing local inputs for duty exemption; this led to dramatic rises across all export categories and to a huge rise in FDI from Asia to use Mexico as an export base for the USA. Mexico now accounts for more manufactured exports than the rest of Latin America put together. South Africa is an outlier in SSA for more obvious reasons. It accounted for 55 per cent of manufacturing value-added in SSA in 1998, and for 45 per cent of its manufactured exports. 13. 'Revealed comparative advantage' is defined as the share of a country's exports of a par ticular product, say clothing, divided by the share of its total exports in world exports. Thus, an RCA ratio of over I for clothing shows that the country has a revealed com parative advantage in clothing: its global share of clothing exports is higher than the share of all its exports. 14. Given its relatively low wages, for instance, the Philippines is performing poorly in tra ditionallabour-intensive exports (Lan, 2001). See Chapter lOon the Philippines for an analysis of the shallow base of its high-technology exports. 15. However, by 1997 Sweden had overtaken Korea and Japan. see UNIDO (2002). 16. However, this may be due to measurement errors and the real figure is likely to be some what higher, taking it ahead of Indonesia but still behind Malaysia. I am grateful to Peter Brimble for this comment. 17. K-means duster analysis is used to cluster large numbers. The number of groups has to be specified in advance: we specified five. Using the squared Euclidean distance (the sum of the squared differences over all of the variables), we identify 'five initial cluster centres' as a reference point for the other cases. Once the cases are classified. we obtain the final cluster centres, which are simply the average values of the variables lor cases in the clusters. 50 Competitiveness, FDI and technological activity in East Asia REFERENCES Best, M. (1990), The New Competition: Institutions of Industrial Restructuring, Cambridge: Polity Press. Hobday, M.o. (1995), Innovation in East Asia: The Challenge to Japan, Aldershot, UK and Brookfield, US: Edward Elgar. Krugman, P. (1995), Development, Geography and Economic Theory, Cambridge, MA: MIT Press. Krugman, P. (1996), 'Making sense of the competitiveness debate'. Oxford Reviell: of Economic Policy, 12(3), l7~25. Lall, S. (1992), 'Technological capabilities and industrialization', World Development, 20(2), l65~86. Lall, S. (1993), 'Understanding technology development', Development and Change, 24(4), 719~53. Lall, S. (1996), Learningfrom tlte Asian Tigers: Studies in Technology and Industrial Policy, London: Macmillan. Lall. S. (1999), 'Competing with labour: skills and competitiveness in developing countries', Geneva: ILO, Issues in Development discussion paper 31. Lall, S. (2000a), 'The technological structure and performance of developing country manufactured exports, 1985~98', Oxford Development Studies, 28(3), 337-69. Lall. S. (2000b), 'Technological change and industrialization in the Asian NIEs', in L. Kim and R.R. Nelson (eds), Technological Learning and Economic Development: The Experience of the Asian NIEs, Cambridge: Cambridge University Press, pp. l3~68. Lall, S. (2001), Competitiveness. Technology and Skills, Cheltenham, UK and Northampton. USA: Edward Elgar. Mathews, lA. and D.S. Cho (2000), Tiger Technology: The Creation of a Semiconductor Industry in East Asia, Cambridge: Cambridge University Press. Nelson, R.R. (ed.) (1993), National Innovation Systems: A Comparative Analysis, Oxford: Oxford University Press. Nelson, R.R. and S.l Winter (1982), An Evolutionary Theory of Economic Change. Cambridge, MA: Harvard University Press. NSF (2000), Science and Engineering Indicators 200f), Washington, DC: National Science Foundation. Pack, H. and K. Saggi, (1997), 'Inflows of foreign technology and indigenous tech nological development', Review of Development Economics, 1(1), 81·98. Radosevic, S. (1999), International Technology Transfer and Catch- Up in Economic Development, Cheltenham, UK and Northampton, USA: Edward Elgar. Stiglitz, IE. (1996). 'Some lessons from the East Asian miracle', World Bank Research Observer, 11(2), 151~77. UNCTAD (1999), World Investment Report 1999: Foreign Direct Investment and the Challenge of Development, Geneva and New York: United Nations. UNCTAD (2000), World Investment Report 2000: Cross-Border Mergers and Acquisitions and Development, Geneva and New York: United Nations. UNESCO (1997), Statistical Yearbook 1997, Paris: United Nations Educational, Scientific and Cultural Organization. UNIDO (2002), Industrial Development Report 2002, Vienna: United Nations Industrial Development Organization. issues and evidence 51 Venables, A.J. (1996), 'Localization of industry and trade performance', Oxford Review of Economic Policy, 12(3), 52-60. Westphal, L. E. (2003), Technology strategies for economic development in a fast changing-global economy', Economics (If innovation and New Technology. World Bank (1993), The East Asian Miracle, Oxford: Oxford University Press. World Bank (1999), World De>'elopment Report 1999, Oxford: Oxford University Press. World Bank (2000), World Development indicators, Washington, DC: World Bank. 52 Competitiveness, FD! and technological activity in East Asia ANNEX Table 2A.1 Leading 13 exporters oj manuJac tures in 1998, values oj exports in 1985 and 1998 (US$ million) 1985 Total RB RBI RB2 LT LTi LT2 mfrs. total total China 6049.2 2349.7 677.7 1672.0 2645.2 2217.4 427.8 Korea 29025.0 2493.6 1016.7 1477.0 12017.6 7892.5 4125.1 Taiwan 29092.5 2883.2 1690.3 1192.9 15381.6 8597.1 6784.5 Mexico 8336.3 1761.4 401.4 1360.0 1097.2 429.3 667.8 Singapore 19014.0 8266.6 1527.8 6738.7 1640.2 764.7 875.5 Malaysia 8626.5 4632.1 3998.9 633.3 692.4 466.6 225.8 Thailand 3657.6 1386.5 1029.8 356.6 1295.1 981.8 313.3 Brazil 17616.8 7744.6 3563.2 4181.4 3757.2 2042.9 1714.4 Philippines 2428.7 1359.1 1047.0 312.0 585.7 352.3 233.4 Indonesia 3856.4 2899.3 1690.4 1208.9 596.2 505.1 91.1 India 6208.9 2518.8 255.] 2263.8 2813.2 2542.2 270.9 Hong Kong 15979.5 504.2 278.1 226.1 10063.3 7045.7 3017.6 Turkey 5790.4 1263.4 590.3 673.1 3075.5 2138.6 936.9 Total above 13 155681.5 37568.9 17766.8 22295.8 43642.7 35976.2 19684.1 Developing 210244.6 79986.4 24308.2 55678.2 63839.8 42194.1 21645.7 world fy<) 74.0 47.0 73.1 40.0 68.4 85.3 90.9 MT MTI MT2 MT3 HT HTI HT2 total total China 738.9 28.6 589.5 120.8 315.4 38.6 276.8 Korea \0807.1 678.6 3020.4 7108.1 3706.7 3287.7 419.0 Taiwan 6124.1 865.3 1668.5 3590.4 4703.6 4430.0 273.6 Mexico 3600.7 766.1 415.6 2419.0 1877.0 1736.4 140.5 Singapore 4445.2 146.4 1066.3 3232.4 4662.1 3916.9 745.3 Malaysia 982.1 20.5 262.9 698.7 2319.9 2167.4 152.5 Thailand 803.8 13.9 289.0 501.0 172.2 147.7 24.5 Brazil 5249.6 1287.2 2079.7 1882.7 865.4 578.4 287.1 Philippines 217.8 21.1 163.5 33.3 266.1 256.5 9.6 Indonesia 246.1 0.9 222.2 23.0 114.8 80.9 33.9 India 624.7 118.9 171.5 334.3 252.1 97.4 154.8 Hong Kong 3050.2 1.9 196.4 2851.8 2361.8 2163.1 198.7 Turkey 1359.6 102.8 752.4 504.3 91.9 64.7 27.2 Total above 13 27442.7 4052.2 10897.8 23299.8 18002.2 18965.5 2743.4 Developing 43369.6 4380.5 14040.5 24948.6 23048.7 19490.7 3558.0 world 0/0 63.3 92.5 77.6 93.4 78.1 97.3 77.1 Issues and evidence 53 1998 Total RB RBI RB2 LT LTl LT2 rufrs. total total China 167681.1 16551.3 7155.4 9395.8 83803.2 52814.7 30988.5 Korea 120700.3 12914.5 4739.6 8175.0 25325.3 13673.1 11652.2 Taiwan 105553.7 5811.3 2761.3 3050.1 32100.7 14291.0 17809.7 Mexico 103681.3 6977.1 3743.6 3233.5 19848.6 9358.2 10490.4 Singapore 103488.5 14588.6 3471.0 11117.6 7254.0 2226.8 5027.2 Malaysia 65940.5 11004.8 9543.2 1461.7 7245.9 3301.7 3944.3 Thailand 44759.5 8657.7 5532.4 3125.3 11345.3 6798.2 4547.1 Brazil 38881.6 15424.7 9319.0 6105.7 5900.6 3158.6 2742.0 Philippines 28118.8 2022.3 1548.6 473.7 4074.3 2988.2 1086.1 Indonesia 26894.8 \0447.6 7154.9 3292.7 8868.8 5511.1 3357.7 India 25855.1 7801.8 847.5 6954.3 12583.4 9977.4 2606.0 Hong Kong 23136.7 1041.7 661.4 380.3 13034.7 11049.2 1985.5 Turkey 22885.2 3339.9 2204.1 1135.8 13236.9 \0276.1 2960.8 Total above 13 877577.3 116583.5 58681.9 57901.6 244621.8 145424.2 99197.6 Developing 996967.5 175130.4 77385.7 97744.7 277435.3 169990.4 107444.9 world 0;;) 88.0 66.6 75.8 59.2 88.2 85.5 92.3 MT MTl MT2 MT3 HT HTl HT2 total total China 33853.9 1864.0 10556.4 21433.4 33472.8 28605.5 4867.3 Korea 46443.7 11354.5 14998.0 20091.3 36016.7 32800.6 3216.2 Taiwan 29044.5 4256.5 9644.3 15143.6 38597.2 37259.0 1338.2 Mexico 45598.6 19200.6 5264.1 21133.9 31257.0 28055.0 3202.0 Singapore 19326.2 861.8 5091.3 13373.0 62319.7 59674.4 2645.2 Malaysia 13360.2 455.2 3107.9 9797.0 34329.6 32276.3 2053.3 Thailand 9165.0 1014.8 2438.8 571!.5 15591.5 14593.9 997.5 Brazil 14363.8 4770.0 4563.9 5029.9 3192.5 1476.4 1716.0 Philippines 3058.9 382.2 346.8 2329.9 18963.3 18673.5 289.8 Indonesia 4972.1 310.0 2647.5 2014.6 2606.3 2381.3 225.0 India 3763.5 735.2 1820.4 1208.0 1706.3 708.5 997.8 Hong Kong 3044.5 0.7 717.9 2325.9 6015.8 4920.1 J 095.7 Turkey 4870.8 761.5 J 992.9 2116.4 1437.7 1156.3 281.3 Total above 13 230865.8 45967.0 63190.3 121708.5 285506.2 262580.8 22925.4 Developing 254289.1 51537.3 75515.3 127236.4 290112.8 265114.5 24998.3 world % 90.8 89.2 83.7 95.7 98.4 99.0 91.7 54 Compelitivenes!>; FD! and technological activity in East Asia Table 2A.2 Leading developing country exporters in 1998, distribution by technological category and growth rates, 1985-98 Shares of LOtal manufactured exports by technological categories 1985 nq Total RB RBI RB2 LT LTI LT2 mfrs. total total China 100 38.8 11.2 27.6 43.7 36.7 7.1 Korea 100 8.6 3.5 5.1 41.4 27.2 14.2 Taiwan 100 9.9 5.8 4.1 52.9 29.6 23.3 Mexico 100 21.1 4.8 16.3 13.2 5.2 8.0 Singapore 100 43.5 8.0 35.4 8.6 4.0 4.6 Malaysia 100 53.7 46,4 7.3 8.0 5,4 2.6 Thailand 100 37.9 28.2 9.8 35,4 26.8 8.6 Brazil 100 44.0 20.2 23.7 21.3 11.6 9.7 Philippines 100 56.0 43.1 12.8 24.1 14.5 9.6 Indonesia 100 75.2 43.8 31.3 15.5 13.1 2,4 India 100 40.6 4.1 36.5 45.3 40.9 4.4 Hong Kong 100 3.2 1.7 1,4 63.0 44.1 18.9 Turkey 100 21.8 10.2 11.6 53.1 36.9 16.2 Shares of total manufactured exports by technological categories 1985 ('X) MT MTl MT2 MT3 HT HTI HT2 total total China 12.2 0.5 9.7 2.0 5.2 0.6 4.6 Korea 37.2 2.3 10,4 24.5 12.8 11.3 1.4 Taiwan 21.1 3.0 5.7 12.3 16.2 15.2 0.9 Mexico 43.2 9.2 5.0 29.0 22.5 20.8 1.7 Singapore 23,4 0.8 5.6 17.0 24.5 20.6 3.9 Malaysia 11.4 0.2 3.0 8.1 26.9 25.1 1.8 Thailand 22.0 0,4 7.9 13.7 4.7 4.0 0.7 Brazil 29.8 7.3 11.8 10.7 4.9 3.3 1.6 Philippines 9.0 0.9 6.7 1.4 11.0 10.6 0,4 Indonesia 6,4 0.0 5.8 0.6 3.0 2.1 0.9 India 10.1 1.9 2.8 5,4 4.1 1.6 2.5 Hong Kong 19.1 0.0 T.2 17.8 14.8 13.5 1.2 Turkey 23.5 1.8 13.0 8.7 1.6 l.l 0.5 Shares of total manufactured exports by technological categories 1998 (%) Total RB RBI RB2 LT LTI LT2 mfrs. total total China 100 9.9 4.3 5.6 50.0 31.5 18.5 Korea 100 10.7 3.9 6.8 21.0 11.3 9.7 Taiwan 100 5.5 2.6 2.9 30.4 13.5 16.9 Mexico 100 6.7 3.6 3.1 19.1 9.0 10.1 Issues and evidence 55 Singapore 100 14.1 3.4 10.7 7.0 2.2 4.9 Malaysia 100 16.7 14.5 2.2 11.0 5.0 6.0 Thailand 100 19.3 12.4 7.0 25.3 15.2 10.2 Brazil 100 39.7 24.0 15.7 15.2 8.1 7.1 Philippines 100 7.2 5.5 1.7 14.5 10.6 3.9 Indonesia 100 38.8 26.6 12.2 33.0 20.5 12.5 India 100 30.2 3.3 26.9 48.7 38.6 10.1 Hong Kong 100 4.5 2.9 1.6 56.3 47.8 8.6 Turkey 100 14.6 9.6 5.0 57.8 44.9 12.9 Shares of total manufactured exports by technological categories 1998 (%) MT MTI MT2 MT3 HT HTI HT2 total total China 20.2 l.l 6.3 12.8 20.0 17.1 2.9 Korea 38.5 9.4 12.4 16.6 29.8 27.2 2.7 Taiwan 27.5 4.0 9.1 14.3 36.6 35.3 1.3 Mexico 44.0 18.5 5.1 20.4 30.1 27.1 3.1 Singapore 18.7 0.8 4.9 12.9 60.2 57.7 2.6 Malaysia 20.3 0.7 4.7 14.9 52.1 48.9 3.1 Thailand 20.5 2.3 5.4 12.8 34.8 32.6 2.2 Brazil 36.9 12.3 11.7 12.9 8.2 3.8 4.4 Philippines 10.9 1.4 1.2 8.3 67.4 66.4 1.0 Indonesia 18.5 1.2 9.8 7.5 9.7 8.9 0.8 India 14.6 2.8 7.0 4.7 6.6 2.7 3.9 Hong Kong 13.2 0.0 3.1 10.1 26.0 21.3 4.7 Turkey 21.3 3.3 8.7 9.2 6.3 5.1 1.2 Annual rates of export growth 1985-98 (%) Total RB RBJ RB2 LT LTi LT2 mfrs. total total China 29.1 16.2 19.9 14.2 30.5 27.6 39.0 Korea 11.6 13.5 12.6 14.1 5.9 4.3 8.3 Taiwan 10.4 5.5 3.8 7.5 5.8 4.0 7.7 Mexico 21.4 11.2 18.7 6.9 24.9 26.8 23.6 Singapore 13.9 4.5 6.5 3.9 12.1 8.6 14.4 Malaysia 16.9 6.9 6.9 6.6 19.8 16.2 24.6 Thailand 21.2 15.1 13.8 18.2 18.2 16.0 22.8 Brazil 6.3 5.4 7.7 3.0 3.5 3.4 3.7 Philippines 20.7 3.1 3.1 3.3 16.1 17.9 12.6 Indonesia 16.1 10.4 11.7 8.0 23.1 20.2 32.0 India 11.6 9.1 9.7 9.0 12.2 11.I 19.0 Hong Kong 2.9 5.7 6.9 4.1 2.0 3.5 -3.2 Turkey 11.2 7.8 10.7 4.1 11.9 12.8 9.3 Total above 13 14.2 9.1 9.6 7.6 14.2 11.3 \3.2 All developing 12.7 6.2 9.3 4.4 12.0 11.3 13.1 56 Competitiveness, FDI and technological activity in East Asia Table 2A.2 (continued) Annual rates of export growth 1985--98 (%) MT MTI MT2 MT3 HT HTJ HT2 total total China 34.2 37.9 24.8 48.9 43.2 66.2 24.7 Korea 11.9 24.2 13.1 8.3 19.1 19.4 17.0 Taiwan 12.7 13.0 14.4 IJ.7 17.6 17.8 13.0 Mexico 21.6 28.1 21.6 18.1 24.2 23.9 27.2 Singapore 12.0 14.6 12.8 11.5 22.1 23.3 10.2 Malaysia 22.2 26.9 20.9 22.5 23.0 23.1 22.1 Thailand 20.6 39.1 17.8 20.6 4 \.4 42.4 33.0 Brazil 8.1 10.6 6.2 7.9 10.6 7.5 14.7 Philippines 22.5 25.0 6.0 38.7 38.8 39.1 30.0 Indonesia 26.0 56.7 21.0 41.1 27.2 29.7 15.7 India 14.8 15.0 19.9 10.4 15.8 16.5 15.4 Hong Kong 0.0 ··-7.5 10.5 - 1.6 7.5 6.5 14.0 Turkey 10.3 16.6 7.8 11.7 23.6 24.8 19.7 Total above 13 17.8 20.5 14.5 13.6 23.7 22.4 17.7 All developing 14.6 20.9 13.8 13.4 21.5 22.2 16.2 3. Competitiveness and technology: an international comparison Hiroki Kawai and Shujiro Urata 1 1. INTRODUCTION The economies of East Asia achieved remarkable economic growth in the post-World War II period. Indeed, the World Bank (1993) published a study entitled The East Asian Miracle to examine the factors that underlay this export-led growth. It analysed the 1960-89 period and attributed success to several factors including rapid accumulation of physical and human capital, export-oriented trade policy, sound macroeconomic envi ronment and well-functioning institutions. The report also emphasized rapid expansion in productivity as an important factor that contributed to remarkable economic growth. The World Bank study attracted a lot of attention. Economists and pol icymakers were particularly interested in the role of government in eco nomic development, because the study acknowledged successful cases of government intervention in the sectoral allocation of resources. This view was quite different from the minimalist view of the government that had been popular earlier. Some observers argued that the success of East Asia was largely due to active government intervention, while others argued that it was attributable to the market mechanism. Consensus has yet to be reached on the issue and the debate on the appropriate role of the govern ment in economic development goes on. The World Bank study instigated another interesting debate on the sources of economic growth in East Asia. Several researchers questioned the contribution of productivity increase to economic growth in East Asia. The World Bank calculated high total factor productivity (TFP) growth for East Asian economies. Krugman (1994) questioned the validity of these findings by referring to the low TFP growth estimates obtained by Young (1995). Krugman argued that the rapid growth achieved by East Asian economies was largely due to the accumulation of labour and capital inputs, and not to productivity increase. This debate reconfirmed the importance of productivity growth in achieving eeonomic growth; it also 57 58 Competitiveness, FD! and technological activity in East Asia brought out the sensitivity of TFP estimates, which depended on the assumptions and the method applied for the analysis. This chapter examines the factors underlying the growing competitive ness of East Asian economies in the pre-financial crisis period. Recognizing the importance of technology in economic grmvth, it focuses on technolog ical capabilities in East Asian economies and its relation to the use of foreign technology. Section 2 examines changes in East Asian competitiveness from 1970-97. The analysis of competitiveness uses two indicators, per capita GOP and total factor productivity (TFP). The use of TFP as an indicator of competitiveness is based on the premise that productivity increase is a crucial factor in growth. Per capita GOP is used as a proxy for labour pro ductivity. It is a reasonable proxy because of limited availability of informa tion on labour inputs such as the number of workers or the number of hours worked. It may be added that per capita GOP indicates the stage of eco nomic development, a main concern for the countries striving to achieve economic development. Our analysis covers the period up to the financial crisis, and thus extends the period examined by The East Asian Miracle. Section 3 analyses technological factors in East Asia. The first part of the discussion deals with domestic technological capabilities and the second with inflows of foreign technology. These two aspects are, of course, closely related because assimilation of foreign technology, a major source of build ing competitiveness, depends crucially on domestic technological capabil ity. We attempt to identify by regression analysis the determinants of competitiveness as measured by the growth of per capita GOP and TFP. Section 4 concludes with policy recommendations. 2. COMPETITIVENESS OF EAST ASIAN ECONOMIES: ECONOMIC GROWTH AND TFP GROWTH Table 3.1 shows the growth performance of East Asian economies between 1970 and 1997. The average annual growth rate in per capita GOP exceeded 5 per cent in this period. 2 The growth rate accelerated over time, from 5.2 per cent in the 1970s to 6.1 per cent in the 1980s and then to 6.4 per cent in 1990-97. This performance was significantly and consistently better than in other regions. 2.1 Economic Growth There are wide variations in growth rates within East Asia. The newly industrializing economies (Hong Kong, Korea, Singapore and Taiwan) Table 3.1 Level and growth ofpel' capita GDp, 1970-97 Per capita GOP level (1992 US$,PPP) Annual growth rate (°lc,) 1970 1980 1990 1997 197080 1980-90 1990-97 Developed countries 13267 16679 20510 22988 2.39 2.12 1.62 Developing countries 4201 5588 5804 7185 3.39 1.90 3.12 1. East Asia & the Pacific 1998 3622 6611 9703 5.13 6.13 6.36 NIEs 3071 5891 11126 16044 6.46 6.58 5.35 Hong Kong 5438 10324 17205 21540 6.41 5.11 3.21 Korea 2509 4367 9240 13831 5.54 7.49 5.76 Singapore 4498 8722 14873 23296 6.62 5.34 6.41 v. Taiwan 2724 5691 10668 15578 7.37 6.28 5.41 \0 (2) ASEAN4 1547 2389 3378 4728 4.54 3.77 4.69 Indonesia 764 1291 1993 2917 5.25 4.34 5.44 Malaysia 2527 4235 5707 8486 5.16 2.98 5.67 Philippines 2629 3629 3311 3493 3.22 -0.92 0.76 Thailand 1545 2294 4104 5914 3.95 5.82 5.22 (3) China 493 690 1437 2801 3.35 7.34 9.53 2. South Asia 962 1030 1454 1845 0.65 3.43 3.37 3. Latin America and the Caribbean 4626 6515 5966 6971 3.80 -0.73 2.05 4. Europe and Central Asia 3664 4933 5844 6835 2.70 1.89 2.45 5. Middle East and North Africa 11482 12804 7709 7496 2.09 -2.55 -0.37 6. Sub-Saharan Africa 4258 4733 4307 4182 0.96 -0.74 -0.51 Sources: World Bank, World Developmem indicators. 2()()O; Republic of China, Taiwan Statistical Data Book. 60 Competitiveness, FDf and technological activity in East Asia achieved high growth during the 1970s and 1980s, with annual average growth rates of per capita GDP of around 6,5 per cent Their growth rates declined in the I 990s to 5.3 per cent. Each of the four NlEs had similar growth performance, with the exception of Hong Kong, which recorded relatively low growth rates in 1990~97 (3.2 per cent). The ASEAN4 (Indonesia, Malaysia, the Philippines, and Thailand) performed less well than the NIEs overall. However, their performance was substantially better than other developing countries over the 1970-97 period. Growth rates of per capita GDP for the ASEAN4 declined from 4.6 per cent in the 1970s to 3.8 per cent in the 1980s, before rising to 4.7 per cent in 1990~97. Among the ASEAN4 the Philippines performed the worst, particularly in the 1980s and 1990s. China achieved a remarkable acceleration in growth from the 1970s to the 1990s, with the annual growth rate in per capita GDP rising from 3.4 per cent in the 1970s to 7.3 per cent in 1980s and to 9.5 per cent in 1990~97. In terms of PPP (purchasing power parity) in 1992 US dollars, average per capita GDP in East Asia increased almost five-fold over the 27 years, from $2000 in 1970 to $9700 in 1997. The comparable increase for develop ing countries as a whole and for developed countries was only about 70 per cent. Average per capita income in East Asia was only about a half of the average for developing countries in 1970 and was 35 per cent higher in 1997. In comparison with the developed world, per capita income in East Asia rose from 15 per cent of average incomes in 1970 to just over 40 per cent in 1997. Per capita incomes vary greatly in the East Asian region. In 1997 the N IEs recorded the highest incomes ($16000), followed by ASEAN4 ($4700), and China bringing up the rear with $2800. Among the NlEs, Hong Kong and Singapore had very high per capita GDP at around $22000, while the corresponding values for Korea and Taiwan were approximately $14000. In the ASEAN4, the income ranks in 1997 were: Malaysia, Thailand, the Philippines and Indonesia. Unlike the NlEs, the ranking changed over the 1970-97 period: the Philippines fell from leading place in 1970 to third place in 1997. 2.2 TFP Growth The level and growth rate of total factor productivity (TFP) are often used to measure the competitiveness of an economy. This is because, given factor inputs such as labour and capital, there is a positive relationship between TFP and output Table 3.2 presents the results of our estimation of TFP levels and TFP growth rates. 3 To begin with TFP growth, we find that the East Asian economies on average performed favourably compared to other Table 3.2 Level and growth oj total Jactor productivity, 197U97 TFP level (USI980= 1.0) Annual growth rate 1970 1980 1990 1997 1970-80 1980-90 1990-97 Developed countries 0.83 0.87 0.97 1.02 0.56 0.98 0.81 Developing countries 0.44 0.47 0.48 0.54 0.61 0.22 1.78 I. East Asia & the Pacific 0.29 0.36 0.47 0.55 2.14 2.76 2.19 (I) NIES 0.39 0.48 0.70 0.79 2.14 3.67 1.77 Hong Kong 0.45 0.64 1.02 1.06 3.42 4.69 0.54 Korea 0.43 0.47 0.70 0.75 0.84 3.98 1.11 Singapore 0.48 0.64 0.79 1.00 2.88 2.10 3.44 Taiwan 0.32 0.43 0.59 0.71 3.02 3.14 2.75 - 0-, ASEAN4 0.22 0.28 0.32 0.35 2.38 1.41 1.17 Indonesia 0.15 0.23 0.27 0.34 4.07 1.72 3.10 Malaysia 0.26 0.34 0.36 0.43 2.58 0.55 2.33 Philippines 0.28 0.35 0.33 0.33 2.37 -0.74 -0.06 Thailand 0.26 0.28 0.36 0.37 0.69 2.77 0.22 (3) China 0.18 0.21 0.21 0.31 1.86 -0.01 5.35 2. South Asia 0.24 0.22 0.28 0.32 -0.50 2.11 2.06 3. Latin America and the Caribbean 0.61 0.62 0.53 0.58 0.08 1.48 1.26 4. Europe and Central Asia 0.42 0.55 0.49 0.56 2.62 1.04 1.82 5. Middle East and North Africa 0.49 0.65 0.66 0.76 2.78 0.16 1.93 6. Sub-Saharan Africa 0.35 0.42 0.33 0.32 1.91 -2.48 -0.37 Source: Authors' estimates. 62 Competitiveness, FDI and technological activity in East Asia regions, including the developed countries. East Asian TFP growth rates were around 2.1~2.8 per cent per annum from 1970 to 1997. By contrast, TFP growth rates for other regions were low and fluctuated substantially. For example, Latin American TFP growth rates fell from 0.1 per cent in the 1970s to 1.5 per cent in the 1980s and then rose to 1.3 per cent in the 1990~97 period. Similar fluctuations are found for other regions. Among the East Asian economies the NIEs had higher TFP growth than the ASEAN4: the range is 1.8 to 3.7 per cent per annum in the former and 1.2 to 2.4 per cent in the latter. Although there are variations in TFP growth rates among the NIEs, they are not substantial. This is in contrast to the ASEAN4, among which TFP growth rates differ substantially. Indonesia records relatively high rates of around 1.8 and 4.1 per cent per annum, while the Philippines has negative TFP growth for the 1980~97 period. China shows a V-shaped pattern, with TFP growth rates declining from 1.9 per cent in the 19708 to -0.01 per cent in the 1980s, and rising to 5.3 per cent in 1990-97. Rapid TFP growth led to rising levels of TFP in East Asia. TFP levels are estimated here by using the TFP level in the USA in 1980 as a base. The estimates indicate that in 1980 the TFP level in East Asia was 36 and 41 per cent of the levels in the USA and developed countries, respectively. Over time, this gap narrowed to 54 per cent of the level in developed countries in 1997. Among the East Asian economies the NIEs had significantly higher levels of TFP than the ASEAN4 and China. The NIEs as a group had TFP levels of almost 80 per eent of the level for developed eountries in 1997; indeed, Hong Kong and Singapore had comparable levels of TFP to that for developed countries. TFP levels in Korea and Taiwan were lower at approximately 70 per cent of the level in developed countries. TFP levels for ASEAN4 and China were substantially lower, less than half of the level in the NIEs in 1997. Malaysia was the exception, with a somewhat higher TFP level than those in the rest of the ASEAN4. 3. FACTORS AFFECTING COMPETITIVENESS: TECHNOLOGICAL CAPABILITY AND IMPORTED TECHNOLOGY Many factors affect national competitiveness. One of the most important, some would say the most important, is technological capability. Even a country richly endowed with natural or labour resources cannot compete in international markets if it cannot use its resources efficiently in combi nation with appropriate technology. For developing countries the importa An international comparison 63 tion of technology from developed countries is a vital source of technol ogy; however, access to foreign technology is not enough - the country needs the capability to absorb and adapt foreign technology. This section examines technological capability and the absorption of foreign technol ogy in East Asia, setting the stage for the statistical analysis of the determi nants of competitiveness later. 3.1 Domestic Technological Capability Educated and well-trained workers are essential for using technologies effi ciently and improving them over time. Research and development also plays an important role by producing new technology, new production pro cesses and new products and helping in the absorption of very complex new technologies. We examine these two factors for East Asia. Education Although formal education may not constitute technological capability without technical training or experience, it provides the base on which tech nical skills are developed. 4 Many studies have found a positive impact of education on economic growth, supporting thc argument that education plays a role in promoting economic growth. For example, the World Bank (1993) finds that education, particularly primary education, contributed significantly to the increase in per capita GDP in 113 countries in the 1960-85 period. Specifically, it found that a 10 per cent increase in the enrolment ratio in primary and secondary education would lead to a 0.3 per cent increase in the growth rate of per capita GDP. Let us examine the educational levels of East Asian economies, using the indicator for educational attainment estimated by Barro and Lee (2000). The Barro and Lee data set has at least one observation on school attain ment, the highest educational level attained for the population aged 15 or over and for the population aged 25 or over, for 142 economies, of which 107 have complete information at five-year intervals from 1960 to 2000. The data for the period up to 1995 are estimates and those for 2000 are projec tions. In our analysis we interpolate values for in-between years. Table 3.3 shows educational attainment and the average years of school ing for developed countries and developing countries. The figures indicate the proportion of population aged 25 or over that attained secondary and higher levels of education. East Asia has higher educational attainments than other developing regions, with the exception of those in Europe and Central Asia, which have longer average schooling and larger shares of the population with higher educational attainment. Educational attainments have improved significantly for all economies in East Asia from 1960 to Table 3.3 Educational attainment Average years of school Highest education attainment Highest education (secondary) 1960 1970 1980 1990 1997 1960 1970 1980 1990 1997 1960 1970 1980 1990 1997 Developed countries 7.24 7.82 9.00 9.53 9.97 34.0 36.1 45.2 42.1 41.7 8.4 10.9 17.3 24.9 28.7 Developing countries 1.71 2.37 3.15 4.28 4.90 5.3 9.6 14.9 21.2 23.5 0.9 1.5 2.4 4.1 5.2 I. East Asia & the Pacific 1.75 2.84 3.73 5.16 5.67 6.1 15.0 19.8 31.1 32.4 1.3 1.5 2.0 3.4 4.6 (1) NIEs 3.37 4.64 6.56 8.52 9.55 11.9 20.8 31.7 47.6 47.5 3.1 5.2 8.7 12.5 20.5 Hong Kong 4.74 5.11 6.73 8.37 9.40 17.4 22.3 30.5 43.3 47.3 4.7 3.1 7.1 10.6 14.9 Korea 3.23 4.76 6.81 9.25 10.27 10.9 21.8 36.9 53.9 50.7 2.6 5.6 8.9 13.4 23.5 Singapore 3.14 3.74 3.65 5.52 7.97 23.4 20.9 14.6 31.3 49.5 0.0 2.0 3.4 4.7 9. Taiwan 3.32 4.39 6.37 7.44 8.28 11.0 18.1 23.3 37.7 40.7 4.2 5.4 9.3 12.2 17.3 (2) ASEAN4 2.12 3.05 3.88 4.54 5.46 4.5 7.0 11.6 17.9 21.7 1.5 2.5 4.1 6.6 8.9 '" ~ Indonesia 1.11 2.29 3.09 3.30 4.37 1.9 5.1 9.6 16.8 20.5 0.1 0.5 0.8 2.3 4.3 Malaysia 2.35 3.05 4.49 5.54 7.76 7.2 9.4 19.9 27.1 42.9 1.5 1.5 1.4 2.8 7.2 Philippines 3.77 4.81 6.06 7.07 7.48 10.6 14.2 18.9 27.2 30.2 6.2 9.6 15.2 18.7 21.3 Thailand 3.45 3.54 3.77 5.35 5.91 4.9 4.4 6.8 8.0 9.1 0.6 1.1 2.9 7.8 10.4 (3) China 1.58 2.75 3.61 5.23 5.61 6.2 17.2 21.7 34.4 35.2 1.1 1.0 1.0 2.0 2.5 2. South Asia 1.31 1.77 2.52 3.37 4.03 3.2 4.6 13.3 14.3 16.1 0.1 1.3 2.3 3.7 4.2 3. Latin America and the 2.95 3.32 3.97 4.93 5.53 9.4 10.4 12.7 16.5 19.1 1.9 2.5 5.4 8.9 11.1 Caribbean 4. Europe and Central 4.27 4.75 5.62 6.57 7.00 9.9 12.2 19.7 27.1 29.3 1.9 3.4 4.9 6.9 9.0 Asia 5. Middle East and 0.76 1.06 2.00 3.36 4.39 2.3 4.1 7.8 14.8 19.7 0.6 1.2 3.0 5.0 7.2 North Africa 6. Sub-Saharan Africa 1.18 1.32 1.90 2.78 3.50 3.9 4.4 6.1 10.3 15.3 0.3 0.6 0.6 1.8 2.5 Note: Data for 1997 are interpolated from those on 1995 and 1999. Source: Barro and Lee (2000). An international comparison 65 1997, the average years of schooling rising from 1.8 to 5.7 years. The shares of the population aged 25 and over with primary, secondary, and higher education increased from 25.3,6.1, 1.3 per cent in 1960 to 39.3,32.4. and 4.6 per cent in 1997, respectively. There are wide variations in education among the East Asian economies. On average the NIEs have better performance than the ASEAN4, but there are substantial differences within these groups. Korea stands out among the NIEs with the longest years of schooling and the highest proportion of population with higher education, while Singapore appears to fare the worst. In the ASEAN4, the Philippines outperforms the others, particu larly in higher education; one out of every five persons aged 25 and over has higher education there. Indonesia has the lowest years of schooling and also the lowest proportion of population with higher education. Thailand has very low shares of the population with secondary education. Despite a substantial improvement over time, China still trails other East Asian econ omies in educational attainment, particularly in terms of higher education. Research and development The number of researchers in the popUlation can serve as a good indicator of technological capability. Table 3.4 shows the number of researchers per million population for East Asian economies. The average for developing countries was 334 in 1997, compared to 3161 for developed countries. Among the developing countries, the NIEs register much higher figures, comparable to those for developed countries; even more significant is the rapid increase in the number of researchers in the NIEs. In 1970 the number was 198, approximately 10 per cent of the corresponding value for the developed countries; by 1997 the number had increased to 2613, about 80 per cent of the value for developed countries. Within the NIEs, Taiwan has thc highest number at 3530 in 1997, higher than the average for developed countries. The ASEAN4 are far behind the NIEs, and they lag even behind China. R&D is an input for building and improving technological capability. The ratios of R&D to GDP for East Asian and other developing countries are shown in Table 3.4. The picture is fairly similar to that for numbers of researchers per million population. The NIEs perform much better than ASEAN4 or China, spending 2.3 per cent of GDP on R&D compared to 0.14 and 0.66 per cent for the ASEAN4 and China, respectively. Over time, moreover, R&D performance in the ASEAN4 is deteriorating. The NIEs have improved their capabilities greatly over time, and China has also achieved some success. However, ASEAN4 countries are lagging and the gap with the NIEs is likely to widen. Table 3.4 Researchers and R&D Reseachers per million population R&D/GDP ("!.,) 970 1980 1990 1997 1970 1980 1990 1997 Developed countries 1798 2201 3107 3161 1.99 2.01 2.43 2.39 Developing countries 157 208 278 334 0.31 0.43 0.71 0.79 1. East Asia & the Pacific 178 226 373 486 0.33 0.48 1.04 1.27 NIEs 198 523 1826 2613 0.42 0.62 1.71 2.27 Hong Kong nla nla nla n/a nla nla nla nla Korea 176 484 1645 2195 0.38 0.57 1.88 2.82 Singapore 326 417 1426 2323 0.21 0.27 0.94 1.13 0. Taiwan 226 620 2260 3530 0.50 0.72 1.66 1.88 0. (2) ASEAN4 86 104 164 162 0.33 0.35 0.20 0.14 Indonesia 85 102 189 191 0.35 0.35 0.15 0.07 Malaysia 137 165 208 93 0.33 0.34 0.41 0.24 Philippines 93 112 141 153 0.15 0.26 0.20 0.22 Thailand 66 79 94 103 0.39 0.39 0.18 0.13 China 203 243 348 454 0.19 0.37 0.68 0.66 2. South Asia 71 85 132 132 0.32 0.53 0.74 0.69 3. Latin America and the Caribbean 134 203 252 222 0.17 0.27 0.47 0.56 4. Europe and Central Asia 968 1378 798 792 0.71 0.95 0.84 0.56 5. Middle East and North Africa 154 212 296 377 0.24 0.19 0.28 0.32 6. Sub-Saharan Africa 70 109 141 235 0.57 0.56 0.56 0.47 Sources: UNESCO, Statistical Yearbook; World Bank, World Development Indicator,; 2000; Republic of China, Taiwan Statistical Data Book. An international comparison 67 3.2 Inflow of Foreign Technology We consider several channels for accessing foreign technology: foreign direct investment, foreign trade and technology imports. Foreign direct investment In recent years FDI has become an important means of importing technol ogy for developing countries. Intra-firm technology transfer has been ana lysed by several studies. Based on a survey of East Asian affiliates of Japanese firms, U rata (1999) finds that relatively simple technologies such as maintenance and repair of production lines were transferred from parent companies to affiliates. He also finds that relatively sophisticated technolo gies such as development of new technologies and new products were not transferred. Analysing the determinants of the extent of intra-firm technol ogy transfer by Japanese MNCs, Urata and Kawai (2000) find that the capability to absorb technologies reflected in educational level in host coun tries plays a key role in successful intra-firm technology transfer. Their study also points out that intra-firm technology transfer takes time and experience, suggesting the importance of maintaining a stable economic environment in the host country. The results of analyses of the presence of technology spillovers are mixed. Using industry-level data, Caves (1974) finds the presence of tech nology spillover in his study of the Australian manufacturing sector but not in his study of Canadian manufacturing. Using similar methodology, Globerman (1979) finds the presence of the spillover effect of FDI in the Canadian manufacturing sector. Blomstrom and Persson (1983) and Blomstrom and Wolff (1994) also detect beneficial technology spillover in their studies of the Mexican manufacturing sector. In contrast, Haddad and Harrison (1993) and Aitken and Harrison (1994) do not find spillover in their studies of Morocco and Venezuela. One possible reason for not detecting technology spillover in these studies may be the limited presence of foreign firms in these countries. Turning to the impact of FDI on economic growth, one finds that few studies on the subject have been conducted using macroeconomic indica tors. Borensztein et al. (1998) find that FDI has a marginally positive impact on economic growth; the impact is significantly positive when FDI is interacted with the educational levels of host countries. 5 Their finding may be interpreted to mean that education becomes more effective when it is associated with foreign knowledge. Given that educational levels in East Asia are relatively high, it is reasonable to argue that FDI inflows have con tributed to economic growth in East Asia. FDI flows have been growing rapidly in recent years. World inward FDI 68 Competitiveness, FDf and technological activity in East Asia increased more than threefold in eight years from US$203 billion in 1990 to US$680 billion in 1998. 6 This rapid increase is attributable to several factors. Technological progress and deregulation in communication ser vices have reduced the cost of international communication, facilitating the management of far-flung operations by MNCs. Liberalization of FDI pol icies by many countries has also contributed. East Asia has experienced a remarkably rapid expansion of FDI inflows. It has undertaken consider able liberalization in FDI policies,7 with several economies offering incen tives to foreign investors, such as preferential tax treatment, especially to export-oriented investments. Another factor in East Asia has been favour able growth prospects, driven by past performance. One notable recent development in Asia has been the rapid increase in cross-border M&As (mergers and acquisitions). The share of M&As in FDI inflows in South, East and South-East Asia increased from around 3 per cent in 1995 to 16 per cent in 1998. 8 Korea and Thailand, hard hit by the financial crisis, liberalized their M&A policies and experienced a large wave of this kind of FDI. Table 3.5 shows the importance of FDI inflows for East Asian and other economies. The proportion of FDI inflows to GDP in East Asia increased from l.l per cent in 1970 to 2.8 per cent in 1997. In 1997 East Asia was behind Latin America in terms of the FDI inflows to GDP ratio, which reg istered 3.3 per cent. Within East Asia, Singapore had by far the greatest reli ance on foreign investors; its ratio of FDI inflows to GDP was 10 per cent in 1997, even after a notable decline from 14.4 per cent in 1990. Malaysia had the second highest ratio at 5.1 per cent, and China was the third at 4.9 percent in 1997. The increase in the share of FDI inflows inGDP for China was spectacular, since the corresponding ratio for 1990 was only 0.98 per cent. Hong Kong also recorded a relatively high ratio of 3.5 per cent. In contrast to these countries, Korea and Taiwan had ratios of around 0.6-0.8 per cent in 1997. The ratio for Taiwan declined over time, while that for Korea rose steadily. Imports of capital goods Foreign trade has been an important source of foreign technologies for developing countries via the importation of intermediate and investment goods that embody new technology. Reverse engineering is one way of assimilating technology from such imports, for countries that have the capability to carry out this complex task. Table 3.6 shows the share of machinery in total imports. 9 East Asia exhibits a significant upward trend in comparison with other developing regions, the share of machinery in total imports rising from 0.30 in 1980 to 0041 in 1997 compared to 0.31 and 0.35 for other regions. The ASEAN4 consistently have higher shares than An international comparison 69 Table 3.5 FDI inflows (gross) (percentage of GDP) 1970 1980 1990 1997 Developed countries 0.46 0.52 1.06 1.27 Developing countries 0.75 0.78 0.85 2.32 1. East Asia & the Pacific 1.10 1.07 1.59 2.76 (I) NIEs 1.72 1.59 1.59 1.60 Hong Kong nla nla 2.31 3.46 Korea nla 0.01 0.28 0.60 Singapore 5.53 14.00 14.36 10.21 Taiwan 2.45 1.13 0.83 0.79 (2) ASEAN4 0.80 0.91 2.23 2.60 Indonesia 0.86 0.23 0.96 2.17 Malaysia 2.24 3.52 5.45 5.10 Philippines 0.17 0.79 1.20 1.49 Thailand 0.61 0.57 2.86 2.50 (3) China nla nla 0.98 4.92 2. South Asia 0.08 0.08 0.13 0.83 3. Latin America and the Caribbean 0.82 0.78 0.68 3.25 4. Europe and Central Asia 0.21 0.09 0.85 1.73 5. Middle East and North Africa 0.64 1.15 0.55 1.00 6. Sub-Saharan Africa 0.56 0.41 0.28 1.79 Sources: IMF, Balance of Payment Statistics; World Bank, World Development Indicators, 2000; Republic of China, Taiwan Statistical Data Book. the NIEs or China. For instance, in 1997 the share for the ASEAN4 was 0.50, while that for the NIEs and China was 0.41 and 0.36, respectively. This may reflect the relatively high value of electronic component imports for export-oriented assembly in the ASEAN4. Singapore and Malaysia have particularly high shares (0.553 and 0.616, respectively) in 1997, again reflecting the strong presence of electronics MNCs in export-oriented activity. An expansion in imports may lead to improvements in technical effi ciency in domestic firms in different ways. One is increased competitive pressure, which to survive firms must introduce new technologies, products, management methods and so on. Several studies have found a positive impact of greater imports on productivity; Lawrence and Weinstein (2000), for instance, found that import expansion was associated with higher pro ductivity in Japan in the post-World War II period. Other studies have not, however, found a clear statistical relationship between import liberalization and productivity. 70 Competitiveness, FDI and technological activity in East Asia Table 3.6 Imports oj machinery as a share oj total imports 1980 1990 1997 Developed countries 0,256 0.355 0.378 Developing countries 0.310 0.348 0.348 l. East Asia & the Pacific 0.297 0.344 0.406 (I) NIEs 0.254 0.383 0.438 Hong Kong 0.237 0.308 0.371 Korea 0.227 0.351 0.355 Singapore 0.281 0.441 0.553 Taiwan 0.273 0.434 0.474 (2) ASEAN4 0.325 0.421 0.501 Indonesia 0.376 0.411 0.425 Malaysia 0.370 0.525 0.616 Philippines 0.296 0.320 0.487 Thailand 0.260 0.428 0.473 (3) China 0.251 0,351 0.365 2. South Asia 0.257 0.244 0.254 3. Latin America and the Caribbean 0.295 0.339 0.373 4. Europe and Central Asia 0,274 0.320 0.364 5. Middle East and North Africa 0.323 0.341 0.335 6. Sub-Saharan Africa 0.330 0.373 0.325 SOUfees: World Bank, World Development Indicators, 2000; Republic of China. Taiwan Statistical Data Book. An increase in exports may also have a positive impact on productivity. This may be so for several reasons: greater capacity utilization in industries in which the minimum efficient scale is large relative to the domestic market; increasing familiarity with and absorption of new technologies; greater learning-by-doing insofar as this is a function of cumulative output; and the stimulating effects of international competition and the feedback of technical and other information from export markets. 10 Several studies have shown that export expansion, particularly that of manufactured prod ucts, results in higher productivity. The World Bank (1993) finds that the high share of manufactured exports in total exports increased the growth rate of TFP in its study of 69 countries for the 1960-89 period. 1J A case study of Korean firms by Rhee et al. (1984) finds that exporting firms achieved higher productivity by obtaining technologies through contact with foreign firms. Table 3.7 shows the ratio of trade (exports plus imports) to GOP in developing countries. The ratio for East Asia rose sharply over 1970-97, from 47.6 per cent to 91.7 per cent. These values are significantly higher An international comparison 71 Table 3. 7 Trade (exporls+ imports) (percentage of GDP) 1970 1980 1990 1997 Developed countries 28.8 39.2 38.3 42.5 Developing countries 37.4 52.1 53.2 60.8 I. East Asia & the Pacific 47.6 78.6 82.7 91.7 (I) NIEs 76.0 122.0 115.7 118.9 Hong Kong 181.5 180.6 260.1 264.2 Korea 37.5 74.4 59.4 70.5 Singapore 231.6 439.0 397.0 315.6 Taiwan 60.7 106.3 88.5 95.7 (2) ASEAN4 40.1 62.7 74.3 94.9 Indonesia 28.4 54.4 49.9 56.0 Malaysia 79.9 112.6 150.6 185.5 Philippines 42.6 52.0 60.8 108.5 Thailand 34.4 54.5 75.8 94.8 (3) China 3.8 15.5 31.9 41.6 2. South Asia 10.7 19.4 20.5 28.1 3. Latin America and the Caribbean 20.5 26.7 28.9 36.3 4. Europe and Central Asia 29.6 36.1 39.1 60.1 5. Middle East and North Africa 71.9 72.7 68.0 64.8 6. Sub-Saharan Africa 47.2 59.7 51.4 60.1 Sources: World Bank, World Development Indicators. 2000; Republic of China, Taiwan Statistical Data Book. than those for other developing regions or developed countries. Within East Asia, Singapore and Hong Kong had very high ratios, 250-300 per cent in 1997, reflecting their role as entrepot centres and their open trade regimes. Malaysia also had a high ratio, 190 per cent, resulting mainly from the large presence of export-oriented MNCs. By contrast, China, Indonesia and Korea had relatively low ratios, with China having a signifi cant increase and Korea and Indonesia showing fluctuations around the trend. The ratios for Taiwan also fluctuate, but the values are higher than for Korea. The trade-GDP ratios increase steadily for the Philippines and Thailand, but are significantly lower than for Malaysia. Technology imports Imports of technology in the form of licences and patents are an important way of obtaining foreign technology. In the past, technology trade was con ducted largely by independent firms through arm's length transactions, but in recent years intra-firm transactions (between MNCs and their affiliates) 72 Competitiveness, FDI and technological activity in East Asia Table 3.8 Royalty payments (percentage of GDP) 1970 1980 1990 1997 Developed countries 0.124 0.124 0.178 0.231 Developing countries 0.112 0.086 0.137 0.172 I. East Asia & the Pacific 0.075 0.128 0.259 0.292 (I) NIEs n/a 0.195 0.472 0.467 Hong Kong n/a n/a n/a n/a Korea n/a 0.195 0.540 0.507 Singapore nla n/a nla n/a Taiwan n/a nla 0.363 0.405 (2) ASEAN4 nla 0.098 0.162 0.357 Indonesia nla nla n/a n/a Malaysia nla 0.152 n/a nla Philippines n/a 0.058 0.086 0.192 Thailand n/a 0.092 0.200 0.539 (3) China nla n/a n/a 0.060 2. South Asia 0.016 0.008 0.021 0.034 3. Latin America and the Caribbean 0.158 0.055 0.090 0.115 4. Europe and Central Asia 0.050 0.050 0.050 0.227 5. Middle East and North Africa 0.\86 0.177 0.\83 0.2\5 6. Sub-Saharan Africa 0.\81 0.155 0.074 0.131 Sources: World Bank, World Development Indicators, 2000; Republic of China, Taiwan Statistical Data Book. have grown rapidly. This reflects at least two related developments: rapid FOI expansion and the increased preference on the part of MNEs to use FOI as a means of deploying technology abroad, Table 3.8 shows patterns of technology trade as measured by royalties and licence fees paid abroad as a percentage of GOP in the East Asian economies. Note, however, that information for Hong Kong, Singapore, Indonesia and Malaysia is not available, and the total for East Asia is understated. The ratio for available countries in East Asia rose from 0.075 per cent in 1970 to 0.292 per cent in 1997, much faster than for other regions, including the developed countries. The increase was particularly large for Thailand, from 0.092 per cent in 1980 to 0.539 per cent in 1997. The ratio for Korea also increased rapidly but not at the pace achieved by Thailand. By 1997 the ratios for Korea, Taiwan and Thailand were more or less comparable at around 0.4 to 0.5 per cent. The ratios for the Philippines and China were significantly lower at around 0.03-0.06 per cent. An international comparison 73 4. THE DETERMINANTS OF COMPETITIVENESS This section presents the statistical analysis of the effect of technological factors on per capita GDP12 growth and TFP growth. We examined the two technological factors discussed above, domestic technological capability and inflows of foreign technology, using annual panel data for 137 countries over 1970-97. To deal with possible simultaneity problems in the regression analysis, we used instrumental variable estimation. Based on the Hauseman test results, we selected either the fixed effect model or the random effect model for the estimation of the panel data. The statistical analysis was con ducted for all the sample countries as well as for subsets of countries. The countries were divided into the following groups: developed countries, developing countries and East Asian developing economies. Table 3.9 shows the results for per capita GDP growth. Let us start with the whole sample. After controlling for initial per capita GDP (GDPO), domestic investment (IN V) , openness (OPEN), government expenditure (GO V) and inflation (INF), we find that domestic technological capability as measured by educational attainment (HC) and R&D activities (RD), and inflows of foreign technology as measured by inward FDI (FDI) , capital good imports (CAP) and patent and licence payments (PA 1) have a posi tive and significant impact. These findings are consistent with expectations, indicating the importance of the technology factor in economic growth. Interaction terms for foreign technology inflows and domestic technologi cal capability suggest that imported patents contribute to growth when combined with a well-educated workforce. Unlike earlier studies, our results do not find a higher impact for FDI when it is combined with high domestic technological capability. The country groups show interesting differences. To begin with domestic technological capability, we find that both educational attainment and R&D have a positive impact on economic growth for all country groupings but the statistical significance differs: educational attainment has a signifi cantly positive impact for developed and East Asian economies, but not for developing countries. R&D has a significantly positive effect on growth for the developed and developing country groups but not for East Asian econ omies. In view of the importance of education in economic growth, our findings on the insignificant impact of educational attainment on growth for developing countries is puzzling and needs further investigation. Turning to the impact of imported technology, the importation of capital goods is found to contribute significantly to growth for all types of countries. FDI has a statistically significant and positive impact for devel oping countries and East Asian economies but not for the developed coun tries. However, FDI has a positive impact on growth in developed countries Table 3.9 The determinants of growth rates ofper capita GDP Total sample (137 countries, OBS =3836) Developing (113 countries, OBS 3164) coefficien t t-value coefficient t-value coefficien t t-value coefficien t t-value LGDPO -6.475 13.780 -6.604 l3.978 -6.485 -12.380 -6.676 12.614 INV 0.091 5.137 0.095 5.309 0.094 4.751 0.093 4.678 OPEN 0.020 3.182 0.023 3.477 0.019 2.730 0.022 3.060 GOV -0.191 -7.566 -0.190 -7.515 -0.186 -6.644 -0.184 -6.554 INF -0.001 -3.331 -0.001 -3.338 -0.001 -3.041 -0.001 -3.082 RD 1.679 3.350 0.465 1.443 2.061 3.004 1.735 1.083 HC -0.017 1.028 0.018 0.876 0.019 1.012 0.045 1.193 ~ CAP 12.824 8.098 12.562 5.400 12.795 7.173 10.247 3.693 FDI 0.443 1.956 0.987 1.625 0.449 1.963 0.974 1.937 PAT 0.931 1.657 2.744 1.057 0.229 0.236 2.991 0.389 CAP*HC 0.025 0.347 0.030 0.351 FDI*HC 0.044 1.162 0.044 1.036 PAT*HC 0.127 2.170 0.133 1.711 CAP*RD 3.086 1.225 9.974 2.347 FDI*RD 1.701 1.159 2.236 1.203 PAT*RD -2.350 1.480 1.249 -0.435 R2 0.188 0.190 0.182 0.185 LMhet 26.180 [.000] 25.991 [.000] 19.210 [.000] 17.950 [.000] Hausman 207.870 LOOOO] 218.000 LOOOO1 161.490 [.0000] 166.910 [.0000] Developed (24 countries, OBS = 672) Developing (E. Asia, 9 countries, OBS=252) coefficient t-value coefficient t-va1ue coefficient t-value coefficient I-value LGDPO -8.832 -8.719 9.719 -9,475 -0.861 -1.830 -0.099 -1.686 INV 0.064 1.715 0.064 1.738 0.117 2.160 0.068 2.063 OPEN 0.072 5.829 0.077 6.298 0.023 1,486 0.027 1.588 GOV -0.346 -5.946 -0,424 -6.973 -0.002 -0.023 -0.040 -0.345 INF -0.012 ··2.581 -0.012 -2.542 -0.006 -0.184 -0.003 -0.103 RD 0.506 ].890 3.173 3.218 0,441 0,485 5.359 1.168 HC 0.039 1.861 0.058 1.639 0.081 2.105 0.091 1.720 CAP 15.511 5.610 30.742 5.694 17.623 3.839 37.333 3.356 FDI 0.197 0.090 16,401 1.758 19.015 4.218 11.025 2.020 PAT 1.032 3.]66 0.156 1.032 -3.669 1.063 -20.325 1.959 CAP*HC 0.094 0.859 -0.235 -0.802 "l FDI*HC 0.501 2.781 OA57 ].277 '-" PAT*HC 0.076 0.875 0.297 0.972 CAP*RD 5.728 1.865 14.272 -].323 FDI*RD -6.824 -2.386 12.558 1.206 PAT*RD -2.378 1.676 1.214 0.182 R2 0.323 0.355 0.325 0.352 LMhet 2.857 {.091] 2.296 [.130] OAIO [.522] 0.215 [.643] Hausman 64.177 {.OOOO] 76.257 [.0000] 21.064 [.0070] 25.721 [.0012] Notes: I. GDPO= In (initial per capita GDP); INV=investmen1iGDP; OPEN = (exporl+ imporl)/GDP; CAP=machinery import/import; FDJ=FDI inflow/investment; PAT Royalities and licence fees payment/GDP; RD= R&D expenditure/GDP; SC=education attainment rate (secondary [SC2) and higher [SC3)); GOV = government expenditure/GOP; JNF= inflation rate. 2. Instrument variables are lagged independent variables and population. 3. T-values are evaluated by Hey if heteroscedasticity does exist. Source: Authors' estimation. 76 Competitiveness, FDI and technological activity in East Asia when combined with educational attainment. Finally technology imports as measured by payments for patents and licensing have a significant and positive impact on growth for developed countries but not for developing countries or East Asian economies. Our findings suggest that capital goods import and FDI play important roles in acquiring foreign technology by developing countries. Our findings for the other variables are largely consistent with studies such as the World Bank (1993) and Borensztein et al. (1998). The initial level of per capita GDP (GAPO) is significantly negative in many cases, indi cating convergence of income levels across countries. The investment ratio (IN v) has positive and mostly significant effects. Openness (OPEl\T) has a positive impact on growth, supporting the argument that more open trade systems result in better resource allocation. Both high government expen diture (GOV) and high inflation (INF) reduce per capita GDP growth. The former, often associated with high inflation, tends to crowd out private sector activities and lead to inefficiency in the use of resources. Moreover, inflation discourages investment by increasing uncertainty about future economic prospects. So far, we have implicitly assumed that technological factors contribute to economic growth by improving productivity. We now investigate this hypothesis directly by looking at the impact of domestic technological capability and foreign technology on TFP growth. We apply the same methodology as for the analysis of GDP growth. The results for all sample countries suggest that inflows of foreign technology in the form of capital goods and FDI have a significantly positive impact on TFP growth (Table 3.10). The importation of foreign technology through patents and licensing contributes to TFP increase when combined with high educational levels. The estimated coefficients on dome tic technological capability are positive, as expected, but not statistically significant. There appears to be convergence of TFP levels among sample countries over time. The effects of investment and openness on TFP growth are positive, but only the effect of investment is statistically significant. Government expenditure and inflation have negative effects, with only the former being statistically significant. These results, very similar to those for growth, are consistent with our expectations. Turning to the results obtained from different sample groups, we find differences in the impact of domestic technological capability and inflow of foreign technology between developed countries on the one hand and developing and East Asian economies on the other. To begin with domes tic technological capability, educational attainment has a significantly pos itive impact on TFP growth for developing and East Asian economies, while R&D has a significantly positive impact for developed countries. An international comparison 77 These findings suggest that innovative capability as reflected in R&D is important in raising TFP levels in developed countries, where TFP levels are already high, while in developing countries it is educational attainment, which improves the capability to absorb and assimilate imported technol ogy, that is the key to improving TFP. As to the impact of foreign technology, the import of capital goods is found to contribute to TFP growth for all the countries regardless of their levels of economic development. However, the impact of FDI and import of patents differs between developed countries and developing and East Asian economies. For developed countries, the purchase of patents and licences turns out to have a significant and positive impact, while for devel oping and East Asian economies FDI has a significantly positive impact. These findings suggest that developed countries, with high technological capability, can assimilate foreign technology in forms that do not generally come with much technical support from abroad. In contrast, for develop ing countries FDI, which brings not only technology but also management know-how, plays a larger role in improving TFP. The import of patents contributes to the improvement of TFP when combined with educational attainment. The findings for other variables are similar to those for the determinants of economic growth. The results for the TFP gap (GAPO) show the inter country convergence in TFP levels over time. Investment and openness have a positive impact on TFP growth though their statistical significance differs by group. Government expenditure and inflation both have negative impacts on TFP growth, with government expenditure showing statistical significance. 5. CONCLUSIONS The remarkable growth and TFP performance of East Asian economies has been driven by several factors. Our study shows that important among these are strong domestic technological capabilities and the inflow of foreign technology, particularly in the form of capital goods and FDI. To sustain and promote these assets in the future, East Asian countries have to undertake various policies. To expand trade and FDI, they should further pursue liberalization and participate actively in multilateral and regional efforts to implement liberalization measures. They can also promote trade and FDI by adopting international standards and rationalizing customs procedures, disseminating information on overseas markets to domestic firms interested in exporting and on domestic markets to MNCs interested in undertaking FDI. Table 3.]0 The determinants of TFP Total sample (68 countries, OBS = 1904) Developing (44 countries, OBS 1232) coefficient t-value coefficient (-value coefficient t-value coefficient t-value GAPO -0.039 -4.620 -0.039 -4.520 -0.040 -3.629 -0.041 -3.699 INV 0.000 1.829 0.000 1.628 0.000 1.155 0.000 1.077 OPEN 0.000 0.341 0.000 0.267 (l.000 1.801 0.000 0.905 GOV -0.003 -7.330 -0.004 -7.439 -0.004 -5.809 -0.004 - 5.850 INF 0.000 1.464 0.000 1.492 0.000 1.164 0.000 1.221 RD 0.009 1.636 0.014 1.021 0.002 0.252 0.035 0.894 ~ HC 0.001 1.411 0.001 1.209 0.001 2.289 0.001 1.758 CAP 0.106 4.517 0.124 3.551 0.125 3.929 0.105 2.174 FDI 0.069 5.416 0.067 2.908 0.074 4.756 0.081 2.560 PAT 0.005 0.874 0.064 1.024 0.002 0.179 0.087 1.641 CAP*lIC -0.001 -0.650 -0.001 -0.386 FDJ*lIC 0.000 0.466 0.000 0.247 PAT*lIC 0.002 2.245 0.003 1.783 CAP*RD 0.007 0.192 0.129 1.368 FDI*RD -0.018 -0.617 -0.071 -1.l92 PAT*RD -0.040 1.933 -0.073 -1.648 R2 0.146 0.149 0.146 0.150 LMhet 22.295 [.000] 21.600 [.000] 22.295 [.000] 9.846 [.002] Hausman 77.214 r.00001 79.907 [.0000] 51.060 [.0000] 57.922 [.0000] Developed (24 countries, OBS = 672) Developing (E Asia, 9 countries, OBS = 252) coefficient t-value coefficient t-value coefficient t-value coefficien t t-value GAPO -0.022 -1.769 -0.017 1.328 -0.059 -3.828 -0.099 -3.871 INV 0.000 1.144 0.000 1.309 0.001 1.671 0.002 1.118 OPEN 0.001 4.970 0.001 5.076 0.000 1.964 0.001 1.854 GOV -0.005 -6.916 -0.005 -7.376 -0.001 -1.199 -0.003 1.874 INF 0.000 -2.148 0.000 -1.869 0.000 0.176 0.000 0.938 RD 0.013 2.523 0.039 3.297 0.006 0.803 0.Q78 1.357 He 0.001 1.402 0.001 1.249 0.001 2.190 0.003 2.000 CAP 0.063 2.213 0.219 3.477 0.062 1.725 0.287 2.280 FDI 0.003 0.135 0.175 0.577 0.237 4.574 0.168 1.275 PAT 0.001 2.215 0.040 1.684 -0.Q38 -1.086 -0.264 1.161 CAP*HC -0.001 -0.443 0.000 0.131 'J FDI*HC 0.004 1.724 0.005 1.171 '0 PAT*HC 0.001 0.843 0.007 2.002 CAP*RD -0.073 -2.011 -0.211 -1.590 FDI*RD -0.022 -0.649 -0.147 -1.204 PAT*RD -0.010 -0.587 -0.055 -0.713 R2 0.208 0.225 0.132 0.253 LMhet 0.244 [.622] 0.549 [.459] 3.182 [.074] 1.764 [.184] Hausman 55.103 [.0000] 59.917 [.0000] 11.551 [.1724] 16.465 [.0362] Notes: 1. GAP=Gap in TFP level vis-ii-vis US TFP level; GDPO=ln (initial per capita GDP); INV=investment/GDP; OPEN = (export + import)/GDP; CA P= machinery importlimport; FDI = FDI inflow/investment; PAT= Royalities and licence fees payment/GDP; RD = R&D expenditure/GDP; SC = education attainment rate (secondary [SC2] and higher [SC3j); GOV= government expenditure/GDP; INF=inllation rate. 2. Instrument variables are lagged independent variables and population. 3. T-values are evaluated by Hev if heteroscedasticity does exist. Source: Authors' estimation. 80 Competitivenes.l; FDI and technological activity in East Asia The liberalization of FDI policies is a necessary condition for attracting FDI but it has to be supplemented by other measures. Among these are strengthening the infrastructure, both hard infrastructure such as transpor tation and communication facilities and soft infrastructure such as the governance and legal systems, education and training. 13 Furthermore, one cannot over-emphasize the importance of maintaining a stable macroeco nomic environment with low inflation, sound fiscal policy and stable, realis tic exchange rates. The reduction of business transaction costs and effective investment promotion are also important in competing for FDI. To improve technological capabilities. East Asian economies have to improve the quality of human resources and promote R&D. An up-to-date technology infrastructure. with strong institutions for quality, standards, testing and R&D support, as well as for SME extension services, is essen tial to support capability building. Universities and public R&D institu tions have to link up with industry, particularly in countries that are reaching the stage (as in the NIEs) of autonomous innovation. NOTES I. The authors are at Keio University. Tokyo. and Waseda University and Japan Center for Economic Research. Tokyo. respectively. An earlier version of the paper was presented at the Workshop on Technology Development in East Asia, sponsored by the World Bank. Bali. Indonesia. December 14-15, 2000. 2. The figures are lor East Asia and the Pacific rather than lor East Asia alone. 3. TFP growth is estimated by applying the following formula: In (TFP(t)l1nTFP(t-I)} = In { YU)I Y(t-I)} - k,s,{t)ln {Xi(t)/ X,(t-I)} where Y is value-added. Xj is input and s; Iron & steel 4 0 0 0 0 0 0 () 0 0 N on-ferrous metals 5 2 2 2 I 0 I 0 0 I General machinery 11 3 6 6 6 0 4 1 I 0 Electric machinery 34 0 6 5 2 0 I I 0 3 machinery 7 0 10 7 4 0 I 0 3 3 Precision machinery I 0 0 0 0 0 0 0 0 0 Other manufacturing 5 0 8 8 " ') 0 0 1 0 Construction 0 0 0 0 0 0 0 0 0 0 Commerce 1l 0 4 4 3 I 0 2 0 I Services 8 0 0 0 0 0 0 0 0 0 Others 17 0 0 0 0 0 0 0 0 0 Total 143 15 81 77 54 12 7 34 17 Indonesia Malaysia Philippines Thailand China Middle Europe Oceania Africa Total East Agriculture 0 0 0 0 0 0 1 0 0 3 Mining 0 0 0 0 0 0 0 0 0 0 Manufacturing 2 6 2 6 6 0 34 3 0 234 Food 0 I 0 0 0 0 1 1 0 41 Textiles 0 1 0 0 0 0 0 0 0 2 Wood, pulp 0 0 I 0 0 0 0 0 0 2 Chemicals 1 1 0 4 0 0 4 0 0 54 Coal & oil products 0 0 0 0 0 0 0 0 0 0 Iron & steel 0 0 0 0 0 0 0 0 0 4 ...... ...... Non-ferrous metals 0 I 0 0 0 0 I 0 0 10 ...... General machinery 0 0 0 0 0 0 7 0 0 27 Electric machinery 0 2 0 0 0 9 0 0 49 Transport machinery I 0 1 0 0 5 0 23 Precision machinery 0 0 0 0 0 0 2 1 0 4 Other manufacturing 0 0 0 0 6 0 5 0 0 18 Construction 0 0 0 0 0 0 0 0 0 0 Commerce 0 0 I 0 0 0 5 0 0 20 Services 0 0 0 0 0 0 5 0 0 13 Others 0 0 0 0 0 0 4 0 0 21 Total 2 6 3 6 6 0 49 3 0 291 Source: MITf, Klligai Jigyo Katsudo Kihol1 Chosa (Comprehensive Survey of Overseas Activities of Japanese Firms), no. 5, Tokyo, 1993. Table 5.3 Overseas R&D by Japanese firms (million North Latin Asia East NIEs Hong Korea Singapore Taiwan ASEAN4 America America Asia Kong Agriculture 10 6 0 0 0 0 0 0 0 0 Mining 0 60 0 0 0 0 0 0 0 0 Manufacturing 48071 319 6585 5627 4187 15 893 653 2626 1416 Food 130 0 91 91 83 0 1 10 72 8 Textiles 0 2 172 172 43 0 38 0 5 126 Wood, pulp 10 93 130 130 10 0 0 0 10 120 Chemicals 10956 21 827 827 728 0 477 25 226 94 ...... Coal & oil products 0 0 0 0 0 0 0 0 0 0 ...... ~ Iron & steel 764 0 0 0 0 0 0 0 0 0 Non-ferrous metals 477 138 62 62 16 0 12 4 0 36 General machinery 5453 35 209 209 201 0 17 70 114 8 Electric machinery 23664 30 1959 1941 1049 15 309 511 214 892 Transport machinery 3719 0 1157 219 151 0 16 6 129 68 Precision machinery 90 0 151 151 151 0 2 27 122 0 Other manufacturing 2808 0 1827 1825 1755 0 21 0 1734 64 Construction 2 11 1 1 0 0 0 0 0 0 Commerce 631 0 285 285 277 191 0 86 0 8 Services 4737 0 5 5 5 0 0 0 5 0 Others 7088 0 1 1 0 0 0 0 0 1 Total 60539 396 6877 5919 4469 206 893 739 2631 1425 Indonesia Malaysia Philippines Thailand China Middle Europe Oceania Africa Total] East Agriculture 0 0 0 0 0 0 7 0 0 23 Mining 0 0 0 0 () 0 0 0 0 60 Manufacturing 45 984 162 225 24 0 68974 1291 6 125246 Food 0 () 0 8 0 0 80 25 0 326 Textiles 18 l3 0 95 3 0 302 0 2 478 Wood, Pulp 0 0 120 0 0 0 0 0 () 233 Chemicals 1 44 0 49 5 0 1320 0 0 13124 Coal & oil products 0 0 0 0 0 0 0 0 0 0 Iron & steel () 0 0 0 0 0 0 0 0 764 Non-ferrous metals 4 2 10 20 10 () 45 136 0 858 " " General machinery 0 0 0 8 0 0 3286 0 0 8983 '"" Electric machinery 0 844 20 28 0 0 54222 66 0 79941 Transport machinery 22 20 12 14 0 0 2403 1 046 4 8329 Precision machinery 0 0 0 0 0 () 689 18 0 948 Other manufacturing 0 61 0 3 6 0 6627 0 0 11262 Construction 0 0 0 0 I 0 0 0 0 14 Commerce 0 0 8 0 0 () 307 1 0 1224 Services 0 0 0 0 0 0 19 0 0 4761 Others 0 0 0 I 0 0 1248 57 0 8790 Total 45 984 170 226 25 0 70555 1349 6 139326 Nole: I The totals ae for North America, Latin America, Asia, the Middle East, Europe, Oceania, and Africa. Source: MITI, Kaigai Jigyo Kalsudo Killon Chosa (Comprehensive Survey of Overseas Activities of Japanese Firms), no. 5, Tokyo, 1993. Table 5.4 Researchers at overseas affiliates by Japanese firms (number) North Latin Asia East NIES Hong Korea Singapore Taiwan ASEAN4 America America Asia Kong Agriculture 15 16 6 6 0 0 0 0 0 6 Mining 4 633 0 0 0 0 0 0 0 0 Manufacturing 11147 1692 17184 16161 5613 696 1342 1649 1926 9753 Food 183 3 433 433 145 0 10 29 106 281 Textiles 205 64 4892 4892 113 10 93 0 10 4574 Wood, pulp 92 0 16 16 1 0 0 I 0 15 Chemicals 1393 219 2047 2047 971 40 449 168 314 1022 " "- 0, Coal & oil products 5 0 35 35 30 0 0 0 30 5 Iron & steel 107 0 21 21 3 0 0 2 16 Non-ferrous metals 630 109 586 586 81 0 55 22 4 466 General machinery 1047 90 973 973 817 480 52 132 153 154 Electric machinery 3474 570 2733 2592 1258 41 300 405 512 1197 Transport machinery 1688 386 3352 2775 1162 0 212 867 83 1348 Precision machinery 5 0 230 230 210 10 20 18 162 3 Other manufacturing 2318 251 1866 1561 822 llS 151 5 551 672 Construction 220 58 1393 1393 451 76 3 131 241 873 Commerce 768 43 774 774 262 196 0 66 0 512 Services 515 69 72 72 26 0 23 0 3 4 Others 1343 4 624 624 468 0 468 0 0 156 Total 14012 2515 20053 19030 6820 968 1836 1846 2170 11304 Indonesia Malaysia Philippines Thailand China Middle Europe Oceania Africa Total East Agriculture 6 0 0 0 0 0 19 0 0 56 Mining 0 0 0 0 0 35 7 0 434 1113 Manufacturing 1804 1601 312 6036 795 45 7791 492 1106 39457 Food 61 152 0 68 7 0 112 31 0 762 Textiles 437 195 0 3942 205 0 5 0 1096 6262 Wood, pulp 0 2 13 0 0 0 139 0 0 247 Chemicals 544 137 3 338 54 35 695 10 0 4399 Coal & oil products 0 0 0 5 0 0 0 0 0 40 Iron & Steel 0 6 10 0 2 2 0 0 10 140 " ...... Non-Ferrous metals 128 35 25 278 39 0 18 10 0 1353 "I General machinery 10 93 2 49 2 0 870 0 0 2980 Electric machinery 0 592 216 389 137 8 1618 10 0 8413 Transport machinery 481 51 40 776 265 0 399 430 0 6255 Precision machinery 0 3 0 0 17 0 187 0 0 422 Other manufacturing 143 335 3 191 67 0 3748 I 0 8184 Construction 169 117 57 530 69 16 174 0 0 1861 Commerce 0 100 12 400 0 0 337 317 0 2239 Services 0 I 0 3 42 0 71 38 0 765 Others 0 0 134 22 0 0 276 4 0 2251 Total 1979 1819 515 6991 906 96 8675 851 1540 47742 Source: MITl, Kaigai Jigyo Katsudo Kihon Chosa (Comprehensive Survey of Overseas Activities of Japanese Firms), no. 5, Tokyo, 1993. Table 5.5 Functions of affiliates' research centres Information Collaboration Sales Production Product Modification collection with university support support design of imports North America 31.5 8.7 33.7 30.4 21.7 15.2 Latin America 14.3 7.1 14.3 21.4 0.0 21.4 Asia 20.7 0.8 11.6 44.6 2.5 23.1 East Asia 19.8 0.9 11.2 44.8 2.6 13.8 NIEs 2104 1.8 12.5 39.3 1.8 19.6 Hong Kong 20.0 0.0 20.0 40.0 0.0 20.0 Korea 20.0 6.7 6.7 20.0 0.0 13.3 ...... ...... Singapore 25.0 0.0 25.0 58.3 0.0 16.7 Co Taiwan 20.8 0.0 8.3 41.7 4.2 25.0 ASEAN4 18.9 0.0 11.3 47.2 3.8 7.5 Indonesia 22.2 0.0 11.1 33.3 Il.l 0.0 Malaysia 27.3 0.0 13.6 36.4 0.0 4.5 Philippines 0.0 0.0 0.0 50.0 0.0 16.7 Thailand 12.5 0.0 12.5 68.8 6.3 12.5 China 14.3 0.0 0.0 71.4 0.0 14.3 Middle East nJa nla nJa nla n/a nla Europe 8.7 0.0 19.6 32.6 10.9 1704 Oceania 50.0 0.0 50.0 50.0 50.0 0.0 Africa 33.3 0.0 0.0 66.7 0.0 0.0 Total 22.3 3.6 20.5 37.1 lOA 19.1 Development of Development of Development Basic Others Sample size local product new products of exports research North America 41.3 22.8 18.5 16.3 12.0 92 Latin America 64.3 0.0 7.1 0.0 0.0 14 Asia 38.8 4.1 7.4 4.1 5.0 121 East Asia 38.8 6.0 8.6 2.6 5.2 116 NIEs 41.1 7.1 7.1 3.6 5.4 56 Hong Kong 20.0 20.0 20.0 20.0 0.0 5 Korea 53.3 20.0 13.3 0.0 0.0 15 Singapore 0.0 0.0 0.0 0.0 16.7 12 Taiwan 58.3 0.0 4.2 4.2 4.2 24 .... .... ASEAN4 39.6 5.7 11.3 0.0 5.7 53 \C Indonesia 33.3 22.2 22.2 0.0 ILl 9 40.9 0.0 13.6 0.0 9.1 22 Philippines 50.0 16.7 0.0 0.0 0.0 6 Thailand 37.5 0.0 6.3 0.0 0.0 16 China 14.3 0.0 0.0 14.3 0.0 7 Middle East nla nla nla nla nla nla Europe 52.2 6.5 8.7 2.2 2.2 46 Oceania 0.0 0.0 0.0 0.0 0.0 2 Africa 66.7 0.0 33.3 0.0 0.0 3 Total 43.2 10.4 11.5 7.6 6.5 278 Source: MITI, Kaigai Jigyo Kalsudo Kihon Chasa (Comprehensive Survey of Overseas Activities of Japanese Firms), no. 5, Tokyo, 1993. 120 Competitiveness, EDI and technological activity in East Asia base of scientists and engineers, competitiveness lies in developing new products and conducting basic research. 3.2 The Determinants of Overseas R&D Activities by Japanese Firms This section examines the determinants of overseas R&D by Japanese MNCs, starting with previous studies of this subject. Odagiri and Yasuda (1996 and 1997) conduct detailed analyses of overseas R&D by Japanese firms, using industry and firm-level data. They find that Japanese MNCs active in overseas R&D activities have the following characteristics: large size, high export sales ratios, high overseas to total (parent and affiliates) sales ratios, large numbers of overseas production units, high R&D-sales ratios and low technological advantage. Their findings suggest that R&D activities are subject to economies of scale and scope, and that firms with high sales dependence on overseas markets tend to undertake more R&D in foreign countries. Overseas R&D is conducted to provide support for local manufacturing and firms set up overseas R&D units to acquire technology. Their analysis of overseas R&D in Asia is broadly similar, except that the relationships are not statistically significant except for one variable, the technological level of the host country. Unlike the finding for the sample as a whole, Japanese firms set up R&D units in Asia in countries where they have a technological advantage. This seems to indicate that Japanese firms set up R&D units to deploy their technology in Asia rather than to acquire new technology. As far as affiliates are concerned, Odagiri and Yasuda (1996) find that affiliates with R&D have large sales, high dependence on local sales and high local input content. Furthermore, they find that parent firms with high R&D-sales ratios are also active in R&D activities overseas. These findings are consistent with the findings from their firm-level study (1997), where they find that firms with large sales, high R&D intensity, high dependence on local or overseas market for their sales as well as purchase of their inputs tend to be active in overseas R&D. We now present the result of regression analysis using firm-level data from the MITI survey. One advantage of this data set is that we can link information on parents to overseas affiliates. We first analyse the determi nants of the decision to set up overseas R&D units, and then the determi nants of the level of R&D activities by affiliates. for the analysis of the decision to establish R&D units we use a binary variable that takes the value I for the presence of an R&D unit and zero otherwise. For the analy sis of the determinants of level of R&D activities in overseas affiliates, we use the value of R&D expenditure and the number of researchers as indi cators. R&D and intra~firm technology transfi:r: Japanese MNCs 121 For both analyses, we use the characteristics of parent firms and affili ates as explanatory variables. Specifically, we use the following variables as the characteristics of parent firms (with the variable names and expected signs in parentheses): R&D~sales ratio (PRD, +), the number of overseas affiliates (POA, +), export-sales ratio (PEX, +), overseas production ratio (POP, +), royalty receipts~sales ratios (PRS, +), and worldwide sales (PWS, +). For overseas affiliates we use the following variables: affiliates' sales/ worldwide sales (ASW, +), local sales ratio (ALS, +), local procurement ratio (ALP, +), local procurement ratio for investment goods (ALP!, +), number of years in operation (AYR, +), share of equity held by parent company (AEQ, +) and technology transfer requirement (ATT, +). The expected signs are derived mainly from Odagiri and Yasuda (1996 and 1997). Basically, firms with large-scale, active R&D and high dependency on overseas sales are assumed to be active in overseas R&D. An explanation of the expected signs for two of the variables is in order. The share of equity in an affiliate is likely to influence positively the level of affiliate R&D since a parent with high equity participation would spend more on R&D where it can appropriate a larger share of the returns. The parent would also be willing to spend more on transferring technology effectively to such affiliates. Technology transfer requirements imposed by host governments will also tend to raise affiliate R&D, since MNCs would presumably seek to comply with local policies. In addition to the characteristics of parent firms and affiliates, we include four characteristics of host economies as independent variables: the quality of human capital (HC), the level of R&D (RD), the share of manufactur ing in GOP (MANF) and the cumulative number of Japanese investments (CFDl). The quality of human capital is measured by educational attain ment ratios, taken from Barro and Lee (2000), while the level of R&D is measured by the share of R&D expenditure to GOP. The expected impact of all these variables on R&D by overseas affiliates is positive. The results of the analysis are shown in Tables 5.6-5.8. On the determi nants of the decision to establish overseas R&D, we find that Japanese parents with active R&D (PRD), high technological capability (PRS) and large worldwide sales (P WS) tend to set up R&D units in developed coun tries (Table 5.6). However, this relationship is not significant for developing countries. Turning to foreign affiliates, we find that affiliates with high local procurement (ALP) and long periods of operation (AYR) tend to have R&D units in developed countries. In deVeloping countries, affiliates with large shares in worldwide sales (ASW) and those subject to technology transfer requirement (ATT) are likely to set up R&D units in East Asia. Technology transfer requirements also have positive effects on affiliate Table 5.6 The determinants of establishment of overseas R&D units (probit estimation) Total Developed countries Developing countries East Asia coefficient t-statistic coefficient t-statistic coefficient t-statistic coefficient t-statistic Characteristics of parent firm PRD 1.738 1.227 5.491 2.624*** -6.117 -1.250 -7.323 1.186 POA -0.002 -0.742 -0.002 -0.573 -0.001 -0.282 -0.002 -0.417 PEX -0.269 -0.637 0.497 0.854 -0.407 -0.600 -0.364 -0.426 .... ~ POP -0.265 -0.411 -1.463 1.637 -0.019 -0.017 -1.240 -0.795 ~ PRS 24.473 2.015** 27.746 1.702* 22.900 1.455 40.935 1.015 PWS 0.154 2.793*** 0.169 2.201 ** 0.087 1.000 0.185 1.S55 Characteristics of affiliates ASW 1.185 1.275 -2.012 -0.995 1.568 1.I48 3.159 1.670* ALS 0.135 0.774 0.073 0.313 0.363 \.150 0.257 0.662 ALP 0.578 3.244*** 0.737 3.078*** 0.440 1.341 0.024 0.056 ALP! 0.007 1.443 0.009 1.443 O.OlD 0.923 0.001 0.070 AYR 0.347 2.485** 0.346 1.872* 0.267 1.083 0.373 1.186 AEQ 0.000 1.566 0.000 1.241 0.000 1.067 0.000 -0.271 ATT 0.326 1.888* -0.130 -0.327 0.715 3.220*** 0.662 2.448** Characteristics of host countries HC 0.008 1.518 0.013 1.946* 0.007 U51 0.069 1.342 RD 0.211 1.078 0.204 0.480 0.041 1.215 0.280 1.770* MANF 0.102 0.672 1.871 1.227 0.030 0.131 0.080 0.207 CFDI 0.008 0.116 0.295 1.204 0.219 1.835* 0.732 1.310 R2 0.163 0.176 0.244 0.255 InL -246.616 -145.234 -82.512 -54.131 ..... OBS 1493 650 843 710 t...) "" Notes: Characteristics of parent firm: PRD: R&D/sales ratio; POA: number of overseas affiliates; PEX: exports/sales ratio; POP: overseas production ratio; PRY: royalty receipts/sales ratio; PWS: In (worldwide sales). Characteristics of overseas affiliates: ASW: sales/worldwide sales; ALS: local sales ratio; ALP: local procurement ratio, ALP1: local procurement ratio of investment goods; AYR: number of years in operation: AEQ: equity participation ratio; ATT: technology transfer requirement. Characteristics of host countries: HC: education attainment rate; RD: R&D expenditure/GOP; M ANF: log (GOP of manufacturing sector); CFD/: log (accumulative FDI); ***, **, and * indicate statistical significance at I, 5. and 10 per cent, respectively. Source: Authors' estimation. Table 5.7 The determinants of R&D expenditure at overseas affiliates of Japanesejirms (dependent variable =/n (R&D expenditure in affiliates) (Tobit Estimation) Total Developed countries Developing countries East Asia Censored 1305 544 761 642 Unsensored 190 106 84 68 coefficient t-statistic coefficient t-statistic coefficient t-statistic coefficient t -statistic Characteristics of parent firm PRD 13.984 2.187** 26.638 2.710*** 9.938 0.879 11.267 0.879 POA -0.027 -2.542** -0.047 -2.358** -0.010 -1.059 -0.008 -0.787 PEX -0.440 -0.261 2.316 0.841 -1.023 -0.577 -0.4\0 -0.208 -- ~ "" POP -0.739 -0.278 -9.204 -2.108** 4.953 1.651 * 3.310 0.977 PRS 61.694 3.014*** 92.594 3.115*** 26.248 1.052 31.526 1.236 PWS 0.921 3.969*** 1.307 3.324*** 0.342 1.383 0.380 1.349 Characteristics of affiliates ASW 9.012 2.357** 0.721 0.083 3.447 0.937 5.358 1.294 ALS -0.272 -0.370 1.192 -1.021 0.240 0.302 0.155 0.165 ALP 4.654 5.747*** 6.483 4.991 *** 3.043 3.367*** 2.136 2.130** ALP! 0.029 1.243 0.045 1.335 0.019 0.678 -0.002 -0.071 AYR 2.639 4.276*** 1.599 1.695* 2.892 4.124*** 3.658 4.340*** AEQ -0.001 1.381 -0.001 -0.614 -0.001 1.255 -0.001 -0.692 ATT 1.931 2.554** 1.869 1.121 1.732 2.616*** 1.931 2.540** Characteristics of host countries HC 0.054 1.975* 0.003 1.049 0.043 1.529 0.189 1.445 RD 0.818 1.631 1.677 1.811 * 0.323 1.661 * 0.239 1.402 MANF 0.127 0.202 2.824 0.417 0.606 1.025 1.372 1.719* CFDI 0.053 0.193 1.747 1.447 0.603 1.767* 2.081 2.352** R2 0.085 0.076 0.110 0.125 lnL -864.170 -488.574 353.828 -279.097 ..... N V, NOles: Characteristics of parent firm: PRD: R&D/Sales ratio; PDA: number of overseas afliliates; PEX: exports/sales ratio; POP: overseas production ratio; PRY: royalty receipts/sales ratio; PWS: In(worldwide sales). Characteristics of overseas afliliates: ASW: sales/worldwide sales; ALS: local sales ratio; ALP: local procurement ratio, ALP/: local procurement ratio of investment goods; A YR: number of years in operation; AEQ: equity participation ratio; ATT: technology transfer requirement. Characteristics of host countries: HC: education attainment rate; RD: R&D expend iture/G DP; MA NF: log (GDP of manufacturing sector); CFD/: log (accumulative FDI); ***, **, and * indicate statistical significance at I, 5, and 10 per cent, respectively. Source: Authors' estimation. Table 5.8 The determinants of the number of researchers at o)lerseas affiliates of Japanese firms (dependent )lariable In (number of researchers in affiliates) (tobit estimation) Total Developed countries Developing countries East Asia Censored 1335 556 779 662 Uncensored 160 94 66 48 coefficient t-statistic coefficient t -statistic coefficien t t -statistic coefficient t-statistic Characteristics of parent firm PRD 9.091 2.119** 16.583 2.741*** 19.359 1.847* 15.034 1.297 POA -0.015 - 2.127** -0.022 -2.024** -0.006 -0.716 -0.005 -0.482 " PEX 0.012 0.010 1.019 0.616 0.460 0.274 0.193 0.097 "" 0\ POP -2.069 1.096 -7.053 -2.674*** 2.024 0.708 1.719 0.519 PRS 39.741 2.906*** 53.606 3.118*** 35.709 1.627 44.686 1.966* PWS 0.855 4.914*** 0.958 3.916*** 0.390 1.576 0.503 1.778* Characteristics of affiliates ASW 9.538 3.580*** 0.456 0.083 6.299 1.808* 5.776 1.437 ALS 0.364 0.706 0.425 0.608 0.282 0.371 -0.300 -0.337 ALP 2.506 4.339*** 3.211 4.138*** 1.957 2.226** 0.607 0.596 ALPI 0.024 1.472 0.020 1.009 0.039 1.389 0.020 0.625 AYR 2.127 4.805*** 1.377 2.446** 3.020 4.347*** 4.421 4.819*** AEQ - ().()O 1 -1.684* 0.000 -0.198 -0.002 -2.544** -0.002 -2.242** ATT 0.734 1.342 -0.630 -0.545 1.458 2.354** 1.787 2.469** Cbaracteristics of bost countries HC 0.004 1.206 0.021 1.570 0.010 1.125 0.029 1.250 RD 0.967 2.622*** 0.248 1.215 0.492 2.040* 0.287 1.549 MANF 0.549 1.281 1.440 0.376 0.275 0.507 0.002 0.003 CFDI 0.230 1.201 0.991 1.462 -0.014 0.043 0.219 0.311 R2 0.101 0.086 0.152 0.175 InL -667.499 387.988 -260.664 -191.916 h.... I-v '..j Notes: Characteristics of parent firm: PRD: R&D/sales mtio; POA: number of overseas affiliates; PEX: exports/sales ratio; POP: overseas production ratio; PRY: royalty receipts/sales ratio; PWS: In(worldwide sales). Characteristics of overseas affiliates: ASW: sales/worldwide sales; ALS: local sales ratio; ALP: local procurement ratio, A LPI: local procurement ratio of investment goods; AYR: number of years in operation; AEQ: equity participation ratio; ArT: technology transfer requirement. Characteristics of host countries: HC: education attainment rate; RD: R&D expenditure/GOP; MANF: log (GOP of manufacturing sector); CFD/: log (accumulative FOT); ***. **, and * indicate statistical significance at 1,5, and 10 per cent, respectively. Source: Authors' estimation. 128 Competitiveness, FD! and technological activity in East Asia R&D in developing countries. Although this suggests that technology transfer requirements are effective, it should be noted that such a policy may discourage foreign firms from investing in countries with such policies in the first place. 12 Let us turn to the determinants of the level of overseas R&D by Japanese MN Cs (Tables 5.7 and 5.8). The variables affecting the decision to establish R&D, not surprisingly, also affect the level of affiliate R&D. However, for parent firms the level of globalization (the number of overseas affiliates, POA) has a negative impact on the level of R&D by affiliates in developed countries. One explanation for this unexpected result may be that R&D is concentrated in a few locations by the parent firm to maximize the benefits of scale economies in R&D activity. In developing countries, we find that, in addition to technology transfer requirements, affiliates with strong dependence on local inputs (ALP) and long periods in operation (AYR) are more active in R&D. The characteristics of host countries also significantly affect the R&D activities of Japanese affiliates. The availability of human capital (He) and R&D resources (RD) in the host countries are positively related to R&D by affiliates in developed and developing countries, as is the share of the man ufacturing sector in the economy (MANUF). The presence of Japanese firms in the host country (CFDI) similarly promotes R&D by affiliates. These findings suggest that affiliates undertake R&D to meet the demand for high quality products in the host market. 4. INTRA-FIRM TECHNOLOGY TRANSFER 4.1 Patterns of Intra-Firm Technology Transfer by Japanese Firms FDI has, as noted, become one of the most important means of transfer ring technology. Governments in host countries have sometimes used poli cies like technology transfer requirements, to promote such transfers by MNCs. Of course, efficient technology transfer is as important for MNCs as it is for the host country, since the performance of affiliates depends largely on the technological capabilities developed in the affiliates. However, there may be differences between host countries and MNCs on the depth of technology transfer desired. This section investigates the extent of technology transfer by Japanese MNCs, at least as far as the data allow. We noted earlier the two types of MNC technology transfer: intra-firm (within the MNC) and technology spillovers to 10caJ firms. Intra-firm technology transfer is achieved, among other means, by training (on the job, in parent companies and so on). R&D and intra-firm technology transfer: Japanese MNCs 129 Technology spillovers take place in various ways: when local workers, who have acquired technology and skills by working at foreign affiliates, move to local companies, when local firms imitate technology and managerial know-how in foreign affiliates, and so on.13 We examine intra-firm technology transfer because of data availability. Before conducting an analysis of technology transfer, we investigate train ing programmes for local employees given by Japanese affiliates, because training programmes are likely to play an important role in technology transfer and also because such programmes tend to indicate the attitude of Japanese affiliates toward technology transfer in that those affiliates with training programmes are eager to undertake technology transfer. Training programmes in Japanese MNCs vary widely according to the nature of the programmes and the characteristics of participants. We examine the presence or absence of training programmes and the extent of adoption of the four types of programmes: training at the parent firm, training in Japan but not at the parent firm, training in the affiliate, and training in the host country but not in the affiliate. Table 5.9 presents the relevant information. Some 9 per cent of Japanese affiliates have training programmes. Those in Asia have the highest percentage (13.4 per cent); within the region, the ASEAN4 and China exhibit high shares. In terms of the types of pro grammes, training in parent firms is the most popular, with three out of four affiliates with training programmes using this method. In the absence of comparable information on other MNCs, it is impossible to judge whether training at parent firms is particularly important for Japanese firms. However, it seems that Japanese firms regard training at parent firms as particularly important for employees to learn the corporate culture. Training in affiliates is the next most popular mode, one out of three firms with training programmes using this mode. Training outside the firm either in the host country or in Japan is less used (one out of five affiliates). Technology transfer is difficult to measure. 14 Some researchers of Japanese MNCs have used evaluations of the transfer by the personnel involved, like Yamashita (1991) who asked about the extent of the tar geted level of technology transfer achieved, that is, 100 per cent, 50 per cent and so on. One of the problems of this approach is the subjectivity of the evaluation. A manager in charge of technology transfer is likely to give a high evaluation, while a person actually engaged in production may give a low evaluation. To overcome this problem, Urata (1999) used infor mation from a questionnaire survey on the nationality of staff given responsibility for such specific tasks as the maintenance of production lines and product development. If local staff, rather than expatriates, are in charge, it is assumed that the technology has been transferred. We l30 Competitiveness, FDI and technological activity in East Asia Table 5.9 Training programmes at overseas affiliates (%) Of the affiliates with training programme Total Training Training Training Training Others affiliates at parent in Japan at affiliate in host with training firm country programme North America 6.6 67.5 8.3 50.0 29.2 3.3 Latin America 8.5 82.9 36.6 31.7 19.5 0.0 Asia 13.4 85.0 22.4 30.9 13.9 4.6 East Asia 13.3 86.1 21.9 30.7 14.2 4.0 NIEs 9.6 80.3 12.4 35.8 19.0 2.9 Hong Kong 1.8 40.0 10.0 70.0 10.0 0.0 Korea 15.2 82.1 10.7 32.1 17.9 0.0 Singapore 9.2 82.5 12.5 27.5 27.5 5.0 Taiwan 15.9 84.7 l3.6 37.3 15.3 3.4 ASEAN4 17.4 89.6 31.9 25.8 11.5 4.9 Indonesia 19.2 81.8 45.5 30.3 3.0 0.0 Malaysia 18.6 90.2 26.2 24.6 23.0 8.2 Philippines 13.8 84.2 31.6 31.6 5.3 0.0 Thailand 16.8 94.2 30.4 23.2 7.2 5.8 China 17.9 90.9 6.1 36.4 9.1 3.0 Middle East 2.4 0.0 0.0 100.0 0.0 0.0 Europe 7.0 69.3 4.4 50.0 41.2 2.6 Oceania 3.3 63.6 9.1 45.5 54.5 0.0 Africa 2.9 50.0 0.0 75.0 25.0 0.0 Total 9.1 78.2 17.2 38.4 22.5 3.7 Source: MITr. Kaigai Jigyo Katsudo Killon Cllosa (Comprehensive Survey of Overseas Activities of Japanese Firms), no. 5, Tokyo, 1993. adopt a similar methodology, one that is admittedly very simplistic and possibly inaccurate. However, in the absence of workable methodology largely because of lack of necessary information, our methodology may be justified. The MIT! survey asked Japanese MNCs whether local staff were respon sible for R&D in affiliates; the results are shown in Table 5.10. On average this was the case in one out of every four affiliates, with very small varia tions between regions except for Africa, where the figure was only one out of ten affiliates. There are variations in the figures among East Asian econ omies, with a very high share (61.3 per cent) in Korea and lower shares for the affiliates in Singapore, Indonesia, Hong Kong and Thailand. On sector al patterns, no clear pattern emerges. R&D and intra~firm technology transfer: Japanese MNCs 131 There is a clear need to conduct a more detailed analysis of the determi nants of technology transfer. The measures used above are very simple and partial, and do not capture the depth of technology transfer and the costs involved. For instance, they do not allow us to judge whether the transfer is 'truncated', with MNCs transferring production know-how rather than innovation capabilities (R&D may, as noted, be largely geared to produc tion support). However, in the absence of better measures and data, this analysis is a useful step forward. 4.2 Determinants of Intra-Firm Technology Transfer This section analyses the determinants of successful intra-firm technology transfer. As noted earlier, there are very few studies of this subject for Japanese MNCs. One is by Urata (1999), who analysed 133 cases of intra firm technology transfer by Japanese firms to Asian affiliates. He found a positive correlation between the extent of technology transfer and the share of equity held by the parent company, when the technologies involved were simple (for example, for equipment maintenance). The opposite was found when sophisticated technologies (such as design technologies, development of new machines and development of new technologies) were involved. He concluded that Japanese MNCs were reluctant to transfer sophisticated technologies to affiliates, transferring such technologies only when there was pressure from joint-venture partners. Urata also found that technology transfer was successful only when MNCs adopted specific measures to promote the transfer, such as providing manuals in local languages and seminars in local areas. Before we take up the determinants of intra-firm technology transfer we should identify the determinants of training programmes, which are impor tant in determining the outcome of intra-firm technology transfers. We use the same explanatory variables as for the determinants of the adoption of training programmes, but we include additional variables for the determi nants of successful technology transfer. The variables and expected signs are: presence of R&D units (ATRU, +), presence of training programmes (ATRN, +), R&D expenditure (ARDE, +) and the number of researchers (ARES, +). The expected signs are based on the assumption that R&D as well as training programmes are effective in transferring technology. The probit estimation method was used for both analyses, since in both cases the dependent variables are binary. The results of the analysis are presented in Table 5.11. Parents with high R&D-sales ratios (PRD) and low export ratios (PE%) tend to have train ing programmes in overseas affiliates. Parents with high R&D-sales ratios tend to set up training programmes in affiliates in developing countries and Table 5.10 Technology tramfer at overseas affiliates of lapanesefirms North Latin Asia East NIEs Hong Korea Singapore Taiwan ASEAN4 America America Asia Kong Agriculture 57.1 53.8 63.6 60.0 0.0 0.0 0.0 0.0 0.0 66.7 Mining 11.1 20.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Food 34.2 71.4 39.2 37.5 33.3 0.0 100.0 28.6 14.3 29.4 Textiles 0.0 23.5 28.9 28.9 60.9 45.5 83.3 0.0 66.7 12.2 Wood, pulp 71.4 0.0 21.4 21.4 14.3 0.0 0.0 0.0 25.0 16.7 Chemicals 47.1 55.6 46.4 45.7 50.0 28.6 91.7 23.5 52.9 41.4 Coal & oil products 0.0 0.0 33.3 33.3 100.0 0.0 0.0 0.0 100.0 0.0 ...... Iron & steel 44.0 75.0 33.3 34.8 37.5 100.0 100.0 25.0 0.0 30.8 '" '"" Non-ferrous metals 40.7 50.0 21.6 21.6 27.3 0.0 66.7 0.0 33.3 20.0 (Jenera 1machinery 32.0 30.8 24.6 25.0 28.6 0.0 83.3 26.7 16.7 13.6 Electric machinery 40.6 47.8 22.5 21.9 25.0 10.0 44.4 13.2 29.4 14.2 Transport machinery 13.8 43.8 25.4 22.3 35.9 0.0 37.5 33.3 35.7 15.2 Precision machinery 66.7 0.0 33.3 33.3 35.3 0.0 83.3 0.0 16.7 40.0 Other manufacturing 34.1 7.1 25.0 25.8 36.3 43.8 57.1 18.8 22.6 14.8 Construction 17.1 44.4 18.9 18.4 21.4 16.7 100.0 20.0 10.0 17.0 Commerce 17.6 11.5 13.4 13.4 12.4 11.3 20.0 11.2 17.5 16.5 Services 14.4 12.5 28.0 27.3 26.2 19.0 62.5 22.0 41.7 30.0 Others 10.5 6.7 11.8 12.1 15.0 30.0 0.0 0.0 0.0 0.0 Total 24.1 25.3 25.0 24.5 26.6 17.9 61.3 15.8 29.3 20.1 Indonesia Malaysia Philippines Thailand China Middle Europe Oceania Africa Total East Agriculture 50.0 0.0 100.0 0.0 0.0 0.0 16.7 40.0 100.0 52.3 Mining 0.0 0.0 0.0 0.0 0.0 0.0 6.3 11.1 33.3 10.9 Food 0.0 0.0 50.0 44.4 71.4 0.0 45.5 75.0 0.0 42.6 Textiles 0.0 20.0 100.0 15.8 32.0 0.0 50.0 33.3 0.0 27.6 Wood, pulp 50.0 0.0 0.0 0.0 100.0 0.0 50.0 0.0 0.0 36.0 Chemicals 30.8 42.1 28.6 48.4 45.5 100.0 57.4 83.3 100.0 50.2 Coal & oil products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 16.7 Iron & steel 0.0 60.0 100.0 0.0 50.0 0.0 0.0 0.0 0.0 39.3 Non-ferrous metals 20.0 22.2 100.0 10.0 0.0 0.0 42.9 0.0 100.0 32.1 ..... t.." General machinery 25.0 16.7 0.0 11.1 50.0 0.0 47.8 50.0 0.0 33.5 t.." Electric machinery 12.5 14.8 25.0 45.5 11.6 100.0 39.8 33.3 33.3 31.2 Transport machinery 21.1 12.5 26.7 4.2 14.3 0.0 45.7 45.5 0.0 26.4 Precision machinery 0.0 33.3 0.0 50.0 20.0 0.0 52.9 50.0 0.0 47.7 Other manufacturing 13.3 8.0 12.5 20.0 27.3 50.0 35.8 44.4 0.0 28.7 Construction 0.0 17.6 20.0 23.8 16.7 25.0 15.4 50.0 0.0 20.0 Commerce 0.0 21.9 37.5 10.0 0.0 0.0 17.1 23.2 0.0 16.1 Services 22.2 35.7 50.0 26.1 25.0 0.0 15.0 7.1 0.0 17.1 Others 0.0 0.0 0.0 0.0 25.0 0.0 15.4 7.1 0.0 9.6 Total 16.1 20.2 30.6 19.1 34.5 25.9 25.7 25.0 9.0 24.7 Note: The figures indicate the percentage share of affiliates where major responsibility in R&D seetion is taken by local staff. ,<,'ource: MrTl, Kaigai Jigyo Katsudo KillOn Chosa (Comprehensive Survey of Overseas Activities of Japanese Firms), no. 5, Tokyo, 1993. Table 5.11 The determinants of training programmes at overseas affiliates of lapanesefirms {dependent variable = training programme at overseas Total countries Developing countries East Asia coefficient t-statistic coefficient t-statistic coefficient t-statistic coefficient t-statistic Characteristics of parent firm PRD 1.841 1.960* 0.148 0.114 5.394 2.972*** 5.677 2.824*** POA -0.001 -0.829 -0.003 -1.205 0.000 -0.019 0.000 0.213 PEX -0.644 -2.959*** -1.166 -3.222*** -0.405 ] .388 -0.357 1.122 .... POP 0.046 0.132 0.578 1.062 -0.325 -0.660 -0.200 -0.366 v., PRS -6.742 1.146 -6.328 -0.794 -6.911 -0.800 -8.159 -0.854 -t.. PWS 0.003 0.125 0.120 2.451 *** -0.073 -2.014** -0.078 1.961 * Characteristics of affiliates ASW 0.762 ].395 1.661 1.479 0.590 0.849 0.416 0.551 ALS 0.126 1.399 0.537 3.663*** -0.103 -0.857 -0.133 -0.972 ALP 0.248 2.364** 0.390 2.348** 0.158 1.102 0.194 l.220 ALP! 0.007 2.349** 0.011 2.415** 0.006 1.266 0.000 0.019 AYR 0.382 4.611 *** 0.363 2.949*** 0.432 3.698*** 0.384 2.990*** AEQ 0.001 5.107*** 0.001 4.2] 7*** 0.000 2.865*** 0.000 2.392** ATT 0.533 5.176*** 0.497 2.049** 0.557 4.798*** 0.511 4.034*** Characteristics of host countries HC 0.005 1.498 0.000 0.026 0.022 1.627 0.072 1.795* RD 0.015 1.256 0.556 2.425** 0.032 1.415 0.060 1.707* MANF 0.200 1.848* 0.755 0.985 0.126 1.461 0.102 1.899* CFDI 0.003 1.105 0.243 1.917* 0.042 1.770* 0.383 1.597 R2 0.080 0.116 0.087 0.098 InL -930.804 -378.768 -528.545 -440.639 .... OBS 1493 650 843 710 t.,; v, Notes: Characteristics of parent l1rm: PRD: R&D/sales ratio; POA: number of overseas affiliates; PEX: exports/sales ratio; POP: overseas production ratio; PRY: royalty receipts/sales ratio; PWS: In(worldwide sales). Characteristics of overseas affiliates: ASW: sales/worldwide sales; ALS: local sales ratio; ALP: local procurement ratio; ALP!: local procurement ratio of investment goods: AYR: number of years in operation: AEQ: equity participation ratio; ATT: technology transfer requirement. Characteristics of host countries: He: education attainment rate; RD: R&D expenditure/GDP; M ANF: log (GDP of manufacturing sector): CFDI: log (accumulative I'DI); ***, **, and · indicate statistical significance at 1,5, and 10 per cent, respectively. Source: Authors' estimation. 136 Competitiveness, FDI and technological activity in East Asia East Asia, but not in developed countries. This suggests that Japanese MNCs transfer technology needed for production to developing countries. Parent firms with high export-output ratios are less likely to set up train ing programmes in affiliates in developed countries, indicating that export oriented firms do not need to train local stafr because exporting from the parent firm is the main means of serving foreign markets. The magnitude of worldwide sales has different impacts on affiliates in developed and developing/East Asian countries. Parents with large worldwide sales tend to train in affiliates in developed countries but not in affiliates in develop ing and East Asian countries. This is unexpected, and may reflect that Japanese firms with large worldwide sales tend to use affiliates in develop ing countries as assembly bases where only limited training is needed. As for affiliates in developing and East Asian countries, those with longer periods in operation (AYR), high equity participation by the parent firm (AEQ) and subject to technology transfer requirement (ATT) tend to have training programmes. In affiliates in developed countries, in addition to these characteristics, high local sales ratio (ALS), high local procurement ratio (ALP), and high local procurement ratio for investment goods (A LPl) are also likely to have training programmes. These findings are consistent with expectations and suggest the follow ing. Parent firms tend to put greater efforts in transferring technology to affiliates in which they have higher equity stakes. Technology transfer requirements seem to induce affiliates to meet the requirements by training local staff. In developed countries, affiliates with local market orientation in sales and procurement feel the need to improve the quality of local employees by training. This is consistent with the earlier observation that parent firms with strong export orientation were less likely to have training programmes. Turning to the characteristics of the host countries, we find that host coun tries with good human resources (HC), R&D (RD), large manufacturing sectors (MANE) and a large presence of Japanese MNCs (CFDI) tend to have training programmes. This suggests that both supply and demand-side factors are important in deciding on setting up training programmes. On the supply side, firms set up training programmes in host countries with well educated popUlations and strong R&D capabilities. On the demand side, large manufacturing sectors and a strong presence of Japanese firms induce affiliates to improve the quality of local employees to meet their needs. Table 5.12 presents the results of the analysis of the determinants of suc cessful intra-firm technology transfer. MNCs successful in transferring technology tend to have high R&D-sales ratios (PRD) and large numbers of overseas affiliates (POA), particularly those with affiliates in developing countries and Asia. Parent firms with high R&D-sales ratios tend to be R&D and intra-firm technology transfer: Japanese MNC~ 137 particularly successful in transferring technology to affiliates in developing countries and Asia. This may indicate that MNCs with a strong technology orientation invest in developing countries to use their technology. The observation that MNCs with many affiliates are sueeessful in transferring technology suggests the importance of accumulated experience. It is some what surprising to find that parent firms with large worldwide sales seem to be unsuccessful in transferring teehnology to affiliates in developing eoun tries or Asia. However, this finding appears consistent with the earlier observation that large Japanese MNCs are not active in conducting R&D nor adopting training programmes in affiliates in developing countries or in East Asia. As to the characteristics of affiliates in developing countries, the results suggest that those with a strong local market orientation in sales have a good record in intra-firm technology transfer: these affiliates have to improve technological capabilities to modify or develop technology to meet local market needs. A somewhat surprising result is a negative and statisti cally significant coefficient on equity participation by the parent (AEQ) for affiliates in developing and East Asian countries. This may indicate that dependence on high equity participation for the supply of technology by parent companies is characteristic of affiliates that are not able successfully to absorb technology. The presence of R&D and training programmes has a positive impact on technology transfer to affiliates in developing countries. In developed host countries, affiliates with large worldwide sales, strong dependenee on local markets for the purchase of investment goods and active R&D contribute to successful technology transfer. These findings are consistent with our expectations. Finally, findings on the characteristics of host countries suggest that the availability of skills and R&D resources, and the presence of a large man ufacturing sector promote intra-firm technology transfer by Japanese firms in East Asian countries. 5. CONCLUSIONS Acquisition and assimilation of technology have played an important role in promoting economic development. Among various channels for acquir ing technology, technology transfer involving foreign direct investment has increased its importance in recent years. In light of these developments, this chapter has attempted to discern the patterns and the determinants of R&D activities which contribute to technology transfer and intra-firm technology transfer at overseas affiliates of Japanese firms. Table 5.12 The determinants of the extent of technology transfer achieved (dependent variable = affiliates where local staff takes the main responsibility in R&D) (probi! estimation) Total Developed countries Developing countries East Asia coefficient t-statistic coefficient t-statistic coefficient I-statistic coefficien t t-statistic Characteristics of parent firm PRD 2.161 1. 731 * 1.920 1.276 4.331 1.975** 6.737 2.613*** POA 0.004 2.865*** 0.006 2.801 *** 0.003 2.063** 0.003 1.995* PEX -0.084 -0.340 0.213 0.544 -0.031 -0.093 -0.140 -0.365 POP -0.893 -2.198** -2.573 -4.029*** -0.Q35 -0.060 OA08 0.607 PRS 7.995 1.227 6.060 0.781 9.634 1.101 8.973 0.981 ...... ......, PWS -0.Q35 1.099 -0.019 -0.345 -0.080 1.861 * -0.135 -2.662*** 00 Characteristics of affiliates ASW lA45 2A04** 2.263 1. 768* 0.859 1.l20 0.611 0.697 ALS 0.273 2.663*** 0.361 2.191** 0.233 1.661 * 0.095 0.579 ALP 0.132 !.I 15 0.083 OA40 0.221 1.349 0.274 IA86 ALP! 0.008 2.266** 0.007 1.302 0.007 1.426 0.005 0.845 AYR -0.053 -0.560 -0.212 -IA91 0.121 0.893 0.127 0.814 AEQ 0.000 -2.922*** 0.000 -2.132** -0.001 -2.857*** -0.001 -2.680*** ATT 0.008 0.065 -0.656 -2.054** 0.136 1.023 0.135 0.888 ARDU 0.099 OA72 -0.272 -0.916 0.575 1.718* 0.599 IA71 ATRN 0.157 1.892* 0.215 1.620 0.197 1.740* 0.173 1.348 ARDE 0.172 4.331 *** 0.244 4.802*** 0.062 0.731 0.115 1.290 ARES 0.128 1.825* 0.143 lA05 0.082 0.724 0.060 OA68 Characteristics of host countries HC 0.011 1.548 0.005 0.583 0.026 1.629 0.048 2.064** RD 0.311 4.428*** 0.148 0.599 0.280 3.114*** 0.324 3.343*** MANF 0.082 1.018 1.213 1.398 0.213 2.085** 0.216 1.944* CFDI 0.046 1.287 0.107 0.806 0.055 0.844 0.244 1.906* R2 0.132 0.204 0.129 0.160 InL -678.960 -288.802 -362.847 -280.129 DBS 1493 650 843 710 ...... ....., \Q Notes: Characteristics of parent firm: PRD: R&D/sales ratio; POA: number of overseas affiliates; PEX: exports/sales ratio; POP: overseas production ratio; PRY: royalty receipts/sales ratio; PWS: In (worldwide sales). Characteristics of overseas affiliates: ASW: sales/worldwide sales; ALS: local sales ratio; ALP: local procurement ratio; ALP!: local procurement ratio of investment goods; AYR: number of years in operation; AEQ: equity participation ratio; ATT: technology transfer requirement; ARDU: R&D unit; ARTRN: training programme; A RDE: R&D expenditure; A RES: researchers. Characteristics of host countries: HC: education attainment rate; RD: R&D expenditure/GDP; MANF: log (GDP of manufacturing sector); CFDI: log (accumulative FDI); ***. **, and * indicate statistical significance at 1,5, and 10 per cent, respectively. Source: Authors' estimation. 140 Competitiveness, FDI and technological activity in East Asia Our analysis revealed a number of useful and important observations, Japanese firms were found to undertake R&D activities at their affiliates in developing and East Asian countries mainly to support production and develop local products, As to the determinants of R&D activities at their affiliates in developing countries, we found that the affiliates with strong dependence on local markets for their procurement of intermediate inputs as well as those with long operation periods actively undertake R&D, Concerning the determinants of successful intra-firm technology transfer at affiliates in developing and East Asian countries, we found that the avail ability of educated people and R&D resources as well as the large manu facturing sector and a large number of Japanese affiliates contributed to intra-firm technology transfer. Our findings have several important policy implications for developing countries that are interested in obtaining technology by attracting foreign firms. First, it is important to have supporting industries, which supply parts and components to the assemblers, in order to promote R&D activ ities by foreign firms. Second, developing countries have to improve quality of labour, engineers and researchers to be able to successfully assimilate technology from foreign firms. Finally, developing countries should be reminded of the importance of having an FDI-friendly environment to attract FDI in the first place, before thinking about R&D by foreign firms or intra-firm technology transfer. An FDI-friendly environment is com prised of various elements including a liberalized economic environment, a well-functioning legal system, a well-developed infrastructure and a sound macroeconomic environment. To achieve these objectives, govern ments in developing countries are advised to utilize effectively economic and technical cooperation from various sources such as multilateral and regional organizations as well as bilateral schemes. NOTES I. Waseda University and Japan Center for Economic Research, Tokyo and Keio University, Tokyo, respectively. An earlier version of this paper was presented at the Workshop on Technology Development in East Asia, sponsored by the World Bank, Bali, Indonesia, 14--15 December 2000. The authors thank tbe workshop participants for helpful discussions and comments 2. The discussion of Japanese FDI in this section draws on Kawai and U rata (1998). 3. In terms of value, these two regions absorbed more than 60 percent of Japan's FDI, indi cating tbat the average size of Japanese FDI in terms of value in developed regions is greater than that in developing regions. 4. The NIEs are Hong Kong, Singapore, Korea and Taiwan. 5. The Asian NIEs are Hong Kong, Korea, Singapore and Taiwan, while the members of ASEAN are Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Vietnam. R&D and intra-firm technology transfer: Japanese MNCs 141 6. The ASEAN4 countries are Indonesia, Malaysia, the Philippines and Thailand. 7. See Urata (1993) for a detailed discussion. 8. The figures for 1995 are not shown in Table 5.2. In terms of cumulative value from 1986 to 1995, electric machinery has the largest share among the manufacturing subsectors, followed by textiles and chemicals. 9. The MIT! has conducted a comprehensive survey of overseas activities of Japanese firms every three years starting in 1980. The results of the survey are published under the title of Kaigai Toshi Kokei So ran (Comprehensive Survey of Overseas A ctivities of Japanese Firms). In the 1992 survey, a questionnaire was sent to 3378 Japanese multinationals, 1594 of whom responded. The respondents covered the activities of 7108 overseas affil iates. The MIT! has also conducted an annual survey with the more limited number of questions except for the years of Comprehensive Survey. The results of the survey are published under the title of Wagakuni Kigyo no Kaigai Jigyo Katsudo (Overseas Activities of Japanese Firmsj. Motives behind FDI are asked for in the comprehensive surveys but not in the annual surveys. 10. The figures shown here refer to R&D ratios for all affiliates including those without R&D expenditure. The R&D-sales ratios for affiliates with R&D expenditure are significantly higher, 1.1 per cent for all affiliates, 1.1, 2.5 and 0.2 per cent for the affiliates in North America, Europe, and Asia, respectively. II. Our finding is consistent with the findings by Odagiri and Yasuda (1996), who found in their study of Japanese firms' overseas R&D activities that the main purpose of R&D at affiliates in developing countries is to support local manufacturing by transferring tech nology from Japan, while the main purpose of R&D at affiliates in developed countries is to gain access to the leading scientific and technological knowledge. 12. See Fukao and Yue (1997) for the discouraging impact of performance requirements on Japanese firms' decision on FDI. 13. See Navaretli and Tarr (2000) for a review of studies on technology spillover. 14. There exists a vast amount of literature on international technology transfer. However, the main issue has been the mode and costs of international technology transfer rather than the extent of the transfer. See for example, Reddy and Zhao (1990) for a detailed survey and U rata and Kawai (2000) for a brief survey of recent studies. REFERENCES Barro, Robert and Jong-Wha Lee (2000), 'International data on educational attain ment: updates and implications', NBER working paper 7911, September. Fukao, Kyoji and Ximing Yue (1997), 'Foreign direct investment of Japanese firms: an empirical analysis of market and firm specific factors', Mita Gakkai Zasshi, The Keio Economic Society, 9()(2), 11-39 (in Japanese). Kawai, Masahiro and Shujiro Urata (1998), 'Are trade and direct investment sub stitutes or complements? An empirical analysis of Japanese manufacturing industries', in Hiro Lee and David W. Roland-Holst (eds), Economic Development and Cooperation in the Pacific Basin: Trade. Investment, and Environmental Issues, Cambridge: Cambridge University Press. Navaretti, Giorgio Barba and David G. Tarr (2000), 'International knowledge flows and economic performance: a review of the evidence', World Bank Economic Review, 14(1), 1-15. Odagiri, Hiroyuki and Hideto Yasuda (1996), The determinants of overseas R&D by Japanese firms: an empirical study at the ind ustry and company levels', Research Policy, 25, 1059-79. Odagiri, Hiroyuki and Hideto Yasuda (1997), 'Overseas R&D activities of Japanese 142 Competitiveness. FDI and technological activity in East Asia firms', in Akira Goto and Hiroyuki Odagiri (eds), Innovation in Japan, Clarendon Press: Oxford, pp. 204-28. Reddy, N. Mohan and Liming Zhao (1990), 'International technology transfer: a review', Research Policy, 19,285-307. Urata, Shujiro (1993), 'Japanese foreign direct investment and its effect on foreign trade in Asia', in T. Ito and A.o. Krueger (eds), Trade and Protectionism, Chicago: University of Chicago Press for NBER. Urata, Shujiro (1999), 'Intra-firm technology transfer by Japanese multinationals', in Dennis J. Encarnation (ed.), Japanese Multinationals in Asia: Regional Operations in Comparative Perspective, Oxford: Oxford University Press. Urata, Shujiro and Hiroki Kawai (2000), 'Intra-firm technology transfer by Japanese manufacturing firms in Asia', in Takatoshi Ito and Anne O. Krueger (eds), The Role of Foreign Direct Investment in Economic Development, Chicago: University of Chicago Press. Yamashita, Shoichi (1991), 'Economic development of the ASEAN countries and the role of Japanese direct investment', in Sohichi Yamashita (ed.), Tran,lj'er of Japanese Technology and lvfanagement to the ASEAN Countries, Tokyo: lJniversity of Tokyo Press, pp. 3-22. 6. The dynamics of technology development: lessons from the Korean experience Linsu Kim INTRODUCTION No nation has come as far and as fast, from handicrafts to heavy industry and from poverty to prosperity, as the Republic of Korea (Vogel, 1991). Korea has transformed itself from a subsistence agrarian economy into a newly industrialized one in the space of only four decades. In 1961, Korea exhibited most of the symptoms of underdevelopment that poor countries show today. From 1962, the Korean economy grew at an average annual rate of almost 9 per cent, raising GNP per capita in current prices from US$87 in 1962 to US$10550 by 1997 (Kim, 1997a). Despite undergoing the worst economic crisis since the Korea War in 1997, it bounced back impressively in 1999 with a growth rate of 10 per cent, followed in 2000 with 8 per cent. Korea also achieved phenomenal growth in its exports, which increased from a mere US$40 million in 1963 to US$143 billion in 1999. The structure of manufactured exports changed radically over time (see Table 6.1). The share of primary products fell from 64.4 per cent in 1960 to 2.7 per cent in 1999, and that of manufactures rose from 17.6 per cent to 91.5 per cent. The share of simple manufactures shows an inverted U-shape, rising from 17.3 per cent in 1960 to 54.9 per cent in 1980 and then declining to 28.7 per cent by 1999. Within manufactured exports, the share of simple products decreased steadily from 98.5 per cent in 1960 to 63.7 per cent in 1980, and to 31.4 per cent in 1999. The share of complex manufactures in total exports increased steadily from 0.3 per cent in 1960 to 62.7 per cent in 1999; within manufacturing, their share rose from 1.5 per cent to 68.6 per cent. High technology products accounted for 52.4 per cent of complex manufactures by 1999, indicating a significant technological upgrading of Korean exports. Moreover, unlike most developing countries, which depend heavily on multinational firms for technology-intensive exports and technological upgrading, in Korea the main agents of change were domestic firms. 143 Table 6.1 Korea's exports by technology intensity Classification 1960 1965 1970 1975 1980 1985 1990 1995 1999 Primary products 64.42 35.75 21.28 15.34 7.21 5.03 3.51 2.48 2.68 Manufactured products 17.56 61.10 68.68 66.58 86.19 90.11 90.60 91.69 91.47 products 17.30 52.56 62.51 53.60 54.93 43.86 41.07 26.02 28.73 Resource-based 8.17 15.80 12.51 9.16 10.64 6.51 5.24 6.14 9.75 - -l>.. -l>.. Low-tech 9.12 36.76 50.00 44.44 44.29 37.36 35.83 19.88 18.98 manufact 0.27 8.55 6.16 12.98 31.26 46.25 49.53 65.68 62.74 Medium-tech 0.27 8.55 6.16 12.98 21.63 34.95 29.55 35.55 29.85 High-tech 0.00 0.00 0.00 0.00 9.63 11.30 19.98 30.13 32.89 Others 18.01 3.14 10.04 18.09 6.60 4.85 5.89 5.82 5.84 Total 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 Sources: Uniled Nations. Year Book o!'fnternatlonal Trade Stellisties; Korean Foreign Trade Association (I999)./·()reign Trade Slatisties. Lessons/rom the Korea experience 145 Such phenomenal growth and structural change may be attributed to many social, economic, and technological factors, but most important is probably rapid technological learning by domestic firms (Kim, 1997a). This chapter addresses the following related issues: (1) the initial conditions, (2) the role of technology transfer and technology spillover effects of multina tional corporations (MNCs), (3) domestic technological activities, (4) the role of education and technology institutions, (5) the efficacy of public pol icies, and (6) lessons of the Korean experience for other developing countries. THE INITIAL CONDITIONS Unlike the great majority of developing nations, Korea was a unified, inde pendent state for more than 1200 years since the Silla Dynasty. But sur rounded by big powers China in the west, Mongolia and Russia in the north and Japan in the east -- Korea was frequently subject to foreign inva sion. The most recent was by the Japanese, culminating in colonial rule from 1910 to 1945. Under Japanese colonial rule, manufacturing growth averaged 9.7 per eent per annum between 1910 and 1941 (Suh, 1978). Japanese capital accounted for 94 per cent of the authorized capital of manufacturing estab lishments. There were some 1600 Korean technicians in the manufacturing sector, but they accounted for only 19 per cent of all technicians in Korea. Nearly 300000 Koreans had some experience of mining and manufactur ing by the time Korea gained independence in 1945, leaving it with a larger base of industrial experience than in most countries gaining independence around that time. However, most Korean workers had experience of low level jobs. Modern education, first introduced to Korea by American missionaries, was expanded by the Japanese colonial government. However, the Japanese limited Koreans mainly to the primary grades, seeking a source of labour for simple tasks in agriculture and industry. At the end of Japanese rule, only 2 per cent of the Korean population over the age of fourteen had com pleted secondary schools and the illiteracy rate stood at 78 per cent. The chaos following the end of Japanese rule, the division of the nation into North Korea and South Korea and the ensuing civil war, all between 1945 and 1953, left Korea a 'nation with little left of its past and facing a bleak future' (Mason et al., 1980, p. 58). US aid helped bring Korea back to its pre-war economic level, but the economy suffered from almost all the problems facing most resource-poor, low-income countries today. Korea started on industrial development with a far lower economic and technical base than other newly industrializing economies (NIEs). 146 Competitiveness, FDI and technological activity in East Asia THE ROLE OF MNCS Lacking domestic technological capabilities, Korea had to depend heavily on foreign technology. However, it chose to rely as little as possible on FDI. It encouraged domestic firms to build extensive global networks, with foreign firms providing technology via licensing, capital goods and original equipment manufacture (OEM) contracts. These networks were a major source of technological learning for Korean firms. Intra-Firm Technology Transfer Korea restricted inward FDI when technology was not a critical element and the relevant mature technologies could be acquired more easily by other modes than FDL These arm's length mechanisms included reverse engineering, OEM, and foreign licensing (FL). There was some FDI in this restrictive policy environment (Table 6.2), rising from a mere US$45.4 million in I 962~66 to over US$8.4 billion in 1992~96. This steady increase indicates a gradual relaxation of FDI policy over the years. The value of FDI and its share of external borrowing were, however, sig nificantly lower in Korea than in the other NIEs. For example, Korea's stock of FDI in 1983 was only 7 per cent of Brazil. 23 per cent of Singapore, and less than half of Taiwan and Hong Kong. The proportion of FDI to total external borrowing was only 6.1 per cent in Korea, compared with 92 per cent in Singapore, 45 per cent in Taiwan and 21 per cent in Brazil (KEB, 1987). As a result, FDI played a relatively small role in the Korean economy. Thus, the contribution of foreign firms to the growth of Korean GNP was only 1.3 per cent in 1972~80; its share of total value-added was only 1.1 per cent in 1971 and 4.5 per cent in 1980. Its contribution to manu facturing value-added was only 4.8 per cent and 14.2 per cent during the same periods (Cha, 1983). Manufacturing MNCs set up foreign plants to optimize their global sourcing of inputs and production of outputs. To do this, they transfer the production and management capabilities needed to ensure efficient produc tion. Some MNCs also undertake R&D in host countries, but this is mainly in order to adapt products to local or regional needs. Very rarely do they transfer advanced engineering and innovation capabilities. A comparative analysis of technological learning and market perfor mance in Hyundai Motor, an independent domestic firm, and Daewoo Motor, a joint venture with GM (the world's largest company, with the highest R&D expenditures), is illustrative. Hyundai licensed technologies from several sources and integrated them into an efficient mass-production system. This was a very risky venture, but it forced and motivated Hyundai Table 6.2 Foreign technology transfer to Korea, 1962-99 Source 1962-66 1967-72 1971-76 1977-81 1982-86 1987-91 1992-96 1997-99* Total direct investment Japan 8.3 89.7 627.1 300.9 876.2 2122.3 1548.3 3260.0 8832.7 USA 25.0 95.3 135.0 235.7 581.6 1477.7 2551.8 9910.5 15012.5 All others 12.1 33.6 117.3 184.0 309.6 2035.9 4305.1 18194.1 25194.3 Total 45.4 218.6 879.4 720.6 1767.7 5635.9 8405.1 31364.6 48859.4 Foreign licensing Japan 5.0 58.7 139.8 323.7 1383.6 2437.0 1461.7 5809.5 " USA 0.6 7.8 21.3 159.2 602.7 2121.9 3687.5 4271.4 10872.4 "" '1 All others 0.2 3.5 16.6 152.4 258.5 853.9 1193.3 1458.1 3936.5 Total 0.8 16.3 96.6 451.4 1184.9 4359.4 7317.8 7l91.2 20618.4 Capital-goods Japan 148 1292 4423 14269 20673 54641 80775 40574 216795 USA 75 472 1973 6219 12434 33098 64681 43129 162081 All others 93 777 2445 7490 17871 33213 75387 50529 187805 Total 316 2541 8841 27978 50978 120952 220843 134232 566781 Note: '" The data for 1999 covers only through November. Source: Ministry of Industry and Energy. 148 Competitiveness, FDI and technological activity in East Asia to assimilate foreign technologies as rapidly as possible. In the process, Hyundai invested heavily in R&D in order to build design and innovation capabilities. That was how in 1975 Hyundai developed its first indigenous car model, the 'Pony', making Korea the second nation in Asia (after Japan) to man ufacture its own automobile. The car had 90 per cent local content, and, by dint of R&D, improved rapidly in terms of quality over time. As a result, Hyundai's local market share for passenger cars increased from 19.2 per cent in 1970 to 73.9 per cent in 1979. Hyundai exported nearly 63000 Ponys (to Europe, the Middle East and Asia), accounting for 67 per cent of Korea's total auto exports in 1976-80 and 97 per cent of total passenger car exports in 1983-86. The Pony accounted for 98 per cent of Hyundai's exports in these periods (Kim, 1998). In contrast, Daewoo, constrained by GM's global objectives, relied solely on the MNC for technology and did relatively little to develop its own tech nological capability and even less to design its products. Technology trans fer by joint venture, where the supplier guarantees the performance of the technology transferred, is apt to lead to a passive technological attitude on the part of the recipient. Investments in product and process improvement by Daewoo in 1976-81 were only 19 per cent of those undertaken by Hyundai, although its production capacity, on average, was approximately 70 per cent as large. As a result, though their products were comparable in engine size and price, in 1982 Daewoo was able to operate only at 19.5 per cent of installed capacity as compared with 67.3 per cent for Hyundai (Table 6.3). The differential in labour productivity was as stark: only 2.61 cars per head at Daewoo compared with 8.55 cars per head at Hyundai. Consequently, Daewoo had only 17 per cent passenger car market share compared to 73 per cent for Hyundai. Daewoo improved its product/process development and market perfor mance only a year after the local partner took over managerial control from GM in 1983. The management established a fully-fledged R&D depart ment, adopted the Japanese 'kanban' system, streamlined production, insti tuted a quality control programme and strengthened its marketing drive. Nevertheless, conflicts between the partners continued to plague the joint venture, giving the smaller Kia a chance to outpace Daewoo. In 1992 the partners finally separated, allowing Daewoo to set its own global strategy and recapture the second position after Hyundai. The semiconductor industry shows a similar picture: MNCs transferred production capabilities but not design or innovation capabilities. Several MNCs - Signetics, Fairchild, Motorola, Control Data, AMI and Toshiba - began to assemble discrete devices in Korea in thc 1960s and I970s, taking advantage of cheap local labour. The operations involved simple packag Lessons from the Korea experience 149 Table 6.3 Basic parameters and performance between Hyundai and Daewoo, 1982 Hyundai Motor Daewoo Motor A Capital (million won) 64.4 44.5 B. Number of workers 9129.0 5675.0 C. Sales (billion won) 4.3 1.9 D. Capacity (cars) 116000.0 76000.0 E. Production (cars) 78071.0 14845.0 F. Exports (cars) 13573.0 114.0 Capacity utilization (E/D) 67.3 19.5 Labour productivity (E/B) 8.55 2.16 Capital productivity (E/A) 1212.0 333.6 Export coefficient (FIE) 17.4 0.8 Market share 73.0 13.0 Source: Adapted from Amsden and Kim (1989). ing processes in bonded operations by the wholly-owned subsidiaries, with all parts and components imported from the parent companies and re exported to them. The operations required only about six months' training of unskilled workers, transferring little design or engineering capability to Korea. The largest chaebols Samsung, Hyundai and LG marshalled the resources needed to enter into very large scale integrated (VLSI) chip design and production. Leading foreign producers refused to license VLSI technology to the chaebols. For instance, Texas Instruments and Motorola in the USA and NEe, Toshiba and Hitachi in Japan, refused to license Samsung 64K DRAM technology. But the chaebo/s were able to find small distressed semiconductor companies in the USA that were ready to sell chip designs and processes to find cash for survival. Samsung licensed 64K DRAM design from financially troubled Micron Technologies, and bought a high speed MOS process for US$2.1 million from Zytrex. To master the licensed technologies, Samsung set up an R&D outpost in Silicon Valley in 1983 and hired five Korean-American PhDs in electronics engineering with semiconductor design experience at IBM, Honeywell, Zilog, Intel and National Semiconductors. The outpost also provided opportunities for Korean engineers to participate in training and research in the USA, ena bling them to enhance their learning of VLSI technology (Kim, 1997b). The Asian crisis, however, forced technologically sound but financially weak Korean firms to invite FDI to cope with pressing cash flow prob lems. They placed not only peripheral but also core businesses on sale. 150 Competitivenes5~ FDI and technological activity in East Asia Consequently, unlike China and Southeast Asian countries that witnessed sharp falls in FDI (for example, Singapore 24.8 per cent and Taiwan and Malaysia 19 per cent in 1998), Korea had a sudden increase in FDI. Thus, FDI in manufacturing rose from US$2.3 billion in 1997 to US$8 billion in 1998 and to US$15.5 billion in 1999. The lion's share of the new FDI took the form of mergers with and acquisitions of existing Korean firms. Hewlett-Packard purchased a 45 per cent stake in its Korean subsidiary from its joint venture partner, Samsung Electronics, for US$36 million. Dow Chemical took over Ulsan Pacific Chemical by purchasing a 20 per cent stake. Philips purchased a 50 per cent stake in LG's highly profitable flat panel display business for US$IA billion. Volvo purchased Samsung's construction machinery division for US$730 million. If assets sales are included, the top five chaebols in Korea raised over US$7.4 billion in the year after the crisis. The Korean economy is, there fore, now linked to a far greater extent with foreign multinationals than before. But in most recent cases the FDI transfers neither new processes nor new product technologies. They do transfer managerial capabilities, which introduces transparent and accountable management systems, which Korean firms previously lacked. Some MNCs have also started to conduct R&D locally. Thirty-nine MNCs, or 1.4 per cent of the total number of MNCs operating in Korean manufacturing, have set up R&D centres. Thirty-three of these were estab lished in the 1990s, after Korea had developed a significant R&D base. MNC R&D units, however, account for less than I per cent of the total number of corporate R&D centres. Most of MNC R&D involves adapting products to local markets, suggesting that local innovation by MNCs is fairly insignificant compared to domestic firms. More recent investors, such as Motorola and Lucent Technologies, have come to Korea to tap Korea's leading edge technologies in semiconductor memory chips, flat panel displays, and code division multiple access (CDMA) mobile telecommunications, in which Korea is ahead of Japan and the USA. Motorola acquired a Korean venture firm, Appeal Technology, to tap its advanced design and innovation capabilities and to source a highly compact mobile telephone set for global markets. Lucent Technologies is in the process of establishing an R&D centre in Korea to tap local capabilities in telecom munications. In this sphere, therefore, FDI in Korea is very similar to that in highly advanced economies. Inter-Firm Technology Transfer In contrast to the minimal contribution of FDI to Korea's acquisition of foreign technologies, arm's-length methods such as reverse engineering, Lessonsfrom the Korea experience 151 original equipment manufacturing (OEM), and foreign licensing (FL) have been critical to transferring technologies and supplementing local efforts. In the 1960s and 1970s Korea promoted technology transfer through reverse engineering of turnkey plants and capital goods. The rapid growth of the Korean economy required commensurate growth in investment. Government policy favoured the import of turnkey plants and capital goods to strengthen the competitiveness of local industries. This led to the massive import of equipment at the cost of retarding the development of the local capital goods industry. Protection of the machinery industry was relatively low until the mid-1970s. The chemical, cement, steel and paper industries, established in the 1960s and early 1970s, all resorted to the import of turnkey plants and machinery for their initial setup. Such turnkey plants and foreign capital goods served as an important means of technology transfer for two reasons. First, they embodied new technolo gies. Second, their use led to technological capabilities and to the develop ment of similar products through reverse engineering (Kim and Kim, 1985). Korean firms assimilated the technologies so rapidly that they could undertake subsequent expansions and improvements with little assistance from foreign suppliers (Kim, 1997a). Of the three categories of technology transfer listed in Table 6.2, capital goods imports far surpassed other means in terms of value. Through 1996, capital goods imports were 24.7 times the value of FDI and 32.2 times the value of FL. The total value of capital goods imports was 14 times that of the other two categories combined. Although the values of different modes of technology transfer are not strictly comparable since they measure differ ent things, they are useful indicators when compared with other countries. Among NIEs, the proportion of capital goods imports to total technology transfer was highest in Korea. Korea clearly acquired more technology through capital goods imports than such NIEs as Argentina, Brazil, India and Mexico (Kim, 1997a). The contribution of reverse engineering cannot be quantified, but in-depth studies reveal that such practice was dominant and widespread in electronics (Kim, 1980), chemicals (Westphal et al. 1985), machinery (Kim and Kim, 1985), computers (Kim et aL 1989) and pharma ceuticals (Kim et aL 1989). Several other instruments promoted the inflow of foreign equipment. For example, the slight overvaluation of the local currency, tariff exemp tions on imported capital goods and the financing of purchases by suppli ers' credits (with low rates of interest relative to the domestic market), all worked to increase the attractiveness of these imports. The current account pressures generated were overcome by rapid export expansion based on the efficient use of the imported equipment. OEM contracts were another major mechanism through which Korean 152 Competitiveness, FDI and technological activity in East Asia firms accelerated their technological learning. No official statistics are available on OEM, as OEM sales are not separated from other exports, but several studies reveal that OEM accounts for a significant share. According to the Korea Trade Promotion Corporation, OEM accounted for over 40 per cent of Korea's total exports in 1985. The Korea International Trade Association records that 61 per cent of all exports to Europe were on an OEM basis. In electronics, OEM accounted for 70 to 80 per cent of exports in 1990 (Ernst, 1996). In personal computers, OEM accounted for about 80 per cent of exports in 1989, compared to about 40 per cent in Taiwan. In Samsung Electronics, most reverse engineering projects were followed for several years by OEM exports (Cyhn, 1999). OEM buyers offered various forms of technical assistance to Korean producers to ensure that the products met their technical specifications. The buyers provided not only blueprints but also training for engineers and technicians. They also helped producers improve facilities, manufacturing systems and quality control systems. When a Korean firm provided a proto type to an OEM buyer on the basis of its reverse engineering, the buyer made significant contributions to improving quality to meet its specifiea tions. For instance, when Daewoo undertook OEM manufacture of colour television sets for NEC in Japan, the latter identified over 80 problems ranging from poor sound quality to faulty control knobs and helped Daewoo correct them. Intensive interactions between engineers of the two sides in ten different meetings (lasting more than 30 days per year) Jed to the transfer of a significant amount of design, production, packaging, styling and quality control knowledge (Cyhn, 1999). Such learning allowed Korean firms to progress from OEM to more demanding stages: own design manufacturing (ODM) and own brand manufacturing (OBM) (Hobday, 1995). By the late 1980s, Korean firms had acquired enough capabilities to design their own products. MNCs tested these products and placed large orders. Technology transfer, however, has been far less in ODM arrangements than in OEM. Korean reliance on FL has increased significantly over time as the economy has approached international technology frontiers. Korean policy on FL was quite restrictive in the 1960s. The I 970s, however, saw a signifi cant change as attempts were made to access sophisticated technologies that were mostly protected by intellectual property rights. As a result, royalty payments for FL increased significantly (Table 6.2) from US$0.8 million during the first Five Year Economic Development Plan (1962·66) to US$451A million in the fourth Plan (1977-81). The 19805 and 1990s saw much larger increases. Most foreign licensing in the early years was asso ciated with technical assistance to train local engineers to run turnkey plants (Kim, 1997a). Lessonsfrom the Korea experience 153 In sum, Korea restricted FDI but promoted technology transfer through other means such as capital goods imports, OEM and FL in the early years. Capital was acquired through foreign loans. Such a policy, designed to maintain independence from MNCs, was effective in forcing Korean firms to take the initiative in learning from externalized foreign sources such as reverse engineering, OEM and FL. The high debt ratio, however, contrib uted to the financial crisis in 1997. DOMESTIC TECHNOLOGICAL ACTIVITIES As Korea entered progressively more technology-intensive industries, loeal R&D became more important to sustaining international competitiveness. As a result, R&D investment has seen a quantum jump in the past three decades. Table 6.4 shows that the total R&D increased from US$28.6 million in 1971 to US$3.4 billion in 1990 and to US$9.5 billion in 1998 (1998 was the first year after the Asian crisis and R&D expenditure then was about 13 per cent less than in 1997). Though the Korean economy recorded one of the world's fastest growth rates, R&D rose even faster. R&D as percentage of GDP increased from 0.32 per cent to 2.52 in this period, surpassing many West European countries. The government launched various programmes to increase private R&D. Spurred partly by these programmes and partly by increasing competition in the international market, the number of corporate R&D laboratories rose from 1 in 1970 to 3760 in 1998 and 4810 in 1999, a clear reflection of the seriousness with which Korean firms pursued high-technology develop ment. There were significant structural changes in R&D, the government dominating in early years but private firms taking an increasing role over time. The private sector accounted for only 2 per cent of the total R&D in 1963 and over 80 per cent in 1994, one of the highest in both advanced and newly industrialized countries. The growth rate of total R&D as a share of GDP in Korea (24.2 per cent in 1981 ~91) was the highest in the world, compared to 22.3 per cent in Singapore, 15.8 per cent in Taiwan, 11.4 per cent in Spain and 7.4 per cent in Japan. The annual growth rate of business R&D as a share of GDP was also the highest in Korea (31.6 per cent), compared to 23.8 per cent in Singapore, 16.5 per cent in Taiwan, 14.0 per cent in Spain and 8.8 per cent in Japan (DIST, 1994). Two factors ~ the 'reverse brain drain' and the chaebol- played an impor tant role in this R&D growth. Korean firms recruited trained scientists and engineers of Korean origin resident in the USA. Many had left Korea over a decade earlier, doing PhDs in leading US universities and rising through Table 6.4 Research and development expenditures, 1965-98 ($ million and percentage) 1965 1970 1975 1980 1985 1990 1995 1998 R&D expenditure 2.1 10.5 42.7 282.5 1237.1 3349.9 9440.6 11336.6 Government 1.9 9.2 30.3 180.0 306.8 651.0 1780.9 3051.8 Private sector 0.2 1.3 12.3 102.5 930.3 2698.9 7659.7 8276.4 Govt vs private 61:39 97:03 71:29 64:36 25:75 19:81 19:81 27:73 University R&D n1a 0.4 2.2 25.9 118.8 244.3 770.9 1265.1 Govt research institutes R&D nla 8.9 28.1 104.5 367.2 731.0 1766.7 1979.2 Corporate R&D 0.2 1.3 12.3 81.4 751.0 2374.5 6903.0 8092.3 R&D/GNP 0.26 0.38 0.42 0.77 1.58 1.95 2.51 2.52 Manufacturing sector .... R&D expenditure n1a nla 16.7a 76.0 688.6 2134.7 5809.9 6439.2 v, .... Percent of sales nla nla 0.36 a 0.50 1.51 1.96 2.72 2.64 Number of researchers (total)h 2135 5628 10275 18434 41473 70503 128315 129767 Govt research institutes 1671 2458 3086 4598 7542 10434 15007 12587 Universities 352 20ll 4534 8695 14935 21332 44683 51 162 Private sector 112 1 159 2655 5141 18996 38737 68625 66018 R&D expenditure per researcher 967 1874 4152 15325 27853 47514 73574 87361 (W 1000) Researcher per 10000 population 0.7 1.7 2.9 4.8 10.1 16.4 28.6 27.9 Number of corporate R&D centres 0 Ie 12 54 183 966 2270 3760 Notes: a for 1976. h The ligures do not include research assistants technicians and other supporting personnel. c for Source: Ministry of Science and Technology. Lessons/rom the Korea experience 155 the ranks of major firms such as IBM, Fairchild, Intel and National Semiconductor. The number of scientists and engineers recruited abroad by corporate R&D centres was substantial, 427 in 1992 alone. The outflow of technical graduates posed a serious problem to Korea in the 1970s, when 97 per cent of scientists and 88 per cent of engineers who received training abroad (mainly in the USA) remained there, compared with 35 and 30 per cent for all countries (Hentges, 1975). However, this helped form an impor tant international technical network and a high calibre pool of skills for Korea's development. It later helped it to become a serious contender in several cutting-edge technologies. The chaebols were key players in technological learning in Korean indus try: (l) they were, as noted, in the best position to attract high quality human resources from abroad; (2) they developed the organizational and technical resources to identify, negotiate and finance foreign technology transfer; (3) they had the resources to expand and deepen industrial R&D; and (4) highly diversified chaebols applied experience in one field of busi ness to others, resulting in rapid diffusion of technological capabilities across subsidiaries. As a result, the 20 largest chaebols now account for 72 per cent of corporate R&D in Korea. While increasing in-house R&D, Korean firms began to globalize their innovation activity. LG Electronics, for instance, built a network of R&D laboratories in Tokyo, California, Chicago, Germany and Ireland. These laboratories monitor frontier technological change, develop strategic alli ances with local firms and develop state-of-the-art products through advanced R&D. LG Technology in California, for instance, plays a pivotal role in designing the latest personal computers, display terminals and high resolution monitors, while the laboratory in Chicago concentrates on HDTV, digital VCR and telecommunications equipment. Samsung, Daewoo and Hyundai Electronics have similar R&D outposts. Samsung has R&D facilities in San Jose, Maryland, Boston, Tokyo, Osaka. Sendai in Japan, London, Frankfurt and Moscow. Daewoo has two in France, one in the UK, and one in Russia. Hyundai has outposts in San Jose, Frankfurt. Singapore and Taipei. Another indicator of Korea's growth in R&D is patent registrations at home and abroad. Patenting in Korea has grown significantly in the past two decades. It rose by 48 per cent in the 14 years from 1965 to 1978, then almost tripled in the next 11 years (1979-89) and almost tripled again in the next four (l989~-93), reflecting the increasing importance of innovation and the declining significance of reverse engineering. There is still a large patent gap in comparison with advanced countries, but the gap is closing rapidly. The share of Koreans in local patent registration increased from 11.4 per cent in 1980 to 69.2 per cent by 1999 (see Table 6.5). Korea ranked fifth in 156 Competitiveness, FDI and technological activity in East Asia Table6,j Patent applications and patents granted 1981 1985 1990 1995 1999 Applications National 1319 2703 9082 59236 55970 Foreign 3984 7884 16738 19263 24672 Total 5303 10587 25820 78499 80642 Patents granted National 232 349 2554 6575 43314 Foreign 1576 1919 5208 5937 19321 Total 1808 2268 7762 12512 62635 Source.' Korea National Statistics Office. the world in 1999 in the number of domestic industrial property applica tions, after Japan, the USA China and Germany. In terms of industrial property applications by local residents per head of population, Korea ranks second after Japan. Patent registration in the USA is often used as a measure of international competitiveness. The number of patent registrations in the USA by Koreans is far below that by Taiwanese, let alone that by advanced coun tries. The cumulative number of patents granted to Koreans by the USA between 1969 and 1992 was 1751 compared to 4978 for Taiwan. However, Korea jumped from thirty-fifth place in the number of patents in the USA (among 36 countries listed in an NTIS report) in 1969 to eleventh in 1992, giving an average annual growth rate of 43 per cent (NTIS, 1993). This growth rate was the highest of the countries in the report. A more recent report shows that Korea jumped to sixth place in 1999, with 3679 patents, after only Japan, Germany, Taiwan, France and the UK. Samsung Electronics, the most R&D-intensive firm in Korea, ranked fourth with 1545 US patents, coming only after IBM, NEC and Canon. These figures again indicate how rapidly Korea has gained in technological competitiveness. The crisis of 1997 appears to have made a significant dent in Korean R&D. In order to improve liquidity, the large chaebols reduced R&D by some 13 per cent in the year following the crisis (Kim, 1999). At the same time, there was a surge of small technology-based firms as well-trained sci entists and engineers laid offby chaebols set up on their own, with a strong emphasis on technology. The promotion of venture capital by the govern ment also played a role in fostering this surge. As a result, the number of R&D laboratories increased from 3060 at the time of the crisis to 5200 two years later. Small and medium-sized enterprises (SM Es) account for 95 per Lessons Fom the Korea experience 157 cent of this increase. This shift may signify a structural change in the skewed Korean industrial structure. Korea now leads the world in several advanced technologies such as memory semiconductors, flat panel displays and CDMA mobile telephone sets in terms of product development and production. It has the largest world market shares in satellite receiving systems and videotapes. The International Standardization Organization (ISO) has recently adopted fifteen new technologies in the multimedia area developed by Korean firms as global standards. Nine more are under review. In conclusion, Korean firms acquired mature technologies through imi tative reverse engineering in the early years, but then invested heavily in R&D to enhance their competitiveness in the face of rapidly changing tech nologies. Korea is now one of the most R&D-intensive countries in the world, in terms of the share of GDP devoted to innovation. THE ROLE OF EDUCATION AND TECHNOLOGY INSTITUTIONS The government and the private sector invested heavily in order to achieve rapid expansion in the education sector in Korea. Many other developing countries attained an equally rapid growth in elementary education; what was unique in Korea was the well-balanced expansion at all levels. This rapid expansion created short-term unemployment problems among the educated, but the formation of human capital laid the foundations for the acquisition and assimilation of mature labour-intensive technologies in the 1960s and 1970s. As Korea underwent major structural change in the I 980s, three sets of institutions - universities, government-supported research institutes (GRTs) and SME technical extension services - became important for local technology development. Universities Universities played a fairly marginal role in the early years of Korean industrialization. They were primarily undergraduate teaching-oriented institutions that undertook little research. University R&D, only US$I.3 million in 1970, constituted about 3.5 per cent of the nation's total R&D while accounting for 35.7 per cent of the number of researchers. Frustrated in its efforts to reform the universities, the government founded a research oriented S&T school· the Korea Advanced Institute of Science and Technology (KAIST) in 1975. KAIST played a pivotal role in training a large number of high calibre scientists and engineers. 158 Competifivenes3~ FIJI and technological activity in East Asia Another effort to upgrade university R&D capabilities began in the late 1970s and 1980s. The government established the Korea Scientific and Engineering Foundation in 1977, and later the Korea Research Foundation in 1981, to fund basic research in universities. The government also enacted the Basic Research Promotion Law in 1989, explicitly targeting basic research as one of the nation's top technological priorities. As a result, uni versity research has also expanded significantly as shown in Table 6.4, almost tripling in eight years from US$341.2 million in 1990 to U S$1.06 billion in 1998. The number of university researchers more than doubled from 21332 to 51162 in this period. In addition, emulating the US experi ence, the government started in 1989 to establish Science Research Centres (SRCs) and Engineering Research Centres (ERCs) in the leading univer sities. The number of SRCs and ERCs increased from 13 in 1990 to 84 by 2000, each receiving grants from the government for nine years. There are also encouraging signs on the quality of university research. The number of scientific publications by Koreans cited by the Science Citation Index increased slowly from 27 in 1973 to 171 in 1980, then more rapidly to 1227 in 1988, to 3910 in 1994, and to 10918 in 1999. This meant a rise from thirty-seventh place in the world in 1988 to twenty-fourth in 1994 and sixteenth in 1999. The ranking, however, is still low compared with its eleventh place in gross national product. The financial crisis prompted the government to formulate an ambitious reform programme to transform a dozen leading universities into first-class research-oriented institutions. The government earmarked about US$1.4 billion over seven years for this programme. It is too early to assess the outcome, but if implemented properly the programme is expected signifi cantly to upgrade the quality of scientists and engineers that Korean uni versities will produce. In addition, leading universities have established techno-parks and business incubators in order to link their research with leading firms and to foster technology-based small enterprises spun off from university R&D laboratories. This means that universities will play an increasingly important role in Korea's pursuit of high-technology indus tries in the future. Government Research Institutes Given the weak research base in universities, the government set up GRIs by recruiting overseas-trained Korean scientists and engineers. GRIs dom inated R&D in Korea in the early years, accounting for 83.9 per cent of total R&D expenditures and 43.7 per cent of researchers in 1970. GRIs, however, faced numerous obstacles in the 1970s. They had poor linkages with industry. Most Korean scientists and engineers in the GRIs came from Lessons/rom the Korea experience 159 academic institutions or R&D organizations. There was little demand from industry for the services GRIs offered. Expertise was particularly lacking in manufacturing know-how and the development of prototypes, which were in great demand in the early years. This does not mean that GRls did not produce any results. The problem was that their usable research output was ignored or distrusted by the private sector, which preferred turnkey plants or licences from experienced foreign firms. Large projects like fertilizer, chemical and cement plants relied completely on turnkey projects, while consumer electronics resorted to licensing. Over time, however, the situation changed. The government set up specialized GRIs (for example in chemical, machinery, electronics, ocean, standardization, nuclear energy, biotechnology, system engineering, aerospace and so on) to serve the increasingly diverse and complex needs of the private sector. GRIs began to play an important role in strengthen ing the bargaining power of local enterprises in acquiring foreign technol ogies. For instance, when Corning Glass refused to transfer optical fibre technology to Korea in 1977, two large copper cable producers in Korea entered a joint R&D project with a GR!. After seven years of R&D, the locally deVeloped optical cable was tested successfully on a 3~~km route in 1983. Although this effort eventually ground to a halt, due mainly to slow progress in R&D, it helped local firms in bargaining for foreign technology. Four firms entered licensing agreements with multinational enterprises in 1984 (Kim, 1993). Another important effect on local technology development of GRIs was a sharp cut in import prices. For example, no sooner had a GRI successfully developed Betamethasone in 1979 than its import price went down from Wonf30000/kg to Wonf15000/kg. When the GRI developed Rifamycine in 1982, its import price dropped from Wonfl OOOOOO/kg to Won/450000/kg (KIST, 1994). The most important but unintended role of GRIs in the early years of industrialization in Korea was the creation of experienced researchers. When the private sector began to invest in R&D in the 1980s, the research ers trained in GRIs moved to corporate R&D centres as well as to new GRIs and universities. For example, over 2800 experienced researchers left from KIST, a leading GRI: 420 went to private R&D centres, 784 to uni versities, and 1594 to newly-established GRIs (KIST, 1994). In addition, the government introduced two major sets of national R&D projects: the Industrial Generic Technology Development Project (IGTDP) and the National R&D Project (NRP). IGTDP concentrated on current problems in 'existing' technologies with high externalities. NRP projects focused on future problems in 'new' (to Korea) technologies with a high risk of failure or with high externalities, thus warranting public support. 160 Competitiveness, FDI and technological activity in East Asia Although these projects were also open to universities and corporate R&D centres, GRIs played a dominant role in these projects through the mid 1990s. Nevertheless, in the face of the rapid expansion of private R&D and the increasing intensity of university R&D, the role of GRls weakened rela tively over time. This was so for two reasons. First, GRls, being under government bureaucratic control, were less dynamic than corporate R&D centres, which responded more dynamically to market signals and techno logical change. Second, GRls had difficulties in retaining competent researchers. The best research staff tended to leave for academic institutions for prestige and freedom or for corporate R&D laboratories for higher sal aries. The reform of GRls has been under discussion for some time, but inertia and the labour unions have made it difficult to implement. SME promotion The government has been so preoccupied with mission-oriented projects that it has failed to develop an effective infrastructure for SME promotion. The technical extension networks developed in the 1980s have not been ade quate to meet the technology development needs of SMEs. A few industry specific R&D institutes for SMEs were (belatedly) established in the 1990s, but their effectiveness remains to be seen. In conclusion, despite the rapid growth of demand for technological innovation, Korea's network of technological institutions was not adequate for industrial needs. GRIs played a useful role in some areas in the early decades, but their role needs to be redefined now. The role of universities has become increasingly important, but they have yet to become first-rate research institutions. The incumbent government is determined to reform educational systems and raise R&D capabilities, but it will take a decade or longer before substantial results can be seen. THE EFFICACY OF PUBLIC POLICIES The Korean government has adopted an array of policy instruments to facilitate technological learning in industry and so strengthen international competitiveness. This history can be best understood from the perspective of market demand and supply (Kim and Dahlman, 1992). This involves analysing three aspects: (I) policies to create market needs for technology development (demand side); (2) policies to increase S&T capabilities (supply side); and (3) policies to provide effective links between demand and supply (linkage). Lessons/rom the Korea experience 161 Demand Side Demand-side policies can cover three areas: export promotion, competi tion policy and government procurement. Export promotion, by pushing firms into highly competitive international markets, has been more influen tial than other policies in forcing firms to expedite technological learning. Export-oriented industries accounted for the bulk of foreign licensing, capital goods imports and R&D in Korea. Exporters also created capacity in excess of local market needs to achieve economies of scale; this led to crises and forced them to accelerate technological learning to maximize capacity utilization. Export promotion brought in many OEM buyers, who provided valuable help in acquiring capabilities through interactive tutorial processes. As a result, firms in export-oriented industries learned signifi cantly more rapidly and grew faster than firms in import-substituting industries. Competition policies such as antitrust, trade liberalization and intellec tual property protection also increased the need for technological effort. In response to the increasing economic power of the chaebol, the government enacted the Fair Trade Act in 1980 to prohibit unfair practices in the market and to restrict the growth of the chaebols. At the same time, the government began to liberalize the local market, bringing down tariff and non-tariff barriers, so forcing Korean firms to compete against multina tional firms not only in exports but also in the domestic market. In 1986, the government introduced legislation to protect intellectual property rights, pre-empting the reverse engineering of foreign products. These pol icies forced Korean firms further to intensify technological effort. Government procurement is often mentioned in the literature as an important tool in creating local demand for technological effort. However, except for significant government procurement of personal computers at the formative stage of the industry in the early I 980s (Kim et aI., 1987), this policy did not playa significant role in Korea in creating demand for tech nological effort. Supply side Major supply-side policies cover human resource development, technology transfer and domestic R&D. Given its lack of natural resources, Korea invested heavily in human resource development in the 1950s and 19608 to prepare for industrialization. As a result, given its relative per capita GNP, Korea achieved the highest educational attainment among NIEs. The for mation of human resources enabled it to master mature production tech nologies through reverse engineering in the early years. However, the 162 Competitiveness, FDI and technological activity in East Asia Korean government made a critical mistake in neglecting to invest in research-oriented tertiary education in preparation for knowledge intensive industries, creating a major bottleneck in innovative technologi cal learning in the 1990s. Foreign technology transfer generally plays a major role in technological learning, providing tacit and cxplicit knowledge as well as interactions with foreign suppliers. Korea restricted reliance on FDI. enabling local firms to retain managerial independence and allowing them to set the direction of technological learning. The restriction on foreign licensing also enabled Korean firms to strengthen their bargaining power in negotiating the trans fer of mature but complex technologies; however, it may have resulted in slow learning owing to the restricted inflow of new foreign technologies. The promotion of capital goods imports forced Korean firms to rely heavily on reverse-engineering foreign goods in early years. Well-trained and hard working employees were motivated to maximize technological learning from readily available foreign goods; they had sufficient tacit knowledge to reverse-engineer them successfully in the early years. The government gradually relaxed restrictions on licensing in the 19705, as Korean industries progressed into more complex technologies. Policies had to adapt to the changing economic environment and to facilitate the inflow of more sophisticated foreign technologies. The government's role in R&D was small relative to other countries, accounting for only about 20 per cent of total R&D in the 1990s. The gov ernment's R&D was largely directed to keeping increasingly weak GRIs afloat (in the mid-1990s over 80 per cent of public R&D expenditures went to GRls) and to mission-oriented national projects. Some national projects had significant results, such as the development of electronic switching systems and COMA mobile telephone systems, making Korea the first suc cessfully to commercialize COMA technology. In general, however, R&D policy neglected diffusion-oriented projects like upgrading the quality of tertiary education and university research. Consequently, Korea tended to produce half-baked human resources while spending heavily on R&D, resulting in relatively low R&D productivity. However, the government's initiative in establishing the first venture capital firm facilitated the growth of the private venture capital industry in subsequent years. Linkage Preferential financing and tax incentives are the major instruments that lubricate the linkage process between demand and supply. During the 1970s the interest rate on R&D loans was one of the highest, reflecting the low priority of R&D in government policies. At the same time, preferential Lessonsfrom the Korea experience 163 financing was largely ignored by industry because of its lack of a felt need to invest in R&D (given the ease of acquisition and assimilation of foreign technologies). It was only in the early 1980s that preferential R&D loans became more important for financing private R&D. The impact of prefer ential financing on facilitating R&D activities, however, is dubious. Its interest rates, ranging from 6.5 to 15 per cent, were far higher than similar loans in other countries. Tax incentives were another indirect mechanism for making funds avail able for corporate R&D. In Korea, tax incentives fell into five categories according to objectives served. The most important were tax incentives for corporate R&D, reduced tariffs on the importation of R&D equipment and supplies, deduction of annual non-capital R&D expenditures and human resource development costs from taxable income, and the exemp tion of real estate tax on R&D related properties. In conclusion, preferential financing and tax incentives definitely pro vided funds for corporate R&D activities and lowered their costs, but were peripheral in promoting R&D in Korea. IMPLICATIONS FOR OTHER DEVELOPING COUNTRIES Many political leaders, economic planners and corporate managers in other developing countries have shown a keen interest in learning from the Korean experience. To what extent can and should they emulate the Korean experience? Many parts of Korea's experience may be emulated. First, the expansion of education, particularly at the secondary and tertiary levels, provides the essential base for subsequent industrialization. Many studies support this argument. For instance, Baumol et al. (1991) conclude that the quantity and quality of education are a major influence on whether an economy can catch up with advanced countries. Many developing countries have achieved parity with advanced countries in terms of enrolments in primary schools: it is the provision of secondary and higher education that explains differences in national wealth. A small cadre of highly-educated elites is not sufficient for industrialization; what is required for rapid industrialization is the provision of quality secondary and tertiary education to the whole population. Second, countries can emulate the programmes that the Korean govern ment used to facilitate technological learning. Most important of all was the creation of a competitive market, particularly through export promo tion. Competing in world markets forces firms to undergo a continuous 'life 164 Competitiveness, FDI and technological activity in East Asia or death' struggle for survival, for which they have to accelerate learning by importing and rapidly assimilating foreign technologies and deepening local efforts to innovate new technologies. And the government has a deci sive role to play in turning an economy into an export-oriented one. Third, countries can adopt a liberal policy with respect to the brain drain. High calibre human resourees abroad can provide valuable overseas technical networks and skill pools for subsequent development. Taiwan's surge in high-technology ventures can largely be attributed to Chinese engi neers in Silicon Valley. In Korea, the government offered a very attractive incentive package for Korean~American scientists and engineers to return home through the 1970s. The chaebols foHowed the same strategy after the mid-l 980s. There are also aspects of Korean policies that are difficult to emulate. First, the new rules of international trade under the World Trade Organization (WTO) make it difficult to protect infant industries. The growing pressures to liberalize domestic markets for products, services and investment make it more difficult now to stay independent of multinationals. Second, stronger intellectual property rights protection restricts the imi tation of foreign technologies. China, for instance, faces enormous pres sures from the USA to honour intellectual property rights, which Japan, Korea and Taiwan did not face during their early industrialization. Third, not all developing countries can emulate the cultural and histori cal conditions that inculcated an entrepreneurial and hard-working spirit in Koreans. These include: (1) a neo-Confucian culture that emphasizes disci pline, learning and harmonious interpersonal relations; (2) the Korean War that transformed Korea from a rigid, closed and class-based society into a dynamic, flexible and classless society; (3) the adversity imposed by several climatic conditions; (4) the social competition caused by a dense population; and (5) the adaptability resulting from frequent foreign invasions. There are also two aspects of Korea's experience that other countries should not emulate. The first is Korea's promotion of large conglomerates, which, despite their strengths, stifled the healthy growth of SMEs. It is better to have well-balanced growth of both large and small firms, as in Japan and Germany. Second, the Korean government was so preoccupied with short-term production and export goals in the 1970s and 1980s that it failed to invest in building the infrastructure needed for the future. One of the most strik ing examples is under-investment in upgrading the quality of tertiary edu cation. Public policies may be introduced overnight, and technology and capital may be imported relatively quickly, but creating human resources requires long-term investment. The Korean experience also offers other lessons. Foreign technology Lessonsfrom the Korea experience 165 transfer should not be viewed as a substitute for in-house efforts or vice versa. The two are complementary. Foreign technology transfer can provide new knowledge and serve as a catalyst for technological change, enabling firms in developing countries to make quantum jumps in techno logical learning. In-house efforts can, on the other hand, raise local capa bilities, strengthen bargaining power in transfer negotiations and enable recipients to rapidly assimilate imported technology. FDI transfers production capability quickly and efficiently, but does not necessarily transfer design or innovation capability, particularly when the parent company uses affiliates to exploit the local market in host countries. Should firms in developing countries then go independent or enter into joint venture with technology suppliers? When firms in developing coun tries invest aggressively in technological learning and can deepen their capabilities, it is better to remain independent of foreign equity participa tion (quadrant I in Figure 6.1), particularly when they have a global vision. Even if some equity participation is allowed, management independence should be maintained. Otherwise (quadrant 3) conflicts can arise between the joint-venture partners. When technology recipients are not aggressive in technological learning, a joint-venture arrangement is preferable (quad rant 4); learning is, however, at the pace set by the foreign parent company. As a result, recipients may remain dependent for technology on the parent. Foreign licensing in a 'packaged' form from a single source involves little risk to the technology recipient, as the supplier guarantees the performance of the transferred technology. However, it leads to a passive attitude by the recipient in the learning process. In contrast, when the recipient unpack ages technologies, acquires them from multiple sources and is responsible Strategy for technological learning Aggressive I Not aggress~.ve § Slow initial learning Slow learning <.::: but dynarnic throughout Q learning in long run .~ Independent <.+::; (1) (1) -= .~ Q Rapid initial learning Learning at the .9 but conflicts restrict pace of the parent a Joint venture firm's strategy. .~ dynamic learning in long run Dependency I '" 1.8 Construction materials 55 4 10 23 38 58 1.9 Others 304 216 173 164 20 180 2. Financial institutionsh 664 72 112 842 247 107 3. Trade 1591 545 1035 1052 1042 75 4. Construction 964 70 164 192 152 -19 5. Mining & quarrying 439 19 20 22 -42 -309 6. Agriculture 34 2 I I 2 I 7. Services 312 125 292 276 485 726 8. Investment 59 -22 26 364 571 273 9. Real estate 2547 753 112 28 149 75 10. Others -202 -3 47 161 -9 -383 Total 9245 2271 3627 5142 3562 2447 Shares in total (%) Sector 19911995 1996 1997 1998 1999 2000 I. Industry 30.7 31.2 50.1 42.9 35.6 77.6 1.1 Food & sugar 2.5 2.0 6.2 1.4 2.6 6.6 1.2 Textiles 1.8 2.2 l.l 2.4 0.6 1.4 .3 Metal & non metallic 4.2 5.0 6.0 6.6 7.4 4.5 1.4 Electrical appliances 11.0 10.6 16.6 5.1 11.9 11.3 .5 Machinery & transport equipment 3.8 4.8 10.9 12.9 11.0 22.2 1.6 Chemicals 5.9 8.1 4.5 4.4 0.2 20.5 1.7 Petroleum products -2.4 11.0 0.3 6.4 0.2 1.5 1.8 Construction materials 0.6 0.2 -0.3 0.5 l.l 2.4 1.9 Others 3.3 9.5 4.8 3.2 0.6 7.4 2. Financial institutions/; 7.2 3.2 3.1 16.4 6.9 4.4 ..... 24.0 20.4 3.1 ~ '4 3. Trade 17.2 28.5 29.3 4. Construction 10.4 3.1 4.5 3.7 -4.3 -0.8 5. Mining & quarrying 4.8 0.9 0.6 0.4 -1.2 -12.6 6. Agriculture 0.4 0.1 0.0 0.0 0.1 0.0 7. Services 3.4 5.5 8.1 5.4 13.6 29.7 8. Investment 0.6 -0.9 0.7 7.1 16.0 11.2 9. Real estate 27.6 33.2 3.1 0.5 4.2 3. 10. Others -2.2 -0.1 1.3 3.1 -0.2 -15.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 Notes: a The flgures cover investment in non-bank sector only. b Direct investment = equity investment plus loans from related companies. Source: Bank of Thailand. by Economic Research Department. 348 Competitiveness, FDI and technological activity in East Asia of FDI. Once the banking sector essentially reached its limits for foreign participation, FDI dropped to 6 and 4 per cent in 1999 and 2000, respec tively. One of the favourite sectors for FDI in the early to mid-1990s was rea) estate, which peaked at 33 per cent of FDI in 1996, but since the prop erty bubble burst in 1996 and 1997, the sector has struggled to aUract investment. Within the manufacturing sector, the electronics industry used to attract the largest share of FDI but was overtaken by machinery and transport equipment in 1998 and 2000. The surge in machinery and transport equip ment derived mainly from the automotive industry, as many Japanese auto motive parent companies injected capital to assist thcir subsidiaries and suppliers in Thailand. The chemical industry is also challenging electronics for FDI following its large increase in 2000, taking in over 20 per cent of FDI. Sources of FDI in Thailand have traditionally been quite diversified, including Japan, the USA, Europe, Taiwan, Hong Kong, and Singapore as shown in Table 12.6. Japan had been the largest source of FDI since the late 1980s until being overtaken by the USA in 1999 and 2000. Japanese FDI dropped sharply in 1999 and 2000 as a result of the weak economic condi tion in the home economy. In 1998, Singapore ranked third behind the USA due to its major investments in Bangkok Bank and Thai Danu Bank. There was a significant increase in FDI from Europe in 1998 and 1999, especially from Dutch investors that have taken control of the Bank of Asia and invested in several power plants. It should be noted that FDI from the UK and Hong Kong rose significantly in 2000 as they respectively ranked third and fourth after the USA and Japan. Closer to home, Thailand's ASEAN neighbours remain an important source of FDI despite the regional eco nomic difficulties. Table 12.7 shows that there has been a decline in investment interest in Thailand as the total planned investment of foreign projects approved by the BO I dropped by 58 per cent from 326 billion baht in 1996 to 136 billion baht in 1999. :However, the number of approved projects increased slightly from 490 to 517 projects, reflecting the fact that foreign projects have become smaller. This was because since the economic crisis, domestic demand has shrunk resulting in oversupply in many industries. However, foreign approved projects increased suddenly in 2000 by 48 per cent over the previous year, with investment rising by 56 per cent. This was partly due to expansion investments of export-oriented projects which performed very well after the baht devaluation and partly to a change in the BOI incentive package in August 2000 which prompted many investors to apply for pro motion status before tax incentives were reduced. The number and investment capital of Japanese approved projects dropped in 1998 and 1999 as their parent companies were busy injecting FDI, technolol5.V and competitiveness in Thailand 349 capital to help existing subsidiaries rather than investing in new projects. However, Japanese investment significantly increased in 2000 especially in automotive and electronic sectors. Total planned investment from other sources, such as Singapore and Taiwan, also fell in 1998 and 1999, while 1998 was the peak year for Europe. Investment from the USA and Canada also increased remarkably in 1999. It should be noted that investment from Taiwan, Hong Kong, Singapore, Malaysia, India, the UK and Germany also expanded in 2000. As domes tic demand continued to shrink, most new investment projects tended to be export-oriented, especially in electronics and light industries. The role of FDI in exports has increased quite significantly over the years. Sibunruang (1986) estimated that foreign firms accounted for at least 25 per cent of Thailand's total manufactured exports by the end of the 1970s, and that the export propensity of foreign firms rose over time from 10 percent to 33 per cent in 1984. In a later study, based on BOI survey data covering 777 companies, Sibunruang and Brimble (1992) found that foreign firms accounted for 24 per cent of Thailand's total manufactured exports. However, this figure understated the true contribution of FDI, because 777 firms accounted for only 40 per cent of BOI promoted manufacturing firms and the sample excluded non-promoted exporters. Therefore, it was esti mated that the real contribution lay between 30-40 per cent. The 983 foreign promoted firms surveyed by the BOI in 1996 accounted for 35 per cent of Thailand's total manufactured exports. Total FDI exports could have amounted to as much as 50 per cent of the totaL Export shares of foreign firms were relatively large in high-tech industries such as fabricated metal products, machine tools, electronic products, chemicals and petro leum products. Indeed, by 2000, the electronic and automotive sectors, which were dominated by foreign firms, already accounted for 43 per cent of Thailand's total manufactured exports. Therefore, the current export share of foreign firms could be as high as 60 per cent. 3.3 FDI Impact on Skill Formation and Technology Development A recent study found that foreign firms on average utilized labour and capital 50 per cent more efficiently than Thai firms, although a group of highly productive Thai firms also performed as well as their foreign counterparts. 13 This indicates strongly the critical role that FDI can play in contributing to overall productivity (Dollar et. aL, 1998). It has also been found that foreign enterprises in Thailand are becoming more involved in innovative programmes in training and in undertaking technological act iv ities. 14 While not yet constituting a statistically significant quantitative trend, there are a number of interesting stories of such activities. Table 12.6 Netflows o.f foreign direct investment in Thailand by region/country Million US$ Region/country 1991~1995 1996 1997 1998 1999 2000 Japan 1942 524 1348 1485 488 582 USA 1401 429 781 1283 641 844 European Union (EU) 998 168 360 912 1368 404 UK 399 57 124 102 186 425 Germany 150 42 60 100 288 67 ..... v, COl France 304 30 3 279 240 6 Netherlands 149 -41 155 332 643 110 Newly industrialized countries 3250 653 877 1115 899 768 South Korea 62 25 30 73 6 -4 Taiwan 424 138 134 106 122 192 Hong Kong 1832 215 442 394 234 325 Singapore 932 275 271 542 538 254 ASEAN (less Singapore) 56 33 27 28 32 20 Other countries 1597 464 234 320 134 170 Total 9245 2271 3627 5142 3562 2447 Shares in total Regionlcountry 1991-1995 1996 1997 1998 1999 2000 Japan 21.0 23.1 37.2 28.9 13.7 23.8 USA 15.2 18.9 21.5 25.0 18.0 34.5 European Union 10.8 7.4 9.9 17.7 38.4 16.5 UK 4.3 2.5 3.4 2.0 5.2 17.4 Germany 1.6 1.8 1.6 2.0 8.1 2.7 Francc 3.3 1.3 0.1 5.4 6.7 0.2 Netherlands 1.6 -1.8 4.3 6.5 18.1 -4.5 w Newly industrialized countries 35.2 28.8 24.2 21.7 25.2 31.4 v. ....... South Korea 0.7 1.1 0.8 1.4 0.2 -0.2 Taiwan 4.6 6.1 3.7 2.1 3.4 7.8 Hong Kong 19.8 9.5 12.2 7.7 6.6 13.3 Singapore 10.1 12.1 7.5 10.5 15.1 10.4 ASEAN (less Singapore) 0.6 1.5 0.7 0.5 0.9 0.8 Other countries 17.3 20.4 6.5 6.2 3.7 -7.0 Total 100.0 100.0 100.0 100.0 100.0 100.0 Note: The ligures cover investment in non-bank sector only. Direct investment equity investment plus loans from related companies. Source: Bank of Thailand, by Economic Research Department. Table 12.7 Foreign investor interest in Thailand: EOl approvals (million baht) 1995 1996 1997 1998 1999 2000 Country No. Total No. Total No. Total No. Total No. Total No. Total Invest. Invest. Invest. Invest. Invest. Invest. Total foreign invest. 561 397168 490 326335 516 301596 485 255070 517 136060 763 212866 100% foreign invest. 136 36856 142 75109 188 36846 204 79977 264 77226 380 123231 Asia Japan 267 190569 233 143693 220 147619 158 54113 188 27042 282 107382 Asian NIEs ""-' 'v, Taiwan 91 39945 61 69135 56 11931 69 10029 86 7910 120 17632 ""' Hong Kong 12 2032 8 1675 9 1389 16 5064 25 1899 31 6241 Korea 14 42247 19 22189 20 3965 13 1836 19 981 17 1394 Singapore 36 32033 41 41798 43 59028 49 10647 52 7003 84 19910 PRC 5 196 4 889 I 45 2 69 7 560 8 367 Malaysia 23 5121 23 1730 33 4713 21 4129 27 3418 43 6096 Indonesia 3 712 3 634 3 559 2 480 5 1149 4 1300 Philippines 2 220 0 0 0 0 0 0 I 72 0 0 India 9 8658 II 8307 5 180 10 10157 6 1374 II 10166 North America USA 45 62613 46 64780 4 88366 62 18646 53 46351 73 37916 Canada 4 542 2 56 6 310 9 2631 3 26002 6 1089 Australia Australia 7 14775 6 1026 16 4733 13 2756 10 1177 21 2705 All Europe 72 53592 87 58021 95 88813 123 134326 83 34007 144 31175 UK 18 6067 22 9952 24 28460 33 31380 17 3919 38 5815 Germany 12 4352 19 7775 19 9425 22 8606 12 1868 39 6394 Switzerland 7 1980 9 2630 to 898 11 1548 10 3170 10 2283 France 5 558 8 4389 9 1698 12 181 11 2829 13 1097 "'" i.:Z Belgium 5 927 7 3498 3 1720 8 948 7 858 2 316 Italy to 1235 2 38 7 935 4 783 3 106 9 425 Netherlands 9 1749 15 17476 12 4258 22 88066 18 22481 22 6381 Note: Firms with investment from more than one country are double counted. Foreign projects are those with a foreign component of 10'% or more, Source: Board of Investment. 356 Competitiveness. FDI and technological activity in East Asia build surface ships for the Royal Thai Navy in 1994, has been transferring advanced technology developed in Australia to Thailand. Local welders and metal workers have been continually trained in new methods of metalwork ing that enable them to meet the strict specifications required by the produc tion needs. ACS has also assisted with training local subcontractors by sponsoring off-site programmes and through the provision of Australian engineers and skilled technicians. Phoenix Pulp and Paper in northeast Thailand has been very proactive in subcontracting. In order to supply the two production lines, the company has utilized local villagers to produce individual plantations of eucalyptus, bamboo and kenaf plants. These villages raise enough wood and are paid nearly one billion baht a year. The company estimated that over 1000 villagers gain employment either directly or through sub contracting. Many MNCs have recently participated in the BOI's VMC programme (as a part of the BUILD initiative) by allowing potential suppliers to visit their plants and explore the possibility of supplying parts and components to them. Foreign companies have also been very active with the matchmak ing activities of the BUILD programme since its inception in the early I 990s. This resurgence of interest and initiative in BUILD is a very positive indication of greater technology transfer from foreign firms in the future. The benefits of foreign firms go beyond training and technology devel opment. One additional impact they have had in Thailand is to promote higher standards of employment practices such as safety and sexual dis crimination. Foreign firms have also played an instrumental role in address ing issues such as AIDS in the workplace. 17 3.4 A Summing Up FDI has played a significant role in Thailand's economic development, especially since the late 1980s as the country liberalized and sought expanded markets overseas. The onset of the economic crisis actually gen erated record levels of FDI in Thailand, with M&A being the preferred mechanism for investment. The current patterns of FDI show a continuing strong interest in the manufacturing sector, and it does not appear that this will change in the near future as long as Thailand's economy remains based heavily on exporting manufactured, labour-intensive goods. The surge of FDI in the financial sector in 1998 probably represents an anomaly in response to the change in regulations for foreign participation, and other sectors such as trade and services have failed to show consistent gains over the past few years. FDI has made important contributions to the Thai economy beyond FDf, technology and competitiveness in Thailand 357 generating employment. It has also saved countless jobs during the crisis by helping to capitalize failing local industries. Other less evident benefits include bringing in new technology and industries to spur competitiveness, improving corporate governance and standards for working conditions, strengthening local capabilities through linkages, and assisting with policy rcforms and industrial restructuring. 4. SCIENCE AND TECHNOLOGY DEVELOPMENT: POLICIES AND TRENDS 4.1 The Policy Approach Attention to science was paid in Thailand more than 100 years ago when the Department of Science Service was established. But it was not until 1956 that the government established the National Research Council of Thailand (NRCT) to encourage R&D to systematically upgrade the country's scientific and technological capabilities. NRCT sets research policy and provides research funding, albeit limited, to a wide range of dis ciplines including science, engineering, social science and humanities. To conduct and support research, the Applied Scientific Research Cor poration of Thailand, now known as the Thailand Institute of Scientific and Technological Research (TISTR), was created in 1963. Its broad objectives are to implement research in applied sciences and to promote the utilization of natural resources for industrial and national develop ment. An important milestone in the development of science and technology policies was the establishment of the Ministry of Science, Technology, and Energy (MOSTE) in 1979 to play the central role of coordinating science and technology policy and planning at the national level by bringing six key departments belonging to the Office of the Prime Minister and the Ministry of Industry under its wing. They are NRCT, TISTR, the National Energy Board, the National Environment Board, the Office of Atomic Energy for Peace and the Department of Science Service. In 1992, MOSTE was renamed the Ministry of Science, Technology, and Environment with the restructuring of the National Energy Board and the National Environment Board, and the addition of two new departments on environment issues. The Fifth National Economic and Social Development Plan (1982~86) was the first plan with a chapter exclusively devoted to science and tech nology. The key policies were to screen and adapt imported technology to local conditions and to develop Thailand's own technology. The country's scientific and technological base was to be strengthened by emphasizing Table 12.8 Rankings 0/ selected technology indicators, 2000 Indicator Unit Thailand Singapore Malaysia Philippines 1. Technological sophistication Bus. Per. 3.4 5.9 4.2 2.9 2. Average years of schooling No. of years 6.10 8.09 7.88 7.62 3. Secondary school net enrolment Percentage 56 90 59 77 4. Maths and science education Bus. Per. 3.9 6.5 4.6 3.9 5. Approach to human resources Bus. Per. 3.5 5.5 4.5 4.0 6. Management education Bus. Per. 4.6 5.8 4.9 5.9 7. Research institutions Bus. Per. 2.7 5.1 3.5 3.0 8. R&D spending °AlofGDP 0.13 1.13 0.24 0.22 9. Private sector spending on R&D Bus. Per. 2.5 4.1 2.8 2.6 10. Research collaboration Bus. Per. 3.6 4.2 2.5 2.8 11. Technology development Bus. Per. 2.8 4.2 3.2 2.9 t..,; C\ c 12. Licensing of technology Bus. Per. 5.2 5.5 5.2 5.2 Notes: Bus. Per.: Business perceptions on a scale of 1 (strongly disagree) to 7 (strongly agree with the lollowing statements). J. Your country'S position in technology ranks among the world leaders. 2. Average years of schooling by population age 25 and up. :I. Secondary education enrolment indicator, 1997. 4. The school system excels in maths and basic science education. 5. Companies invest heavily to attract, develop. motivate, and retain staff. 6. Management education is locally available in first-class business schools. 7. Scientific research institutions are truly world-class 8. Research and development spending in percent of GNP, latest available year. 9. Companies invest heavily in research and development relative to their international peers. 10. Companies collaborate c1ose1y with loca1 universities in their research and development activities. 11. Companies obtain technology pioneering their own new products or processes. 12. Licensing of foreign technology a common means to acquire new Source: WEF, The Global Competitiveness Report 2000. FDI, technology and competitiveness in Thailand 361 Unsurprisingly, Singapore is well ahead of Thailand in every category. However, Thailand is also ranked behind Malaysia in nearly every category, and in several items fails to keep pace with the Philippines. Both have much smaller economies than Thailand. Some interesting features of Table 12.8 are: · Thailand comes well behind Singapore, Malaysia and the Philippines in most areas of basic education. Particularly worrying is the net sec ondary school enrolment rate indicating that nearly half of Thailand's children are not enrolled in secondary school. · The perceived emphasis on staff training in Thailand falls one full point below that of Malaysia and two points behind Singapore. · The stronger overall scores of the Philippines in these key HRD areas likely explain why it has been able to maintain its competitiveness position, despite other problems. · Private sector spending on R&D is the lowest of the four countries. Thailand also lags far behind its neighbours in key information technology indicators, as shown in Table 12.9. Some of the shortcomings of Thailand here are dearly a result of the language barrier. Singapore, the Philippines and Malaysia have a distinct advantage over Thailand because of the extensive use of English, with the former two countries using it as the national language of business and administration. Thailand spends, according to Table 12.8, a mere 0.13 per cent of its GDP on R&D. A recent study on Thailand's innovation and R&D suggests a higher figure, 0.29 per cent. 20 According to the latter survey, of the total of US$358 million spent on R&D in 1999, the manufacturing sector accounted for US$147 million, non-manufacturing firms and SMEs for US$26 million and the public sector for US$185 million. Even this more optimistic estimate, however, falls far short of the target of 0.75 per cent of GDP of the Eighth National Economic and Social Development Plan. A detailed comparison of R&D in Asia (Table 12.10) shows that Thailand's R&D spending per capita of US$5.8 is only ahead of India, Indonesia and the Philippines, but is less than half that of Malaysia. While Thailand's total R&D exceeds Malaysia's, the amount spent by the private sector on R&D is well below. The gap is even more pronounced between Thailand and Malaysia in terms of private sector R&D per capita: Thailand's US$2.8 is approxi mately one-third of Malaysia's US$8.57. Thailand also fares poorly in its ratio of business R&D in total R&D. Its ratio of 48.3 per cent is higher than only the Philippines and India. In terms of R&D personnel, despite Thailand's population of 62 million 364 C()mpetitivenes.\~ FDl and technological activity in East Asia and Malaysia's 22.7 million, the former has only 5300 R&D workers in manufacturing business enterprises for a per capita ratio of 0.086 (full time equivalent per I 000 people). 21 Malaysia by comparison employs 3500 R&D workers in the private sector for a ratio of 0.16 (full time equivalent per I000 people). 4.3 Human Resource Development Thailand lags behind many of its regional competitors in human resource development, raising serious concerns about its ability to sustain competi tive growth in the future. The English language issue is a structural handi cap, but this could have been compensated by investments in education (after alL Korea and Taiwan also suffered this handicap). In early stages of economic development, Thailand concentrated on developing lower levels of education and relied on foreign expertise for higher education and research. Before the economic crisis, the education and skill levels of the population lagged behind many neighbouring countries. In the mid-1990s, only 40 per cent of workers had completed secondary or post-secondary education. A major factor in Thailand's lagging human resources is the insufficient number and quality of S&T students, particularly at the postgraduate level. Thailand has a paucity of scientists and engineers who can perform high quality R&D. It had only 119 engineers and scientists per million popula tion before the economic crisis, compared to more than 2500 each in Korea and Singapore and 350 in China (UNESCO, 1997). Private firms in Thailand are forced to rely heavily on foreign skilled labour, managers, sci entists, and engineers. Thailand's higher education system is partly to blame. Thai universities, like those of most other East Asian nations, were established primarily as teaching institutions and research was considered secondary. In addition, the archaic university system and the government's meagre funding for R&D dissuade academics from conducting R&D. This makes it very difficult for students to acquire the skills and attitudes needed for R&D. Thai universities also suffer from staffing problems. In 1997, around 55 per cent of academic staff in public universities had a masters degree, while 27 per cent held doctorates. Breakdowns of staff qualifications for science and engineering were not available, but there are strong indications that staff-student ratios in S&T university faculties have been deteriorating as academic staff leave for the private sector while student enrolments con tinue to rise. This trend may well be exacerbated by the recent moves to support early retirement of public university lecturers. A major problem is the wide disparity between academic salaries and salaries in the private FDI, technology and competitiveness in Thailand 365 sector. In areas like telecommunications the gap can be as wide as five times or more. The disparity in basic wages is too wide to be covered by the addi tional income academics might earn from private consulting. The univer sity wage structure is very inflexible. High-level S&T skills can be acquired abroad, but very heavy reliance on foreign training is undesirable. It is more expensive, may have less direct application to Thai industry and raises the risk of a brain drain. One of the major initiatives to improve the quality and quantity of S&T in Thai universities is the Higher Education Development Project financed by the Asian Development Bank (ADB). This project intends to raise S&T human resources by funding Thai university training at the postgraduate level. Various consortia have been formed among Thai universities and in partnership with industry to develop postgraduate S&T projects in seven fields: chemistry; environmental science, technology, and management; environmental and hazardous waste management; energy and environmen tal technology; petroleum and petrochemical technology; agricultural biotechnology; and post harvest technology. The project is expected to produce S&T postgraduates to meet the public and private sectors' labour demands, stimulate research and technological development, and facilitate Thailand's structural shift up the technology ladder in manufacturing and services. 4.4 Technological Development at the Firm Level Thai companies have focused on acquiring strong manufacturing capabil ities through imports of foreign technology, with little attention being paid to engineering, design, development and research. Yet, as noted, these capabilities will be essential for Thailand's future growth as lower-cost com petitors emerge and the world market becomes more competitive. Most studies of the technological capabilities in Thai firms indicate that growth has been based on low-technology activities with little attention paid to strengthening long-run technological capacities and technological learning (Intarakumnerd et aI., 2002). A recent study of technology development supported by the World Bank (Arnold et aI., 2000) uses the framework pre sented in Figure 12.2 to analyse technological activity among different types of Thai firms. Most large MNC subsidiaries, some large domestic firms and a few SMEs are generally able to acquire and assimilate technologies reasonably well, and are therefore on the threshold of technology upgrading and reverse engineering. Relatively few such firms have actually gone further into research and technology development. 22 The situation for the remain ing large firms and SMEs is more worrying. In many of these firms, even 368 Competitiveness, FDI and technological activity in East Asia 5. CONCLUSIONS AND RECOMMENDATIONS 5.1 Main Conclusions (a) Thai policymakers have generally not been aware of the importance of technology and the need to devise policies supportive of R&D and S&T. Government expenditures in these areas have been insufficient, as have policies to encourage private R&D. The coordination of tech nology support policies has been weak, especially in linking such pol icies to broader economic policies and institutions. (b) Thai industrial firms have not recognized the importance of investing in R&D and human resources. Many have diversified into peripheral businesses rather than consolidate their position in regional or global markets. They generally lack a strategic vision of moving up the value chain, and do not engage in collective action (through business asso ciations and groupings) to undertake technological activity. (c) The education and training system is not geared to providing the human resources needed for technology upgrading and innovation. Despite rising enrolment rates, Thailand's educational system, partic ularly the higher education system, remains weak. The resources pro vided are inadequate, curricula are outdated and teachers poorly paid and under-trained. Technical subjects receive little emphasis. At the firm level, employee training is low, partly because of employers' atti tudes but also because of the lack of suitable institutions to provide the necessary training. (d) The full potential of FDI has not been realized. Despite the billions of dollars of inflows, little attention has been given to raising linkages between foreign and domestic firms, inducing affiliates to deepen tech nological activity and parents to transfer innovative activities and skills. 5.2 How to Kill some Demons among Policymakers There are some features of Thai science and technology policymaking that have to be addressed: · There is insufficient realization that industrial firms are not only users of S&T services but also the major generators of technology firms are in fact at the centre of the national innovation system. Evidence from developed countries shows that up to 80 per cent of technolog ical effort is expended by firms. The perception in Thailand remains that the government's role is to support technological activities through public sector research institutions. This leads the govern- FDI, technology and competitiveness in Thailand 369 ment to neglect the role of measures to stimulate firm-level techno logical effort. · Policymakers must understand that the main challenge today is to build knowledge, not just buildings and machines. Most incentives in place are aimed at capital investments of one kind or another. While there has been much talk of the 'knowledge economy' and the value of information, policies to support the acquisition, utilization and development of such assets have yet to be developed and imple mented. · The prevailing perception is that MNC strategies are completely determined by the head office, and that there is little is to be gained by closer collaboration with local affiliates. In fact, MNCs often give autonomy to their affiliates to devote resources to a range of techno logical activities and increasingly regard this as crucial to their global competitiveness. This trend is also evident in Thailand and needs to be exploited. · A common perception in Thailand is that large or foreign firms do not require assistance from the government. International evidence suggests that the government can use incentives to encourage large and progressive firms to raise their technological activity and link up more closely to domestic suppliers and institutions. · There is insufficient appreciation of the significance of clusters, net works, partnerships and the like in enhancing competitiveness. 5.3 An Agenda for .'DI Policies Promoting linkages and spillovers Efforts to enhance spillover benefits from FDI should be an intrinsic part of government strategies to enhance competitiveness and restructure industry. There is a strong case for govern ment interventions because of the widespread externalities and informa tion problems involved in building local linkages (UNCTAD World Investment Report, 2001). Such programmes are weak or absent in Thailand. It is recommended that: · The government set up mechanisms to interact with MNCs in formu lating policies and getting feedback on the impact and implementa tion of policies. The recent activities of the foreign business community to meet regularly with key government agencies have been fruitful and should be intensified. 26 · The government should support local firms through vendor develop ment programmes and other activities such as cluster and supply chain development. UNCTAD's World Investment Report, 2001 372 Competitiveness, FDI and technological activily in East Asia NOTES 1. This paper was prepared by Peter Brimble of The Brooker Group pic, as an input into a World Bank project on Competitiveness, FDI and Technological Activity in East Asia. Substantive inputs were received from A tchaka Sibunruang on Section 3 and from David Oldfield on parts of Section 2 and 4. The research in this paper is based mainly on sec ondary sources, with a few selected interviews and industry meetings. 2. Brimble et al. (1999) present a comprehensive overview of these issues. 3. As documented in the World Bank Atlas of 1996. 4, See Development Evaluation Division (1998). 5. A constellation of factors led to the rapid decline in exports in addition to declining com petitiveness, including a slowdown in world trade, the emergence of China in global markets, EU restraints on certain Thai exports, and fluctuations in global electronic markets. 6. See Siamwalla (1997). 7, In this context, the increasing anti-foreign feelings that have characterized the policy approach in the first half of 2001 are a cause for concern, especially if they eventually extend beyond simple rhetoric to actual measures to roll back the critical reforms and to introduce anti-foreign regulations into the business environment. 8. See Brimble et al. (1993) for an analysis of the role of the private sector in the infrastruc ture sector. Annex I-E summarizes the private sector initiatives of the Seventh Plan, 9. This comparison is developed in more depth in Arnold et at. (2000). 10. Indeed, a committee has been established to review the law, and preliminary indications are that a more restrictive definition of foreign ownership will be recommended as an amendment to the law. The time frame for the consideration and possible implementa tion of this are uncertain. ) 1. It should be noted that the Bank of Thailand did not include foreign capital inflows for banking capitalization in FDI statistics. The figure was about US$2 billion in 1998, which is when most of the capital injections into the banking sector occurred. 12. See the World Investment Report 1992, published by UNCTAD, for a more detailed anal ysis of the productivity of foreign firms versus that of domestic firms in developing coun tries. 13. These results were calculated from firm-level data of more than 1000 tirms from which a simple production function was estimated. More details are provided in Dollar et al. (1998). 14. See Brimble et aL (1999) for additional examples, More such case studies are presently being compiled by The Brooker Group for a number of research projects related to com petitiveness. 15. Storage Technology essentially includes all elements of the technology that is embodied in the products of the HDD industry. 16. See UNCTAD World Investment Report (2001) for a detailed profile of the TMT sup plier initiative. 17. An innovative position paper presented in Joint Foreign Chambers of Commerce in Thailand (1FCCT) (2001) contains additional examples of the beneficial impacts of MNCs and identifies. from the perspective of the MNCs themselves, ways and means of enhancing these impacts. 18. Arnold et al. (2000) consider these issues in some depth. 19, It deserves to be emphasized that the figures in Table 12.8 that are derived from business perceptions must be treated with caution, However, they are included here as they rep resent the perceptions of business and are often used by foreign investors when making investment decisions, 20. See the draft final report of the NSTDA R&D/Innovation Survey in Brooker Group (200 I), 21 See the draft final report of the NSTDA R&D/Innovation Survey in Brooker Group (2001). Since the survey only covered the manufacturing sector, no results are available for the entire business sector. FDI, technology and competitiveness in Thailand 373 22. A recent survey of R&D in Thailand confirms this, finding that only around IS per cent of medium to large manufacturing firms carried out some form of R&D (Brooker Group, 2001). 23. '. . given Thailand's present level of economic development and its desire to continue rapid growth based on greater internationalization and competitiveness ... The country needs to strengthen considerably its technological capability in industry. While most firms in the modern sector have reasonably adequate capability to operate their existing technology, they are weak in searching for, acquiring and adapting foreign technology. They are even weaker in developing their own technology. Local R&D e/l'orts are minimal in the private sector .. .' (Dahlman and Brimble, 1990, p. 41). 24. Brooker Group interviews with private firms. 25. Arnold et al. (2000). 26. See Joint Foreign Chambers of Commerce in Thailand (JFCCT) (2001) for more details on the content and nature of these meetings and rclatcd activities of the foreign business community. 27. See Brooker Group (2001). The proposal has been widely accepted, yet not acted on to date. BIBLIOGRAPHY ADB (2000), Key Indicators ofDeveloping Asian and Pacific Countries 2000, Volume XXXI, Manilla: Asian Development Bank. Arnold, Erik, Martin Bell, John Bessant and Peter Brimble (2000), 'Enhancing Policy and Institutional Support for Industrial Technology Development in Thailand The Overall Policy Framework and The Development of the Industrial Innovation System', paper prepared for the National Science and Technology Development Board with support from the World Bank, Bangkok, December 2000. Brimble, Peter (1999), Building Partnerships for Better Development: Outlook for Partnerships in Thai/and, Conference Proceedings of The 1st International Outlook Conference on Community Development in Asia-Pacific, Bangkok, Thailand. Brimble, P., S. Greer and A. Brimble (1993), The private sector in Thailand: Towards enhancing ADB's private sector strategy', paper prepared for the Asian Development Bank, Bangkok. Brimble, P., J. Sherman, A. Sibunruang and W. Rachatatanun (1999), The Broader Impacts of Foreign Direct Investment on Economic Development in Thailand: Corporate Responses', paper prepared for the High-Level Roundtable on Foreign Direct Investment and its Impact on Poverty Alleviation, Singapore, 14-15 December 1998; Revised April 1999. Brimble, Peter and Chatri Sripaipan (1994), 'Science and Technology Issues in Thailand's Industrial Structure: The Key to the Future', prepared for the Asian Development Bank, June. Brimble, P" C. Sripaipan, S. Vanichseni, and Y Mukdapitak (1997), Towards a Technological Innovation Strategy for Thailand', paper prepared for the First International Conference on Technology Policy and Innovation, Macau, 2-4 JUly. Brooker Group (2001), NSTDA R&DlInnoJ!ation Survey, Draft Final Report, July. Dahlman, Carl and Peter Brimble (1990), 'Technology Strategy and Policy for International Competitiveness: A Case Study of Thailand', Industry and Energy 376 Competitiveness, FDI and technological activity in East Asia development and the policies that supported this. Section 4 discusses the lessons from Indonesia's technological development. 2. INDICATORS OF INDONESIAN TECHNOLOGICAL DEVELOPMENT Technological development cannot be captured in a single measure; it is necessary to use several complementary indicators to get a clear picture (Hill and Wie, 1998: 13). It can be evaluated from 'inputs' ~ R&D, educa tion or 'outputs' export competitiveness or productivity.2 Let us start with one output indicator: exports. Figure 13.1 shows total and manufac tured exports in Indonesia for 1980~98. Total exports grew very rapidly, at around 12 per cent per year; the performance of manufactured exports, at 23 per cent per year, was even more impressive. As a result, the share of manufactured products in total exports increased from 4 per cent in 1980 to 56 per cent by 1997. At the sectoral leveL most industries enjoyed high export growth (Table 13.1); with the exception of wood and steel, other industries had annual growth rates of over 20 per cent in dollar terms. Moreover. although resource-based and labour-intensive industries (such as food, garments, footwear, and toys) still account for over half of total manufactured exports, Indonesia did diversify its export base significantly. For instance, exports of textiles and electronics increased rapidly, their shares in manu factured exports rising from 10.8 and 3.2 per cent, respectively, in 1990 to 14.5 and 11.2 per cent by 1996. However, despite this strong performance, Indonesian exports remain dominated by resource-based and labour-intensive products. In compari son to other countries in the region, its export structure is relatively unso 60000,-------------------------------------~ .§ 50000 - 40 000 Total exports ·s 30000 ~ 20000 ;:l 10 000 OL-~========~~------------------~ 0 N 00 M 0'> - V) 00 \0 r- oo 0'> 0 0'> 00 0'> 00 0'> N "" M 00 0'> V) \0 r- oo 00 0'> 00 0'> 00 0'> - 00- 0'> 00 0'> ,..... 0'> 0'> ..... 0'> 0'> 0'> ..... 0'> 0'> 0'> "" 0'> 0'> 0'> 0'> 0'> 0'> 0'> 0'> - 0'> 0'> Year Source: Asian Development Bank (1999). Figure 13.1 Total and manufactured exports in Indonesia, I 980~98 Technology development in Indonesia 377 phisticated (Table 13.2), even relative to other industrial latecomers like Malaysia and Thailand. For instance, about 50 per cent of Thailand's export is categorized as being technologically complex, whereas the corre sponding figure for Indonesia is around 23 per cent. It should be stressed that export figures are an imperfect measure of tech nology. For instance, classifying industries at different levels of technolog ical intensity may be misleading: some low-technology products may use relatively sophisticated technological processes; and conversely, apparently high-technology activity may consist of simple assembly of imported com ponents. In Indonesia, the latter is illustrated by electronics. The share of high-technology electronics products in manufactured exports was above II per cent in 1996, while its share of manufacturing value-added is below 4 per cent. This suggests that the industry consists largely of assembling imported inputs rather than making complex components. Of course, it should be expected that a country like Indonesia, with abundant labour and plentiful natural resources, would specialize in products that intensively use these factors rather than in high-technology products. At the same time, maintaining dynamic export growth does require a steady move up the technology ladder, and other resource-rich countries, such as Malaysia, have supplemented healthy growth of resource based products with far more dynamic high-technology ones. Other measures of technological performance seem to confirm Indonesia's technological lags. Take some 'input' measures. Table 13.3 shows R&D expenditure as a percentage of GNP in some Asian countries. Indonesia allocated only 0.07 per cent of GNP to R&D in 1994, substan tially less than any other country in the region, even those with a similar or lower GNP per capita (Philippines and India). Similarly, Indonesia has very low patent applications between 1981 and 1990: 12 compared to Korea's 6629. Malaysia's 406, the Philippines' 141, Singapore's 812, and Thailand's 144.3 The stock of skilled professionals shows, indirectly, the ability of a country to undertake technological effort. Indonesia possesses one of the smallest pools of skilled workers in the region, attributable to the low pro portion of Indonesians enrolled in, or graduating from, science and tech nology courses at tertiary level (Hill, 1995: 94). International comparisons of measures such as R&D and education may, however, underestimate technological efforts in Indonesia. They do not capture the kinds of informal technological activity that predominate in early stages of development, when most new technology is likely to be imported rather than created locally (Evenson and Westphal, 1995: 2249). The following discussion offers a more complete picture of technology development in Indonesia. Table 13.1 Average annual growth rates of manufactured exports 1990-96 and sectors' shares of total manufacturing export (%) Sector International Export growth Sector's share of total export Standard Industrial Classification 1990 1996 Resource-intensive Food, beverage, tobacco 31 26.6 11.5 11.7 Wood and furniture 33 18.4 31.3 19.3 w Coal and oil refineries 353,354 nla 0.0 0.0 0; Rubber 355 20.9 1l.9 8.5 Non-metallic minerals 36 21.6 2.0 1.5 Labour-intensive Garment 322 25.0 7.7 7.1 Leather and footwear 323,324 40.8 3.2 7.6 Toys, etc. 39 43.0 0.4 1.2 Scale-intensive Textile 321 31.3 10.8 14.5 Paper and printing 34 27.0 2.3 2.4 Chemicalsa 351,352 21.8 4.7 3.6 Steel and non-ferrous met. 37 9.2 7.1 2.5 Transport 384 57.9 0.5 3.6 Differentiated Metal products 381 29.3 1.7 2.1 Machinery 382 56.3 0.1 0.6 Science-based Pharmaceuticals" 33.1 0.2 0.3 Plastics 356 32.6 1.2 1.7 Electronics 383 47.4 3.2 11.2 Precision 385 68.6 0.0 0.5 ....... ~ Total 26.4 100 ]00 Notes: Exports calculated in US dollar terms. a except 35221, 35222, 35223. 35224. b = 35221,35222.35223 and 35224. Source: Calculation based on data from the Indonesian Central Bureau of Statistics. 380 Competitiveness, FDI and technological activity in East Asia Table 13.2 Distribution of manufactured exports by technological categories, 1995 Country Technologically complex High-technology Korea 65,2 44.1 Taiwan 59.9 49.3 Singapore 79.2 73.7 Indonesia 23.1 14.8 Malaysia 78.7 72.8 Thailand 53.6 45.9 Notes: 'Technologically complex' includes three categories: scale-intensive, differentiated, and science-based products. 'High-technology' includes two categories: differentiated and science-based products, Source, LaB (1998: 142-3), Table 13.3 R&D expenditures as percentage of GNP in Indonesia and other selected Asian countries Country Year R&D expenditure as % of GNP Indonesia a 1994 0.1 India 1994 0.7 Korea 1994 2.6 Malaysia 1994 0.4 Philippines 1992 0.2 Singaporeh 1995 1.1 Thailand 1995 0.1 Notes: a Figures refer to the prod uctive sector only, h Not including R&D in social sciences and humanities. Source: UNESCO (1999). Technology development in Indonesia 381 3. SOURCES OF TECHNOLOGICAL CAPABILITY AND POLICY ISSUES 3.1 Sources of Technological Capability There are two (complementary) ways for a country to improve its techno logical capability. The first is to undertake domestic innovation, by build ing a strong R&D capability. The second is to import technology from other countries by inward FDI, licensing, subcontracting and original equipment manufacturing (OEM) arrangements. Both have been used in Indonesia. Domestic R&D R&D is, as noted, relatively low in Indonesia throughout the manufactur ing sector (Table 13.4). Even highly R&D-intensive industries internation ally spend little on R&D in Indonesia (R&D is only 0.24 per cent of sales in pharmaceuticals). However, some Indonesian firms have started to conduct R&D, with the share of establishments engaged in R&D higher in scale-intensive, differentiated and high-technology industries. Over time, this should increase as industry moves into more complex technologies where local R&D is essential even to absorb and adapt new technologies. The public infrastructure supporting industrial R&D is weak (Hill and Wie, 1998; Lall, 1998; Thee, 1998). There are 12 national public R&D insti tutes and several regional R&D centres. According to Hill and Wie (1998: 50), most institutes face two problems. The first is a weak funding base, as a result of which they have difficulties in hiring qualified staff. The second is the lack of ties with the private sector: R&D programmes tend to be supply rather than demand driven, carrying out programmes devised by their managers and politicians rather than in response to industrial needs. ror example, linkages between public R&D institutes and the successful export-oriented textile, garment and electronics firms have been very weak (Thee, 1998: 125). Many public R&D institutions have been involved in projects like aircraft manufacturing in Bandung, a strategic initiative of the government rather than a commercial venture. :Foreign direct investment Indonesia has traditionally relied on capital inflows through external borrow ing rather than FDI. One reason has been the widespread suspicion in Indonesia of foreign involvement in general and of FDI in particular. However, falling oil prices in the early 1980s led to a liberalization of the FDI regime, and this continued throughout the 1980s and 90s. These policy changes led to a large influx of FDI and the number of foreign establishments Table 13.4 R&D intensity hy industry, 1995 Group Sector International R&D as a share Share of establishments Standard of output ('Yo) engaged in R&D (%) Industrial Classification Resource-intensive Food, beverage, tobacco 31 0.02 6.8 Wood and furniture 33 0.03 5.4 Coal and oil refineries 353, 354 0.15 13.5 Rubber 355 0.05 14.4 Non-metallic minerals 36 0.02 6. Labour-intensive Garment 322 0.01 3.0 Leather and footwear 323, 324 0.03 9.6 Toys, etc. 39 0.04 7.4 Scale-intensive Textile 321 0.02 5.5 '-'.; Paper and printing 34 O.OS 7.0 ~ Chemicals" 351,352 0.08 17.4 Steel and non-ferrous met. 37 0.02 13.3 Transport 384 0.05 12.4 Differentiated Metal products 381 0.04 7.7 Machinery 382 0.05 12.5 Science-based Pharmaceuticalsb 0.24 31.1 Plastics 356 0.0 I 4.6 Electronics 383 0.09 13.8 Precision 385 0.02 8.1 Notes: " except 35221, 35222,35223,35224. h 35221, 35222, 35223 and 35224. Source: See Table 13.1. Technology development in Indonesia 383 increased by 80 per cent between 1980 and 1990 (Table 13.5). However, the foreign share of value-added and employment decreased for some time because of the rapid expansion of domestic enterprises, perhaps because they were more aware than foreign investors of policy changes and better able to take advantage of them. After 1990, however, FOJ rose even faster, with the number of foreign-owned establishments increasing by more than 120 per cent between 1990 and 1996. The share of foreign enterprises in manufactur ing employment rose from to to 16 per cent and in manufacturing value added from 22 to 30 per cent over 1990-96. Table 13.5 shows that the foreign share of value-added exceeds the foreign share of employment: value-added per employee is higher in foreign than in domestic establishments. Part of this higher labour productivity can be explained by the larger size and greater capital-intensity of foreign affil iates. Part is likely to be explained by the use of more advanced technolo gies by foreign-owned firms. Okamoto and Sj6holm (2000a) find that the pace of technological change, as measured by TFP growth, is relatively high in foreign establishments. In fact, foreign establishments, despite account ing for relatively small shares of sector output (Table 13.5), generate the bulk of TFP growth in many sectors. One plausible explanation for this is that foreign affiliates have access to superior technology from parent firms. Foreign firms have played a vital role in Indonesia's recent manufactured export growth. Exporting is a difficult process, requiring detailed knowl edge of foreign markets, distribution systems, regulations and other char acteristics; multinationals have a clear advantage in collecting such information. It also needs specialized skills, technologies and capabilities that MNCs generally possess to a greater extent than local firms. It is not surprising, therefore, that foreign firms in Indonesia export more in relation to output than domestic firms (Ramstetter, 1999). Moreover, even the foreign firms that start by producing only for the Indonesian market tend to be relatively flexible and seek out foreign markets when this is profitable (Sj6holm, 1999a). Figure 13.2 shows the correlation between inward FOI flows and manufactured exports. While MNCs have certainly introduced new technologies and manufac turing capabilities to Indonesia, their impact on domestic firms is more ambiguous. There are likely to be both positive and negative effects on domestic firms from the entry of FDL Some domestic firms face decreases in their market shares and may even be forced out of business; in contrast, firms supplying foreign affiliates may enjoy larger markets. Some firms may gain access to the new technologies brought in by MNCs and so raise their productivity and competitiveness. Others may respond to increased compe tition by investing in new technologies. There may be other spillover bene fits from MNCs from labour turnover or demonstration effects. 386 Competitil'eness, FDI and technological activity in East Asia the wider dissemination of technology. Hence, FOI attraction should be complemented by policies to increase the degree of competition and raise the level of local skills and capabilities. Other external private sources: the case of local textile companies Formal technology transfers through FOI and licensing agreements are clearly important in acquiring and mastering new technology. However, a case study of some local Indonesian firms reveals that these mechanisms are only part of technology inflows from abroad. Table 13.6 shows various sources of technology in seven major Indonesian textile firms.4 All these firms are successful exporters. Two have licensing agreements with Japanese textile companies, buying proprietary technologies from Japanese compa nies in return for royalty fees. However. this formal channel is only one source of technological acquisition. Another source is imported capital equipment. All companies surveyed purchased capital equipment from abroad, mainly from Japan. As Evenson and Westphal (1995: 2264) note, imports of capital goods played an impor tant role in the industrial development of many successful export-led economies. Indonesian trade statistics show that eq uipment imports are rising in importance. As Figure 13.3 shows, the total value of Indonesian imported capital equipment has increased sharply since the mid-l 980s. Japanese trading companies often played an important role in facilitating such imports, for instance by providing loans. In many cases, trading com panies also provided valuable marketing know-how and information on foreign markets. Foreign technical experts have also been an important source of techno logical upgrading. All the surveyed companies used foreign, mainly Japanese, experts. Many of these experts were financed by Japanese aid. It is important to note, however, that tied aid was not the main reason for hiring Japanese experts; these experts already have long-standing and extensive contacts with Indonesian firms, and employment contracts have been generally made through private contacts. Korean and Taiwanese experts may be more important in the future, as the number of Japanese experts is falling with the decline of Japan's textile industry. Japanese and Korean affiliates in Indonesia provide technical assistance in the textile industry through inter-company linkages. Almost all the sur veyed textile exporters purchase synthetic fibres mainly from foreign affili ates, which provide technical assistance as part of the business arrangement. This case study shows the importance of various sources of technologi cal capability in addition to explicit channels such as FOI and licensing. The importance of such sources cannot often be inferred from conven tional data on technology purchases or innovation. Our findings are Table 13.6 Sources of technological capability in seven Indonesian textilefirms Company Founded No. of Ellport markets Licensing Imported capital Foreign trading Foreign Technical assistance employees agreements equipment companies and technical through inter- buyers experts company linkages A nla 1200 Japan None Made in Japan JTC Japanese Japanese and Korean Indian merchants Korean affiliates in Indonesia Non-pribumi firms (Indian) B 1968 1552 Malaysia, Brunei, Yes nla nla Japanese Mainly a Canada, Australia, From 1985-prescnt affiliate in Others (Japan) C 1976 2400 Middle East (55'%) None Made in Japan, Indian merchants Japancse Japanese and Korean Europe (20r:!o) purchased in 1989 affiliates in Indonesia Others through JTC Non-pribumi firms (Indian) v., Q:) D 1962 1600 Europe None Made in Japan, Chinese buycrs Japanese Japanese and Korean '..j Middle East purchased between affiliates in Indonesia 1986-96 through JTC E 1976 1500 Europe (60%) None Made in Japan JTC Taiwanese Mainly a Japanese Middle East (10%) Japanese affiliate in Indonesia USA (5 Others F 1969 1230 Middle East None Made in Japan nla Japanese Japanese affiliates Europe Made in Korea in Indonesia G 1970 2500 Asia (30°ltl) Yes Yes JTC Japanese Mainly a Japanese Europe (30')!o) ( 1994-present) KTC Taiwanese atliliate in Indonesia Middle East (20'Vo) (Japan) TTC Korean Others Note: JTC Japanese trading company; KTC = Korean trading company; TIC ~ Taiwanese trading company. Source' Based on research material providt.'d by Mr Abc, Institute of Developing Economies, Japan. 388 Competitiveness, FDI and technological activity in East Asia 20000~------------------------------------~ 18000 16000 ~ 14000 '8- 12000 10000 8000 6000 4000 2000 O~----------------------~----~-----------~ Year Source. Asian Development Bank (1999). Figure 13.3 Machinery imports in Indonesia, 1981-97 consistent with another study on Indonesian textiles (Pangestu, 1997), according to which the main sources of technological capability are imports. Imports of equipment and provision of support services by foreign firms are some of the observed channels of technology transfer. Other external private sources: the case of local SMEs in three industries What about technology development in small enterprises? We draw upon a survey of small and medium-sized enterprises (SMEs) by Berry and Levy (1994), who studied successful Indonesian exporters in the garment, rattan furniture and carved wooden furniture industries. The findings are similar to those for large textile companies, with private external channels being pivotal in acquiring technology. Foreign buyers played a dominant role in establishing and sustaining export market linkages for SMEs. Local firms and foreign buyers had a strong mutual interest in improving the quality of the product; the buyers consequently provided designs and technical assis tance to producers. In addition, expatriate employees were critical to acquiring new technical know-how. In particular, experts from Hong Kong and Korea facilitated the introduction of advanced technical know-how. SMEs have also been supported in technology development by industry association and public agencies. This is particularly the case for pribumi (indigenous Indonesian) firms, which lack access to international networks. In general, however, private channels tend to be the most important sources of technology in all industries surveyed. Technology development in Indonesia 389 3.2 Public policy issues Studies on technology development stress the importance of incentives, education and training, well-developed technical infrastructure and science and technology (S&T) policies. 5 This subsection evaluates policy and insti tutional arrangements in Indonesia in these four areas. Incentives Most channels of technology transfer, formal or informal, are related to trade and FDI. Thus, the adoption of more outward-oriented economic policies in the mid-1980s led to a rise in technology inflows into Indonesia. Export-oriented investments were stimulated by the duty exemptions and drawback scheme introduced in 1986, which allowed exporters to buy inputs and capital equipment at international prices (Pangestu, 1997:45). Several other measures to reduce the earlier anti-export bias were also introduced. Effective rates of protection for manufacturing feU signifi cantly, from 59 per cent in 1987 to 16 per cent in 1995 (Fane and Condon, 1996: 40). Economic reforms were extended in 1994, with further liberal ization of FDI and international trade (Pangestu, 1998: 5). It may be argued that Indonesia's economic reforms in the 1980s and early 1990s did more to promote technology development than any explicit technology policy (Hill, 1995: 110, Hill and Wie, 1998:40). Indonesia's stable and growing economy, together with policies to promote interna tional trade and FDI, facilitated technology inflows and induced indige nous firms to upgrade existing technologies and adopt new ones. Education and training Education and training provide most of the skills needed for technology development (Thee, 1998: 119; Lall, 1996). Educational requirements change over time: basic literacy and numeracy may be sufficient for the first phase of industrialization, but more advanced and specialized skills become necessary as the industrial structure becomes more complex and diverse. Indonesia expanded basic education rapidly in the 1970s. Over 60000 new schools were built; real expenditures spent on education more than doubled; primary education was made compulsory; and school fees were abolished (Duflo, 2000). As a result, a near 100 per cent enrolment ratio was achieved in primary education by the 1980s. Secondary school enrolments increased from 35 to 48 per cent for male students and from 23 to 39 per cent for female students between 1980 and 1993 (Thee, 1998: 121). The expansion of primary education received widespread international recog nition and it is likely to have facilitated Indonesia's first stage of industrial development. 390 Competilivenes~; FDI and technological activity in East Asia However, despite the expansion, the Indonesian education system remains weak. Government education expenditures comprise only 2 per cent of GDP, substantially lower than most neighbouring countries (Booth, J999). Though junior secondary school was made compulsory in 1994, enrolment rates are still only about 70 per cent (Booth, 2000: 154). The quality of education at all levels remains poor.6 Indonesian 9-10 year olds perform below the international average in comparative tests (World Bank, 1997: 120) and most university graduates are said to require months of extensive on-the-job training (Booth, 1999: 30 I). The poor quality of education is attributable to large classes, poorly trained teachers and a lack of proper school materials. There are additional problems at the tertiary level. The system seems to emphasize cheap education rather than the provision of the science and engineering skills needed by the economy. This has resulted not only in a weak skill base but also in high rates of unemployment among university graduates. The 44 state universities, 24 state polytechnics and 5 state fine arts academies have not been able to meet the demand for higher education (Mukhopadhaya, 2000). As a result, more than a thousand private insti tutes have been established to meet this demand, but inadequate monitor ing has led to widespread quality problems. The government has recently abolished university subsidies, which may exacerbate skill shortages. Although poor education could be partly offset by extensive on-the-job training, Indonesian firms spend relatively little on human resource devel opment (HRD). Table 13.7 shows HRD expenditures as a percentage of sales and the proportion of establishments with HRD programmes in 1995. Expenditures on employee training are low. similar in size to R&D.7 The proportion of establishments with HRD is small apart from a few indus tries like rubber, steel, pharmaceuticals and electronics. A weak skill base may not be a significant problem in the early stage of industrial development when low wages and natural resources provide the basis for exports and investment. However, more sophisticated production technologies require a higher quality of education and training. The weak industrial skill base is perhaps the single most important deterrent to tech nology development in Indonesia. Technological infrastructure The technological infrastructure is another important determinant of tech nology development. Many forms of technical information and support, with public good characteristics, cannot be provided by the market and have to be provided by the government (Lall, 1996: 44). The government's role includes the setting of industrial standards, the promotion of quality awareness, the provision of metrology services, testing or information Table 13.7 Human resource develofJment (HRD) in the Indonesian manufacturing sector, 1995 Group Sector International Standard HRD expenditures as a Share of establishments Industrial Classification share of output (%) inHRD Resource-intensive rood, beverage, tobacco 31 0.02 9.5 Wood and furniture 33 0.02 11.9 Coal and oil refineries 353,354 0.02 10.8 Rubber 355 0.10 23.2 Non-metallic minerals 36 0.02 7.7 Labour-intensive Garment 322 0.01 5.6 Leather and footwear 323, 324 0.04 12.9 Toys, etc. 39 0.02 11.6 Scale-intensive Textile 321 0.02 9.2 w Paper and printing 34 0.06 14.7 \0 ..... Chemicalsa 351,352 0.03 19.6 Steel and non-ferrous metals 37 0.03 22.1 384 0.04 18.9 Differentiated Metal oroducts 381 0.03 14.3 382 0.04 15.1 Science-based Pharmaceuticalsh 0.26 35.1 Plastics 356 0.01 8.4 Electronics 383 0.06 27.4 Precision 385 0.03 14.9 NOles: 1/except 35221,35222,35223,35224. h = 35221, 35222. 35223 and 35224. Source: See Table 13.1. 392 Competitiveness, FDI and technological activity in East Asia search for firms that lack the facilities or skills, undertaking contract research, and extension services for SMEs, The public provision of information and support is particular important for SMEs. In a study of SMEs in Japan, Korea, Indonesia and Colombia, Levy (1994) concluded that although the leading sources of marketing and technical support were private, public initiatives in this area were also important. More precisely, public provision of technical information, tech nological training and assistance, and joint and contract technology devel opment were of significant help to SMEs. Moreover, the need for public support tends to be high when technological requirements are complex or when private technology networks are weak (Levy, 1994: 30). This suggests that as Indonesia moves up the quality ladder, the role of government will tend to increase. Indonesia has not yet developed an effective technological infrastructure. There are numerous examples of institutions failing to address the tasks for which they were set up (Thee, 1998: 127). Metrology, standards, testing, and quality assurance services are inadequate. This is partly because the impor tance of such services is not fully understood, and partly because compre hensive industrial standards are lacking (Lall, 1998: 154). Indonesia lacks an effective productivity centre to provide industry-wide technical training and assistance, especially for SMEs. Organizations such as the Institute for Machine Tools, Automation and Production Technology, established to provide such services, tend to meet the demands of only a few 'strategic' industries (Lall, 1998: 155). To sum up, a technical infrastructure is needed to adopt, disseminate, and upgrade technology, but presently such an infrastructure is poorly developed in Indonesia. Specific S&T policies One of the most notable, and controversial, of all government interven tions in Indonesia since the late I 970s, has been the aggressive high technology programme. The government targeted and tried to promote indigenous technological capability in ten high-technology industries, including aircraft, shipbuilding, railroads, telecommunications, electronics, steel and machine goods. The best known part of the program is the devel opment of an aircraft industry (IPTN), which has been the government's largest and most ambitious programme of technology development (LalI 1998: 158). Despite the accumulation of some engineering and production knowl edge, IPTN has never been financially viable. According to McKendrick (1992: 64), managerial weakness, the absence of an independent aviation agency and Indonesia's weak scientific and engineering infrastructure Technology development in Indonesia 393 account for its poor performance. The industry has few linkages with other sectors of the economy, and there are few noticeable spillovers (Lall, 1998: 158). Indonesian S&T policy since the late 1970s illustrates that high technology projects do not always lead to a broad-based and efficient tech nology development, particularly when the underlying research, education and technical infrastructure is weak (Hill, 1995: 118). 3.3 Technology Policy after the Crisis No country in Southeast Asia was more severely hurt by the financial crisis than Indonesia. GDP fell by about 15 per cent, unemployment and poverty increased, many businesses closed down and there were dramatic social and political changes. As a result, the scope and potential for technology policy is different from the pre-crisis period. Technology seems to be of less imme diate concern than raising employment and revitalizing firms forced out of business; thus, efforts are likely to focus on labour-intensive rather than technology-intensive industries. 8 The government's ability to fund technol ogy policy is seriously constrained, as there are many other pressing eco nomic needs. The crisis in the financial sector is estimated to cost the government somewhere between 60 and 100 per cent of GDP (spread over several years). Considering the limited benefits of previous public technology projects, perhaps it does not matter very much that the government cannot continue with these projects. It seems more important that the government concen trate its efforts on maintaining high enrolment rates in primary and secon dary education. Rising rates of school dropout in poor urban areas following the crisis alarmed the government, which responded by abolish ing school fees (Booth, 2000: 151-4). The situation seems to have improved, but new problems lie ahead. Most importantly, the new decentralized struc ture of Indonesia may affect the education system. Two new laws on polit ical and fiscal decentralization were introduced in 2001. As a consequence, the districts and regions were allowed to keep a larger share of taxes and revenues from oil and gas, which had previously been transferred to Jakarta. On the other hand, they will also receive fewer transfers from Jakarta and will be responsible for certain expenditures, including educa tion. The new system will benefit a few resource rich provinces as well as West Java and Jakarta, but other parts of Indonesia will suffer substantial reductions in resources. There is an obvious risk that these regions will find it increasingly difficult to obtain the resources adequately to educate their citizens. Liberalization of the FDI regime will probably continue since Indonesia 394 Competitiveness, FDI and technological activity in East Asia is likely to be more dependent on FOI in the future. The financial crisis has wiped out large parts of domestic capital and foreign banks are reluctant to offer further loans. Financing current account deficits by external bor rowing has been found to be much riskier than by relying on FOI. Unfortunately, FOI inflows have declined in the last few years, falling from a peak of US$6 billion in 1996 to minus US$3 billion in 1999 (Ramstetter, 2000). To attract new FOI and cope with intensified competition in the region, a robust investment regime is needed. If Indonesia can attract more FDI, there is the potential for increasing technology diffusion by improv ing the skill base and increasing competition in the economy. 4. SUMMARY AND LESSONS FOR OTHER DEVELOPING COUNTRIES In Indonesia's present economic predicament, it is easy to forget its remark able economic performance during previous decades. Indonesia achieved not only high growth rates and improved living standards for the popula tion, it also underwent substantial structural transformation. At least four lessons can be learned from Indonesia's technology devel opment. First, several external sources of technology are important in the early stages of industrial development when technological requirements are relatively low. Foreign buyers, trading companies, and foreign experts are channels through which new technology and know-how are acquired. This implies that openness to trade, investment and skilled labour will enhance industrial technology development. Second, the public sector has to make greater efforts to acquire, upgrade, and disseminate technology and know-how as a country moves up the tech nology ladder. This is particularly the case when private industrial linkages and networks are weak. Hence, government intervention in technology development may increase over time. Third, FDI can play an important role in introducing new technology, generating employment and expanding production and exports. FDI is crucial to technology development when there are constraints on domestic efforts, notably insufficient government funding. However, FOI does not automatically generate substantial spillovers and linkage effects, nor does it necessarily lead to technological upgrading. Complementary efforts to maximize benefits from FDI are important in enhancing technological development. Finally, policies targeting high-technology industries tend to fail when the technological, managerial and institutional infrastructure is underde veloped. Hence, a micro-level intervention to promote technological devel Technology development in Indonesia 395 opment might be useful, but only when sufficient skills are available and the technical infrastructure is in place. NOTES I. Asian Development Bank (1999). 2. For instance, Lall (1998: 138) uses the technological structure of manufactured exports (divided between low. medium and high-technology products) as an indicator of techno logical sophistication. 3. See table 1 of Hill (1995: 92). 4. More specillcally, they are producers and exporters of woven fabric of polyester filament yarn or polyester filament fabric. We are grateful to Mr Makolo Abe, Institute of Developing Economies in Tokyo, for providing the information in Table 13.6. 5. Lall (1996, 1998), Thee (1998), Nelson (1993), and Evenson and Westphal (1995). 6. See, for instance. Hill (1995). Hill and Wie (1998), Lall (19981. Thee (1998), and Booth (1999). 7. See Table 13.4 for figures on R&D. 8. However, technology, in the sense of market knOWledge, will continue to be important. In fact, exports will be crucial in any expansion of production since domestic demand is con strained by the recession. Trade contacts, foreign trading houses, FDI and other channels will have to be utilized to achieve the necessary knowledge for this export expansion. REFERENCES Asian Development Bank (1999), Key Indicators of Developing Asian and Pacific Countries, 1996, Hong Kong: Oxford University Press. Aswicahyono, H., C. Basri and H. Hill (2000), 'How not to industrialise? Indonesia's automotive industry', Bulletin of Tndonesian Economic Studies, 36, 209~241. Berry, A. and B. Levy (1994), 'Indonesia's small and medium-size exporters and their support systems', World Bank policy research working paper No.1402. Blomstrom. M. and F. Sjoholm (1999), 'Technology transfer and spillovers: does local participation with multinationals matter?', European Economic Review, 43, 915~23. Booth, A. (1999), 'Education and economic development in Southeast Asia: myths and realities', ASEAN Economic Bulletin, 16(3), 290~306. Booth, A. (2000), 'The impact of the Indonesian crisis on welfare: what do we know two years on?', in C. Manning and P Van Diermen (eds), Indonesia in Transition: Social Aspects of Reformasi and Crisis, Singapore: Institute of Southeast Asian Studies. Duflo, E. (2000), 'Schooling and labor market consequences of school construction in Indonesia: evidence from an unusual policy experiment', NBER working paper No. 7860. Evenson, R.E. and L.E. Westphal (1995). 'Technological change and technology strategy'. in J. Behrman and T.N. Srinivasan (eds), Handbook of Development Economics Vol. III, Amsterdam: Elsevier. Fane, G. and T. Condon (1996), Trade reform in Indonesia, 1987~95', Bulletin of Indonesian Economic Studies, 32(3), 33~54. 396 Competitiveness, FDI and technological activity in East Asia Hill, H. (1995), 'Indonesia's great leap forward? Technology development and policy issues', Bulletin of Indonesian Economic Studies, 31(2), 83-123. Hill, H. and Thee Kian Wie (eds) (1998), Indonesia's Technological Challenge, Singapore: Institute of Southeast Asian Studies. Lall, S. (1996), Learningfrom the Asian Tigers: Studies in Technology and Industrial Policy, London: Macmillan Press. Lall, S. (1998), 'Technology policies in Indonesia', in H. Hill and Thee Kian Wie (eds), Indonesia's Technological Challenge, Singapore: Institute of Southeast Asian Studies. Levy, B. (1994), 'Technical and marketing support systems for successful small and medium-size enterprises in four countries', The World Bank policy research working paper No. 1400. McKendrick, D. (1992), 'Obstacles to 'catch-up': the case of the Indonesian aircraft industry,' Bulletin of Indonesian Economic Studies, 28(1),39-66. Mukhopadhaya, P. (2000), 'Income gap and educational opportunity in some Southeast Asian countries: government efforts in bridging inequality', mimeo, Department of Economics, National University of Singapore. Nelson, R.R. (1993) (ed.), National Innovation Systems: A Comparative Analysis, New York: Oxford University Press. Okamoto, Y. and F. Sj6holm (2000a), 'FDI and the dynamics of productivity: micro economic evidence', APEC discussion paper No. 34, Nagoya University. Okamoto, Y. and F. Sj6holm (2000b), 'Productivity in the Indonesian Automotive Industry', A SEAN Economic Bulletin, 17,1-14. Pangestu, M. P. (1997), The Indonesian textile and garment industry: structural change and competitive challenges', in M. E, Pangestu and Y. Sato (eds) Waves of Change in Indonesia's Manulacturing Industry, Tokyo: Institute of Developing Economies. Pangestu, M. P. (1998), The Indonesian textile and garment industry: structural change and corporate responses', in Y. Sato (ed.), Changing Industrial Structures and Business Strategies in Indonesia, Tokyo: Institute of Developing Economies. Ramstetter, E.D. (1999), 'Trade propensities and foreign ownership shares in Indonesian manufacturing', Bulletin of Indonesian Economic Studies, 35,43-66. Ramstetter, E.n (2000), 'Survey of recent developments', mimeo. Kitakyoshu, International Centre for the Study of East Asian Development. Rock, M. T. (1999), 'Reassessing the effectiveness of industrial policy in Indonesia: can the neoliberals be wrongT, World Development, 27(4),691-704. Sj6holm, F. (1999a), 'Learning to export: the role of foreign contacts', working paper No. 326, Stockholm School of Economics. Sj6holm, F. (I 999b), 'Technology gap, competition and spillovers from direct foreign investment: evidence from establishment data', Journal ol Development Studies, 36, pp, 53-73. Sj6holm, F. (I999c), 'Productivity growth in Indonesia: the role of regional char acteristics and direct foreign investment', Economic Development and Cultural Change, 47, 559-84. Thee Kian Wie (1998), 'Determinants of Indonesia's industrial technology devel opment', in H. Hill and Thee Kian Wie (eds), Indonesia's Technological Challenge. Singapore: Institute of Southeast Asian Studies. UNESCO (1999), Statistical Yearbook, Lanham, MD: Beman Press. World Bank (1997), Training and the Labor market in Indonesia: productivity gains and employment growth', report No. I6990--IND, Washington DC. Index Abrenica, Joy V. 285 automotive industry 24, 30, 283, 348, academic journals, Japan 87 385 aerospace repair, Singapore 212 Aw,BeeYan 176, 178, 182, 188, 189 affiliates in developing countries 110, 120, 136 Bangladesh, challenge to China in agricultural biotechnology 365 textiles 265 goods, China 242,246 Bank of Japan 94 agriculture 358 Bankruptcy Law, Thailand 343 graduates 274 bankruptcy laws in Singapore 220 agro-forest products 24 banks, nationalization of 343 air conditioners, Japan 89 Barro, Robert 63,81,121,141 aircraft manufacture 86,381,392 Baumol, William 1. 163, 166 Aitken, Brian 67, 81,176,188 benchmarking, international, Allen, T. 174,188 Singapore 223 America, North, Japanese affiliates 110 Bernard, Andrew B. 176,189 Amsden, Alice H. 166, 236, 313, 331 Berry, A. 388, 395 Anti-Monopoly Law, Japan 92, 94 Bertschek, Irene 96, 101 antitrust policy 161 Best, M 14,50,326 Anuwar Ali 320, 325, 331 Biomedical Research Council, Appeal Technology, Korea 150 Singapore 222, 230 Argentina, automobile industry19 biotechnology 159, 358 Arnold, Erik 367 Singapore 204 ASEAN4 Blomstrom, Magnus 81, 176, 189,385 education in 65 blueprints, purchase of, in Japan 87, 88 per capita growth 60 Blumenthal, T. 83, 101 Asian Development Bank 365, 395 Bo, Yibo266 Asian financial crisis 149,336, 375, 394 Booth, A. 390, 393 Singapore 219 Borensztein, E.67, 76, 81 Asian Free Trade Agreement (AFTA) 312 borrowing, external, Indonesia 381 Asian Institute of Technology 354 Brimble, Peter 349, 355. 367, 372. 373 assets price rise in Japan 106 broadband, Singapore 211 Aswicahyono, H. 385, 395 Brown. John Seeley 83, 101 Automation Application Centre, bubble economy in Japan 106 Singapore 223 business networking, Thailand 354 automation technology, China 254 automobile industry Capabilities Development Division. Argentina 19 Singapore 223 Brazil 19 capital competitiveness in 83 equipment 386 Japan's technological agreements in goods imports, Korea 146, 151, 161, 88,98 162 Korea 148 producers 18" 68~71, 77 397 398 Competitiveness, FDf and technological activity in East Asia capital (continued) computer products investment 369 China 249 liberalization 96 Japan 91 mobility I3 Korea 151, 161 cars, passenger 148 Philippines 270 Caves, R.E. 67, 81 Singapore 203, 211, 222, 230 cement plants 159 Confucian culture, Korea 164 Census of Manufactures, Taiwanl77 copyrights, Singapore 197 Center Satellite Factory system, cosmetics imports, Japan 89 Taiwan 175 cotton export, China 243, 249 Cha, Dong-Sae 166 crude oil export. China 243, 249 chaebols, Korea 149, 150,153,161, 164 Cultural Revolution. China 242 in technology 156 customs procedures, rationalization 77 Changi International Airport Service Cyhn, Jin 152, 166 211 Changzhou. China 260 Daewoo cars, Korea 146, 148, 149, Chantasmary, M. 307,331 152 Chartered Semiconductor Dahlman, Carl 355, 373 Manufacturing, Singapore 216 economy of Taiwan 168, 172, 189, chemical industry 348, 365 194,236 in China, R&D 242, 252 Defence Science and Technology Japan 86,89, llO, 141 Agency, Singapore 230 Korea 151, 159 Deng Xiaoping 242 Singapore 212 Department of Science and Chen. Baochun 266 Technology (DOST) 285 Chen, Huiqin 266 design technologies. Japan I3 I China developed countries 20, 76 and FDr 44, 60 and China 262 education in 65 education in 65 exports 30-31 developing countries 313 technological development 239-66 defined 49 Chinese Academv of Science 251 education in 65 Chinese immigration to Taiwan 169, and export 23 172 importing technology 62-3,136, Chuang, Yih-Chyi 178, 179, 186,189 137 civil engineering graduates 274 R&Din 128,140 Clerides, Sofronis 176, 189 technological activity 13, 37--42 Coe, David T. 176, 189 diversification, Malaysia 328 Cohen, Wesley M. 83, lO I Dollar, D. 349, 373 commodity diversification in Malaysia domestic 306 economy 282 communications industry, Singapore enterprises, expansion 383 191,212 firms in Korea 143 competition policies 161, 199 industries and inward FDI 96 competitiveness in Malaysia 310 in Asia 334, 340 market in China 262 determinants 73-7 research and development 161 for firms 95, 103 support institutions 315-27 and technology 12--49, 57-81 technological activities in Korea 145, in Thailand 334-73 153-7 Index 399 Doms, M.189 electrical Doner, Richard 373 and electronic products, Malaysia Dosi, G. 195,305, 331 309310 Dow Chemical 150 equipment industries 86, 91, 98, 110, Duflo, E. 389, 395 141 machinery, exports 309, 314 East Asia electricity industry, China 240 affiliates of Japan 11 0, 140 electronic exports 46 components import 69 East Asian economies, competitiveness data interchange, Singapore 207, 58-62, 77-80 211,212 East Asian Miracle (World Bank) 57, data processing industry 286 58 engineering, Korea 149 Eastern Europe, imports to China 240, firms 286 241 electronics 24,80,377,390,392 e-commerce industry, Singapore 216 applied 358 Economic Development Board 370 and communications 270, 271, 273 economic policy and development and computers 358 in China 239, 242-3 industry 348 in East Asia 57, 58-60 China R&D 243,252 in Indonesia 375 in Korea 151 in Japan 86 in Taiwan 169, 171-74, 178, 182 in Singapore 191 markets 319 in Taiwan 168, 169 employment practices 356, 357 of Thailand 336 engineering 355, 357, 364, 392 education and technology, role of, in in education 37 Korea 157-60 exports 24, 30 education and training 80.361,368, graduates 278 376,389 knowledge 87,99 basic, expansion of 389 Engineering Research Centres (ERCs) in China 239,241 Korea 158 in economic growth 73 engineering institutions 197, 198 science, Singapore, training in 218 in Japan for industry 98, 99 technologies 16 poor quality 390 engineers in Taiwan 169, 175 in developing world 40-41 for technology 63-6 foreign, in Singapore 223 education, formal 273-5 in Korea 155,157,158 higher, in China 65, 243 English language in Korea 145, 163. 164 in the Philippines 279 primary. its role 63. 222, 223 problem in Thailand 361, 364 public spending on 274 enrolment rates in education 37, 389, in Singapore 222, 223 390,393 technical 317 enterprises, state-owned 198, 225 tertiary, inadequate 275, 315 entrepreneurs 329 educational environmental attainment 63-5, 64, 73, 77 compliance 175 in Korea 161 and hazardous waste management ratios 121 365 system in Philippines 274 science 365 400 Competitiveness, FDf and technological activity in East Asia equipment foreign direct investment (FDI) 12~49, manufacture, original, Korea 146, 42-5,281, 368, 381~6 151 in Chinese manufacturing sector purchase 197 259~64,263 Ernst, Dieter 166 dependence on 35, 36, 47 Esso, in Japan 88 and exports 349 Europe, Japanese affiliates 110 falls in Korea 150, 165 Evans, D. 189 industrialization in Malaysia 306, Evenson, R.E, 173, 189,377,386,395 313-15, 320, 328~9 exchange rates. stability 80 inward 73, 74, 83-4 export in Japan 95~ 7 competitiveness in East Asia 23~34, policies 342~57, 369~70 46 recent trends 343-49 expansion 329 in Taiwan 168, 170~73, 178~9, 186 market, Taiwan 173-4, 178, 185 in technology transfer 18-23,67-8, performance 14, 35, 36, 70 280~85 promotion, Korea 161 in Thailand 334-73, 345, 346, 347 sales ratio. high 120, 121 foreign enterprises in Thailand 349 structure in Thailand 339 Foreign Exchange and Foreign Trade exports Control Law, Japan 93 by country 52~6 foreign expertise 383, 388, 394, and export patterns 48 need in Singapore 234 by foreign multinational companies presence in Indonesia 384 42 foreign firms high, low and medium technology 23 hosting 176 and imports, China 243, 244, 247, in Japan 88 248 Foreign Investment Act 278 of Korea 144 Foreign Investment Advisory Service in Malaysia 328 373 Exxon 88, 307 Foreign Investment Board 94 Foreign Investment Law, Japan 93-4 famine in China 242 foreign Fan, X. 261, 266 investment policies, Singapore Fane, G. 389, 395 199 fertilizer plants, Korea 159 licensing, Korea 146.151,161 fibres, synthetic 88 loans, for capital, Korea 153 financial markets 177 sector liberalization 342 multinationals. Korea 150 services, Singapore 191 ownership 372 financing, private R&D Korea 163 suppliers. Korea 150 firms technology 68,71,73.74, 77, 182, foreign 314 365 local 129 transfer, Korea 147, 162, 164, 165 Five Year Economic Development franchising 18 Plan (Korea) 152 Frank, A.G. 313, 331 Ford Company 96, 97 Freeman, C. 305, 331 foreign French Electrical/Electronic Industry affiliates 35 Federation 223 business community 369 French-Singapore Institute 222-3 competition in domestic market 92 Fu, 1. 260, 266 Index 401 Fuji Electric 88, 98 Hewlett-Packard, Korea 150 Fujimoto, Takahiro 97,101 Higher Education Development Fukao, Kyoji 141 Project 365 funding sources. Singapore 204 high-tech exports in Japan 84 garment production. China 243, 249, industrial parks in China 240 252 investment 204,216,226 GDP per capita growth 18,59,60 Singapore Gee, San 173, 189 program 392 General Motors. Japan 97 specialization, East Asia 46 genetic engineering 358 zones in China 256, 257 Genome Mapping project, Singapore Hill, H. 283, 376, 381, 389, 396 219 Hiroshige, Toru 87, 101 German Agency for Technical Hitachi, Japan 149 Cooperation, Singapore 223 Hobday, Michael 30, 50, 166, 170, German-Singapore Institute, 189 Singapore 223 Honda 92, 98 Gerschenkron, A. 305, 331 Honeywell, Korea 149 Global Competitiveness Report. Hong Kong 30,37,60 Singapore 208 and China 262 globalization 19,25,46 electronic consumer goods 31 of Japanese industry 99, 128 Hou, CM. 168, 189 in Korea 146, 155 Hsieh, CT. 231, 236 Globerman, Steven 67,81,96, 101 Hsinchu Science and Industrial Park, Goto, Akira 83. 86, 94, 95,101 Taiwan ) 71, 172, 326 government Hua Guofeng, China 242 domination of technology in China human capital 121, 128. 136, 240 315~18,329 expenditure 57, 73 in Philippines 268 initiatives in China 254-7 humanities 357 in Japan 98 Human Resource Development Fund institutions in Singapore 198 316 policy in Taiwan 175, 186 human resources 269 in technology 93~5 development 161, 344, 359, 364-5, research institutes, Korea 157, 390,391 158~61 investment 368 restrictions on inward FDI 99 Hyundai Motor, technological learning Government of Singapore 236 146, 148, 149 Guangdong, China 259 Guangxi, China 259 IBM data storage 203 Haddad, Mona 81, 96, 101 in Japan 88, 96 Hall, B. 189 immigration policy, Singapore 223 Hamilton, C 313, 331 import hard disk drive industry 275,277,278, liberalization, Japan 89 283, 354 -substituting industries 36 He, B. 266 importing in China 259 healthcare 230 imports Heckman, James 1. 189 of capital goods 68~71, 386 Henges, H.A.166 of technology 71 402 Competitil'eness, FDI and technological activity in East Asia import-substitution 36 Information Society Index, Singapore in Malaysia 328 208 regime 287-8 Information Technology Institute, incentives Singapore 218 export-oriented investments 389 innovation 18,20,22,83,305,306,323 in technology 389 in China 240 incineration technology, Singapore 211 sectors, linkages, 224-6 income tax returns, internet filing, system, Singapore 195-200 Singapore 211 Inoki. Takenori 101, 102 Indonesia 31, 44, 60 Institute of Systems Science, Singapore technology development 375-95 223 industrial institutions, public 198-200 development 336-42 Intarakumnerd 365, 367, 374) in Taiwan 169 Integrated Microelectronics Inc 286 graduates, fewer 274 Intel Philippines 149,277 growth of technology 375 intellectual property link with universities 80 rights protection, Korea 152, 161, policies 28, 47, 286-90 164 skills in China 239, 241 Singapore 197,200,222,225 technology transition 307-313 inter and intra-firm technology in Malaysia 305-33 transfer, Korea 146-50, 150~53 Industrial Coordination Act 320 international Industrial Finance Corporation of agencies 47 Thailand 338 conferences, Singapore 198 Industrial Generic Technology investors 22 Development Project, Korea production 19 159 spillovers 177-8 Industrial Master Plan 326 standards 77 Industrial Technology Research technology linkages, Singapore 197 Institute (IRSI) Taiwan 171 trade, Korea 164 industrialization International Finance Corporation 266 of countries 35,40-41,58 Internet, Singapore 204, 211,220 in Japan 98 intra-firm technology, 197 in Korea 166 transfer 104, 136, 137, 140 industries, supporting Japan 140 determinants 131-7 industry Japanese firms 128-37 and education linkages, Singapore transactions 71, 72 222,224 invasion, foreign of Korea 145 heavy and light, China 240, 241, 252 investment since the Revolution in China 242 in China 239,242 rural in China 243 domestic and foreign 43, 338, 342 structural changes 338-42 in education inflation, low 80 in Japan 93 information 15 in Taiwan 176 collection R& DIll flows 15,76 sharing 354 growth in Korea 151 technology (IT) 327 private in technological capability, Singapore 191, 204, 218, 228 Taiwan 179, 184 Information Communications Institute promotion 80 of Singapore 223 ratio 76 Index 403 in R&D Singapore 218-19, 235 knowledge-intensive industries 162 Investment Promotion Law 339 Komiya, Ryutaro 88, 102 inward direct investment - Japan Korea 31, 35, 44,143-66 83-101 education 37, 65 Ionics Inc. 286 electronics 20 iron and steel industries Japan 89, Korea Advanced Institute of Science and Technology 157 Jaffe, Adam B. 176, 189 Korea Exchange Bank 166 Japan Korea Institute of Science and affiliates 129 Technology 167 in hard disk drive industry 275 Korea International Trade Association high productivity in 69 152 inward direct investment 83-102 Korea Research Foundation 158 research and development in 84-6 Korea Scientific and Engineering supplies from 284 Foundation 158 technological capabilities 83-102 Korea Trade Promotion Corporation Japan Productivity Centre 88-9 152 Japan Steel 92 Krugman, P. 48, 50, 57, 81 Japanese Kuo, Cheng-Tian 173, 190 colonial rule of Korea 145 labor of Taiwan 169 force, educated, Taiwan 168, 170, firms 67 177 foreign direct investment (FDI) intensity, China 243, 246 104-7, 105 productivity 148 kanban system 48 shortage in Japan 106, 140 multinational companies (MNCs) source in Korea 145 103-41 in China 262 training for Philippines 278 in Korea 148 universities 99 in Malaysia 312 Japanese Ministry of International in Singapore 199,201 Trade and Industry (MITT) division of 316 104 English-speaking 278 Japan-Singapore Artificial Intelligence labor-intensive Centre 223 goods 270 Jiang, Junlu 266 manufacturing in Singapore 191 Jiangsu, China 259,260 technologies K157 .lin, Pei, 266 Lai, Y.w. 314, 331 Johnson, C. 306, 331 LaB, Sanjaya, 12-49,50,81,305,331, Jomo, K.S. 307, 331 381,389,392,393,396 Lawrence, R.Z.69, 81 Kaldor, N. 305, 331 learning-by-using in Singapore firms Kawai, Hiroki 57-82, 103-42 197,201 Kawai, Masahiro 141 learning process 83, 86-90 Keppel Group 212 Lee, J.S. 231, 331 Key Technology Project, China 256 legal systems 80 Kim, Linsu 143-66,195,237,306,331 Levy, B. 169,173,190,388,392,396 knowledge Li, Jingwen 266 sources 177 liberal arts in university curricula, spillover 176 Singapore 224 404 Competitiveness, FDI and technological activity in East Asia liberalization managerial capabilities, transfer of, of domestic markets ]64 Korea 150 of FDI policies 68, 393-4 manpower in industry 21 development institutions, Singapore in trade 77, 80 222-4,228,229 in Japan 88, 94, 140 training 175 of world markets 329 Manpower master plan, Singapore 219 licence imports 71, 72, 77 manufacturing licensing agreements exports in Indonesia 376,378, 379, in Japan 87, 95, 386 380 in Korea 146 growth in Malaysia 307 in Singapore 225 industries 22, 44 in Taiwan 168 Japan 83, 84, 136 Lichtenberg, Frank 97, 102 share in GDP 121 Life Sciences Fund, Singapore 219 Mao, Zedong 242 life sciences, Singapore 204, 207, 219, maquiladoras, Mexico 28, 49 230 market light industry 212 demand and supply Korea 160, 163 Lin, Y Justin 178, 179, 186,241,242, forces 36 267 over-dependence on 268 List, E 305,331 institutions, Singapore 198 Liu, Yigao 264, 267 mechanism and government policies loans for small firms 323 in Japan 90-95 LundvaIl, BA 195,237,305,331 market-based economy 240 marketing machine experience 177 goods 392 internationa144 tool industrv 285 markets, domestic and international tools imported, Japan 83, 88 171 machinery Mason, B.S. 145, 167 imports and exports 68, 70, 80, 309 materials science, Singapore 218 Korea 151 McKendrick, David 231, 237, 373, 392, sector, China 246 396 and transport equipment industry media and broadcasting industry, 348 Singapore 230 machines, development of new 131 medicine industry, China 252 macroeconomic Meiji Restoration, Japan 83, 98 environment 67,80 mergers and acquisitions 21, 68, 150 factors 106 metal indicators 335, 336, 337 casting industry 284 mismanagement 268 products 246, 284 policies in Singapore 199,235 Metal Institute of Research and Malaysia 30, 44, 60 Development Centre 284 electronics industry 19 Mexico 28 hard disk drive industry 231 automobile industry 19 industrial technology transition electronics industry 305-30 micro-data studies Taiwan 176 Malaysian Industrial Development microelectronics, training in, Singapore Authority (MIDA) 315 218 Malaysian Mining Corporation 307 Micron Technologies, Korea 149 Index 405 military National Computer Board, Singapore and civilian goods 86, 87, 98 216,218,228 projects, China 241,242 National Information Teaching Plan mining (NITP), Singapore 217-18 in Korea 145 National Information Technology R&D in China 252 Agenda 327 Ministry of Defence, Singapore 229-30 National Research Council of Ministry of Education, Singapore 217, Thailand 357 227 National Science Foundation, China Ministry of Health, Singapore 230 254 Ministry of Information and the Arts, National Science Foundation, Singapore 230 Singapore 222 Ministry of International Trade and National Science and Technology Industry (MITI), Japan 94 Board, Singapore 194, 195,207, Ministry of Manpower, Singapore 218,227 219 National Semiconductors, Korea 149 Ministry of Science and Technology, National Technical Information Singapore 217 Service, Korea 167 Ministry of Science, Technology and National University of Singapore 220 Energy (MOSTE) 357 Natsteel Group 212 Ministry of Trade and Industry, natural sciences 99 Singapore 217,227,237 Navaretti, Giorgio Barba 141 Mitsubishi Electric 88,319 Nelson, R.R. 15, 16,50, 195,236,237, Mitsubishi Heavy Industries 92 305, 332, 396 Mitsui Petrochemical process 88, 93 Nestle in Japan 88 Miyazaki, Masayasu 86, 102 Nippon Steel in Japan 98 Mobil oil in Japan 88 Nissan Motor 88, 91, 97, 98 mobile telephone systems, Korea 157, nuclear power 162 China 239 molecular and cell biology Sing 221 Korea 159 Morocco 28 technical spillovers 67 Odagiri, Hiroyuki 83, 99, 102, 120, Motorola (Philippines) Inc 277 141 Motorola, USA/Korea 149,150 Odaka, Konosuke 91, 102 Mowery, D.C. 174, 190 Oji Paper, Japan 92 Mukhopadhaya, P. 390, 396 Okamoto, Yumiko 285, 383, 385, 396 Multimedia Super Corridor 326,327 Olivetti in Japan 88 multinational companies (MNCs) 14, optical and laser technology training, 23,36,306,314,383 Singapore 218 business units, Singapore 191, 197, optical fiber technology, Korea 159 203 Organisation for Economic Co in electronics industry 277 operation and Development role of, in Korea 146-53, 161 (OECD) 18, 20, 195, 237 sellers of licensed technology 18, 20 original equipment manufacture, in Taiwan 170, 172 Korea 146 have technology skills 34, 35 outsourcing, Singapore 203 transfer of technologies 265 overseas affiliates 120, 121, 136, 137 National Automation Masterplan, investments, Japan 104 Singapore 218 production ratio 120, 121 406 Competitiveness, FDI and teclmological activity in East Asia Pacific Semiconductors 286 population movements in China 241 Pack,fI. 21, 50, 80n, 81.173,186,190 Port Authority of Singapore 211 Palm Oil Research Institute of Porter, M. 237 Malaysia 307 precision engineering industry, Pangestu, M. P. 389,396 Singapore 212, 222 Paopongsakorn, Nipon 374 pribumi (indigenous Indonesian firms) partnerships, Singapore and Europe, 388 Japan, US 234 price competitiveness, Japan 106 patent primary see also education applications 318. 318-20, 377 education 37 licences 84. 88, 89 schools, enrolments, Korea 163 registration Korea 156 private firms in Taiwan 186 studies, Taiwan 176 sector in Japan 91 patenting, Singapore 197,216, 225 Private Universities Bill 316 patents and licences 71, 73, 74, 77 privatization in industry 21 China 257-8 product Pavitt, K. 306, 332 engineering 17 Penang Development Corporation quality in Japan 88 (PDC) 315 production per capita high-tech 14, 76 GDP growth rate 63 techniques 176, 177 income 60 Productivity and Standards Board 228 Perodua, automobile manufacturers Proton automobile manufacturers 312, 312 319 petrochemicals in Japan 91, 92 public institutions and R&D, China petroleum 307 249-59 and petrochemical technology 365 public refining, Singapore 31 policies pharmaceuticals 390 Korea 160-63 Japan 99 impact 231-2 Korea 151 in Philippines 286--97 Philippines 44, 60, 268-304 sector 230 electronics industry 19 support for start-ups, absence of 285 skills education 37, 65 purchasing power parity 60 textile products 31 Philips consumer electronics, quality control 22, 152 Singapore 203 Philips Semiconductors, Philippines Radosevic, S. 50 277 railroads 392 Phoenix Pulp and Paper Ramstetter, E.D. 383, 396 subcontracting 356 Rao B.V.V. 237 plant and equipment investment in Rasiah, Rajah, 307-32 Japan 91 rattan furniture 388 plastics industry, Taiwan 169 Reddy, N. Mohan 142 policies, public, Singapore 198-200 refined oil exports, China 243, 249 polyethylene process 93 regional patterns 25, 26 polytechnic regression analysis 120, 121 engineering courses, Singapore 201 research and development (R & D) 73, and industrial training links, 305, 306, 307, 376 Singapore 203. 224 China 249-54, 260, 264 Index 407 domestic 381 science and technology in education 65 in China 241. 242, 257~8 funding 323-4 comparative performance in Asia by industry 382 276 in Japan 84-9, 93, development policies 357~68 Korea 30, 153, 154, 163 Singapore 194, 195, 198, 208, measuring 40 217-32 overseas, in Japanese multinationals education 37 99, 103~142 graduates, supply in 316 in Philippines 286 networking, Singapore 234-5 private 228 policies 392~3 and researchers 65, 66 students insufficient, Thailand 364 statistics 323 at tertiary level. Indonesia 377 support 320-3 training at postgraduate level 365 Taiwan 30,1746,186 Science and Technology Agency, Japan research associations, Japan 87, 95, 99 101 research grants, awarding, Singapore Science Research Centers (SRCs) 222 Korea 158 research institutions 18 scientific China 258~9, 265 instruments 314 Korea 157 papers in Japan 84, 87 Singapore 233-4 scientists 464 researchers in Asian economies III see in developing world 40-41 also R&D and engineers in R&D, Philippines reverse engineering 68 275 Korea 146, 150, 157 foreign, Singapore 223 Reynolds, P.D. 237 in Korea 155, 157,158 Rhee, YW 70,81 Sease D. 173, 190 Roberts, M.1. 179, 190 Second National Science and robotics, training in Singapore 218, 222 Technology Plan, Singapore 218 Rock, M.T. 375, 396 secondary schools, Korea 145 Romer, P. 197, 236, 237 see also education rubber products 314, 390 Sekiguchi, Sueo 102 Selvaratnam V. 332 sales, local and world-wide, Japan Sembawang Group 212 121 semiconductor Samsung Electronics, Korea 149, 150, assembly, Philippines 275 152,155 industry, Japan 83 San, Gee 168.190 Korea 148, 157 satellite wafer fabrication, Singapore 213, launching, China 239 223 receiving systems, Korea 156 Shell Oil 307 Schein, E. 235, 237 shipbuilding 392 Schive, Chi 171, 190 Japan 86 science and engineering Siamwalla, A. 373 base 335 Sibunruang, A. 349, 373 graduates 274 Siemens, 1. 88, 98 skills, poor 390 silk exports, China 243, 249 Science and Engineering Research Silla Dynasty Korea 145 Council 222 Sim Wong Hoo 220 408 Competitiveness, FDI and technological activity in East Asia Singapore Sunward Technologies 275 creating technology 191-36 system engineering Korea 159 education in 65 electronics industry 19 Taiwan 44, 60, 168-90 hard disk drive industry 231 and China 262 high rank in technology 209, 210 electronics 20 national innovation system 191-236 exports 35 petroleum refining 31 skills education 37, 65 skills education 37, 200, 201, 223 technology acquisition 168-88 stock exchange, national 226 Taiwanese Census of Manufactures Singapore Airlines 211 178 Singapore Broadcasting Authority 230 tariff Singapore Institute for Standards and barriers 385 Industrial Research 221 Korea 161 Singapore Science Council 217 protection 339 Singapore Technology Group 202, 212 tax Singapore Telecoms 211 credits for R&D, Taiwan 175 Sjoholm, Fredrik 176, 383, 385, 396 incentive schemes, Singapore 226 skilled labor, foreign 364 in China 239, 256 skills training 15,36-7,47,329,359-64 in demand and supply Korea small and medium enterprise (SMEs) 162-3 promotion, Korea 160 Japan 94, 98 support 325-6, 388, 392 teaching in universities Korea 157 Taiwan 169, 171,185 technical Small Industry Finance Office 338 change 14,47 Sony, Japan 92, 98 competence 37 Soon, TW 222, 224, 225, 237 education, Soviet Union, technological aid for in Korea 152 China 240,241 poor, in Malaysia 315 spillover effects of FDI tertiary 39 in China 264 enrolments 37 on productivity Taiwan 179, 182 expertise import 223, 386 spillovers 394 institutes, scholarships 355 promotion 335, 369 support and development 324-5 Star Project China 256 training state sector in China 239, 240 public provision 392 State Statistics Bureau, China 267 technicians steel 390, 392 in Korea 145 China 240, 244 in Philippines 278 imports in Japan 83, 86, 88, 91, 92 technological making, Singapore 216 activity 37-42 Stiglitz, IE. IS, 50 risk capital provision 226 Strategic Pragmatism (Schein) 235 agreements, Japan and US and subcontracting 282, 283 Europe 88 Taiwan 175, 184 capabilities 35,44,77,80,83-102 Sub-Saharan Africa 28 domestic 63-7, 76 subsidy programs for R&D 95 in Philippines 269, 280-86 Suh, I 237 in Singapore 231, 235 Suh, Sang Chuh 167 sources 381-94 Sunpino Technology 284 change 383 Index 409 characteristics of Taiwanese firms learning 15 17888 licensing 197 classification 23~5 in China 243 deepening, Singapore 200,201 market, international 48 development 17,376-81 policy after the financial crisis 393-4 in China 239, 240,246-9,254, promotion 217 259-66 requirements, Japan 121, 128, 129, effort, local 285-6 131,134,136 factors on per capita GDP 73-7 spillover 67 infrastructure 390~92 between firms 176-8 knowledge 83, 87, 97 from foreign firms 385 lag 377 Japan 104, 128, 129 learning 16, 17 trade 71 in Korea 145, 152 journals 87 and skill base of economy 340 transfer 21,34,44,67,278,344 national 47 agreements 320,321 technologies China 259, 260-61 new 12-13,15,20,34 and FDI, Japan 137 technology intra-firm 103-142 acquisition, in Taiwan 168-89 Japan 87-9, 89-90, 97, 103, 104 advanced, in Singapore 208 Korea 145, 148, 153, 161 balance in China 265 from MNCs, Singapore 194, 197 alliances between firms. Singapore technopreneurship promotion 219, 197,208-10 220 and competitiveness 12-56,57-80 Tecson, Gwendolyn R. in Japan 103 telecommunications 392 in Thailand 334-73 China 244 contracts in China 257-8 and IT 327 courses, enrolments, Singapore 203 Singapore 211. 216, 230 deployment promotion 230 television sets 152 development 12-49 China 243 by education and training 389 color, Japan 89 in Indonesia 375-95 Temic Telefunken Mieroelectronics 277 in Korea 143-66 tertiary education 37 in Taiwan 168-190 Korea 162 and economic development. Japan manpower from Malaysia, 131,137-40 Singapore 201 in education 37 training institutions, links 224 foreign 58, 63 Texas Instruments, Philippines 149, in Japan 84, 89 277 Taiwan 185 textile companies 28, 386~8 importance of 368 and garments exports 309, 310 imported 62-72, 73, 77 Japan 86,110,141 in Japan 89, 95 products 246, 249, 252, 259, 270 Taiwan 175, 186 Taiwan, exports 169, 173 imports 67-8 Thailand 44, 60 in China 243 education in 65 industrial in Malaysia 305-30 and FDI 334-73 institutes 17 hard disk drive industry 231 intensive activities 174 supplies from 284 410 Competitiveness, FDI and technological activity in East Asia Thailand Asset Management Ulsan Pacific Chemical 150 Corporation 343 United Nations Conference on Trade Thailand Institute of Scientific, and Development (UNCTAD) Industrial and Technological 19,21,42,50,373 Research 338, 357 unemployment 393 Thee, Kian Wie 381,389,392,396 universities in China, R&D 250, Tian'anmen Square 246 258-9 timber 307 university curricula expansion in Timmons, lA. 324, 332 Singapore 224 tobacco 270, 314 engineering courses 20 I Toh, M.H. 232, 238 graduates 390 Tokui, Joi 96, 102 research tool and die making 222 in Japan 99 Torch Project, China 254, 255, 256 in Korea 157-8, 160 Toshiba in Japan 88,98, 149,284 university-industry linkages, Singapore total factor productivity growth (TFP) 197,199 57,58,60-62 Urata, Shujiro 57-67, 80, 82, 129, China 244, 245,246 131,142 Toyota, Japan 97, WI Toyota Motor, Thailand 355 Venables A.I 48, 51 trade Venezuela, technical spillovers 67 and GDP70 venture capital and industry 269 funding for high-tech start-ups 220 inter-firm and intra-firm 42 Korea 156 international 15 Singapore 198, 199,204,205,226 and investment flows 14 Veugelers, Reinhilde 96, 102 in Japan 87,88,91,95,97 videotapes, Korea liberalization 36, 161 Vietnam, challenge to China in textiles shows, Singapore 197, 198 265 Trade Development Board 228 Vijaya, Letchumy 332 trading vocational training, Singapore 222 companies 394 Vogel, Ezra E143, 167 in-house 316 Volkswagen Santana, China 264 institutions 17, 316 on the job 390 Wade, R. 306, 313, 333 programs wafer fabrication plant 326 Japan 128, 129 wages, rising in Japan 106 at overseas affiliates 130, 134~5 Walmart, Taiwan 173 and university linkage 355 Wang, Yueping 261, 262, 264, 267 transnational companies 197 Westphal, Larry E. 15,51,167,173, transport 174, 190, 377, 386 business, Singapore 191,212 White, H. 190 equipment 282, 309 Williamson, G.E. 198, 238 machinery, Japan 110 wireless communications, Singapore transportation 80 211,221 industry, Taiwan 178 Wong, P. K. 191,196,200,201 Tsinghua University, China 260 wooden furniture 388 turnkey projects, Korea 151, 152, worker 159 quality, Malaysia 315 Tybout, James R. 179, 190 skills in Japan 86, 99, 129 Index 411 workforce Xichang Satellite Launch Centre, in China 241, 256, 257 China 241 well-educated 73 World Bank 51, 82, 173, 190,238,283, Yamashita, Shoichi 129, 142 333,396 Yamazawa, I. 80, 82 skill shortages 315 yen study on imports 70 depreciation of 106 World Competitiveness Report, exchange rate 106 technology 208 You, Guangrong 267 World Economic Forum 238, 373 Young A. 57, 82, 231, 238, 326, 333 world trade 20, 42 World Trade Organization 164 Zaibatsu (conglomerates in Japan) World War II 69 92 Japan 83, 86, 98 Zhou, Enlai 242 Wuzhou, China, textiles 259, 260