The Internationalisation of Business R&D

ESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS EVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH NTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT USINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH ESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION USINESS INTERNATIONALISATION DEVELOPMENT RESEARCH NTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS EVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH ESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT USINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION EVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS R NTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOP NTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DEV EVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT The Internationalisation of Business R&D EVIDENCE, IMPACTS AND IMPLICATIONS RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSIN USINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATION ESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIO ESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION USINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION EVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS NTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOP INTERNATIONALISATION BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEARCH DE DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTERNATIONALISATION DEVELOP BUSINESS RESEARCH DEVELOPMENT INTERNATIONALISATION BUSINESS RESEA RESEARCH BUSINESS INTERNATIONALISATION DEVELOPMENT RESEARCH BUSINESS INTE RESEARCH DEVELOPMENT INTERNATIONALISATI BUSINESS INTERNATIONATIONALISATION DEV DEVELOPMENT IN

The Internationalisation of Business R&D EVIDENCE, IMPACTS AND IMPLICATIONS

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT The OECD is a unique forum where the governments of 30 democracies work together to address the economic, social and environmental challenges of globalisation. The OECD is also at the forefront of efforts to understand and to help governments respond to new developments and concerns, such as corporate governance, the information economy and the challenges of an ageing population. The Organisation provides a setting where governments can compare policy experiences, seek answers to common problems, identify good practice and work to co-ordinate domestic and international policies. The OECD member countries are: Australia, Austria, Belgium, Canada, the Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, the Slovak Republic, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States. The Commission of the European Communities takes part in the work of the OECD. OECD Publishing disseminates widely the results of the Organisation s statistics gathering and research on economic, social and environmental issues, as well as the conventions, guidelines and standards agreed by its members. This work is published on the responsibility of the Secretary-General of the OECD. The opinions expressed and arguments employed herein do not necessarily reflect the official views of the Organisation or of the governments of its member countries. Corrigenda to OECD publications may be found on line at: www.oecd.org/publishing/corrigenda. OECD 2008 OECD freely authorises the use, including the photocopy, of this material for private, non-commercial purposes. Permission to photocopy portions of this material for any public use or commercial purpose may be obtained from the Copyright Clearance Center (CCC) at info@copyright.com or the Centre français d'exploitation du droit de copie (CFC) contact@cfcopies.com. All copies must retain the copyright and other proprietary notices in their original forms. All requests for other public or commercial uses of this material or for translation rights should be submitted to rights@oecd.org.

FOREWORD 3 Foreword The OECD Forum on the Internationalisation of R&D of March 2005 recognised the internationalisation of research and development (R&D) as a key dimension of globalisation with a major impact on economic development and public policy and as an extremely complex phenomenon, the importance of which was not clearly understood. Three years on, some aspects of the internationalisation of R&D are better understood, but many questions remain and new ones have arisen. Globalisation is reshaping the international economic landscape and creating increasingly complex relationships between countries and companies. As global competition intensifies and innovation becomes more costly and risky, the business sector is internationalising knowledge-intensive corporate functions, including R&D. This clearly has important implications for policy making, given the role of (business) innovation in OECD countries economic growth. This volume analyses trends in the internationalisation of business research, examines its drivers and motivations and identifies implications for innovation policy. It examines the internationalisation of R&D through foreign direct investment (FDI) by multinational enterprises (MNEs), which account for the bulk of business R&D in the OECD area. It also discusses complementary aspects of the global innovation landscape such as the internationalisation of science, the growing importance of international technology co-operation and the growing international mobility of researchers. The study was prepared by Jean Guinet and Koen De Backer of the OECD Secretariat, in close co-operation with Peter Teirlinck (Belgium) and Arie van der Zwan (the Netherlands). It builds on information in the background report for the Forum on the Internationalisation of R&D and a chapter on the internationalisation of R&D in the OECD Science, Technology and Industry Outlook 2006.

TABLE OF CONTENTS 5 Table of Contents EXECUTIVE SUMMARY... 7 INTRODUCTION... 11 CHAPTER 1. THE ACTIVITIES OF MULTINATIONALS ABROAD... 13 Globalisation on the rise: the emergence of global value chains... 13 The increasing internationalisation of R&D... 17 The role of multinationals in global R&D... 20 Most R&D investments abroad still go to OECD countries... 22 Increased R&D activity by multinationals in emerging countries... 27 Patenting by multinationals abroad... 31 Drivers and motivations of foreign direct investment in R&D... 38 Outward FDI in R&D by emerging countries... 47 Organising and managing R&D internationalisation within MNEs... 51 CHAPTER 2. THE INTERNATIONALISATION OF INNOVATION... 55 The changing landscape for R&D investment... 55 The internationalisation of R&D and technology collaboration... 59 The internationalisation of science... 64 The international mobility of human resources in science and technology... 70 The international exploitation of technology... 78 CHAPTER 3. POLICY IMPLICATIONS... 83 Challenges and opportunities in the internationalisation of business research and development... 83 Policies relating to the internationalisation of R&D by MNEs... 87 Policies to encourage the internationalisation of innovation... 91 Concluding remarks... 98 BIBLIOGRAPHY... 101

EXECUTIVE SUMMARY 7 Executive Summary Because it increasingly affects competitiveness and industrial structure, globalisation is high on the policy agenda in many OECD countries. The globalisation process is currently characterised by the rapid integration of large emerging countries such as China and India, by the increasing international tradability of services, and by a growing specialisation of production in global value chains. The increasing internationalisation of business research and development (R&D) is another important dimension. Until recently, the technological capabilities of firms were far less globalised than activities such as marketing and production, but firms now increasingly offshore R&D activities to other countries. Large multinational enterprises (MNEs) are the main drivers in this process. While internationalisation of R&D is not entirely new, it is now taking place at a much faster pace. Moreover, it is spreading more widely, including to developing countries, and involves more than adapting technology to local conditions. It is linked to changing motivations for outward investment in R&D. Traditional cross-border R&D largely involved adapting products and services to the needs of host countries and to local conditions close to lead users. R&D activities were also undertaken abroad in order to support MNEs local manufacturing operations. Today, MNEs seek not only to exploit knowledge generated at home in other countries, but also to source technology internationally and tap into worldwide centres of knowledge. Intensified global competition has forced companies to innovate and develop commercially viable products and services faster. The knowledge to do so has become more multidisciplinary and more broadly located, making innovation more expensive and riskier. Innovation strategies therefore increasingly depend on global sourcing to sense new market and technology trends worldwide. This has become a major reason for locating R&D outside the home country. MNEs geographic dispersion is also increasingly viewed as a basis of knowledge creation.

8 EXECUTIVE SUMMARY Most R&D investments still go to OECD countries, and the United States is the most important receiving country. However, non-oecd countries have attracted an increasing amount of R&D investment in recent years. Surveys indicate that China and India, among others, are now considered very attractive locations for future investment both because of their large and rapidly growing markets and their large pools of qualified workers and their relatively low, though rising, labour costs. However, they also indicate important drawbacks, such as inadequate enforcement of intellectual property rights (IPR). While these may not deter companies from investing, they may affect the type of R&D undertaken in these countries. Changes in the investment behaviour of MNEs reflect the changing landscape of innovation and the increasingly global supply of science and technology (S&T) resources and capabilities. China and India, for example, have taken their place as important players with a growing capacity for research and innovation. While they lag OECD countries in terms of investment in R&D per capita, their capabilities are already large in absolute terms. The internationalisation of R&D is part of the broader process of internationalisation of innovation. Business R&D has become increasingly internationalised, but so have science, human resources and technology cooperation. Complex policy issues therefore arise, since innovation policy instruments such as R&D support, education and training policies, and infrastructure policies are predominantly national in scope. The challenge for governments is to adjust national policies in light of increasingly international innovation networks. Overall, the internationalisation of a firm s R&D promises substantial benefits: a more cost-efficient innovation process, better ability to learn about R&D conducted by other companies/institutions, a quicker road to commercialisation, and a positive impact on the firm s own innovation capacity. At the same time, many countries are concerned about the possible erosion of home-based R&D and thus a reduced capacity to absorb knowledge and technologies developed abroad. This is particularly true for smaller countries with small markets that may lack the critical mass for research. Some OECD countries and many developing countries fear being marginalised in this way. The policy response to these challenges needs to take account of the current nature of the globalisation process and to build on individual countries strengths. The key elements for action include:

EXECUTIVE SUMMARY 9 Excellent framework conditions. Political stability, public infrastructure, market size and development, tax rates and labour market conditions are key factors in locating R&D. An excellent innovation system based on local strengths. A strong and vibrant research base, effective protection of IPR and a well-trained workforce are major determinants of MNEs investment in R&D and promote the growth of domestic enterprises. Stronger international linkages. This involves supporting the internationalisation of public research organisations, fostering the international mobility of researchers (inward and outward), and linking domestic firms to foreign sources of innovation. It may also require opening R&D funding and programmes in OECD countries to foreign firms and research institutes. Policy coherency. Policy approaches need to be better integrated and more coherent. This involves horizontal co-ordination across various policy areas (education, science and innovation, but also macroeconomic, trade, fiscal, competition, development and employment policies) as well as vertical co-ordination at regional, national and international levels of governance. Effective policies should respond not only to national concerns in terms of attractiveness and competitiveness, but also to global challenges. Turning the internationalisation of R&D into a global win-win situation will require a stronger policy focus on: Encouraging brain circulation and brain connection. Embedding inward and outward foreign direct investment in R&D in the local environment and thus fostering inward and outward spillovers. Enhancing the exploitation of home-based knowledge in developing countries in response to global challenges and development objectives. Strengthening the relevance of international collaboration by focusing on thematic priorities in fields of worldwide importance.

INTRODUCTION 11 Introduction As countries become more interconnected and worldwide flows of information, technology, capital, goods, services and people are expanding a megatrend that is likely to be sustained over the next decade. This is typically regarded as a force for world economic growth, higher living standards and deepened global interdependence. At the same time, it creates many economic, social and environmental challenges. The overall benefits of globalisation are expected to be positive, but they will not necessarily be equally shared. It is generally believed that the greatest benefits will accrue to countries and individuals able to access and adopt new technologies. This risks increasing the distance between those with the ability to access and adopt new technology and those without. Science and innovation are an area in which globalisation is having a large impact. If the Triad (the United States, the European Union, Japan) have so far been the leaders in this respect, countries such as China and India have a growing research and innovation capacity. The scope for global collaboration is increasing as more of the world's regions possess important science and technology capabilities. A related major change is the increasing internationalisation of business R&D, with increasing cross-border flows of R&D and significant shares of domestic R&D performed by affiliates of foreign firms. Companies headquartered in OECD countries now perform increasing amounts of R&D away from their home base as they relocate or expand R&D abroad to benefit from knowledge capabilities in other countries, either in partner companies or in public-sector knowledge infrastructures. The growing complexity of industrial and service sector knowledge bases is forcing companies to adopt global strategies for accessing the knowledge they need. While the internationalisation of R&D is not entirely new, the current phase has three distinguishing characteristics: it is gathering pace, it is spreading to more countries, including developing countries, and it goes beyond adapting technology to local conditions (OECD, 2006a). In the 1980s, R&D investments mainly took place between developed countries through mergers and acquisitions (M&As), but in the 1990s developing

12 INTRODUCTION countries became increasingly attractive locations for R&D investments. The internationalisation of R&D and the redistribution of R&D capabilities are in many cases part of firms broader strategic decisions on production, marketing and M&As. These developments raise complex policy issues. They are not without consequences in terms of competitiveness and employment and raise the question of how the internationalisation of business R&D will influence national economic performance. Issues for policy makers include the possible erosion of home-based R&D and thus a diminished capacity to absorb knowledge and technologies developed elsewhere; a decrease in downstream business activities; and reduced national influence on business decision making. The policy response typically aims to transform the national innovation system so as to embrace the global division of labour in R&D and to specialise in local strengths. This can be achieved by strengthening the local innovation system and by supporting capabilities to acquire and assimilate knowledge and technologies developed elsewhere in the world. This study examines these issues in three chapters. The first analyses the internationalisation of R&D by MNEs, which are considered the main drivers of this process. MNEs increasingly relocate and carry out abroad R&D and other activities higher on the value chain. Changes in R&D investment behaviour are analysed in the context of companies motivations for setting up R&D facilities abroad. Specific attention is paid to the increasing importance of emerging countries, given the increasing attraction of countries such as China, India and Brazil as destinations for R&D. The growing internationalisation of R&D requires profound changes in the management of innovation activities by MNEs, since it has important implications for the role of subsidiaries in recognising and exploiting the potential for innovation. The focus on business should not detract from complementary aspects of this process, such as the internationalisation of science, the growing importance of international technology co-operation and the growing international mobility of researchers. Successful innovative firms are typically part of a system of formal and informal links with other firms, public research institutes, universities and other knowledge-creating bodies. Governments also play a role because policies for R&D, education and infrastructure affect the structure and functioning of innovation systems. These questions are discussed in Chapter 2. Chapter 3 then discusses policy challenges and opportunities presented by the internationalisation of R&D and describes policy initiatives adopted by governments in OECD countries to respond to these challenges.

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 13 Chapter 1 THE ACTIVITIES OF MULTINATIONALS ABROAD Globalisation on the rise: the emergence of global value chains The unprecedented pace and scale of today s globalisation process has attracted much attention in recent years. Since the second half of the 1990s, globalisation has expanded beyond the Triad the United States, the EU and Japan to new global players such as China, India, Brazil and Russia (OECD, 2007a). Because of these countries large populations and lower incomes, their impact on globalisation is significant. China and India, in particular, have enormously increased the global labour supply with workers at a fraction of the cost (and also of the productivity) of more advanced countries (Figure 1.1). Figure 1.1. The spread of globalisation % Population of integrating economies as a ratio of that in advanced countries GDP per capita gap between integrating and advanced economies 250 250 200 200 150 150 100 100 50 50 0 1870 1950 2000 Entry of North America Entry of Japan Entry of China and India and peripheral Europe 0 Source: Maddison (2007) in OECD (2007b).

14 1. THE ACTIVITIES OF MULTINATIONALS ABROAD Trade and foreign direct investment (FDI) are the two main channels of economic globalisation; both have grown faster than GDP over the past decade (Figure 1.2). While economic linkages between countries have long existed, the scale and complexity of transactions have increased substantially (OECD, 2007a). Multinational enterprises (MNEs) are the major driver behind this globalisation process. They have set up large numbers of plants abroad and increasingly export intermediate and final goods between their foreign subsidiaries. The movement of labour across borders has also contributed to international economic integration; foreign workers make up an increasing share of the labour force in most countries. Figure 1.2. Trends in international trade and investment Direct investment (right axis) Trade in services Trade in goods GDP 280 700 260 240 220 200 180 160 140 120 100 1990 1991 1992 1993 Source: OECD (2007a). 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Globalisation has been driven in the first place by the growing openness of the international trading system and sharp reductions in duties on manufactured products. The lowering of trade barriers has been greatest in regional blocs such as the European Union (EU) and the North American Free Trade Area (NAFTA). The gradual lowering of non-tariff barriers has also facilitated international trade of goods and services. The liberalisation of capital movements has further assisted international integration by 600 500 400 300 200 100 0

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 15 gradually eliminating the restrictions on FDI imposed after the Second World War. In addition to lower tariff and non-tariff barriers, drastically reduced transport and communication costs induced by technological progress have also made the integration of markets across borders easier. The decline in these costs has helped reduce economic distances and smoothed economic interaction among countries. Hummels (2007) presents empirical evidence showing that technological changes in air transport and the declining cost of rapid transport have been crucial to today s globalisation process (see Box 1.1). Moreover, advances in computing power and the emergence of the Internet have sharply cut the costs of processing and transmitting information, thereby further facilitating international transactions and trade (OECD, 2007b). Box 1.1. Transport and communication costs have fallen Over the past 50 years, passenger air travel costs, measured by the ratio of airline revenues to miles flown, have been reduced fourfold in real terms. The decline in international communication costs has been even more dramatic. For instance, expressed in 2005 US dollars, the charge for a three-minute New York-London call fell from USD 80 in 1950 to USD 0.23 in 2007. Sea freight (1) Passenger air transport (2) International calls (3) 120 1930 = 100 100 80 60 40 20 0 1930 1940 1950 1960 1970 1980 1990 2000 1. Average international freight charges per tonne. 2. Average airline revenue per passenger mile until 2000. Spliced into US import air passenger fares afterwards. 3. Cost of a three-minute call from New York to London. Source: OECD (2007b).

16 1. THE ACTIVITIES OF MULTINATIONALS ABROAD Technological advances in information and communication technology (ICT) have benefited economic globalisation not only by lowering communication costs but also by enlarging the number of goods and services that can be traded internationally. Improvements in technology, standardisation and infrastructure and the decrease in data transmission costs have all facilitated the sourcing of services from abroad. Rapid advances in ICT have increased the tradability of many service activities and created new kinds of tradable services. In particular, knowledge work, such as data entry and information processing services and research and consultancy services, can easily be carried out via the Internet and e-mail, and through tele- and video-conferencing. Increasingly, activities such as call centres are offshored. ICTs have also made it possible to slice up the value chain and to fragment the production of goods and services globally. Production processes are increasingly geographically distributed, and intermediate and final production is increasingly outsourced abroad. This leads to an increase in exports and imports. Grossman and Rossi-Hansberg (2006) and Baldwin (2006) argue that the current globalisation process increasingly implies trade in individual tasks instead of trade in whole products. In the past, rapidly falling transport costs enabled the unbundling of production and consumption and resulted in growing trade of goods ( the first unbundling ). Until recently, production generally took place in one location, in line with countries comparative advantages, but the strong decline in communication and co-ordination costs has facilitated the spatial distribution of production ( the second unbundling ). Within global value chains, MNEs play a prominent role; their global reach allows them to co-ordinate production and distribution across many countries and to shift activities according to changing demand and cost conditions (OECD, 2007a). The globalisation of value chains is motivated by the search for greater efficiency as growing competition in domestic and international markets forces firms to become more efficient and lower costs. One way of achieving this goal is to source inputs from low-cost or more efficient producers, either domestically or internationally, and either within or outside the firm s boundaries. However, entry into new markets is also important since demographic shifts and rapid growth in several large non-oecd economies mean that an increasing share of global economic activity takes place outside the OECD area. If firms wish to benefit from these growth centres, they need to be present. Finally, firms may wish to gain access to so-called strategic assets, whether skilled workers, technological expertise, the presence of competitors and suppliers, or the possibility of learning from their experience. Tapping into foreign knowledge has become especially important in this context (OECD, 2006a).

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 17 The fragmentation of the production process across various countries has led to the restructuring of firms to include outsourcing and offshoring (Figure 1.3). Outsourcing can be defined as the purchase of intermediate goods and services from outside specialist providers at arm s length; offshoring refers to purchases by firms of intermediate goods and services from foreign providers at arm s length or the transfer of particular tasks within the firm to a foreign location (Kirkegaard, 2004). Offshoring includes both international outsourcing (activities are contracted out to independent third parties abroad) and international insourcing (to foreign affiliates). The crossborder aspect is the distinguishing feature of offshoring, i.e. whether goods and services are sourced within the domestic economy or abroad, not whether they are sourced from within the same firm or from external suppliers. Figure 1.3. Outsourcing and offshoring Location National International Between firms (outsourcing) Domestic outsourcing International outsourcing Sourcing Offshoring Within firms (insourcing) Domestic supply International insourcing Within countries Source: Van Welsum and Vickery (2004). Between countries The increasing internationalisation of R&D Because of its complex and tacit nature, R&D has long been one of the least mobile activities of MNEs. Firms by and large performed R&D and undertook patenting at home. Because much technological knowledge is tacit and embodied in persons and therefore not easily transferable, it is often sticky and linked to a specific location. Because of economies of scale and scope in R&D, the geographical dispersion of R&D also prevented R&D laboratories from reaching critical mass and reduced synergies among a company s various R&D projects. Finally, a company s competitive advantage is often directly related to that of its home country and as such is strongly shaped by that country s industrial specialisations and national innovation systems, including its accumulated research and labour force skills (Pavitt and Patel, 1999).

18 1. THE ACTIVITIES OF MULTINATIONALS ABROAD These centripetal forces seem to be increasingly counterbalanced by centrifugal forces favouring the geographical decentralisation of R&D (Table 1.1). While some R&D has traditionally been undertaken abroad because of demand-driven factors (e.g. proximity to local markets), more supply-driven factors have now become important. The rising cost of R&D (because of its increasingly multidisciplinary character) and the growing global competition in innovation have led to moves to reduce R&D costs while speeding up the development process. This has led firms to rely more on external sources of innovation and a range of strategic instruments such as strategic alliances, mergers and acquisitions (M&As), corporate ventures and R&D outsourcing. Table 1.1. The determinants of R&D internationalisation Centrifugal forces Demand-driven factors: Need for proximity to local customers Need to adapt products to local market Supply-driven factors: Access to highly skilled scientific personnel Proximity to renowned university and private R&D laboratories Proximity to potential partners (customers and suppliers) Centripetal forces Economies of scale and scope in R&D Fear of leakages of key technology High co-ordination and control costs Strong basis in home country comparative strengths and historical inertia Access to low-cost supply of R&D personnel Source: Criscuolo (2005). At the same time, rapid developments in ICT have lowered some barriers to the dispersion of R&D. While economies of scale and scope remain important, greater codification and standardisation of R&D processes have increased possibilities for segmenting and dispersing R&D over a number of locations. Advances in ICT have also facilitated the management of dispersed R&D and the general trend towards outsourcing and offshoring of services. Economies of scale and scope within the home national innovation system have often been considered an important centripetal force. However, the increasing globalisation of science and technology capabilities and the larger number of locations with attractive S&T bases have also widened the possibilities for economies of scale and scope. Integration into local innovation networks and agglomeration have often become more important for R&D than central economies of scale (Cantwell, 1989 2005). Following the fragmentation of the value chain (discussed above) and the resulting inter-

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 19 nationalisation of manufacturing, MNEs now increasingly establish R&D facilities at many locations worldwide. While corporate R&D activities are still carried out predominately in the home country, MNEs are changing how they innovate and building globally distributed R&D (and innovation) networks. Instead of simply adapting technology to local conditions, firms now source knowledge from worldwide centres of knowledge. In the 1960s and 1970s, the tendency was to centralise R&D in companies laboratories, but since the 1980s firms have tended to outsource a larger part of their R&D internationally (i.e. offshoring, see Figure 1.3). Information on the companies that spend most on R&D reveals that, on average, nine out of ten outsource 15% of their R&D; two-thirds of it to other companies and one-third to public research organisations (European Commission, 2005b, p. 6). This clearly indicates that outsourcing of R&D has become an integral complement to in-house R&D as part of companies innovation strategies (Figure 1.4). Figure 1.4. Towards greater outsourcing of business R&D Outsourced 20% 15% 10% 5% 0% 1 st generation 2 nd generation Golden age of corporate R&D labs 3 rd generation 4 th generation During the late 19 th and the early part of the 20 th centuries, practically all research was conducted outside the firm in stand-alone research organizations 1900 1920 1940 1960 1980 2000 Roughly 3% of research is bought outside the firm EIRMA study Importance of innovation networks as source of knowhow Balance between outsourced R&D and in-house capacity Now on a global scale Source: European Commission (2005b).

20 1. THE ACTIVITIES OF MULTINATIONALS ABROAD A recent study for the European Commission analysed the effects of internationalisation of R&D, based on a survey of 158 EU companies which have recently offshored R&D (LTT-Tutkimus Oy, 2007). The results indicated that nearly 70% of the companies had increased their R&D offshoring over the last five years (the base year was 2004) and almost 75% intended to do so in the next five years. While some companies perceived R&D abroad as complementary to domestic R&D, others indicated that internationalised R&D may come at the expense of R&D at home. The main benefits of offshoring R&D were considered to be increased cost efficiency in the innovation process, the ability to learn about R&D conducted by other companies/ institutions, more rapid commercialisation and a positive impact on the firm s innovation capacity. Moreover, 87.3% of respondents estimated that R&D offshoring had a positive effect on the general innovation capacity of EU firms. The role of multinationals in global R&D FDI plays a major role in the internationalisation of R&D, and MNEs are the main actors. More than 95% of the 700 firms with the largest R&D expenditure are MNEs; they account for close to half of the world s total R&D expenditure and more than two-thirds of the world s business R&D (UNCTAD, 2005). The top R&D-performing MNEs often spend more on R&D than many countries and their presence is felt not only through activities in their home countries but also increasingly abroad (European Commission, 2007a). An analysis of the top spenders reveals that over 80% come from five countries: the United States, Japan, Germany, the United Kingdom and France. Only 1% are from emerging countries but their importance is growing, especially the MNEs from China, Korea, Chinese Taipei, Brazil and South Africa. In 2004 expenditure on R&D by the top MNEs grew much more in the rest of the world (+17%) than in the Triad (up around 4%) (UNCTAD, 2005; European Commission, 2005b). The activities of these MNEs are concentrated in a few industries. Information technology (IT) hardware (21.7%); automotive industry (18.0%); pharmaceuticals and biotechnology (17.5%) and the electronics and electrical industry (10.4%) account for over two-thirds of R&D by the top 700 spenders (UNCTAD, 2005). Their industry composition varies considerably by region The EU, the United States and Switzerland specialise strongly in pharmaceuticals and biotechnology. R&D in automobiles and parts is concentrated in the EU and Japan. The United States, Chinese Taipei and Korea specialise in IT hardware. Asian companies (Japan, Korea and Chinese Taipei) are strong in R&D in electronics and electrical equipment. In

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 21 software and computer services, US companies are responsible for more than 85% of global R&D spending (European Commission, 2005b). Recent empirical evidence shows that the top spenders on R&D have increasingly invested in R&D outside their home country. A survey of the largest investors in R&D, undertaken from November 2004 to March 2005, suggests that the pace of internationalisation in R&D is accelerating (UNCTAD, 2005). As many as 69% of the responding firms stated that their share of foreign R&D is set to increase (only 2% indicated a decline and the remaining 29% expected the level of internationalisation to remain unchanged). The momentum appears to be particularly strong among companies in Japan and Korea, which have so far been less aggressive in this respect. Nine out of ten Japanese firms in the sample and about 80% of Korean firms planned to increase their foreign R&D; 61% of the European firms indicated similar intentions. The average firm in the UNCTAD survey spent 28% of its R&D budget abroad in 2003, including in-house expenditure by foreign affiliates and extramural spending on R&D contracted to other countries. These survey results are confirmed by more systematically gathered data on outward R&D investment of MNEs. For countries for which data on outward investment are available in the OECD Activities of Foreign Affiliates (AFA) database, R&D performed abroad has increased since 1995 relative to R&D performed at home (Figure 1.5). The only exception is Switzerland which has seen a slight decline, but Swiss affiliates abroad do as much research as all firms inside Switzerland. The share of R&D investments abroad is smaller in other countries but still over 20% in Germany, Finland and Sweden. However, some authors have pointed out that R&D establishments abroad may be acquired incidentally through mergers and acquisitions. Ronstadt (1978) and Håkanson and Nobel (1993) noted that many R&D investments were the result of acquisitions by the parent company which did not have access to R&D as their primary objective. Assessing the importance of such incidental internationalisation of R&D is difficult as data on R&D facilities abroad that distinguish between M&As and greenfield investment are not readily available.

22 1. THE ACTIVITIES OF MULTINATIONALS ABROAD Figure 1.5. Business sector R&D expenditure by affiliates abroad as a percentage of domestic R&D in selected R&D countries % 1995 2005 120 100 80 60 40 20 0 Switzerland (1,2) Sweden Finland (3) Germany Belgium United States Japan (2,5) Italy (3,4) 1. 1996. 2. 2004. 3. 1998. 4. 2003. 5. 1997. Source: OECD, AFA database, January 2008. Most R&D investments abroad still go to OECD countries In the OECD area, the largest cross-border flows of R&D take place between the three main regions: the United States, the European Union (here EU15) and Japan. Figure 1.6 shows that in 2005, US multinationals placed over USD 17 billion of their foreign investment in R&D in the European Union and USD 1.7 billion in Japan. The European Union invested USD 19.1 billion in the United States and USD 3.9 billion in Japan. The United States was a net exporter of R&D to the EU in the late 1990s, but in the early 2000s European firms established more foreign R&D affiliates in the United States. Japan invested only USD 1.1 billion in the United States and USD 0.7 billion in the EU.

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 23 Figure 1.6. R&D flows between EU15, the United States and Japan, 2005* Millions of PPP USD 1 115 USA Total business sector 225 984 19 112 733 EU Total business sector 140 231 17 044 3 925 Japan Total business sector 93 062 1 762 * Figures for Japan are for 2004. Source: OECD, AFA database, January 2008. A survey of 209 R&D-intensive MNEs confirms that Japanese MNEs are less likely to set up R&D facilities abroad. Table 1.2 shows that the R&D budget spent outside Japan by Japanese MNEs is a fraction of those of EU and US MNEs. Other studies have found that EU-based firms are more inclined to generate technological knowledge abroad and to engage in international R&D activities than companies in the United States, which in turn are more internationally active than those in Japan (Roberts, 1995a, 1995b; von Zedtwitz and Gassmann, 2002). MNEs from smaller advanced countries with a limited domestic market and S&T base internationalise R&D faster. Table 1.2. Share of R&D budget spent outside the home country 1995 1998 2001 2004 (estimated) Western Europe 25.7 30.3 33.4 43.7 Japan 4.7 7.0 10.5 14.6 North America 23.2 28.4 31.7 35.1 Note: Based on a survey of 209 MNEs. The geographic zones refer to the origin of the MNEs. Source: Reger (2002).

24 1. THE ACTIVITIES OF MULTINATIONALS ABROAD European companies have frequently established foreign R&D outposts, often through the acquisition of companies. Japanese companies have largely relied on R&D at home and licensing and listening posts to acquire technology rather than on R&D abroad. US companies have tended to support new business opportunities by setting up overseas R&D units staffed with US researchers (von Zedtwitz and Gassmann, 2002). MNEs cross-border flows of R&D tend to be concentrated in certain sectors. Industries characterised by greater technological complexity (computers, scientific instruments, aeronautics, motor vehicles) tend to keep more of their technological activities at home, while more innovative activities take place outside the home base in traditional sectors (drink and tobacco, food, building materials, other transport, mining and petroleum industries). Pharmaceutical and medical firms also have an above-average engagement in international innovation. European R&D investment in the United States is mainly in the chemical and pharmaceutical industry (50%), computers and electronics (13%) and petroleum distribution (10 %). Investment by US MNEs in the European Union is largely in three sectors: automobile (33%), pharmaceuticals (26%) and computers and electronics (14%). Japanese R&D investments in the United States are concentrated in services (69%), especially in wholesale trade and professional/scientific services, rather than in manufacturing (31%). US MNEs R&D investment in Japan is principally in pharmaceuticals (63%) and computers (20 %) (OECD, 2006a). Table 1.3 combines outward R&D investments by MNEs of the larger OECD countries in various countries with their counterpart inward R&D investments to show where the flows between countries are concentrated. It indicates that the United States receives the largest share of R&D investments from other large OECD countries. A significant part of the outward R&D investments of US MNEs goes increasingly outside the OECD area to emerging countries (see also below). However, these overall geographical patterns may conceal important differences among industries.

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 25 Table 1.3. Share of R&D expenditures of foreign affiliates abroad by country of destination, 2003 Country of origin United States Japan Germany France United Kingdom United States 47% 69% 35% 63% France 9% 5% 10% 2% United Kingdom 18% 9% 5% 16% Country of destination Japan 8% 4% 20% 2% Italy 4% 2% 3% 2% 2% Belgium 2% 3% 2% 4% 2% The Netherlands 3% 8% 1% 2% 2% Germany 19% 5% 18% 11% Sweden 4% 0% 0% 0% 15% Other 33% 19% 2% 1% 1% Total 100% 100% 100% 100% 100% Source: OECD, AFA database, January 2008. The rise in MNEs R&D investments abroad has raised the importance of foreign affiliates in host countries R&D. Between 1995 and 2005, R&D expenditure by foreign-controlled affiliates in OECD countries rose by USD 46.6 billion in purchasing power parity (PPP). Although the share of the United States decreased between 1995 and 2005, it continues to attract 37.9% of R&D expenditure by foreign affiliates in the OECD area. Other countries that attract important R&D investments are Germany, the United Kingdom and, to a lesser extent, Japan, France and Canada (Figure 1.7). The three largest EU R&D performers (Germany, the United Kingdom and France) together attract 38% of foreign R&D investments in the OECD area.

26 1. THE ACTIVITIES OF MULTINATIONALS ABROAD Figure 1.7. R&D expenditure under foreign control in the business sector in selected OECD countries, 1995 and 2005* PPP USD Billion USD 90 85 83.6 billion USD 80 75 70 65 60 37.9% United States 55 50 45 40 35 30 25 47.4% 37.0 billion USD 15.8% 11.4% 6.2% Germany United Kingdom Japan 20 15 10 5 0 12.8% 11.5% 3.7% 10.1% 4.7% 9.8% 10.8% 3.7% 14.1% 1995 2005 France Canada Other OECD (1) * Figures for Japan are for 2004. 1. The Czech Republic, Finland, Hungary, Ireland, Poland, the Netherlands and Sweden. Source: OECD, AFA database, January 2008. The growing role of foreign MNEs in host countries R&D has raised concerns about the dependency and vulnerability of the local R&D base (OECD, 2006a). Such concerns are greater in countries such as Ireland and Hungary where the ratio of R&D expenditure to turnover is higher in foreign affiliates than in domestically controlled firms, an indication of these firms relative lack of investment in R&D (Figure 1.8). There is some evidence that firms in these countries tend to buy the bulk of their technology abroad rather than develop it at home (see also below).

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 27 Figure 1.8. Share of affiliates under foreign control in total business sector R&D expenditures, 1995 and 2005 1995 2005 % 80 70 60 Japan Greece United States Finland Turkey Slovak Republic Italy Spain Germany Netherlands Poland France Portugal Canada United Kingdom Australia Sweden Austria Czech Republic Belgium Hungary Ireland Note: Czech Republic: 1996; Finland, Hungary, Netherlands, Turkey: 1997; Portugal: 1999; Hungary: 2003; Austria, Canada, Italy, Japan, Netherlands: 2004. Source: OECD, AFA database, January 2008. Increased R&D activity by multinationals in emerging countries While most internationalisation of R&D by MNEs still takes place within the main OECD regions, emerging countries increasingly attract R&D investments although these remain relatively small in a global perspective. A 2007 study by the Economist Intelligence Unit of more than 300 senior executives identified India (26% of respondents), the United States (22%) and China (14%) as the most attractive overseas locations for R&D. The Asia-Pacific region, in particular, is expected to attract more offshore R&D over the next three years: 30% of respondents planned a substantial increase in their investment there. 50 40 30 20 10 0

28 1. THE ACTIVITIES OF MULTINATIONALS ABROAD Other surveys on the location of R&D centres by MNEs provide similar results. In the UNCTAD survey of the largest R&D spenders worldwide, China ranked third and India sixth as current locations for R&D (Figure 1.9). Singapore and Brazil also appeared high in the ranking. Likewise, information on new greenfield and expansion FDI projects involving R&D over the period 2002-04 reveals that of the 1 773 projects identified, 1 095 were undertaken in developing countries, eastern Europe and the Commonwealth of Independent States (CIS) (LOCO-monitor of OCO-consulting, cited in UNCTAD, 2005). 70 Figure 1.9. Current foreign R&D locations % of responses OECD country Non-OECD economy 60 50 40 30 20 10 0 United States United Kingdom China Source: UNCTAD (2005) in OECD (2006a). France Japan India Canada Germany Singapore Italy Brazil Spain Belgium Sweden Switzerland Australia Finland Norway Russian Federation Netherlands Ireland Poland Chinese Taipei Austria Israel Korea Thailand

1. THE ACTIVITIES OF MULTINATIONALS ABROAD 29 These survey results are confirmed by official data for the United States, one of the few countries to have recently published detailed information. Table 1.4 shows the main trends in the geographical distribution of US R&D investment abroad. The main change between 1995 and 2005 is the decline in the European Union as a destination and the increased importance of Asia-Pacific, especially China. The pattern of investment in other geographical zones has not changed significantly. In spite of the relative decline in Europe s share, it continues to attract over 60% of US MNEs R&D investment. Latin America, Eastern Europe, the Middle East and Australia together attract only 8.5% of total US R&D investment. The decline in US R&D investment in Europe mainly concerns Germany and France; investment in the United Kingdom and Sweden doubled in value. Table 1.4. R&D expenditure of affiliates of US parent companies abroad by country or zone of destination Percentage of total expenditure 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Canada 8.5 11.1 12.5 11.9 9.3 11.4 10.8 10.8 10.7 10.6 10.2 European Union 1 70.4 66.9 66.4 68.6 65.6 61.0 58.8 61.4 62.5 62.1 61.0 Eastern Europe 2 0.1 0.3 0.3 0.5 0.3 0.4 0.2 0.3 0.3.. 0.6 Latin America 3.1 3.9 4.5 5.1 3.4 3.2 2.9 3.7 3.0 2.8 3.2 of which Brazil 2.0 2.5 3.0 3.0 1.6 1.2 1.0 1.4 1.4 1.4 1.5 Africa 0.2 0.1 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Middle East 0.8 1.2 1.4 1.0 2.1 3.1 3.7 3.5 3.0 3.4 3.7 Asia-Pacific 14.8 14.8 12.8 10.9 17.8 19.2 21.3 18.0 17.8 17.8 18.2 of which Japan 10.2 9.5 7.5 6.6 8.4 8.0 7.6 7.3 7.2 6.3 6.2 China 0.1 0.2 0.2 0.4 1.8 2.5.. 3.1 2.5 2.2 2.5 Australia 2.3 2.9 2.5 2.0 1.6 1.7 1.5 1.5 1.8 1.8 1.8 Total 100 100 100 100 100 100 100 100 100 100 100 Total in billion USD 12582 14039 14593 14664 18144 20457 19702 21063 22793 25840 28316 1. EU15 up to 2003, EU25 from 2004, 2. From 1999, Eastern Europe only includes the Czech Republic, Hungary, Poland and Russia. Source: OECD, AFA database, January 2008. The large increases in foreign R&D investment in developing Asia and particularly in China and India have attracted much attention in recent years. According to official Chinese statistics, some 750 foreign R&D centres had been established in China by the end of 2004, most of them after 2001. Other surveys and studies present lower figures (e.g. OECD, 2007d). Japanesebased manufacturing companies increasingly base their foreign R&D centres