VATT-TUTKIMUKSIA 110 VATT RESEARCH REPORTS

Reino Hjerppe and Jaakko Kiander (eds.)

TECHNOLOGY POLICY AND KNOWLEDGE-BASED GROWTH IN SMALL COUNTRIES

Valtion taloudellinen tutkimuskeskus Government Institute for Economic Research Helsinki 2004

Based on contributions by: Cees van Beers, Elina Berghäll, Joe Cogan, Juha Kilponen, Vladimir Lavrac, James McDevitt, Mihaly Simai, Peter Stanovnik and Lena Tsipouri.

ISBN 951-561-495-3 ISSN 0788-5008 Valtion taloudellinen tutkimuskeskus Government Institute for Economic Research Arkadiankatu 7, 00100 Helsinki, Finland Email: first [email protected]

Oy Nord Print Ab Helsinki, May 2004

Technology Policy and Knowledge-Based Growth in Small Countries

KNOGG Deliverable 5 May 2004 Reino Hjerppe and Jaakko Kiander (eds.)∗

Project No: STPN-2000-00104 Project Co-ordinator: VATT

Funded by the European Commission FP5 Improving Human Potential and the Socio-Economic Knowledge Base, IHP-1999-5.1

∗

Government Institute for Economic Research, Helsinki, Finland.

The KNOGG Consortium Government Institute for Economic Research (VATT), co-ordinator, Finland Technische Universiteit Delft (TUD), The Netherlands Institute for World Economics, Hungarian Academy of Sciences (IWE), Hungary National University of Ireland (NUID), Ireland Institute for Economic Research (IECOR), Slovenia National and Kapodistrian University of Athens (UAT), Greece

HJERPPE, REINO – KIANDER, JAAKKO (eds.): TECHNOLOGY POLICY AND KNOWLEDGE-BASED GROWTH IN SMALL COUNTRIES. Helsinki, VATT, Valtion taloudellinen tutkimuskeskus, Government Institute for Economic Research, 2004, (B, ISSN 0788-5008, No 110). ISBN 951-561-495-3. Abstract: The rapid diffusion of technology is a prerequisite for small countries to catch-up or remain at the fore-front of the technology race. The critical issue is how innovation policies can be framed so that they encourage the diffusion of technology and consequently upgrade the domestic technology structure of the host country. The range of available policy options to a small country as host to foreign R&D intensive trans-national companies (TNC) is defined by the way the country’s productive capacities can link to the TNCs global production structure. These abilities and the degree and quality of the integration can be improved with the help of foreign direct investment (FDI) and science, technology and innovation (STI) policies. While financial incentives aimed at attracting R&D intensive FDI may be necessary for a small country, such schemes need to be supplemented with policies that stimulate learning and investments in domestic firms. R&D intensive TNCs do not necessarily upgrade the technological knowledge base since high-tech industries also incorporate low-tech phases in their production processes. It is particularly important to strengthen the technological capacities of local firms to learn from foreign TNCs and to create knowledge spillovers. Key words: Small countries, knowledge-based growth, national innovation systems, foreign direct investments

Tiivistelmä: Nopea uuden teknologian omaksuminen on edellytys sille, että pienet maat voivat päästä teknologisen kehityksen eturintamaan ja pysyä siellä. Keskeistä on se, miten innovaatiopolitiikalla edistetään teknologian siirtoa ja parannetaan teknologisia valmiuksia. Pienen maan mahdollisuudet toimia tutkimusja tuotekehitysintensiivisten monikansallisten yritysten isäntämaina määräytyvät sen perusteella miten hyvin maan tuotantokapasiteetti soveltuu yritysten tuotantorakenteeseen ja alueelliseen työnjakoon. Näitä ominaisuuksia sekä integraation astetta ja laatua voidaan parantaa tiede-, teknologia- ja innovaatiopolitiikalla ja suoria ulkomaisia sijoituksia suosivan politiikan avulla. Ulkomaiset yritykset eivät välttämättä kohota pienten maiden teknologista osaamista, koska myös korkean teknologian yritysten tuotantoon liittyy matalaa teknologiaosaamista edellyttäviä prosesseja. Kotimaisten yritysten teknologinen osaaminen, oppiminen ulkomaisilta yrityksiltä sekä teknologian leviämisen edistäminen ovat keskeisiä menestyksen edellytyksiä. Asiasanat: Pieni kansantalous, osaamisperusteinen kasvu, kansallinen innovaatiojärjestelmä, suorat sijoitukset ulkomailta

Foreword This report is the fifth deliverable of the EU financed research project “Knowledge, Growth and Globalisation – Science and Technology Policy as a Growth Factor in Smaller Economies” (KNOGG), which focuses on the experiences of small European countries. The KNOGG partners consist of research units from Finland, Greece, Hungary, Ireland, the Netherlands and Slovenia. The first report of the KNOGG project “The Role of Science and Technology Policy in Small Economies” was published in the VATT Research Reports series with the number 91 in 2002, the second “Science, Technology and Innovation Policies in Selected Small European Countries” with the number 96, the third “The Role of Foreign Direct Investments on Small Countries’ Competitive and Technological Position” with the number 100 and the fourth “Practical Guide for Active National Policy Makers – What Science and Technology Policy Can and Cannot Do?” with the number 103, all in 2003. This survey concludes the KNOGG thematic network working process. It discusses the best solutions and the best practices in the field of science, technology and innovation policies from the point of view of small European economies. We believe this report concerns and discusses most of the crucial elements for small economies and different regions in the European Union, in the process of creating and constructing the European Research Area.

Helsinki, May 2004

Reino Hjerppe

Acknowledgements This report was edited for the EU financed thematic network called KNOGG “Knowledge, Growth and Globalisation, The Role of Science and Technology Policy in Smaller Economies” by Reino Hjerppe and Jaakko Kiander from the Government Institute for Economic Research (VATT) in Finland. The editors have benefited from supporting material and comments from the other partners of the KNOGG project, particularly Paavo Löppönen from the Academy of Finland, Elina Berghäll, Tuomo Heikkilä and Juha Kilponen from the Government Institute for Economic Research (VATT) in Finland, Cees van Beers and Tom Poot from Delft University of Technology (TUD) in the Netherlands, Mihaly Simai from the Institute for World Economy in the Hungarian Academy of Sciences (IWE) in Budapest, Joe Cogan and Jim McDevitt from the National University of Ireland (NUID), Vladimir Lavrac and Peter Stanovnik from the Institute for Economic Research (IECOR) in Slovenia and Lena Tsipouri from the University of Athens (UAT), Greece. Our special thanks belong to all speakers, commentators and participants including experts in the European Commission in numerous workshops and seminars arranged in all participating countries and in Brussels during the work.

i

Executive summary Technological development in an open economy can either be based primarily on imitation or on domestic knowledge creation. Large countries enjoy a criticalmass advantage in innovating with their own research, technology and development (RTD) investments. The innovation growth path is predicated upon a superior level of technological knowledge, derived from basic research, and a commensurate stock of human capital. Yet, publication and patent statistics reveal many small countries to have been innovators in important fields of science and technology, such as microelectronics, pharmaceuticals, biotechnology, and material sciences. Due to globalisation and increased mobility of factors of production, small countries may be discouraged from increasing their own R&D efforts in knowledge creation and generation of new technologies, and focus more on absorption capacities and commercialisation of innovations produced elsewhere. The alternative “imitation” growth path, for economies without an adequate knowledge base, is not self-sustaining and requires external policy intervention to close the technology gap. Imitation entails relying on absorbing existing technologies, which are imitated or reverse engineering to produce existing products (or upgraded versions of them) for competitive markets. In reality, all countries have elements of both alternatives. The KNOGG framework demonstrates this choice. Both ways require an adequate pool of human and social capital (the knowledge base). Many accession countries are heirs to a strong tradition in science and basic research but struggle to compensate for a deficit in technological universities and industry-oriented research institutes, which could bridge the gap between research output and commercialisation. On the basis of the main characteristics, goals and instruments of science, technology and innovation (STI) policies in the KNOGG countries, four socalled policy patterns were identified deriving from an interventionist – noninterventionist regime to the KNOGG framework dichotomy of imitation vs. innovation. (1) Innovation-intervention favours the active promotion of technological diffusion e.g., through foreign direct investment (FDI), imports of technology intensive products, networking, and cooperation. Ireland and to a lesser extent, Hungary are well-known successful examples. As with interventionist measures, in general, policy harmonisation within the EU and other international organisation constrain the scope for adopting discriminatory measures and using subsidies.

ii (2) In the second interventionist policy pattern, the country invests heavily in various aspects of its national innovation system to create the capacity for homegrown innovation to complement and improve absorptive capacities vis-à-vis technology imports. Serious investments in the education system, the research community, bridging institutions, and R&D subsidies are needed with a longterm commitment. Despite the obvious risks, this presents an option to countries such as Finland, which have encountered serious difficulties in attracting large amounts of FDI. Public R&D expenditure is exceptionally high also in the Netherlands, but the country differs in regard to FDI. (3&4)The third and fourth policy patterns pertain to countries that are unwilling or unable to improve their developments by public policy measures. Reliance on market-based technology diffusion may be the only alternative for a small country without sufficient resources to establish large-scale support programs and without the political willingness to improve the incentives for inward FDI. However, without any specific policies, it is difficult to see how knowledgebased activities would locate in a country that is on the periphery or has a small market. Greece and Slovenia fall closest to these categories, but mainly because of the small scale or ineffectiveness of their policies, rather than due to an absence of efforts. The KNOGG framework clearly points to the alternative “imitation” growth path for Hungary, Slovenia, Greece, Ireland and, to a more limited extent for the Netherlands. These economies could leverage the potential of their STI policies to make the transition to the higher path of 'innovation' growth. Explicit policies must be put in place to acquire and absorb technology from external as well as internal sources. The role of bridging institutions and intermediaries is critical. A relatively high proportion of the industry-oriented segment of the national STI budget should be allocated to knowledge diffusion measures. There is a canon of STI wisdom that can be applied without attempting to homogenise national innovation systems: • STI must have a strong voice at Cabinet • Major stakeholders must have a voice in national STI Councils • Co-ordination across Ministries is vital • Policy implementation must be separated from policy making • Implementation must be close to the needs of clients (targeted to specific categories of firms, regions, technologies etc) • Regular evaluations leading to corrective action are necessary.

iii The institutional advantage of small countries can be found in the close contacts of decision-makers and facility of knowledge flows. As argued in WP1 (Berghäll et al. 2002), an imperative for them is flexibility and rapid adjustment to global trends and shifts in the business cycle. This requires small countries to implement institutions and systems which are flexible to reform and adapt with ease to new conditions and demands placed on them. At the same time, stable and predictable operating conditions in the long run are a prerequisite to reduce uncertainty and risks for the enterprise sector. The outcome of such balancing acts and policy reform are, to a large extent, determined by political will and long-term commitment. The rapid diffusion of technology is a prerequisite for small countries to catch-up or to remain at the fore-front of the technology race. The critical issue is how innovation policies can be framed so that they encourage the diffusion of technology and consequently upgrade the domestic technology structure of the host country. The range of available policy options to a small country as host to foreign R&D-intensive TNCs is defined by the way the country’s productive capacities can link to the TNCs global production structure. These abilities and the degree and quality of the integration can be improved with the help of FDI and STI policies. While financial incentives aimed at attracting R&D intensive FDI may be necessary for a small country, such schemes need to be supplemented with policies that stimulate learning and investments in domestic firms. R&D intensive trans-national companies (TNCs) do not necessarily upgrade the technological knowledge base since high-tech industries also incorporate low-tech phases in their production processes. It is particularly important to strengthen the technological absorptive capacities of local firms to learn from foreign TNCs and to create knowledge spillovers. Such fertile conditions would most likely result in superior results relative to targeted industrial policy of picking winners or stimulation of selected high technology sectors. Even if successful in attracting large foreign TNCs or building their own, small countries need to constantly keep building, diversifying and broadening the R&D base of their small and medium sized enterprises (SMEs). Focusing on the large TNCs should not entail the neglect of local SMEs that have difficulties in following the pace of technological progress. In other words, FDI alone is not likely to guarantee a long-term solution. This is one of our main arguments, i.e., to generate effective technological diffusion, an efficient national innovation system and stimulation of indigenous technological progress are required. Domestic firms are important factors to attract R&D intensive strategic asset seeking TNCs. Countries with strong domestic technological capabilities, such as Finland, are better positioned to attract strategic asset seeking TNC affiliates with R&D facilities than countries, such as Ireland or the Netherlands, with lower levels of technological activity. In addition to the promotion of indigenous technological progress through

iv technology policies aimed at building or improving the national innovation system, another central course of action is the encouragement of cross-border R&D cooperation, such as inter-country research programmes within the EU (e.g. ESPRIT). The discussions about the European Research Area (ERA) strategy of promoting the emergence of European poles of excellence have highlighted the risks of marginalisation of small and peripheral countries in science and technology. It is particularly important for these countries that there remain some room for national initiatives and priorities in STI policy. First, there is a well known pervasive tendency for all factors of production, including physical and human capital, to congregate together and concentrate all economic activity on all geographical levels; globally, nationally, regionally and for metropolitan areas, as well as for ethnic groups. In an integrated market, these forces are strong enough to operate even without the support of the EU Commission STI policy. The larger European countries and urban centres may gain disproportionately from the reallocation of research resources, notably of high skilled research jobs, without major acceleration of technological progress. Second, while fragmentation may be reduced, so will technological diversity leading to a dwindling of the resource pool, as Matthews and McGowan (1992) have argued; “Any tendency to eliminate technological diversity represents a diminution of the technological resource pool, even if, via such processes as gains from economies of scale, short term efficiency is increased by convergence”. In science and technology, Europe enjoys a strong historical tradition and firmly founded institutions. While reforms towards the American system may be necessary, the reform process should not be carried out by starting with the dismantling of the national strengths of the member countries, but on building on them. Undoubtedly, the concentration of resources around new poles of excellence can allow for the creation of the critical mass necessary to compete internationally in high fixed cost industries. But to achieve this, the market-based approach may be the best. This would entail the increased use of competitive technological and research procurement. While funding is central for producing research and development results, it is more of a push factor without a clear direction. Meanwhile, demand allocated funding has been crucial to breakthroughs made in creating many radical and complex innovations and general purpose technologies. The US enjoys the defence industry advantage as a technology pull factor. Demand led technology policy may eventually lead to the build-up of large centres of excellence, but they would be forced to be competitive due to the challenges posed to them by competition.

v In the new STI landscape, it is important for small countries and countries far away from the economic core of the EU to foster their own research platforms and poles of excellence, not only to maintain cohesion and to avoid brain drain, but also to supply Europe with a network of variety and diversity from which innovations emerge. Small countries (and regions) should be supported in their efforts to build on local strengths and create local centres of excellence as nodes in the ERA network. Within the countries, this may require reallocation, restructuring and relocation, as well as higher public investment in the promotion of research and technological development. Better databases are needed to facilitate quantitative evaluations of STI and enterprise policies, as well as more emphasis on monitoring and evaluation. The quality and productivity of R&D hinges on it. Policy-makers need to know the impacts of R&D subsidies in order to target them effectively. In fact, better databases and access to data for empirical analysis are required almost across the scale in R&D in order to create a truly knowledge-based economy guided by informed policy-makers. At the European level, the provision of reliable comparable data would benefit research and policy comparison efforts. Further research using more comprehensive European data sets is needed to clarify our view of the state of European research, innovation systems and entrepreneurship. We need a better understanding of knowledge flows between universities, research institutes and firms, and a better understanding of the role of entrepreneurship and of the factors affecting it.

Contents

1. Introduction 1.1 Small country issues 1.2 The growing importance of knowledge 1.3 The role of Community level RTD policy 2. Analytical framework 2.1 Knowledge-based growth 2.2 Small country constraints 2.3 The KNOGG framework 2.4 National innovation systems and alternative policy patterns 3. Comparative analysis of KNOGG countries – Innovation systems and Industry-oriented STI policies 3.1 Introduction 3.2 Innovation Systems in KNOGG countries 3.2.1 Finland 3.2.2 The Netherlands 3.2.3 Ireland 3.2.4 Hungary and Slovenia 3.2.5 Greece 3.3 Industry-oriented innovation policies in KNOGG countries 3.3.1 Finland 3.3.2 The Netherlands 3.3.3 Ireland 3.3.4 Greece 3.3.5 Hungary 3.3.6 Slovenia 3.4 Policy Implications 3.4.1 Innovation systems 3.4.2 Industry-oriented innovation policies 3.4.3 The Trend Chart on Innovation 3.5 Using the Innovation Trend Chart

1 2 4 6 7 7 9 14 17 23 23 24 24 25 26 26 27 27 28 30 32 34 37 42 46 46 47 48 50

4. Small countries, FDI and multinationals 4.1 Motives for the internationalisation of R&D 4.2 FDI related policies in KNOGG countries 4.3 Outward FDI 5. STI policy instruments on EU level 5.1 General objectives set by the EU 5.2 The EU research programmes 5.3 Trend in SME Participation in Framework Programmes (FP 3 to FP 6) 5.3.1 SME Participation in FP 3 (1990–1994) and FP 4 (1994–1998) 5.3.2 SME Participation in FP 5 (1996–2002) 5.3.3 SME Participation in FP 6 (2002–2006) 5.3.4 Evaluation of SME Participation 5.3.5 Barriers to SME Participation in KNOGG Countries 5.3.6.Policy Implications 5.4 Other EU policies and knowledge-based growth 5.4.1 Enhancement of RTDI expenditure by the Structural Funds 5.5 ERA and small countries 6. Knowledge-based growth and other policy areas 6.1 Focus on the firm 6.2 Role of innovation policy 6.3 Better understanding of interfaces with other policy areas 7. STI policy and knowledge-based growth: guidelines 7.1 National level: The role of national STI policies in small countries 7.2 The role of Community level STI policy 8. Conclusions and suggestions for further research 8.1 Conclusions 8.2 Suggestions for future research

52 54 56 62 63 63 64 65 67 68 70 71 71 72 74 76 83 87 87 90 91 94 94 96 98 98 103

References

105

Glossary

110

1

1. Introduction KNOGG, an EU financed thematic network active in 2001–2003, studied the role of STI policies in driving economic growth in six small European countries. The project aimed to develop guidelines at the EU level for improving knowledgebased growth in small European economies. The project KNOGG was an interdisciplinary research project on the role of knowledge as a central growth factor in the era of globalisation and regional integration. It emphasised the specific characteristics and problems of small economies and aimed to find policy strategies to improve their innovative capabilities in order to improve the competitiveness of Europe. This report is the summary of the findings of the project. The report contains policy recommendations based on the experiences of small countries. The participants of the project KNOGG represented six small European countries: the Netherlands, Greece, Hungary, Finland, Ireland and Slovenia. The participants of the project represented current member countries which have managed to create success stories based on applied science and new technology on one hand, and a group of accession countries with strong traditions in skillintensive production on the other. Two of the KNOGG countries are accession countries (Hungary and Slovenia), two are or have been cohesion countries (Ireland and Greece), and one was, in the latter half of the 1990s, an exponent of ICT-driven growth (Finland). Only one of the countries was centrally located in the core of Western Europe (the Netherlands). In the main, the project focused on the fundamental factors that would enable countries in the periphery to catch up and converge with those in the centre. The periphery in this context includes eastern enlargement countries. The first report of KNOGG (Berghäll et al, 2002) set the stage for the project by providing a selective survey of the current literature on economic growth from the point of view of small open economies. Based on New Growth Theory and a Schumpeterian view of innovation, the report developed a conceptual framework for the project. The constraints imposed by country size and the availability of resources were highlighted, as well as the implications of these constraints for knowledge-based growth strategies. It was concluded that although STI policies in small countries may be predominantly knowledge-diffusion oriented, sole reliance on imitation strategies and foreign spillovers may lead to underinvestment in domestic R&D and to difficulties in absorbing new technologies. The historic experience of the KNOGG countries varies enormously as does their success in the face of the constraints of size and resource limitation. A study of their development experiences provides very interesting lessons for lagging economies in particular.

2 The second report of the project (Cogan and McDevitt, 2003) described and interpreted the current STI policy regimes and their framework conditions in the six KNOGG countries. It used the EU Innovation Trend Chart database and the associated benchmark and country reports, as a point of departure and extended the EU data to include national research systems, as well as institutions critical to the formulation and implementation of policy. This report also contains empirical findings on the quantitative and qualitative differences in the resources allocated to industry support measures in KNOGG countries. As a result it highlights the differences between small countries pursuing a self-sustaining innovationoriented growth strategy and those following the technology-diffusion or catchup path to economic growth. The third KNOGG report (van Beers, 2003) analysed the role of Transnational Corporations (TNCs) in the KNOGG countries and demonstrated their role in generating spillovers. The efficiency of the national innovation system in the host country was found to be critical in strengthening the technological advantages of local firms and in enabling them to absorb successfully technology from foreign TNCs. The fourth report (Simai, 2003) examined the role of science, technology and innovation policies from the perspective of the national policy process, and commented on the challenges Europe faces in making progress toward a knowledge-based society and economy. Three different policy patterns (or heuristic models) were postulated in the light of KNOGG country experiences. In this fifth and concluding report, the findings of project KNOGG are summarised and their implications for science, technology and innovation policies are highlighted.

1.1 Small country issues Small countries are different from large countries in dimensions other than just scale. The size of their home markets means that they cannot benefit from the economies of scale unless they have ready access to export markets. In order to achieve the capability to compete in world markets they have to specialize in a limited number of sectors and this increases their vulnerability to so called asymmetric shocks, i.e. to sector-specific downturns. In addition to specifically small-country threats, location has become an important growth factor due to the changing role of knowledge in economic competition. There are many prosperous and dynamic regions in Europe which already have achieved the status of being the most competitive in the world in their fields of specialization. Such regions are at the cutting edge of many areas of science and high-tech industries, and they have the capacity to turn ideas into

3 innovative products and services. Typically these regions are wealthy and have low unemployment, and they are growing rapidly. Such regions are, of course, extremely important for small countries. The scale of R&D projects continues to grow in the globalised economy. Small economies can only justify participation in a limited number of such projects. Without sufficient resources they can be marginalized in the global innovation system. The growing complexity of new core technologies, combined with limited resources, impedes the development of an extensive R&D infrastructure. Small country enterprises, with limited domestic markets, are facing high and rising R&D costs, at a time of shortening product life cycles and increased competition. Many small countries are located in the periphery, far from the economic core, and as a result face severe difficulties in attracting knowledge-intensive businesses: small domestic markets are an impediment to external investment. In order to attract valuable knowledge-intensive activities small countries in the periphery need to have specific assets such as an educated workforce or a leading-edge knowledge base. There is evidence that small countries can develop high levels of social capital and common purpose, and achieve high level of economic and social well-being. Well-known examples of such successful small countries are the Benelux countries, the Scandinavian countries, Switzerland and Austria. Consensus is easier to promote between the social partners: government, employers and workers. Interaction and co-operation between economic and social institutions is facilitated by more frequent meetings and exchanges in a smaller geographic space. Many of the successful small countries are located close to the economic core of Western Europe. Figure 1.1 shows the difference in per capita GDP between the US (the richest country in the world) and some other countries: it can be seen that the second largest economy of the world (Japan) is clearly lagging behind the US. Even Switzerland, which for decades has been the most successful small country, has a per capita GDP 10 percent lower than the US. Among the four KNOGG countries presented in the figure there is a large variation. Hungary and Greece have reached only 50 percent of the US income and productivity levels, while Finland, Ireland and the Netherlands have been more successful. Ireland, which traditionally has been close to Greece in terms of GDP per capita, has achieved its high rank only recently, as a result of spectacularly rapid economic growth in the 1990s. The KNOGG project focused also on the problems facing small countries with peripheral location. There is a significant economic and technological gap between new member states of EU and the more developed core member states.

4 To some extent the same applies to Portugal and Greece, too. These lagging economies also face globalisation pressures that impact upon their capacity to pursue an independent policy of technology-driven industry growth and specialisation. The challenge is to devise STI policies at regional, national and Community level that meet the exigencies of this special category of small country. Figure 1.1.

GDP per capita differences with the US in 1999

Percentage point differences in PPP-based GDP per capita with respect to the United States, 1999

Hungary Greece

Percent

Finland

Ireland Japan Netherlands Switzerland -80

-70

-60

-50

-40

Percentage gap to US GDP per capita

-30

-20

-10

0

10

20

Percentage gap to US GDP per hour worked

Source: OECD.

The small country group in the EU increased by 9 new members in May 2004 – Poland is the only large country among the new members – and this has implications for STI policy after enlargement. Historically, STI policies in the accession countries have been top down and if national policies dominate also in the future then an even more fragmented science and technology policy regime will emerge in Europe.

1.2 The growing importance of knowledge A key issue determining the long term economic growth of any country or region is its capability to absorb new knowledge. Today, knowledge is shared globally through business networks and innovation platforms that are formed both spontaneously and systematically. Currently, it is not very well understood what determines the emergence and success of a learning environment and what is the

5 role of human and social capital in this process, although recent research provides some insights. Innovation and knowledge creation are now perceived as complex, interactive, and open-ended processes with a collective dimension. Institutions, human capital, skill formation, social capital and organisations are crucial to the knowledge creation and innovation processes which are also evolutionary and path-dependent. Figure 1.2 benchmarks the level of investment in knowledge by KNOGG countries and shows that, apart from Finland, it is lower than the OECD average. National Innovation Systems (NIS) comprise interacting organisations and institutions within a country’s boundaries. The institutions within a NIS facilitate innovative activities, as well as the management of knowledge in the society. The main difference between the national innovation systems of large and small countries is the greater dependence of the latter on external sources of knowledge. The role of market forces in determining the direction of national research and development effort has increased markedly, and there is a drive for greater efficiency in its implementation. This has resulted in making research more applied, more interdisciplinary, more collective and problem-solving in response to the realisation that knowledge is an increasingly important factor in economic growth. Figure 1.2.

Investment in Knowledge Investment in Knowledge

As a percentage of GDP, 1998

7,00 6,00 5,00 Higher education Software R&D

4,00 3,00 2,00 1,00 0,00 Greece

Hungary

Ireland

Netherlands EU (3)

Japan

OECD (2)

Source: OECD, National Accounts database; Education database; MSTI database and International Data Corporation, March 2001.

Finland United States (1)

6 History and institutions are critical to understanding the distinctive innovation policy regimes of individual countries. The KNOGG project, therefore, focused on country-specific institutional developments, their emergence and maturity and the extent to which they drive the innovation systems. The six countries (“the KNOGG countries”) analysed in the project vary greatly in their paths towards economic enhancement and innovation governance. In some of them, innovation policies have a central role in government policies, while others are struggling to build up political commitment for the support of long term growth-oriented policies.

1.3 The role of Community level RTD policy The EU has adopted targets to increase total R&D spending to 3% of GDP, and to make the EU the leading knowledge-based economy by 2010. The EU is also committed to promoting the Information Society. Among the measures for achieving this are the 6th Framework Program of 17½ billion EUR (targeted to LIFE, INFO, NANO and SPACE technologies) and the creation of a European Research Area, ERA. The EU document ‘Towards a European research area’ has the objective of creating a European border-free zone for research, in which scientific resources will be deployed to create more jobs and to improve Europe's competitiveness. Strong measures are obviously needed, since European R&D spending and innovation activity are currently lagging well behind the US. There are good research platforms in some European countries but in most cases they are not as competitive as the top US institutions, and European research is relatively more fragmented. Technological development is often driven by political rather than by technological concerns. Large countries – most notable the US – tend to focus their efforts on promoting spillovers from defence or space technology. Sometimes even small countries may enjoy the fruits of long term demand pull on technological development (as the Finnish experience with the breakthrough of mobile technology shows). However, such success is rare, and small countries may fare better by focusing on market-oriented technologies and on becoming complementary suppliers in large technological projects. The KNOGG project studied the impact of EU Framework Programs on KNOGG countries. This report includes a short evaluation of the participation of small and medium sized enterprises in these programs, and an assessment of the impact of other EU policies (e.g Community Structural Funds) on economic growth in Member Countries.

7

2. Analytical framework 2.1 Knowledge-based growth The aim of the project KNOGG – Knowledge-based growth and globalisation – was to review the factors of knowledge-based sustainable economic growth and to assess the role of public science, technology and innovation (STI) policies to promote such growth. To lay down conceptual and theoretical underpinnings we used the so called new growth theory (NGT) as a starting point. New Growth Theory combines an old economic idea with a new reality. The theory acknowledges that ideas and knowledge are the real source of economic growth, and while we must invest in ideas, many private firms cannot afford to do so or do not find it profitable enough. It asserts a role for government in subsidising new ideas while at the same time ensuring everyone benefits from that investment. According to the empirical studies, the welfare gains from R&D investments for the whole society are 4–5 times larger than their profitability to individual companies. These positive external effects have been proved by numerous empirical studies. The main reason behind this ”discrepancy” (i.e., a big difference between the private and social gains) are knowledge spillovers, which are at the core of New Growth Theory. The main difference between the New Growth Theory and the old neoclassical growth theory is that the NGT emphasises the role of increasing returns to scale associated with new knowledge while according to the old theory there would be limits to growth determined by eventually decreasing returns to scale. Knowledge has properties that are different from economic goods; knowledge is a non-rivalrous, and can be used by many people at the same time. As a consequence, by increasing its pool of knowledge, an economy greatly improves its economic growth potential. Knowledge can be increased without limits because it does not require the use of limited physical resources. The New Growth Theory rests on two observations. First, the old economic idea of diminishing returns is more related to a lack of imagination than to reality. Second, while the development of a new product may be costly, its production and distribution may not be, and this creates a disincentive to invest in the development of ideas which can be easily copied. But new ideas, and the resulting technology, are the proven mainsprings of economic growth. If they are stifled by a lack of funds government has an important role to play in investing in research and development. In the original model of NGT by Paul A. Romer (see Romer, 1986) four factors of economic growth are of importance, namely physical or tangible capital stock (measured in units of consumption goods), labour (or the skill-weighted sum of

8 labour inputs), human capital (person-specific abilities created by formal education and on-the-job training), and finally an index of the level of the technology. The NGT model clearly differs from the neoclassical Solow growth model, where the inputs were capital and labour only, and where technological advancements were treated as unexplained residuals. Recently many authors have suggested a fifth factor also, namely social capital (see e.g Hjerppe, 2003, Lundvall, 2001). The key factor in Romer’s model is an adequate stock of human capital. He finds that “what is important for growth is integration not into an economy with a large number of people but rather into one with a large amount of human capital” (Romer, 1990, S98). Market power or monopoly power is another concept common to the NGT and to the old Schumpeterian evolutionary theory of economic growth. In this respect it also differs from the neoclassical growth theory which relied on the assumption of perfect competition. Individuals and firms have some market power and may earn monopoly rents on their discoveries. This is central to the birth of innovations. Schumpeter stressed the fact that monopoly power was a motivating force in the process of innovation.1 In “The New Age of Capitalism: Innovation-mediated Production” by Richard Florida and Martin Kenney (1993) the authors describe a system of mass production which uses decentralized decision making, and uses the knowledge and intelligence of all employees, making daily learning important. In such a ‘new economy’ – or learning economy as suggested by Lundvall (1992) – there is a shift in the main source of value creation from physical skill or manual labour to intellectual capabilities or mental labour. This change emphasises the importance of social or collective intelligence (or social capital) as opposed to individual knowledge and skills, and yields an acceleration of the pace of technological innovation. The emphasis is on the increasing importance of continuous improvement at the point of production and the blurring of the lines between the R&D laboratory and the factory. This is the post-Fordist model of production where, in order to stay in business, companies have to use all available resources. There is a possibility of market failure, markets may not produce enough knowledge when innovators cannot capture all the gains associated with creating new knowledge because that knowledge can be reused by others at zero marginal cost. Economies with widespread increasing returns, on the other hand, are unlikely to develop along a unique equilibrium path. Instead, development is likely to be a process of creative destruction, with a succession of 1

One clear example of market power is Microsoft company, which is operating in the knowledge intensive sector.

9 monopolistically competitive technologies and firms, as originally described by Joseph Schumpeter. Markets alone may not converge on a single most efficient solution, and technological and regional development will tend to exhibit path dependence. New Growth Theory, and increasing returns associated with knowledge have many implications for economic development policy. New Growth Theory underscores the importance of investing in new knowledge creation to sustain growth. Policy makers need to pay careful attention to all of the factors that provide incentives for knowledge creation (research and development, the education system, entrepreneurship and the tolerance for diversity, expectations of future growth, openness to trade). Because it undermines the notion of a single, optimal general equilibrium, New Growth Theory implies that economics are less capable of predicting future outcomes. In order to promote sustainable growth, countries’ economic policies should, according to NGT, encourage investment in research, as opposed to encouraging investment in physical capital accumulation. If this is not possible, then policies should at least subsidize the accumulation of human capital. Yet, New Growth Theory should not be understood to imply market intervention in the usual sense. Rather, New Growth Theory requires re-thinking the role of the nation state and that of the EU, particularly regarding intervention policies on those markets which no longer seem to behave according to the neo-classical paradigm. It seems that increasing returns deliver monopolistic competition (segmentation of product markets due to product differentiation), in contrast to perfect competition facilitated by diminishing returns to scale. Monopolistic competition together with increasing returns to scale leads to profoundly different growth dynamics and firm behaviour than traditional neoclassical analysis.

2.2 Small country constraints Another conceptual starting point for the project KNOGG is the small open economy. Small countries are subject to tighter constraints than larger countries. There are also bigger threats from e.g., sector-specific adverse shocks which can have devastating effects on specialised small country economies. In periods of rapid and extensive technological change, however, small country adaptability may be more aptly described as the small country squeeze (Kristensen and Levinsen, 1983). The present squeeze results from two major global trends which erode competitiveness particularly in those countries that have specialized in medium-tech products:

10 1) First, the growing complexity of new core technologies (e.g. microelectronics and biotechnology) requires more fundamental research and therefore a more active role of governments. Due to resource constraints, smaller countries face a dilemma of priority setting: do comparative advantages in the research infrastructure exist? If so, in which niche areas? Who is going to decide upon the destination of public funds? 2) Second, competition in low- and medium technology products is increasingly dominated by Newly Industrialized Countries (NICs). Walsh (1987) showed that the share of small countries in high R&D intensive products declined in the 1980s as a consequence of the growing share of large countries. Van Tulder (1991) concludes that the social, political and economic systems of small countries are particularly geared towards the medium-tech areas mainly as a result of their position in the international division of labour between the larger countries and the developing countries. In other words, there is evidence that the technology gap between large and small countries is increasing. Small countries can tackle the small country squeeze by exploiting the rising internationalisation of technology and globalisation to upgrade their domestic knowledge base. This requires careful policy design, in which foreign direct investments (FDI) can play a central role. Knowledge-based growth strategies are not limited to the acquisition and creation of science and technology. They encompass a whole range of policies relating to the transfer and diffusion of various types of knowledge. The resources and institutions available to the government for the “generation of knowledge” are at centre stage. In small countries, however, domestic resources may be insufficient to capture the critical mass necessary for technological breakthroughs and as a result innovation policies assume added importance. History, institutions and geography all shape the development of knowledgebased economies. History matters because increasing returns generate positive feedbacks that tend to cause economies to “lock in” to particular technologies and locations. Development is in part accidental because small events at critical times can have persistent, long term impacts on patterns of economic activity (Krugman, 1991). The divergence of small country experiences demonstrates this. Institutions also matter because they shape the environment for the production and deployment of new knowledge. Societies that generate and promote new ideas, and that continuously adapt to changing economic and technological circumstances provide the conditions for sustained economic growth. Geography matters, too, because knowledge doesn’t move without frictions among economic actors. Important parts of knowledge are tacit, and embedded in the routines of individuals and organizations in different places.

11 Within the enlarged European Union it is imperative to consider the risks and opportunities of small open economies when integrated into the Single Market and Monetary Union. Economic and political integration is in many aspects beneficial to small countries (they benefit mostly from access to larger markets). At the same time economic integration is likely to lead to increased specialisation, which makes the small economies subject to bigger risks. They also have less independence in their economic policy and less possibilities to support their industries. Figure 2.1

GDP per capita and population in 2000 in OECD countries

40000 Norway 35000

USA Ireland Switzerland Netherlands Iceland Austria Canada Denmark Japan France Finland UK Germany Sweden Australia Belgium Spain Portugal

GDP per capita in USD

30000

25000

20000

New Zealand Slovenia Greece 15000

Czech Rep Hungary Poland

10000

Mexico 5000

0 1

10

100

1000

10000

100000

1000000

Population in thousands (log scale)

Figure 2.1 fails to show any correlation between country size and income level. There are some countries which are small and rich (e.g. Luxembourg, Iceland and Ireland), but there are also rich large countries (US) and relatively poor small countries. It is of course possible to argue that most of the rich small countries have benefitted from their location close to the core of the European economy; that is true with respect to the Benelux countries and also to Denmark and

12 Sweden. It is less common for small countries in the periphery to achieve high income levels. Ireland, Finland and Iceland, however, are peripheral countries which belong to that group.2 On the global scale, current economic indicators provide no basis for arguing that small countries are systematically worse or better off than large countries. Above and below average GDP growth rates occur over a wide range of country sizes: income per capita tends to decrease with the size of population, but the relationship is very weak as there are very rich and very poor large countries. Figure 2.2

Country size and foreign trade in 2000

90

80 Ireland 70

Exports as percent of GDP

Belgium 60 Netherlands Czech Rep

50 Slovenia

Hungary 40 Finland New Zealand

30

Norway Denmark Iceland

20

10

Sweden Switzerland Austria Portugal

Germany

Canada SpainFrance UK Australia Poland Mexico Japan USA

Greece

0 1

10

100

1000

10000

100000

1000000

Population in thousands (log scale)

Small countries are more open than large countries; see Figure 2.2 which presents a scatter diagram of the populations of the OECD countries and a measure of the openess of their economies. It is very clear that large countries are

2

Though Finland has likely benefited from proximity to Sweden and Russia, and Ireland from its closeness to large UK market.

13 less open to internation trade than small ones. Their openness makes small countries more vulnerable to the shocks of international economy. It also makes inward FDI an important and, in many cases, a necessary mechanism for increasing the knowledge base. Small countries have a small domestic market and less ability to invest in risky R&D projects. However, without a strong RTD base, small countries may not be able to attract knowledge-intensive FDI. The relationship between FDI and technology development is a complex and crucial question for many small countries. Economic flexibility and political stability are mutually contingent for small countries. Peter Katzenstein (1985) has argued that small European countries have followed a flexible, reactive, and incremental approach, not only in their economic policy, but also in their innovation policies. They do not attempt to pre-empt change by ambitiously reordening their domestic economies but continually improvise in order to live with change. Figure 2.3

The relationship between R&D expenditure and GDP per capita R&D per GDP and GDP per capita

6

5 Israel (2001)

Sweden (2001)

R&D in % of GDP

4

Finland (2001) Italy (2000)

3

Japan (2001) Iceland (2001)

Korea (2001)

United States Switzerland (2000) Germany (2001) OECD (2001) France (2001) Chinese Taipei (2001) Denmark (1999) Singapore (2001) Netherlands (2000) Canada (2001) Belgium (1999) Austria (2001) Slovenia (2001) EU (2001) United Kingdom (2001) Australia (2000) Czech Republic (2001) Ireland (2001) Russian Federation (2001) New Zealand (1999) China (2001) Spain (2001) Hungary (2001) Greece (1999) Turkey (2000) Poland (2001) Portugal (2001) Slovak Republic (2001) Argentina (2001) Mexico (1999) Romania (2001)

2

1

Norway (2001)

0 0

50000

100000

150000

200000

250000

300000

350000

400000

GDP per capita (thousands of PPP$)

Figure 2.3 shows that there exists a positive relationship between productivity and R&D intensity. Maintaining strong R&D depends on the viability of national

14 innovation system and of government policies. But today it increasingly depends on the decisions of firms. The key to success in a small country is to be attractive. It is particularly important in the context of globalisation, where large transnational companies (TNCs) can choose their location freely. To be able to create or preserve jobs and to raise productivity, a country or a region needs to attract qualified people, capital and knowledge services. However, to be attractive may be difficult for a small and remote country. Figures 2.4

Foreign cooperation in research Percentage of scientific publications with a foreign co-author, 1995-97

60,0

50,0

Percent

40,0

30,0

20,0

10,0

0,0 Japan

EU (3)

OECD United States

Finland

Greece

Ireland

Hungary

Netherlands

Source: OECD.

Figure 2.4 indicates that small countries may have more foreign co-operation in their research work. Researchers in small countries tend to have more foreign coauthors in their scientific publications.

2.3 The KNOGG framework Technological development in an open economy can either be based primarily on imitation or on domestic knowledge creation. Even the biggest countries are nowadays dependent on technology created elsewhere. Yet, large countries enjoy a critical-mass advantage in innovating with their own RTD investments. If a country is small, it has limited resources to do significant R&D. For most

15 countries – but particularly for small countries – there is a choice to be made as to what extent knowledge-based growth should be based on technology acquisition and diffusion from abroad as opposed to relying on own R&D in order to create new innovations. This basic choice can be presented as a “KNOGG framework” (see Figure 2.5), which postulates two separate growth paths, depending on whether a country has the capacity for self-sustaining R&D based innovation, or alternatively, is constrained to catching up through imitation. The innovation growth path is predicated upon a superior level of technological knowledge, derived from basic research, and a commensurate stock of human capital. The alternative “imitation” growth path, for economies without an adequate knowledge base, is not selfsustaining and requires external policy intervention to close the technology gap. In reality, all countries have elements of both alternatives. However, it can be argued that some small countries are more inclined to pursue innovation in some key clusters by supporting national R&D efforts while others rely on ‘imitation’, i.e. on technology diffusion through FDI. The KNOGG framework demonstrates this choice by showing alternative ways to achieve the target, sustained knowledge-based growth. Both ways require an adequate pool of human and social capital (the knowledge base). Imitation entails relying on absorbing existing technologies, which are imitated or reverse engineered to produce existing products (or up-graded versions of them) for competitive markets. Costs will be lower because expensive R&D investments are not needed. The alternative strategy is to seek competitive advantage by developing new products for imperfectly competitive markets. In a Schumpeterian world, only the producers of new products can enjoy some monopoly power. That position will erode eventually because of imitators. If the competitive advantage is to be maintained (and monopoly profits sustained), the country must continue to develop new products and this requires high investments in R&D. It is not a priori clear if both strategies are available for all small country. Maintaining the competitive edge may require such high support for R&D that the benefits do not cover the costs. Due to globalisation and increased mobility small countries may be discouraged from increasing their own R&D efforts in knowledge creation and generation of new technologies. They may instead orient themselves towards imitation, and copy and commercialise the successful innovations of their competitors. Ability to penetrate new global markets may play an increasingly important role for small countries and their ST&I policies. Moreover, small countries are likely to emphasise closer-to-the-market R&D and the STI requirements of foreign companies whose interests may differ from those of indigenous industry.

16 This raises many important challenges for the development of the knowledgebased society in small countries. Will small countries become more and more imitators instead of innovators? Do small countries lose their policy independence, and instruments for policy direction, because of increasing globalisation? Figure 2.5

The KNOGG Framework Knowledge Base

Innovations

New Quality Products

Imperfect Markets

Imitation

New Old Products

Competiti ve Markets

Knowledge-Based Growth

On the other hand, it is clear that in historical perspective a number of small countries have successfully pursued the more difficult innovation growth alternative. Publications and patents reveal that small countries have been innovators in important fields of science and technology, such as microelectronics, pharmaceuticals, biotechnology, and material sciences. Income from innovation can be used to define whether a country operates in an innovation or an imitation technology transfer/diffusion regime. Based on royalties and licence fees per capita in 2001, innovator status can be granted to the US (US $135), the UK (US $135), Sweden (US $160), Finland (US $112), the Netherlands (US $107), and even Ireland (US $90). Meanwhile, Hungary

17 (US $9.4), Slovenia (US $7.2), and Greece (US $1.3) appear to be followers in technology. The question remains whether the potential for technology catch-up is helped or hindered by the onset of globalisation and the era of knowledgebased economies.

2.4 National innovation systems and alternative policy patterns Innovation is now perceived as a complex, interactive, and open-ended process with a collective dimension. It is essentially a process of learning, where both formal and informal institutions and networks play a major role (cf. e.g. Dosi, 1988). Innovation is also an evolutionary and path-dependent process. It is a collective and always uncertain undertaking, which involves various actors with complementary skills and competencies. The concept of national innovation system (NIS) was introduced 15 years ago by Freeman and Lundvall (see Freeman, 1987 and Lundvall, 1988 and 1992). It helps to view the generation of innovations in the framework of national institutions (formal institutional settings and policies and also more informal “rules of game”). The NIS can be understood as an interacting system of national organisations or a meta-institution. Many studies of national innovation systems and their design and management have appeared (see e.g. Edquist, 1997, Andersson, 2001 and Cogan and McDevitt, 2003). In many advanced economies, the development of National Innovation Systems (NIS) have become a central instrument of Science, Technology, and Innovation (STI) policy. The institutions of NIS are used to facilitate innovative activities of firms, as well as the management of knowledge, social capital and social cohesion (cf. Lundvall, 2001). However, due to globalisation and deepening economic and political integration of Europe, there is a need to re-consider the role innovation systems – whether supra-national, national or regional – are playing. In particular, there is a need to better understand processes underlying innovations and the role of emerging knowledge and innovation centres as driving forces of balanced economic growth. Institutions and their history are important and critical in understanding the distinctive innovation policy regimes of individual countries. The six countries analysed in KNOGG vary greatly in their paths towards economic enhancement and innovation governance. Some have placed the innovation imperative at the centre of government and have appointed ministers to sponsor innovation. Others are struggling to get political commitment after episodes of economic and political disruption and ideological change. Infrastructure for the creation and diffusion of knowledge is conventionally referred to as the research system and it is the heartbeat of every country’s

18 innovation system. It includes basic and applied research institutes and, critically, intermediaries and other institutions that provide a bridge between research output and commercialisation. Many accession countries are heirs to a strong tradition in science and basic research but struggle to compensate for a deficit in technological universities and industry-oriented research institutes. This in turn impedes the development of bridging institutions such as technology acquisition instruments and technology transfer services that provide a critical link between knowledge suppliers and knowledge users. On the basis of the main characteristics, goals and instruments of STI policies in the KNOGG countries one may identify some policy patterns for the governance of their STI regimes. The goals of their policies can be different in the framework of these models which offer alternate ways of climbing the ladder of the international technology system. Alternative strategies of national STI policies can be classified by using the alternative routes presented in the KNOGG framework (“Imitation” vs “Innovation”) and public sector interventions (“Interventionist” vs “Laissez faire”). Table 2.1

Alternative policy patterns Interventionist

Imitation

Innovation

Non-interventionist (“laissez faire”) POLICY PATTERN 3A: Reliance on imported technology without any specific supporting policy measures

POLICY PATTERN 1: Support technology diffusion by developing absorptive capacity and subsidising inward FDI POLICY PATTERN 2: POLICY PATTERN 3B: Support national RTDI by Emergence of strong private specific policy programmes sector RTDI without any specific supporting policy measures

(RTDI= research, technology, development and innovation).

In the first policy pattern, a small country depends mainly on external sources of innovation. The aim of STI policy is to promote and speed up the transfer process by supporting with subsidies and incentives the absorbtion, adaptation, diffusion and development of technologies.3 Education and training are an important part of this policy and the country takes advantage of innovations developed in other countries. The strategy used in this model aims to attract trans-national companies (TNCs). It may be necessary to create incentives (e.g. preferential tax 3

On the importance of absorptive capacity, see Cohen and Levinthal, 1990.

19 rates) to speed up FDI investment. This policy is predicated upon the proposition that new technology is usually embodied in new equipment, and also in new work practises imported by foreign investors and especially by TNCs. There are a multiplicity of ways in which TNCs can enhance the innovation potential of the host country: they can make an important contribution to its research system through R&D spillovers; they can incubate high-tech spin-off companies; they can act as conduits for the transfer of codified and non-codified technology through human resource mobility and intensive-interaction with the indigenous sector; and they can stimulate the creation of networks of suppliers of high value-added products, components and services to supply foreign-owned industry. That kind of policy emphasises the business infrastructure (e.g. logistics), availability of an educated and flexible labour force, and financial and fiscal incentives. Ireland is a well-known and very successful example of such a policy. A distinguishing characteristics of model 1 is the presence of high share of inward FDI and supporting policy measures. Policy harmonisation within the EU however, and the rules of WTO, may constrain the future scope for such fiscal and other incentives as have been used in the past. In the second policy pattern (pattern no 2 in the table 2.1), the country aims to develop its own national sources of innovations by deliberate government actions. This makes it necessary for the government to play a fundamental role in the STI process. The aim of the policy is to improve the knowledge creation capabilities of the economy by committing to a national innovation system with an advanced education system, a high level research community and applied research laboratories. Additionally, government agencies provide substantial R&D funding and promote intermediary institutions to bridge the gap between research and industry. Such a system promotes the creation of original innovations, but at the same time it also improves the capacity to adapt and develop new technologies created elsewhere. That choice can be risky for a small country since it requires large investments in a very limited number of sectors. It also requires long term political commitment to a knowledge-based economy. Finland is a prominent example of such a policy pattern. In the past aggressive innovation policies required at least implicit infant industry protection but this is no longer possible in the Single Market; hence historical experiences are not necessarily replicable. In the third policy pattern, the market is playing a decisive role in the size and patterns of R&D activities and in the development of technology and innovations. This pattern can be viewed as laissez-faire policy since there is no role for deliberate government intervention to speed up technological development. Rather the policy is based on the hope or assumption that markets will take care of diffusion process – as they in fact do in most cases. This may also be the only alternative for a small country without sufficient resources to establish large-scale support programs and without the political willingness to

20 improve the incentives for inward FDI. A non-interventionist policy pattern would also obtain for a country which happened to be attractive to investors and the R&D activities of large companies. By relying on market forces this model would be in accord with the basic ideas of the European single market. Laissez faire may be a practical policy choice if there is a risk that policy interventions would be ineffective or wasteful. However, without any specific policies, it is difficult to see how knowledgebased activities would locate in a country that is on the periphery or has a small market. In the past such regions in Europe (or elsewhere) have not, for the most part, succeeded in attracting investments. Figure 2.6 shows that there indeed are significant differences between OECD countries in their stocks of inward FDI and in their R&D intensity. Of KNOGG countries, Finland clearly has the highest R&D intensity and relatively low FDI stock while the opposite holds for Ireland, and to lesser extent also to Hungary. Figure 2.6

Positioning in regard to own R&D – innovation and FDI – technology transfer Inward FDI Stock and R&D Intensity in percent of GDP, 2000

Gross domestic expenditure on R&D as a percentage of GDP, 2000

4,0 Sweden 3,5 Finland 3,0 2,5

Japan United States Korea Germany Iceland France

2,0 1,5 Italy 1,0

Switzerland

Netherlands Denmark United Kingdom Norway Canada Australia Ireland Czech Republic New Zealand Spain

Slovenia 0,5 0,0 0%

Hungary

Portugal Greece

10 %

Mexico

20 %

30 %

40 %

50 %

Inward FDI Stock in % of GDP, 2000

60 %

70 %

80 %

21 Public R&D expenditure is exceptionally high in Finland and in the Netherlands – suggesting that these countries have a strong public policy component in their national innovation systems (see Figure 2.7). Figure 2.7

Public R&D expenditure, in percent of GDP, 1999 R&D expenditure as a percentage of GDP in 1999

1,2 1

% of GDP

0,8 Higher education Government

0,6 0,4 0,2 0 Ireland (1998)

Greece(1997)

Hungary

OECD

United States

EU

Japan

Finland

Netherlands (1998)

Source: OECD, R&D and MSTI database, May 2001.

R&D intensity is also related to patent data. Figure 2.8 shows the number of patents per million inhabitants. Both Finland and the Netherlands have very high figures suggesting that their innovation systems are comparable to the US, and more productive than the EU and OECD averages. There is also a very big difference when these figures are compared to other KNOGG countries, Greece, Hungary and Ireland.

22 Figure 2.8

Patents per population

Number of patents in European Patent Office (EPO), US Patent and Trademark Office (USPTO) and the Japanese Patent Office (JPO) per million population, 1995 60

50

40

30

20

10

0 Greece

Hungary

Ireland

EU

OECD

United States

Finland Netherlands

Source: OECD, Patent database, May 2001.

Source: DG Enterprise (2001), Innovation Policy in Six Candidate Countries [all data relates to 1999 except for 1998 USPTO data].

23

3. Comparative analysis of KNOGG countries – Innovation systems and Industry-oriented STI policies 3.1 Introduction The innovation systems and the industry-oriented STI policy instruments of the KNOGG countries were examined in the project in order to identify characteristics associated with good practice, in so far as this is reflected in the European Innovation Scoreboard and other international competitiveness indicators. The purpose of the exercise was to draw out the STI policy implications for small countries, both EU15 and the new member countries. A particular focus of this project is the daunting challenge facing small countries in the catch-up phase of their economic and technological development. Innovation systems, in the present context, include the education and research sub-system, the political sub-system (policy formulation and implementation bodies) and the STI infrastructure. They also include the intermediaries, brokers and other bridging institutions that facilitate the flow of knowledge between the research and education sub-system and the enterprise sector. Industry-oriented STI policy instruments, on the other hand, comprise all public measures and supports designed to orient the innovation system towards improving the innovation capacity of the enterprise sector. Industry-oriented innovation policies may be resource-based, e.g. R&D subsidies and equity investment, or soft supports such as networking, technology transfer and measures to stimulate technology diffusion. The most pressing need of domestic industry in small countries is frequently to improve its technology absorptive capacity, through a range of soft support measures. A comparative analysis that embraces industry-oriented innovation support measures, as well as the blocks that constitute the innovation system itself, is necessary to obtain a more measured insight into the functioning of the innovation system in KNOGG countries. An analysis of innovation structures and frameworks can highlight the need for a full spectrum of knowledge generation and distribution capabilities. It can identify the key role of intermediary institutions that provide a bridge from the research system to industry and the imperative of effective innovation governance. It can also identify weaknesses in the innovation system but it cannot quantify these deficiencies because there are no metrics to calibrate country performance in areas such as STI institutions, technology infrastructure and culture. On the other hand, an analysis of industry-oriented innovation policies can quantify the resources allocated to specific industry support measures in each

24 country. The purpose of industry-oriented STI policies is to channel the resources and energy of the innovation system towards realising the innovative potential of the enterprise sector. That is important because knowledge-based growth is ultimately a function of industry’s capacity to generate new and improved products and processes. The impact of industry-oriented innovation policies can be assessed, at least partially, using economic and technology output indicators as exemplified by the European Innovation Scoreboard (EIS). The KNOGG project quantified the resources allocated to industry-oriented innovation policies in KNOGG countries using data obtained from project partners, and then calibrated country performance using the EIS and other innovation output data sources. The KNOGG countries were benchmarked on eight industry-oriented innovation policy objectives, representing best practice in promoting innovation across all stages of the innovation process from knowledge generation to knowledge distribution and knowledge application. A table, extracted from the EIS, showing the innovation performance of the KNOGG partner countries, relative to the EU average and extending across 18 innovation indicators, is shown below (see Table 3.3). The complementary and contrasting features of the innovation systems, and of the industry-oriented innovation policies, of the six KNOGG countries were described in detail in Workpackage 2 of this project (Cogan and McDevitt, 2003), and are summarised here. This summary will conclude with a number of STI policy recommendations.

3.2 Innovation Systems in KNOGG countries 3.2.1 Finland The most competitive KNOGG economy4, Finland, has put in place most of the key elements posited as essential features of a successful NIS. These include very effective STI policy institutions and a strong STI research infrastructure. Its powerful Science and Technology Policy Council coupled with a resourceful technology agency, TEKES, serve as the model for government commitment to technological innovation and as the high road to economic growth. Finland’s public commitment to expenditure on research places it at the head of European and OECD scoreboards. Finland’s premier technical research centre, VTT, is the archetypal bridging institution dedicated to transferring advanced technologies to user industries. But the salient factor in the Finnish success was the early progress of its STI policy from an assortment of discrete interventions, typical of most developed economies, towards a holistic approach in which the producers 4

According to many comparative competitiveness studies; see e.g. IMD, 2003 and WEF, 2003.

25 and users of knowledge, together with the interactive relations between them (or mediated through intermediaries) are supported as a single system.5 Co-operation and networking are identified as the active catalysts of knowledge-based development and Finnish STI policy assiduously cultivates key linkages both within and external to the NIS. A pervasive culture of innovation is reinforced by a democratic, corporatist system of policy-making which canvasses the views of the broad community of STI stakeholders. A Foresight Studies unit in the Ministry of Trade and Industry works with external professional experts to systematically conduct technology assessment exercises and thereby plot the most promising development paths for the economy. The Science and Technology Policy Council publishes a science and technology policy review every third year, and issues guidelines for monitoring and evaluation activities in the various ministries, agencies and individual STI programmes. The other KNOGG countries fall conspicuously short of the Finnish NIS standard in more than one dimension.

3.2.2 The Netherlands The closest competitiveness contender, the Netherlands, boasts a powerful STI Policy Council chaired by the Prime Minister. One expects this to lead to a healthy STI budget allocation and, indeed, the Netherlands exhibits a strong public expenditure on research (0.87% GDP).6 But the resources allocated to industry-oriented innovation support measures were only 0.094% to the GDP as opposed to 0.218% in relation to the GDP for Finland (see Table 3.1 below) and this may be a factor in explaining the disappointing level of BERD, less than half the Finnish proportion of GDP.7 A related factor may be the absence of a dedicated technology agency such as TEKES – the Netherlands’ development agencies SENTER and SYNTENS are broader in objectives and scope. The Netherlands is comparatively well served by bridging institutions including the flagship TNO organisation and sectoral centres of excellence (the four Technological Top Institutes), but huge challenges remain for these intermediary organisations in persuading SMEs to avail of available expertise. In assessing NL’s embrace of the NIS approach it should be stated that the Netherlands came to this holistic viewpoint later than Finland. The economic strength of the Netherlands has traditionally derived from its small cohort of domestic, worldclass manufacturing TNCs and from its strategic location as a European distribution centre. The country is now fully wedded to the NIS concept and 5

See also Castells and Himanen, 2002, and Palmberg et al., 1999. Cf. Finland’s public expenditure on R&D was 0.95% GDP [data from European Innovation Scoreboard 1991]. 7 Of course the Netherlands has a different industrial structure to Finland. It specialises in R&D extensive industries rather than R&D intensive sectors. Services account for a large and growing share of the Dutch economy, and service firms typically focus on non-technical innovations, and purchase technology rather than invest in R&D. 6

26 leads the way in promoting innovation networks and clusters. But this is a longterm project and much work remains to be done to inculcate an overt culture of entrepreneurship and innovation.

3.2.3 Ireland The next country in the competitiveness ranking is Ireland. The Irish STI policy institutions are of very recent vintage. A government minister with responsibility for Science and Technology was first appointed in 1987 but the country had to wait until 1996 for an integrated policy formulation and regulatory system. This system was headed by a cabinet sub-committee under the Prime Minister charged with guiding STI policy. Furthermore, there is an Interdepartmental Committee for Science, Technology and Innovation consisting of senior officials in the major STI spending departments. These two committees, however, have not been active and policy direction resides for the most part with ICSTI, the Irish Council for Science Technology. ICSTI comprises experts from industry, academia and government agencies, but it lacks the political authority of its Finnish counterpart to ensure expeditious implementation of policy directives. Because of the country's late industrialisation and a reliance on importing innovation through FDI for nearly half a century, some of the key elements of a national innovation system are missing. There are no stand-alone institutes for basic or missionoriented industry research (except for food/agriculture and health). Likewise, Ireland lacks intermediary institutions to bridge the gap between the research system and industry, and private technology brokers have not prospered in the absence of these industry-oriented research institutes. Efforts during the 1990s to promote dynamic linkages within the Irish innovation system have achieved only limited success. Business sector research is expanding from a low base but is inhibited by the limited amount of R&D undertaken by both domestic firms and foreign-owned branch plants in the host country. The Irish Government has lately been devoting unprecedented levels of public resources to knowledge creation and venture capital markets but the same energy and resources have not been committed to technology diffusion and absorption initiatives at enterprise level.

3.2.4 Hungary and Slovenia The two KNOGG candidate countries, Hungary and Slovenia, have a proud record in scientific and technological research but episodes of economic and political disruption stifled the co-evolution of private enterprise and innovation. The transition to a market economy is understandably arduous and progressive elements in both the policy block and the fledgling enterprise block of the NIS are struggling to overturn the inertia of the research infrastructure and to set in motion a ‘triple-helix’ dynamic. The full voice and vision of the enterprise sector has yet to sweep through the Science and Technology Policy Councils in these countries. STI Policy implementing agencies still lack the necessary autonomy

27 and flexibility to respond to the changing needs of clients and markets. The two countries are making strides in the provision of publicly funded applied research institutions, and these efforts are accompanied by measures to support diffusion mechanisms and technology brokers.

3.2.5 Greece Greece introduced a modernised National Council for Research and Technology in 2001. This council includes representatives from the three social partners, industry, academia and government. The principal agency responsible for STI policy formulation, however, is the General Secretariat for Research and Technology (GSRT), a junior secretariat within the Ministry of Development which lacks the authority to provide tight co-ordination of policy measures across ministries. A second defect in the governance institutions is the fact that STI policy implementation is centralised within the policymaking body. The absence of separate executive agencies results in the typical ‘top-down’ defects of inflexibility and insensitivity. These shortcomings are compounded by a failure to establish systematic evaluation mechanisms. Greece, like Ireland, lacks major, publicly funded, applied research institutes and there are few effective bridging mechanisms to transfer technology from a HE sector that is not predisposed to commercial transactions. The very modest level of Greek BERD reflects the pressing need of SMEs for hands-on support to improve their absorptive capacity, but the NIS is characterised by a very low level of interaction between the companies and other parts of the innovation system.

3.3 Industry-oriented innovation policies in KNOGG countries Data, for the most recent period, were obtained from each KNOGG country, relating to the resources allocated to specific industry-oriented innovation policy objectives. As a result, it was possible to identify what proportion of its industryrelated budget each country allocated to knowledge creation as opposed to knowledge diffusion and absorption. These data are summarised in Table 3.1 and would appear to indicate that the stronger KNOGG economies spend about threequarters, and the weaker ones about half, of their industry innovation budgets on knowledge creation. An initial interpretation of the data in this table would suggest that Finland, the Netherlands and Ireland, all of whom allocate about three-quarters of their industry innovation budget to knowledge creation, are pursuing an innovationbased growth path. But the overall intensity of knowledge creation in the Netherlands (2.0% of GDP) and Ireland (1.4% of GDP) does not justify that conclusion. The low R&D performance of a substantial number of TNCs in the host country contributes to the relatively low level of knowledge creation in Ireland and the Netherlands.

28 Table 3.1

Allocation of Industry-oriented Innovation Budget between Knowledge Creation and Knowledge Diffusion (most recent year) Budget, % of GDP

Finland Netherlands Ireland Greece Hungary Slovenia

0.218 0.094 0.127 n.a. n.a. 0.11

Knowledge Creation Knowledge Diffusion 73.5% 26.5% 71.0% 29% 78.0% 22% 49.6% 50.4% n.a. n.a. 46.3% 53.7%

The data point more clearly, however, to the alternative “imitation” or technology-acquisition growth path for Hungary, Slovenia, Greece and, to a significant extent also for Ireland. The desired strategy for these economies must be to leverage the potential of their STI policies to acquire and absorb technology from external as well as internal sources. The role of bridging institutions and intermediaries will be critical. It is important that, in “imitation strategy” countries, a relatively high proportion of the industry-oriented segment of the national S&T budget is allocated to knowledge acquisition and diffusion measures. We now turn to a critique of the industry-oriented innovation policies that individual countries are pursuing in their search for knowledge-based growth. Two caveats, however, attend our use of allocations of financial resources to measure the relative commitment of a country to R&D and non-R&D industryoriented supports. First, most of the soft non-R&D supports are not resource intensive but require scarce specific intellectual assets and managerial skills. The amount of financial resources expended is an unsatisfactory surrogate for effectiveness in the case of these soft measures. Secondly, many of the integrated industry support packages offered by KNOGG partners make it difficult to disaggregate the effort involved and distribute it across the different categories of innovation policy. It has been necessary, therefore, to describe here the variety and richness of non-R&D support measures in the different KNOGG countries.

3.3.1 Finland The Finnish bar chart (Fig 3.1) purports to show that nearly half of all public support for industry went to promote enterprise R&D and that a further 28.5% was allocated as risk capital subsidy. The residual 26.5% of resources are assigned to “soft” non resource-based innovation policy objectives such as cluster formation, “technology clinics” and technology transfer services.

29 This overstates the R&D proportion because all the Tekes Technology Programmes, which account for nearly half the Finnish industry-oriented innovation support budget, are classified as R&D supports. But the Tekes Technology Programmes were designed, not as a subsidy for individual firms, but as an instrument to develop networks and promote linkages between SMEs and large firms. They have proved to be a very effective package of measures for boosting business sector R&D but they also qualify for inclusion under two other headings, namely R&D Co-operation and Knowledge Diffusion. Technology Programmes promote development in specific clusters of technology or industry and provide a conduit for the flow of research results to business. Figure 3.1.

Finland: Annual Expenditures by Objective in 2001 (€m), Total 287 €m ~ 0.218 % GDP

Finland

150

100

50

0 1

2

3

4

Enterprise High-Tech Risk Capital R&D CoR&D Start-Ups markets operation 44.7%

0.3%

28.5%

11.8%

5

6

Public Knowledge Knowledge Diffusion 0.0%

14.4%

7

8

Human Mobility

Absorptive Capacity

0.0%

0.3%

A well-funded cluster programme has been a cornerstone of Finnish industrial policy since the mid-1990s. Cluster programmes are extensive R&D supports that involve several sectors of the Finnish economy. The aim of these clusters is to accumulate and transfer knowledge, in chosen fields, by promoting cooperation among various actors including both the producers and users of knowledge. They also aim to break boundaries between different sciences and fields of expertise and hence to stimulate innovation. Thus the focus is on strengthening the general business environment for enterprise. Successful clusters have developed in ICT, the forest industry, metals and engineering and in knowledge-based services. Cluster programmes are shown in the bar chart as instruments of technology diffusion but they also qualify as R&D co-operation measures. The Centres of Expertise are part of a complementary programme that enhances competitiveness by addressing weaknesses in the structure of production in the regions.

30 Special steps were taken in Finland to address a difficult problem for all catch-up countries, namely how to increase the capability of SMEs to absorb technology from external sources. Tekes recognised that it had to tackle a capability deficit in both the technology service providers and in the technology service users. The Technology Clinics are an innovative and very successful initiative, focussed not on firms, but on technologies. The technologies targeted are, for the most part, technologies new to Finnish industry. The clinics are delivered to networks of SMEs and this contributes to the learning experience as well as to the costeffectiveness of the programme. Technology diffusion measures are demand-driven and much more difficult to implement than supply-driven measures such as R&D subsidies or venture capital supports. The services of intermediaries, as well as of service suppliers, are normally needed. Finland’s record in designing and implementing successful technology diffusion mechanisms is exemplary and worthy of emulation. In addition to the technology Clinics, well-resourced technology transfer programme (TTRANS) provide intermediary and brokerage services to firms. The distinctive feature of the Finland’s portfolio of industry-oriented innovation supports is that they form an integrated package addressing all aspects of the innovation process, knowledge-creation, knowledge distribution, knowledge acquisition and extending to knowledge application. Many of the initiatives, through their design and delivery mechanism, have a pervasive impact across the innovation spectrum. This is a sphere of activity in which Finland has been uniquely successful.

3.3.2 The Netherlands The bar chart (Fig. 3.2) for the Netherlands is rather similar to that for Finland in so far as more than 70% of the country's industry-oriented innovation support is reported to be directed towards knowledge creation. It is surprising therefore to discover that the value of the BERD indicator, at 1.05% of GDP, is below the EU average. The knowledge-creation capability of Dutch industry is depressed because foreign-owned TNCs in the Netherlands normally locate their R&D function in the country of their corporate headquarters.

31 Figure 3.2.

Netherlands: Annual Expenditures by Objective in 2000 (€m), Total €657m ~ 0.094 % GDP

Netherlands

400 300 200 100 0 1

2

3

4

5

6

7

Enterprise High-Tech Risk Capital R&D CoPublic Knowledge Human R&D Start-Ups markets operation Knowledge Diffusion Mobility 49%

4%

18%

20%

0%

9%

0%

8 Absorptive Capacity 1%

The proportionately high level of support for knowledge creation cannot be interpreted as evidence of a dedicated technology-push strategy because many of the R&D measures also satisfy a technology diffusion policy objective. There is a trend away from single company support and a move towards forming centres of excellence as a focal point for funding. This is linked to a policy of improving the interaction between industry and public sector research organisations. As a result, the R&D supports contribute to inter-firm and firm-to-research centre cooperation, and to the diffusion objective of industry-oriented innovation policy. Industry policy in the Netherlands, like Finland, shows a strong commitment to technology diffusion. Overall there are nine diffusion-oriented programmes, many well established and most tailored to meet specific Dutch needs. Syntens (formerly Innovation Centres) is best known outside the Netherlands as one the earliest SME support measures in Europe. It is an integrated technology strategy initiative, designed to improve the innovative capability of SMEs. The programme is based on a network of 18 regional centres and has been through many evolutions in its 20-year history. On the other hand, the Technology Top institutes are new and their focus is on particular areas of technology. They are well-funded, industry-driven initiatives that co-ordinate and exploit existing sources of expertise. Furthermore, in selected areas of national importance, such as Maritime Research and Environmental Technology, there are specific programmes for stimulating knowledge transfer from research institutes to firms. Research co-operation is a priority funding area in the Netherlands. The two most extensive measures are the Technology Co-operation initiative and a programme,

32 aimed specifically at SMEs, to promote the introduction of new technologies. These two programmes share over 100 million euros per annum. In general, the Netherlands has adopted a bottom-up approach to industry innovation funding: clustering and networking of SMEs are seen as the best avenues to improved competitiveness. The headings Human Mobility and Improving Technology Absorption register only one entry on the Dutch bar chart but the policy objectives of these two categories are, to a considerable extent, met by initiatives appearing under other headings. The intermediary role of Syntens enhances the innovation capability of firms, as does the overall priority given to stimulating the networking of firms. There is a human placement programme (Knowledge Carriers) to transfer tacit and codified knowledge to SMEs, using graduates and post-doctoral personnel from technical colleges and universities. In conclusion, this analysis found that the Netherlands has a comprehensive portfolio of innovation supports, substantially covering all activities of the innovation policy spectrum. The success of Dutch innovation policy is reflected in the EU Innovation Scoreboard and in the World Economics Forum indicators. The emphasis on industry clusters and on the system of innovation philosophy is not as overt as was the case for Finland. But the delivery mechanisms for the industry programmes are oriented towards the support for consortia rather than for single firms, for centres of excellence rather than stand alone projects. There is evidence that a lack of innovation culture may be impeding entrepreneurial high-tech start-ups and that business expenditure on R&D is rather weak in international terms. The latter problem may be related to the dual industry structure, with separate foreign-owned and indigenous segments.

3.3.3 Ireland The Irish bar chart (Fig. 3.3) is unexceptional in so far as it shows that, as was the case for Finland and the Netherlands, Ireland spends more than 70% of its industry-oriented innovation budget on knowledge creation. It could be argued, however, that in Ireland's case, a higher allocation of resources to technology acquisition and diffusion would be appropriate. In other words Ireland should hold an intermediate position between the strong and the weak countries in Table 3.1. Ireland has experienced great difficulty in expanding its base of R&D performing SMEs and this has been attributed to an inherent lack of technology absorption capacity and innovation capability.

33 Figure 3.3.

Ireland: Annual Public Expenditure by Objective €m, National Development Programme 2000–2006 (Average €132m per year ~ 0.127 % GDP)

Ireland

60 50 40 30 20 10 0 Cogan /McDevitt UCD

Enterprise High-Tech Risk Capital R&D CoPublic Knowledge Human Absorptive R&D Start-Ups Markets operation Knowledge Diffusion Mobility Capacity 37.4%

6.5%

34.1%

6.1%

6.4%

4.3%

0.0%

5.2%

Ireland did have a number of well-funded, non-R&D, innovation support initiatives in place during the 1990s. These included a Technology Acquisition (Licensing) Programme, a Technology Partnership Programme and a Programme in Advanced Technologies (PATs) attached to the HE sector. The PATs engaged in industry-oriented research but they were also mandated to assist SMEs with technology transfer services such as consultancy, demonstration and specialised training. A National Technology Audit Programme was launched in parallel with the MINT and IMT initiatives elsewhere in Europe. This was intended to be an integrated technology strategy programme but it had serious limitations in design and implementation. All the programmes mentioned in this paragraph were either discontinued or scaled down at the end of the decade of the 1990s. Ireland has an overt industry policy of encouraging industry, an objective that has been articulated in successive industry policy reviews. But because of the critical role of Ireland's foreign-owned sector and the propensity of that sector not to engage in any dynamic way with the indigenous system of innovation, these recommendations have not been acted upon. A pilot Inter-firm Co-operation Networks Programme was introduced in 1996 and ran for a number of years. Two human mobility programmes also operated during the 1990s. The Techstart Programme placed young technical graduates in SMEs for a year, with a view to giving low capability firms a window on the world of science and technology. The Techman programme placed experienced technologists in SMEs for a 3-year

34 term, to undertake significant product development projects. Both these programmes achieved some success but they tended to isolate the placements in low-technology SMEs and leave them without technical supervision, or the support of technical colleagues. In this respect the Irish programmes contrast unfavourably with better designed human placement programmes, such as the UK Teaching Company initiative. The Irish placement programmes were discontinued in 1999. Table 3.2 shows how the balance of resource allocation between knowledge creation and knowledge diffusion changed for Ireland after 1999. Table 3.2

Ireland’s Industry-oriented Budget Resources to R&D and Non R&D Measures in Successive Periods

Average p.a. (1994–1999) Average p.a. (2000–2006)

Knowledge Creation % of Budget 49.6% 78.0%

Knowledge Diffusion % of Budget 50.4% 22.0%

Ireland is currently pursuing a dedicated policy of improving its commitment to research. R&D, as a percentage of GDP, currently stands at a lowly 1.5%: with an aspiration of reaching 3% of GDP by the end of the decade. There are serious dangers, however, in trying to achieve this goal without a parallel effort to improve the technology absorptive capability of industry. The consequences may well be that the fruits of the enhanced research effort will flow out of Ireland, probably to the US and Japan.

3.3.4 Greece The Greek bar chart (Fig. 3.4) indicates that industry-oriented innovation support measures account for of 0.2% of GDP which in relative terms is in line with the spending of high-performing KNOGG member states. This is distributed fairly evenly between resource-based supports and knowledge diffusion and absorption measures. However, notwithstanding significant transfers from EU Structural Funds, the absolute STI funding involved is still very moderate. The Innovation Scoreboard 2001 indicates a poor Greek performance, falling more than 20% below the EU average in 13 of 15 indicators (the chart below shows Greek performance relative to EU=100).

35 Figure 3.4.

Greece: Expenditures (€m) by Objective 1996–2002, Total for 7 years: 1,389 €m

Figure 3.5

Greece innovation scoreboard 2001 (EU=100)

4.6 1993-97 high tech value added 4.5 Share of ICT markets in GDP 4.4 Home Internet access 4.3 Sales share of new products 4.2 Capitalisation of new markets 4.1 High-tech venture capital 3.3 Innovation expenditures in manufacturing 3.2 Innovation co-operation by SMEs 3.1 In-house innovation by SMEs 2.3 b Patent applications (USPTO) 2.3 a Patent applications (EPO) 2.2 Business expenditures on R&D 2.1 Government R&D funding 1.5 Employment in high tech services 1.4 Employment in high tech manufacturing 1.3 Participation in life-long learning 1.2 Population with tertiary education 1.1 New graduates in Science & Engineering 0

80

100

36 Greece exemplifies the proposition underlying the “technology gap” growth model that catch-up is not automatic. It was seen earlier that there are inherent weaknesses in the country’s national innovation system, in respect to the research system and STI governance. These deficiencies make it difficult to orient the innovation system towards fostering a culture of technology-based innovation in the enterprise sector. Up to recently industry-oriented innovation supports consisting mainly of R&D grants to firms and support for high tech spin-outs from academia. Direct R&D support for industry is provided by the PAVE programme, instituted in 1986, but re-launched as PAVET in 2001 to emphasise its technology dimension. Support for cutting edge research at universities is funded by the EKVAN programme but, in the absence of applied research institutes to bridge the gap, the output of this long-term research does not transfer to a technologically weak domestic industry. Other resource-based initiative include PRAXE and TANEO, that provide venture capital for academic spin-outs, and ELEFTHO that supports the development of incubators and science and technology parks. The failure of these initiatives is reflected in the level of Greek BERD which, at 0.13% of GDP8, is unconscionably low and showing an adverse trend. In a period when substantial Community Structural Funds were available for enterprise R&D, it must be concluded that the deficit in BERD is a function of industry’s poor technical absorptive capability, rather than any lack of R&D subsidy. One manifestation of this lack of capability is the absence of a commitment to enterprise training, highlighted in the Scoreboard as a weakness of the Greek NIS.9 There are signs, however, that the focus may now be moving to non-R&D innovation supports for industry. Greece has recently brought forward a profusion of measures to enhance absorptive capacity. The PAPHOS initiative encourages benchmarking and the assimilation of best practice and world-class technologies. PEPER promotes the participation of Greek firms in large-scale demonstration and TT programmes. Efforts to establish full-time, private intermediaries and technology brokers have not proved successful but the government continues to experiment. A new scheme called AKMON (Research Centres Development and Services Supply Projects with User Participation) was launched by the General Secretariat for Research and Technology (GSRT) in 2001, aimed at encouraging public sector research institutions (PRIs) to provide technical, research and advisory services. Thirteen regional Centres of Entrepreneurial and Technological Development (KETA) were also opened in 8

Greek BERD is 25% of GERD while, on average, the corresponding EU figure is 65%. Training is a Greek example of where lack of policy co-ordination has impeded a measure: enterprise training suffered because it was, until recently, split between the two Ministries, Labour and Education. A new co-ordination body has recently been agreed with the social partners. 9

37 2001 to provide information and consulting services to companies. An emphasis on the information society and the ‘new economy’ is reflected in the new CSF (EPET III) which contains initiatives such as the ‘Network Yourself’ programme, encouraging the uptake of e-business ICTs by SMEs. There is also a placement programme, IRON, which introduces research graduates into enterprise. Clustering is another important policy instrument to facilitate learning in enterprises. EKVAN’s support for R&D in “high economic interest areas” is felt to be the programme closest to this objective. Co-operation among HE, RTO and enterprise was envisaged, but, to date, there has been little evidence of cluster formation. The Technology Brokers initiative includes an incentive for the networking of technology providers. In summary, important steps are now being taken to improve the absorptive capacity of Greek industry to complement the resource-based measures that support enterprise R&D and academic spin-outs. The upgrading of the largely traditional industry sector of the economy is going to be very challenging, given the dearth of industry-oriented applied research institutes and the propensity of the Greek University sector to distance itself from the grind of industry. Commentators have also criticised the lack of clear policy vision and focus from the top. This is compounded by the absence of a strong executive STI agency that can interact directly with firms and encourage them to take up the supports on offer. The absence of such an agency also leaves a vacuum with respect to monitoring the impact of policy initiatives, and inhibits the prospects of getting accurate and swift feedback.

3.3.5 Hungary Hungary was unable to quantify the distribution of industry-oriented innovation support measures in such a way as to construct a country bar chart. Comprehensive descriptions were provided, however, of the policy measures operating under the various categories. From these descriptions it is clear that equal importance is currently given to R&D and non-R&D innovation supports for enterprise. Data were available for seven indicators of the conventional EU Innovation Scoreboard and these are displayed in the chart, relative to a benchmark of 100 for the EU Member States. The BERD figure at 0.26% 0f GDP is disappointingly low despite a 21% share of high tech products in exports. This anomaly derives from Hungary’s sizeable FDI stock which is known to carry out very little research in the host country.

38 Figure 3.6

Innovation Scoreboard Indicators: Hungary (EU=100)10

% 4.5 ICT expenditure 4.1 High-tech venture capital 2.3 b High-tech patents (USPTO) 2.2 Bus ines s R&D expenditure 2.1 Public R&D expenditure 1.4 Em ployed in high tech m anuf. 1.2 Population with tertiary education 0

20

40

60

80

100

120

140

160

Source: DG Enterprise (2001), Innovation Policy in Six Candidate Countries (all data relates to 1999 except for 1998 USPTO data).

The depressed BERD performance comes after a proud historic record of investment in science & technology spanning several decades. 1989 marked not only a year of important political changes in Hungary, but also the first year that GERD dipped below 2% of GDP. In the early transition years both public and business expenditure on R&D fell away sharply and remained at a low level through the later 1990s.11 BERD was particularly badly affected when collapsing markets and shrinking profits forced companies to concentrate on short-term survival. The shrinking BERD resulted in the Government’s share of GERD reaching 65% by 1993. In 2000, when GERD had recovered to 1% of GDP, government funding still accounted for over 60%. This compares with an EU average of 35% for the Government component of GERD. The political transition saw the total number of researchers plummet to 60% of 1988 levels. The enterprise sector took the brunt of the casualties. In 1988, 40% of researchers were employed in enterprise; today the figure has only recovered to 30% of researchers.

10

In this Simulated Scoreboard values for EU are simple averages, rather than weighted averages that take the relative size of country economies into account. 11 GERD reached its nadir in 1996, slumping to 0.67% of GDP.

39 The main R&D funding for firms is allocated by the Ministry of Education (OM) through the Central Technological Development Fund (KMUFA).12 Raising the level of BERD has proved a daunting challenge. Most companies are focused on upgrading their production operations in terms of efficiency and quality and are more interested in consultancy services than in R&D. Tightened profit margins constrain their options and an under-developed capital market makes it difficult to access external funds for R&D projects. Of the large state companies, traditionally prime customers of research institutes, many have disappeared13 and others have switched to sub-contract work for TNCs in which case research ceases to be an imperative.14 The government offered tax concessions to the TNCs to encourage them to perform more R&D in Hungary. Since 1997 a number of TNCs have increased their R&D spending or established new research units. Most research is connected with production in the local affiliate, some is performed for the corporation. The government is trying to encourage more R&D operations by offering favourable loans worth up to €2m for R&D projects that guarantee employment for 30 researchers for a minimum of 3 years. A ray of light in the enterprise R&D story is the fact that the overall number of company research units began to increase again from 1996, and, by 2000, had recovered to 1988 levels.15 In addition to direct support for R&D in firms, high tech start-ups are stimulated. Policy measures to promote the conditions for new technology-based firms (NTBFs) include financial support for industrial parks.16 The leading industrial park, operated by an Austrian-Hungarian joint venture, is situated in Györ near the Austrian border and offers a high quality industrial environment, with dynamic educational institutions, for its current complement of 32 enterprises. In 1999 The National Committee for Technological Development (OMFB) instituted the TECH-START scheme to promote the creation of NTBFs in new growth areas. It is planned to extend this scheme as part of a broader regional innovation programme promoting spin-off companies and cluster development. 12

A Technological Development Fund was originally set up in 1959. This attracted matching funds from industry eager to use the research results. A National Committee of Technological Development (OMFB) was set up in 1961 to manage the Fund. In 2000 this Committee was downgraded to a technology policy advisory body, the R&D Division of the Ministry of Education. 13 A very few (notably, large Pharmaceutical companies) had sufficient strengths in depth to survive by relying on in-house research capabilities. 14 Both the companies and government are aware that such a strategy can lead to entrapment in noninnovative activity if not properly managed. 15 The culling of inefficient industry research units in the early transition period in parallel with the decline of traditional business activities was followed by the development of new independent R&D facilities in other sectors. 16 Technology Parks in the original strict sense of this concept (ie joint ventures between government and private enterprise) do not exist in Hungary.

40 Economic conditions in Hungary (low inflation, dynamic economic growth, developed financial institutions) are becoming favourable for the development of a risk capital market and the establishment of venture capital companies. To date, the scarcity of capital has constrained this development. Another problem for innovation policy makers is that present regulations orient the flow of capital towards lower risk, large company projects. We turn now to non-R&D supports for industry and the larger issue of improving industry capability. In the pre-1989 era, Hungary had strong connections between its industry and its research system. During the transition years, research institutes under the aegis of the Hungarian Academy of Sciences (MTA)17, and university research laboratories, suffered large budgetary cuts. There was also a huge drop in industry demand and the sectoral industrial research institutes were decimated. Today, apart from the Agriculture, Food & Forestry sector, companies have few systematic relationships with public research institutes. The universities sought to maintain their former strong relations with surviving large enterprises which now represent an important alternative source of research finance.18 In an effort to broaden this academia-industry collaboration the R&D division of the Ministry of Education has redesigned its R&D schemes to promote linkage and most HEIs now claim relations with industry. The major government initiative to promote research co-operation is the Cooperative Research Centres (CRCs) programme, launched in 1999. This was based on a similar US policy instrument. Its major aim is to foster strategic, longterm co-operation between higher education institutions, other non-profit R&D units and businesses. Five consortia have been approved to benefit from a subsidy pool of HUF 1083m to undertake 3 to 4 year projects. While hard times help to focus research institutions on market needs there is a danger that excessive curtailment of public funding may force them to concentrate on fee-earning low level, routine technical services. The Zoltan Bay Foundation for Applied Research has sponsored new applied research institutes, based on the German Fraunhofer model. These institutes are guaranteed a substantial level of public funds19 and the latitude to pursue high level applied research. The EU RTD Framework Programmes 4 & 5 have provided another opportunity for Hungarian public research institutions to renew their knowledge assets. These Framework Programmes have also been an important collaboration resource for 17

These institutes traditionally focused on basic research. From the innovation policy maker’s perspective, one benign result of the straitened transition experience has been that both universities and MTA institutes were forced to adopt a more market-driven approach to research. 19 The anecdotal figure is 2/3 of its costs (ie from federal and lander sources combined). 18

41 Hungarian firms that aspire to upgrade their production quality or develop new products. Hungary also participates in other international technology fora including EUREKA, COST, CERN and the scientific programmes of NATO. Mobility of senior staff between research institutions and industry has not been a feature of HE-industry relations in Hungary. The predominant mobility has been within these two sectors.20 There are no specific placement schemes in operation that might compare with the UK teaching company model. Research co-operation between TNCs and the research infrastructure has been minimal. The ICT sector is an exception where firms such as Ericsson, Nokia, Sony and Knorr-Bremse have located their R&D labs in universities such as the Budapest Technical and Economic Sciences University. Some of the large scale mergers and acquisitions, notably General Electric’s acquisition of Tungsram, were patently research motivated. However, the general run of greenfield TNC projects are more interested in recruiting researchers than in carrying out research locally. For these TNCs, the extent of their co-operation has been their contribution to the design of college curricula and to the fast-tracking of graduates in sectors where skills are in short supply. Capability-building in firms is supported by government on a number of fronts. The education system has boosted the output of third level graduates, up from 76K full time graduates in 1990 to 172K in 2000. Efforts are in hand to dismantle the rigid “Prussian” education bureaucracy in order to achieve a more flexible delivery system and a curriculum that develops the skills required for a global, knowledge-driven economy.21 EU Phare funding is being used to introduce new third level courses in Innovation and Technology Management and overcome the difficulties in securing qualified teaching staff. Intermediary organisations such as the Hungarian Innovation Association, the Hungarian Centre of Technology Transfer, and the National Business and Innovation Center have been established to provide information and consultancy services to firms seeking to upgrade their capabilities. The Széchenyi plan (2000) included a proposal for the development of a BUNT-like programme to teach the specific consultancy skills and innovation management techniques (IMTs) that business firms require to integrate their technology and business strategies. Increasing linkages between indigenous sub-suppliers and the burgeoning TNC sector has been high on the policy agenda. The INTEGRATOR scheme was launched in 1999 to build up the capabilities of sub-contracting firms. This is a scheme under which a TNC (the “integrator”) and its potential suppliers (a 20

A worrying statistic is that 60% of researchers who go abroad to complete their studies do not return. MNEs have complained that they had to spend months acquainting recruits with modern production practices. 21

42 minimum of two SMEs) apply jointly as a consortium to undertake a joint technological development project of 1 to 2 years duration. To date 26 consortia have been funded at an average outlay of HUF 20 m. In summary, the economic outlook for Hungarian industry is more optimistic than for late industrialising countries that do not have a historical memory of scientific and technological achievement. The current low level of business R&D is an aberration that is unlikely to persist, given the industrial traditions and the inherent technology absorptive capacity of indigenous industry. The future role of Hungary’s large FDI sector is more difficult to predict. A change of government took place in mid-2002 and new STI measures are in train. It is anticipated that one of the major commitments of the new administration will be an increased effort to improve the indigenous STI capability of TNCs. It may take some time, however, to overturn the inertia of the research infrastructure and to give the emerging enterprise sector a strong voice in the country’s Science and Technology Policy Council.

3.3.6 Slovenia The Slovenian bar chart (fig. 3.7) indicates a total industry-oriented innovation expenditure of 0.11% of GDP per annum. This level of expenditure places Slovenia at the top of the second tier of KNOGG countries. It also reconciles with a GERD of 1.4% of GDP, placing Slovenia within reach of the EU average. The public sector share of GERD, at 37% (1999), is around the EU average. Slovenia has nine entries in the simulated Innovation Scoreboard (2001) and three of these are above the EU average. In this respect Slovenia is best placed of the new EU member countries. The Scoreboard shows a high relative percentage of employment in high tech manufacturing, a high innovation intensity (spending on innovation as a % of sales), and a solid public sector R&D performance. Slovenia and Poland were the only new EU member that published statistics on their innovation performance.22

22

The general absence of data in this ‘transmission and application of knowledge’ section of the Scoreboard is indicative of a lack of emphasis on enterprise innovation in the CCs.

43 Figure 3.7.

Slovenia: Annual Expenditures by Objective 2001 (€m), Total 21.0 €m ~ 0.11 % GDP

Slovenia

10.0 8.0 6.0 4.0 2.0 0.0 1

2

3

4

5

6

7

Public Knowledge Human Enterprise High-Tech Risk Capital R&D CoStart-Ups markets operation Knowledge Diffusion Mobility R&D 40.48%

Figure 3.8

5.84%

0.00%

12.12%

6.93%

15.58%

8 Absorptive Capacity

19.05%

0.00%

Innovation Scoreboard Indicators: Slovenia (EU=100)

% 4.5 ICT expenditure 4.2 New capital raised 3.3 Innovation expenditure 3.1 SMEs innovating in-house 2.3 b High-tech patents (USPTO) 2.2 Business R&D expenditure 2.1 Public R&D expenditure 1.4 Employed in high tech manuf. 1.2 Population with tertiary education 0

20

40

60

80

100

120

140

160

180

Source: DG Enterprise (2001), Innovation Policy in Six Candidate Countries. (all data relates to 1999 except for 1998 USPTO patent data).

44 These results need to be more closely scrutinised in the light of specific country circumstances. In the new member countries (CCs), “high tech” industries tend to be specialised in low value-added segments, a fact borne out, in the case of Slovenia, by the low BERD and patenting scores. IMD data shows that valueadded per employee in Slovenian manufacturing and service industries was only a third of the EU level in the period 1999–2001. World Bank data for 1998 show that the percentage of high tech products in Slovenian exports was only 4% compared with 21% for Hungary (World Bank, 2001). The strong Slovenian innovation intensity indicator is driven by a small number of highly innovative small firms and a few innovative large firms, and masks the gulf between this dynamic group and the majority of non-innovating SMEs. The question that arises here is what story does the bar chart convey about industry-oriented innovation supports in Slovenia? What is the balance between resource-based measures (e.g. R&D subsidies and venture capital for high tech start-ups) and soft measures to improve the technology absorptive capacity of the vast majority of firms? And how does this reflect on the innovation performance of industry? The bar chart shows that some 46% of the STI financial support for industry is allocated to resource-based measures, 40% going direct as R&D subsidy and 6% specifically to high tech start-ups. There is no public allocation for risk capital although the EU study of innovation policy in CCs found that the Slovenian capital market, and its venture capital market in particular, was heavily underdeveloped. The Scoreboard shows that Slovenia lags well behind European countries in respect of the availability of capital for new innovative firms. Banks remain the major source of long-term financing. R&D in the enterprise sector is supported by two principal measures. The first measure is a direct project-based R&D subsidy which encourages firms to set up internal R&D units. The second measure is the Research Group Programme financing scheme, that guarantees a five-year budget (2000–2004) to 300 research groups to carry out rigorously vetted projects. While the bulk of this research is of a basic or strategic orientation it is estimated that 10% of the budget reaches enterprises for industrial research projects. The BERD statistic for Slovenia, at 0.75% of GDP, is in advance of other CCs but only 66% of the EU average. Before the transition Slovenia had a characteristically strong, public-dominated R&D system which carried out research for industry. Many of the public research institutes did not survive, leaving industrial clients to find other sources of research services while endeavouring to build up internal research capability. The major casualties of the transition occurred in the privatised business sector. Large vertically-integrated conglomerates were broken up, and the dismantling of their R&D units, in the

45 early 1990s, led to a massive brain-drain of some 3,000 industrial researchers. The number of researchers in the business sector per 1,000 labour force is currently only 0.7, compared with an EU average of 2.3. The Slovenian government recognises that R&D supports must be complemented by measures to increase the absorptive capacity of industry and to stimulate the diffusion of knowledge from public research to industry. The bar chart shows that 54% of its industry supports currently fall into in this category. A scheme has been established to promote joint projects involving at least three companies and a knowledge institution. Projects can relate to international markets, joint development of new products, processes or services, or programmes to assist firms to develop niches in production chains. In a second initiative, following a pilot study on industrial clusters, a two-year subsidy programme has been put in place to promote the development of clusters. These clusters will embrace large firms, SMEs and research institutes. The 1997 OECD Economic Survey for Slovenia pointed to light, high-tech industries as a potential area of competitive advantage whereas industry has traditionally been concentrated in medium-technology transport and chemicals sectors. The government is endeavouring to upgrade these parts of the industrial structure by sponsoring Technology Parks to assist new technology-based firms (NTBFs) in their early years. The Ljubljana and Maribor parks are the flagship projects. Allied to this, financial support is provide for incubator centres attached to universities and research institutes. The CCs have a tradition of strong investment in education but the emphasis has historically been placed on participation rates at second level and on vocational skills. The Innovation Scoreboard shows that Slovenia lags Europe23 by some distance in the percentage achieving tertiary qualifications, a vital cadre to fill the management and technologist positions in the new economy. The most recent data show a healthy output rate for the third level ‘engineering, manufacturing and construction’ grouping in Slovenia, but a relatively low flow of graduates into the ‘science, mathematics and computing’ stream.24 A Young Researchers Programme was established in the mid-1980s to encourage graduates to proceed to postgraduate research. Today some 300 to 350 fellowships are awarded annually. The mix among MSc degree and PhD holders has shifted. In 1985, of the fellowships awarded, 60% went to PhDs, while today’s proportion approaches 90%. In the early 1990s the overwhelming 23

And the average figure for Europe is less than 50% of the US average. Some data for the thirteen CCs is included in this third edition of the Enterprise Policy Scoreboard and this augments the Innovation Scoreboard picture in several areas: venture capital, entrepreneurship, human resources, and innovation & knowledge diffusion.

24

46 majority of participants went on to fill vacancies at universities or research institutes, where working conditions were perceived to be more attractive than in industry. This poor mobility tradition is beginning to change as new industry sectors offer more attractive careers for qualified research personnel. In summary, Slovenia is making good progress relative to other new EU countries, in adjusting its industry to a knowledge-driven market economy. But increased commitment to R&D and capability-building has not yet been leveraged into enhanced innovation output. The Slovenian innovation survey found that over 60% of manufacturing firms performed no product innovation in the period 1997–1998 and that 84% performed no process innovation. Patent performance was only 20% of the EU average and there is a low incidence of cooperation with external firms or research institutions. Most of the structures are now in place but the National Science and Technology Council has not fully embraced the vision of a new market-driven enterprise sector. Knowledge transfers from research to industry are retarded. Science and Technology are split between two ministries, Education and Economy, and the interface is not entirely satisfactory. R&D efforts, administered by the Department of Education, are heavily focused on academic and human capital objectives to the detriment of applied research and industrial production. The Ministry of the Economy has not adequately addressed the challenge of commercialising locally-produced research results. There is hope, however, that Slovenia’s industry-oriented innovation policies may be reviewed following a recent restructuring of the Department for the Development of the Entrepreneurial Sector and Competitiveness, at the Ministry of Economy.

3.4 Policy Implications 3.4.1 Innovation systems A wide diversity of innovation systems characterises KNOGG countries. This is inevitable given their different histories, cultures and stages of economic development. But this diversity does not justify conspicuous innovation system shortcomings in the face of well established and proven good practice. There is a canon of STI wisdom that can be applied without attempting to homogenise national innovation systems: • STI must have a strong voice at Cabinet. • Co-ordination across Ministries is vital. • Major stakeholders must have a voice in national STI Councils. • Policy implementation must be separated from policy making.

47 • Implementation must be close to the needs of clients (targeted to specific categories of firms, regions, technology clusters etc). • Regular evaluations leading to corrective action are necessary.

3.4.2 Industry-oriented innovation policies The vision and the consistency of a country’s industry-oriented innovation policies are decisive in achieving knowledge-based economic growth. Ultimate responsibility for product and process innovation rests with the enterprise sector of the economy. The innovation performance of enterprise reflects the effectiveness of industry innovation policies. A study of the innovation performance of KNOGG countries, in conjunction with their industry-oriented innovation policies, leads to the following conclusions: • There is a tendency among weaker countries to equate R&D policies with innovation policies • R&D subsidies, and support for firms with R&D capability (including startups), are the most widely used and best-resourced STI policy measures in the KNOGG countries • R&D support does not touch the needs of the great majority of firms, mainly SMEs in medium-tech and low-tech sectors whose stock-in-trade is what Nathen Rosenberg calls “the grubby and pedestrian forms of knowledge that are at the heart of economic progress” (Witness the small minority of R&D performing firms in most KNOGG countries and the failure of R&D subsidies to increase their number) • Improved technology absorptive capability, the first priority for the overwhelming majority of medium and low-tech firms, is best achieved through demand-driven initiatives such as improved inter-firm co-operation, participation in industry clusters, networking, shared skill enhancement and other learning experiences. • The specific assets required for a programme of soft innovation supports are hands-on skills and human resource-based experiential knowledge. Public sector agencies are, for the most part, ill-equipped to deliver such a programme. • The promotion of technology acquisition and diffusion requires the support of technology intermediaries, brokers, technology resource centres and transfer services. These institutions are underdeveloped in the lagging KNOGG economies.

48 • Market-seeking FDI, that operates without dynamic links to the domestic innovation system, can lead policymakers to neglect the critical need to upgrade the technological capability of domestic industry.

3.4.3 The Trend Chart on Innovation The EU Trend Chart has been running since 2000 and tracks the innovation policy developments of Member States and CCs. All six KNOGG countries participate in this venture which is called the Trend Chart on Innovation in Europe25. One of its main services is to compile a database of STI policy measures across Europe. These policy measures, however, are only assembled under three headings: Fostering a Culture of Innovation, Establishing a Framework Conducive to Innovation and Gearing Research to Innovation. This leaves a gap in areas where the benefits of open co-ordination would be of special benefit to weaker Member Countries and it is recommended that the database be expanded to: • Provide information on the policy formulation and implementation institutions and structures in member countries. There should be data. • Provide information on the research systems of member countries, e.g. on the extent to which the spectrum of research activity is appropriate for the particular country. Is the balance between fundamental, applied, industryoriented and technology transfer services appropriate? • Extend the scope of the existing Trend Chart category, “Gearing Research to Industry”, to incorporate non-R&D industry innovation policies. The current focus, as can be inferred from the title, is almost exclusively on measures to support research.

25

www.cordis.lu/trendchart/.

49 Table 3.3 No

European Innovation Scoreboard 2001

Indicator

Yr1 So.2

US

EUtotal

JP

FIN NL

IRL

GR

Hun Slov

EU av

1. Human resources 8.1

11.2

10.4

10.4

5.8

15.6

34.9

30.4

21.2

32.4

25.0

22.2

16.9

1

8.4

19.6

15.6

5.2

1.1

99

1

7.8

7.2

4.7

7.3

2.4

99

1

3.2

4.3

3.6

4.0

1.5

1.1

‰ S&E grads/ 20-29 pop

99

1

1.2

% pop with 3rd education

00 1,2

1.3

Life-long learning

00

1.4

% empl. h-tech manuf

1.5

% empl. h-tech services

9.32 15.6

15.9

23.25 9.65

8.49 10.18

6.29 3.27

2. Knowledge creation 2.1

Public exp. R&D / GDP

99

1

0.56

0.70

0.66

0.95

0.87

0.35

0.38

0.37

0.64

0.62

2.2

BERD / GDP

99

1

1.98

2.18

1.19

2.14

1.05

1.03

0.13

0.26

0.75

1.14

2.3a

EPO h-tech pats /pop

99 1,3

29.5

27.4

17.9

80.4

35.8

13.3

0.5

2.3b

USPTO h-tech pats /pop

98 1,4

84.3

80.2

11.1

35.9

19.6

3.8

0.5

19.14 2.08

1.52

11.65

16.9

41.01

3. Transmission and application of knowledge 3.1

% SMEs innov in-house

96

10

44.0

27.4

51.0

62.2

20.1

3.2

% SMEs innov co-op

96

10

11.2

19.9

13.8

23.2

6.5

3.3

% innov exp /total sales

96

10

3.7

4.3

3.8

3.3

1.6

1.08

1.38

1.62

0.65 0.04

1.1

0.3

5.6

0.9 1.5

6.5

7.3

6.9

8.4

15.42 3.9

3.41

4. Innovation finance, output and markets 4.1

‰ vent capital / GDP 00 1,5

4.2

% new capital / GDP

99 1,6

4.3

% new-to-markt products

96

4.4

% home internet access

00 7, 8

47

28

28.0

44

55

36 12

4.5

% ICT markets / GDP

00

9

5.9

4.3

6.0

6.0

6.6

4.8 6.0

4.6

% h-tech value added 97

1

25.8

13.8

8.2

12.5

7.5

5.6

3.8

4.7

2.9

Summary Index

1.9

10

20.5 1.2 -7.9

0.016

0.09 0.15

1.53 5.40 33.40

6.42

4.31

5.86 9.50

50

3.5 Using the Innovation Trend Chart The Innovation Directorate of DG Enterprise pursues the goal of open coordination of innovation policy across the EU. The Trend Chart project collects, updates and analyses information on innovation policies in the EU15 and in the new member countries, under the following three-way classification: Fostering an Innovation Culture, Establishing a Framework Conducive to Innovation and Gearing Research to Innovation. As of now, the Trend Chart coverage and depth of analysis varies considerably between countries and the project has been tracking innovation policy developments only since January 2000. For the purposes of this study, Trend Chart data were augmented by national innovation policy analysis supplied by the KNOGG partners. The Trend Chart has obvious limitations if used as the only data-source for policy analysis It concentrates on the innovation policies currently being pursued by each country and has little to say about the presence or absence of governance institutions to articulate and implement those policies. It is also silent on the extent to which the prevailing research system, in individual countries, is capable of feeding and sustaining those policies. The infrastructure for the acquisition, distribution and diffusion of knowledge in the partner countries is another area that is not well captured in the Trend Chart policy reviews. The evidence from the Trend Chart, and from the supplementary investigation carried out for this project is that there is a great variety of national innovation systems amongst European countries – and among KNOGG countries. The level of R&D spending and the relative shares of publicly financed R&D vary enormously. There are also large differences in the infrastructure and institutions of NIS, and in STI policy decision-making. The mobility of researchers and engineers between universities and other elements of the research system and between both of these and the business sector, was used as an indicator of knowledge flows. Country differences are very obvious in these areas and also in the extent to which policies exist to improve awareness among key decisionmakers and the general public, of the economic and social returns to ST&I.

3.6. Conclusions There are no metrics to quantify and calibrate country performance in areas such as STI institutions, technology infrastructure and innovation culture. However, in industry-oriented innovation policies the performance indicators are more developed and quantification of resource allocation more feasible. Ultimately knowledge-based growth is a function of a country’s capacity to generate new and improved products and processes. The KNOGG project attempted to quantify the resources allocated to industry-oriented innovation policies and to calibrate country performance in this domain, using the EU Innovation Scoreboard and other innovation output instruments. The KNOGG countries

51 were benchmarked on eight industry-oriented innovation policy objectives, representing best practice in promoting innovation across all stages of the innovation process from knowledge generation to knowledge application. These policies extend well beyond support for R&D in single firms and investment in high-tech start-ups to a range of networking and knowledge diffusion measures. They also include policies to improve the absorptive capacity of firms such as the deployment of innovation intermediaries and schemes to improve the capability of firms to acquire and use technologies and innovations from external sources. The quantification of resources that are allocated to industry innovation in KNOGG counties extends beyond the range and depth of analysis associated with the EU Trend Chart. Each KNOGG country supplied data on the resources it allocated, in the most recent year, to each of eight firm-based policy interventions. The data for the six countries were standardised as share of country GDP and displayed in bar-chart form. These charts give insights into factors that differentiate the firm-level innovation policies of successful innovation regimes from those of less successful ones. All countries encountered difficulty, some more so than others, in disaggregating innovation budgets to the level of detail required. A minority of countries have sophisticated, integrated technology packages that provide support across a range of intervention points in the innovation process. More importantly, resource allocation is not a surrogate for impact or effectiveness. Many of the ”softer”, knowledge-intensive innovation supports are not resource-intensive and have to be qualitatively assessed. The KNOGG framework clearly points to the alternative 'imitation' growth path for Hungary, Slovenia, Greece, Ireland and, to a more limited extent for the Netherlands. These economies could leverage the potential of their STI policies to make the transition to the higher path of 'innovation' growth. Explicit policies must be put in place to acquire and absorb technology from external as well as internal sources. The role of bridging institutions and intermediaries is critical. A relatively high proportion of the industry-oriented segment of the national S&T budget should be allocated to knowledge distribution and diffusion measures.

52

4. Small countries, FDI and multinationals FDI provides an opportunity to raise investment, employment, economic growth and technological capabilities. In technologically more advanced countries, foreign affiliates may not only be a source of foreign technology, but they may actually set up local R&D facilities and innovate as part of the national innovation system. For small countries the complex relationship between underlying motives, policies and outcomes in terms of technological development and growth is all the more an important challenge to tackle and master. In terms of investment and export oriented growth, at best TNCs can integrate local SMEs and research into their global production chains. The impact of multinational enterprises on the technological base of a country can be measured in technology transfer and diffusion, i.e. externalisation of the ownership of technology. The latter is of main interest to the host countries’ policy makers. A distinction should be made between technology transfer (between firms across national borders) and dissemination of technological knowledge. TNCs may well transfer technology and upgrade the production capacities of the acquired plants to globally competitive levels, but disregard other host country firms in their supplier chains or other forms of linkages which might result in knowledge spillovers into the host country technological base. Meanwhile, as reported in KNOGG Working Package 1 (Berghäll et al., 2002), the rapid diffusion of technology is a prerequisite for small countries to catch-up or to remain at the fore-front in the technology race. There are a number of important options at the disposal of small host country governments to encourage the transfer of technology by foreign TNCs, ranging from doing nothing to developing the local innovation system. The range of available policy options to a small country as host to foreign R&D-intensive TNCs is defined by the way the country’s productive capacities can link to the TNCs global production structure. These abilities and the degree and quality of the integration can be improved with the help of FDI and S&T policies. The critical issue is how innovation policies can be framed so that they encourage the diffusion of technology and consequently upgrade the domestic technology structure of the host country. Particularly interesting for small European countries, is to attract the hightechnology production phases of TNCs in order to upgrade their technological knowledge base. The attraction of R&D intensive TNCs does not necessarily lead to upgrading in the technological knowledge base of the country. High-tech industries also incorporate low-tech phases in their production processes. While it is necessary to develop policies that make small countries attractive for TNCs, small countries also need to build a strong national innovation system (NIS) and

53 stimulate indigenous technological progress. Domestic firms are important factors to attract R&D intensive strategic asset seeking TNCs. While TNCs diffuse and create knowledge with positive externalities, the extent of these knowledge spillovers depends crucially on the willingness of the TNC to share information with local firms and S&T institutes. This in turn may depend on the ability of the domestic knowledge base to offer something in return. Therefore, countries with strong domestic technological capabilities, such as Finland, are better positioned to attract strategic asset seeking TNC affiliates with R&D facilities than countries, such as Ireland or the Netherlands, with lower levels of technological activity. While financial incentives aimed at attracting R&D intensive FDI may be necessary for a small country, such schemes need to be supplemented with policies that stimulate learning and investments in domestic firms.26 This is our main argument, i.e., to generate effective technological diffusion, an efficient national innovation system is required. It is particularly important to strengthen the technological absorptive capacities of local firms to learn from foreign TNCs and to create knowledge spillovers. Higher technological spillovers, in turn, may motivate further FDI or its technological upgrading. Such fertile conditions would most likely result in superior results relative to targeted industrial policy of picking winners or stimulation of selected high technology sectors. Even if successful in attracting large foreign TNCs or building their own, small countries need to constantly keep building, diversifying and broadening the R&D base of their small and medium sized enterprises (SMEs). Focusing on the large TNCs should not entail the neglect of local SMEs that have difficulties in following the pace of technological progress. In other words, FDI alone is not likely to guarantee a long-term solution. Instead, whatever the situation, countries need to strive for a strong national innovation system based on local capacities. In addition to the promotion of indigenous technological progress through technology policies aimed at building or improving the national innovation system, another central course of action is the encouragement of cross-border R&D cooperation, such as inter-country research programmes within the EU (e.g. ESPRIT). Jacobs (1989) argues that the logic of private enterprises that arbitrarily move activities from one country to another is fundamentally different from that of nation states, and therefore, the interests of these two entities do not coincide. It is important to make sure that the TNCs are embedded in a national network of high quality small and medium enterprises, which in turn are not too dependent on the larger enterprises for their technological experience, or, more generally, their continuity. Integration in such networks will restrain TNCs from closing down their local operations too quickly. Similarly, other authors have warned of 26

On the economics of FDI, see Blomström and Kokko, 2003.

54 the danger of becoming a branch plant economy dominated by footloose multinationals (Dicken, 1992).

4.1 Motives for the internationalisation of R&D There are four types of motive for the internationalisation of R&D activities within the firm: (1) market driven R&D, (2) national treasure R&D, (3) technology driven R&D and (4) global R&D. Market driven FDI keeps research at home while development activities are dispersed across markets. Business development is dominated by customer demand with a low scientific component. Small market size makes this type of R&D relevant only to favourable geographic locations vis-à-vis larger markets, as e.g. in the case of the Netherlands in regard to Germany or Ireland relative to the United Kingdom. It is therefore connected to the export platform-seeking motive of TNCs, with implications for the nature and extent of technology transfer to the host economy. Host-country technology policy has little relevance to market-driven R&D, as it is primarily influenced by factors within the sphere of market access, customer proximity, local market requirements, as well as by factors affecting the cost structures of the firms, such as country specific cost advantages and economies of scale in manufacturing. Instead, cost factors can at least partly be addressed by financial or fiscal policies. National treasure R&D is maintained close to headquarters, and therefore there are few spillovers to host countries. Technology driven and global R&D seek host economies offering favourable conditions for research at the technology frontier. In this case, financial or fiscal incentives are likely to prove ineffective if not irrelevant. Instead, the quality of the physical and technological infrastructure, R&D facilities, innovation centres, research parks, universities, education, brain drain and the entire national innovation system require constant attention. This focus can also help in hanging on to domestic TNCs and in identifying appropriate high technology niche markets in which to specialize. The emphasis is on strengthening technological progress in local firms to improve their absorptive capacities vis-à-vis foreign TNCs and promote the diffusion of technology.

55 Figure 4.1

Principal determinants and trends in the internationalisation of R&D

Market-Driven Dispersed

Development

Global

Support of local development

Developing business opportunities

Customers and standards

Domestic National Treasure

Domestic

Access to local science

Technology-Driven

Dispersed

Research

Source: Von Zedtwitz & Gassmann (2002).

Table 4.1

Main motives to invest in KNOGG countries and technology implications of inward investments in the 1990s

Country

Motives

Technology implications

Finland Ireland Netherlands

Outward Market/efficiency Market/strategic Market/strategic

Inward Strategic Sourcing Export platform/ efficiency Exploiting Export platform/efficiency Exploiting

Greece

Market

Export platform

Hungary Slovenia

Market Market

Export platform/efficiency Exploiting Export platform/efficiency Exploiting

Exploiting

Inward investment motives in the KNOGG countries are diverse and often determined by the geographical position of the countries. Finnish inward investments have had a strong strategic asset component since the beginning of the 1990s. The strong domestic R&D base – the result of government’s innovation and technology policies – attracted R&D-intensive foreign firms to Finland. The geographic motive for FDI is particularly relevant for Ireland and the Netherlands. Their favourable location in the vicinity of large neighbouring markets allows them to attract market or export-platform-seeking and efficiencyseeking foreign investments. Furthermore, wage moderation policies implemented by the Dutch and Irish governments in the last twenty years has

56 resulted also in efficiency-seeking investments mainly from other EU-countries. Greek, Hungarian and Slovenian inward investments are also dominated by these two motives but to a lesser extent. The Netherlands’ central geographical position relative to large markets such as France and Germany makes it an attractive place for European Distribution Centres (EDC). Although these are characterized by low technology and low value-added activities, in the course of time secondary production activities have followed and increased innovation in these centres. The establishment of EDCs fits well into the existing transport and logistics structure of the Netherlands but not in the S&T system. The rising services sector (particularly the transport sector) in the Dutch economy attracts foreign firms that are innovative, but in non-technological fields such as organisational forms or distribution chains, and in consequence foreign firms in the Netherlands are more innovative in nontechnological fields than local firms. The increasing importance of services in the Dutch economy, and of foreign firms’ non-technological innovations, suggest that the future potential for “soft technology” knowledge spillovers is rising.

4.2 FDI related policies in KNOGG countries The scope of policies that affect FDI is broad. They can be categorised into three groups: (i)

general macroeconomic policies with an impact on FDI flows;

(ii)

other non-FDI targeted policy measures with an impact on FDI flows; and

(iii)

targeted FDI policy measures.

In most KNOGG countries, overall macroeconomic objectives have dominated STI policy decision-making. Targeted STI policies have been formed only gradually as the need for more specific efforts has become obvious. Tax or financial incentives for investment have been used in all KNOGG countries, but often in a regional policy context aimed at raising domestic and foreign investment in lagging areas. None of the KNOGG countries have introduced openly discriminatory practices in regard to foreign firms, although especially Irish tax policies have been very favourable to inward FDI. The role of TNCs in technology transfer and knowledge diffusion is different for each KNOGG country. Finland is the only one of the KNOGG countries that actively promotes S&T policies that enforce the national innovation system, which makes the country an attractive place for technology intensive inward FDI. Although Ireland, the Netherlands and Hungary also try to attract FDI with linkages to their domestic economy, these are often not linkages to the national

57 innovation system. Greece and Slovenia use financial incentives to attract TNCs but do not have explicit policies that aim at increasing linkages. 4.2.1. Finland Finland is an example of a country that attracts relatively little FDI particularly in comparison to Ireland and the Netherlands, but enjoys a higher technology component in what it receives. Finnish inward investments have had a strong strategic asset component in the 1990s. Many high technology foreign firms have been motivated to invest in Finland due to the potential for technology sourcing. While in theory this might reduce technology spillovers, case studies of foreign ICT companies in Finland have revealed extensive co-operation links with Finnish universities and R&D institutes, backward linkages to Finnish suppliers and/or horizontal cooperation with Finnish firms. Both vertical and horizontal linkages with domestic firms and R&D institutes offer avenues for technology diffusion, with its extent depending mainly on the absorptive capacities of the parties involved. National S&T policy in Finland has focused on networking and the creation of centres of innovation and platforms for technology firms. TEKES (the National Technology Agency of Finland) directs R&D funding to companies based in Finland and to Finnish Research Institutes and universities. Finland is the only KNOGG country where policy makers have been committed to a long-term strategy to actively strengthen the national innovation system by building local innovation capacities. Finland’s total R&D expenditure in percent of domestic production is substantially higher than that of other KNOGG countries. Moreover, the local corporate sector accounts for a majority of this expenditure, even with the exclusion of the considerable impact of one single firm, Nokia.

58 Figure 4.2

R&D expenditures by foreign affiliates and national firms as a share of domestic production in the industrial sector

3,5

% of domestic production

3 2,5 2

National firms (%) Foreign affiliates (%)

1,5 1 0,5

H un ga ry

G re ec e

ds N et he

rla n

nd Ire la

Fi nl a

nd

0

Sources: AFA Database May 2001.

4.2.2. The Netherlands The relatively higher R&D expenditures per unit of domestic production for the indigenous sector is true also in the Netherlands, which homes large multinationals like Phillips with huge R&D budgets. At the same time, the presence of large home country TNCs constrains the policy scope to control R&D structures and conduct independent STI policy. In this case, policies need rather to adapt to the circumstances and focus on technological diffusion and spillovers between different types of actors. In the Netherlands, the most important fiscal incentive in regard to R&D is a tax relief aimed at stimulating R&D (see also Cogan and McDevitt, 2003), but its effectiveness has been a matter of dispute. With regard to S&T policies measures also exist to stimulate R&D-co-operation among firms and between firms and S&T institutions. Technological loans for high-risk innovations are provided. The budget cuts in education in the last twenty years have led to a shortage of technologically skilled workers. This trend is expected to continue to finally erode the attractiveness of the country to future high technology investments.

59 4.2.3. Ireland The FDI oriented approach emerged in Ireland as an attempt to tackle macroeconomic imbalances, and has resulted in rapid economic growth, but also technological success. The success has been due to investments by large foreign multinationals from leading technology countries, using Ireland as an exportplatform to EU markets. Combined with an abundant educated local labour force and other advantages, Ireland’s main weapon to attract FDI has been tax incentives not only to FDI but also to R&D. Past discrimination against local firms has been eliminated due to EU regulations on non-discrimination. The widely advertised tax incentive schemes in place in Ireland have indeed brought about tremendous success in terms of economic growth, toppling all other European countries in the end of the 1990’s, but low- and mediumtechnology production phases have continued to dominate. While the high technology content of the investments has been on the rise, local linkages between TNCs and domestic firms have been weak and provided little dynamism to the national innovation system. Irish inward investments are certainly present in high technology firms but they rarely announce new product innovations. Table 4.2

Knowledge characteristics of foreign and indigenous firms in Finland, Ireland and the Netherlands, in percent of all firms % innovator All firms Large firms*

Ireland 1993-1995 Foreign Indigenous The Netherlands, 1994-1996 Foreign Foreign + Indigenous Finland, 1998 Foreign Foreign + Indigenous

% own R&D** All firms Large firms*

67 64

78 78

27 25

42 43

59 37

77 64

35 20

60 50

44 36

54 46

44 34

62 51

* = firms with 100 or more employees ** = own permanent R&D (> € 127,000 spent) Sources: Forfas database, Ireland, Community Innovation Survey Database Finland and the Netherlands (CIS-2) and Dutch Ministry of Economic Affairs (2001).

Despite all this, the table above shows Dutch and Irish firms to be clearly more innovative than Finnish firms, in spite of higher own R&D in Finnish firms and the innovator status of the country. There are several alternative explanations to the numbers above. First, it is possible (though unlikely) that the productivity of R&D is low in Finland. Second, it is also possible that R&D facilities are used to extract technological knowledge from the domestic economy for export to the foreign R&D base in accordance with the strategic seeking motive. The risk with

60 technology sourcing FDI is the negative externality it may place on productivity growth, as argued by Driffield and Love (2002). The reverse side would be countries like Ireland and the Netherlands, where firms may be innovating based on R&D results produced elsewhere. Although the Netherlands is home to large multinationals, like Philips, with huge R&D budgets per unit of domestic production, they may well further complement their innovations with research results produced elsewhere. Reporting practices may also differ and Finnish firms may focus on technological innovations, while innovation may be more broadly reported in other countries. The table 4.2 also shows that in Ireland, foreign firms perform slightly more R&D than indigenous firms, while this difference disappears when only large firms are taken into account. Cogan (1992) found the level of product innovation based on new product announcements published in the trade press developed by Irish manufacturing industry to have been 0.19 innovations per 1,000 employees. This figure referred to new or substantially improved products, excluding process innovations. This compares with a figure of 0.41 for the Netherlands and 0.46 for the US. While FDI is present in local high technology firms, this has rarely resulted in new products, suggesting Ireland to be an important location for hightech FDI, but only in regard to production phases with a lower technology component. Consequently, global TNC production and innovation processes engage the local resource base and provide dynamic linkages to the domestic innovation system to a negligible degree. Since the early 1990’s, developments in Ireland have been rapid. As the table below shows, Irish industry has made considerable leaps in the high-tech content of its manufacturing in just seven years. Table 4.3

Value Added in Manufacturing in Ireland (FDI + indigenous).

Low-tech Medium low-tech Medium high-tech High-tech

1991 46.7 11.6 28.7 13.0

1994 43.0 10.8 31.2 15.0

1998 32.5 8.9 41.5 17.2

Source: Cogan, McDevitt and Godinho (2001).

4.2.4. Hungary In Hungary, the privatisation process of the 1990s brought an influx of foreign investment. The foreign companies included large TNCs, which rapidly integrated their Hungarian plants into their globally integrated production networks. While this accelerated transition made Hungarian value-added globally competitive and contributed to exports, a globally integrated supplier system, by definition, undermines the local supplier network. Consequently, supplier-buyer

61 linkages weakened, together with the knowledge diffusion mechanism. The Hungarian experience shows that while TNCs can rapidly transform the firms they acquire into highly competitive ones, the rest of the economy can be totally excluded from the process. Over the course of time linkages between TNCs and Hungarian firms have strengthened, perhaps due to learning effects on both sides. Part of this improvement can be attributed to the Supplier Target Program of the Hungarian government, which was specifically aimed at intensifying linkages and boosting knowledge spillovers. Yet, TNCs fail to fully exploit local technological capabilities and absorptive capacities. The national innovation system also suffers from a lack of applied R&D institutes, while universities are burdened by massive education responsibilities, which erode their research capacities. One needs to underscore the importance of technology imports, but at the same time emphasize the importance of implementing measures to enhance local absorptive capacities and encourage spillovers. Consequently, the S&T policy aspects of broader growth strategies with FDI components need to be better addressed to bring about technological progress. Tax incentive opportunities for R&D activities exist also for R&D activities sourced out to other organizations like universities and non-profit R&D institutes. Hungarian tax policies include tax breaks, free trade zones and industrial parks as well as grants, interest subsidies and loan guarantees for the SME sector. In particular, the free trade zones have proved attractive for TNCs. It is not clear which of these measures can persist after EU accession. With regard to S&T policies, R&D expenditures have been fully tax deductible since 2001. R&D related support measures in regard to TNCs are discussed in section 3.3.5 in more detail. 4.2.5. Greece and Slovenia No significant technology spillovers could be found in Greece and Slovenia. In both countries some fiscal measures exist but these are not part of explicit S&T policies. In Greece, the labour market is very rigid and a major shortage of engineers exists. Slovenian tax policies are such that corporate taxes are not considered an impediment for FDI. There are no special incentives to encourage S&T cooperation and development. Slovenian investments in the education system are insufficient to make the country an attractive place for high technology components of high technology firms.

62

4.3 Outward FDI Small markets do not enable enterprises to recoup high and rising R&D costs, particularly in a time of shortening product life cycles and increased competition. As a result R&D intensive firms from small countries need export outlets and outward foreign direct investments to achieve scale economies that firms from large countries can achieve in their domestic market. The resulting internationalisation of the production and technology of small country firms is not a weakness of the national innovation system but a mere consequence of being small. FDI is of great importance for small countries to catch-up technologically. While outward foreign direct investment may not need to be stimulated, it should be perceived as an opportunity or pressure to up-grade local production. For instance, according to Moon et al. (2001), Singapore has been the most successful among the NICs, due to the combination of inbound FDI that attracts foreign capital and technology, and outbound FDI that allows access to cheap foreign labour and natural resources. In all KNOGG countries, outward investment flows have been mainly marketseeking, as can be expected when small countries are concerned. A minor part of the outward investments is also motivated by efficiency-seeking (Finland) or strategic asset-seeking (Dutch investments with regard to banking services). Future research should focus on a more detailed investigation of types of R&D co-operation between foreign firms on the one hand and domestic firms and S&T institutions on the other hand in a number of small and large EU- and accession countries. A quantitative analysis of these forms of R&D co-operation could reveal systematic patterns and detailed determinants defining foreign involvement in co-operation networks with domestic firms and S&T institutions. In addition it would make clear what target variables future S&T policies in small countries should aim at, and how such policies can be framed at the national and EU level.

63

5. STI policy instruments on EU level 5.1 General objectives set by the EU The European Commission proposed the creation of a European Research Area (ERA) in January 2000. Soon after that, the Heads of State and Governments of the EU member countries accepted this project at the Lisbon European Council in March 2000. The ERA project was seen as a central component in the establishment of a European knowledge-based society. The European Council also set a series of objectives and an implementation timetable for the period up to 2010. The background to these decisions was a concern about the competitiveness of Europe vis-a-vis the United States and Japan, the technological leaders of the world. Already in the early 1980s there were voices warning that the slowgrowing Europe would fall behind the United States and Japan in major fields of science and technology. In the mid-1980s, Japan’s emerging leadership in information and communication technologies was perceived by Europeans as the threat to their own competitiveness, but today it is the robust US economy that constitutes the new reference model for a majority of policymakers and opinion leaders. Although Europe is still at the top in many areas of science, there is growing evidence that the European higher education and research system fails to attract enough (or the brightest) people and investment, from within Europe and or worldwide. For instance, most highly skilled Asian emigrants turn to the USA, not to Europe. The Commission’s 2000 Communication Innovation in a knowledge-driven economy27 identified five priorities to steer Member State and EU-level actions to promote innovation: they were (1) coherence of innovation policies, (2) a regulatory framework conducive to innovation, (3) creation and growth of innovative enterprises, (4) improved interfaces in the innovation system, and (5) a society open to innovation. Funding for research and education is still lagging in Europe, particularly in the business sector. Indicators show that the United States and Japan lead the EU in terms of investment in the knowledge-based economy. Average public spending on education in the EU as a percentage of GDP remained unchanged at 5 per cent between 1995 and 2000 and that was less than in the US. Only Denmark, Sweden, Austria and Finland matched the American levels. As a result, Europe is lagging in tertiary education (university and non-university third level training); 27

COM(2000) 567.

64 only Finland (32 per cent of the population) approaches the level reached in Japan (close to 40 per cent) or the United States (35 per cent). Also research spending by governments and business is lower in the EU than in the United States and Japan: the figure was only 1.9 per cent for EU compared to 3 per cent for Japan and 2.6 per cent for the US in 1999 (figures in relation to GDP). Within the EU, only Sweden and Finland exceed the 3 percent of GDP share. The absolute difference becomes very large if one takes into account the higher US GDP. This investment gap is mostly due to lower investment by business. Moreover, the average annual growth of business research spending was lower for the EU than for the US although it was slightly higher than for Japan, during the second half of the 1990s. In terms of research carried out in the public sector, the picture is rather different. In higher education research, Europe spends a similar proportion of its GDP as its two main competitors (around 0.4 per cent), while government research is higher in the EU and Japan than in the US. The ERA proposal aims to change this situation. The European Commission proposed in January 2002 that the European Council would endorse action to strengthen European research and innovation by setting a target of 3 per cent of GDP for the overall level of public and private spending on research and development by the end of the decade. Within that total, the amount funded by business should rise to around two-thirds as against 55 per cent today. The Heads of State and Governments of EU countries endorsed this objective in March 2002 in Barcelona. This the first time that a commitment was made to a quantitative target for research, at the EU level.

5.2 The EU research programmes The realisation of the ERA objectives, however, will depend largely on decisions made in the member countries at national and regional levels because only a small fraction (5–6 %) of the total European research spending is channelled through the EU. The EU influence on research policy is articulated through the Framework Programmes and other Community level research spending. In spite of the growth of these programmes, they are still a minor component of the total RTD spending in Europe. The Framework Programmes constitute the main funding instrument of the EU. Their objective is to develop a true European scientific community, equipped with the best skills and know-how, and to support scientific and technical work of the highest quality, conducted through transnational projects, benefiting from mobile researchers.28 The Sixth Framework Programme (FP 6) was launched in 28

On the spillover effects generated by the Framework Programmes, see Dumont and Tsakanikas, 2002.

65 2002. It represents the third largest operational budget line within the EU, after the Common Agricultural Policy and Structural Funds. It comprises 3.9% of the EU’s overall 2001 budget (3.4% of 2002) and 5.4% of all 2001 public (nonmilitary) research spending in Europe. Over the period 2002–2006 FP 6 will distribute €17 billion to the parties involved in European RTD. FP6 (2002–2006) focuses European partnerships on seven key priorities: life sciences, genomics and biotechnology for health; information society technologies; nanotechnologies and nanosciences, knowledge-based multifunctional materials and new production processes and devices; aeronautics and space; food quality and safety; sustainable development, global change and ecosystems; and citizens and governance in a knowledge-based society. The new instruments of FP 6 are likely to mobilise a much greater proportion of national resources than previously because they aim to integrate national research activities. It is thus hoped that the FP 6 will have a noticeable impact on Europe’s research specialisation. After a research agenda is set at the Community level, it influences also national policy agendas. As Tsipouri and Xanthakis (1993) have shown, the influence is greater in those member countries where a structured policy-setting process is lacking.

5.3 Trend in SME Participation in Framework Programmes (FP 3 to FP 6) Small countries, and particularly those in the catch-up phase of their development, can support only a few, if any, large high tech domestic companies. While they make every effort to stimulate their high tech sector, particularly high tech start-ups and spin-outs, the imperative of their industry policy must be to upgrade the technological capability of their mainstream industry that consists over-whelmingly of SMEs in medium tech and low tech sectors29. The KNOGG project explored the extent to which the EU R&D Framework Programmes, “one of the biggest research funding programmes in the world”30, impacts on mainstream industry in small Member States and Candidate Countries. The importance of SMEs is widely recognised, and DG Research has given considerable thought and energy to improving SME participation in its R&D programmes. A number of so-called Technology Stimulation Measures for SMEs (TSMEs) are designed to give SMEs additional opportunities to participate, outside the priority thematic research areas. The most important of these are the Co-operative Research and the Exploratory Awards instruments. Co-operative 29

The impact of high tech manufacturing on GDP was ‘surprisingly small’, accounting for less than 3% of GDP in the US, the OECD country with the largest share of high tech manufacturing. See Keith Smith, 2002. 30 EU Research Commissioner Philippe Busquin.

66 Research (CRAFT) projects enable groups of SMEs which have similar technical problems, but lack the means to do research, to engage third parties (called “RTD performers”) to carry out research on their behalf. The Commission will share with the group of SMEs the cost of the research by funding up to 50% of the total project costs. The RTD performers will be fully paid for the work performed, but the SMEs will own the results of the research. Exploratory Awards allow SMEs to benefit from EU funding (up to 75%) for preparing a better research proposal, be it for a mainstream or a co-operative research project. Table 5.1 gives the trend in the main indicators of SME participation in Framework Programmes from FP 3 (1990–1994) to FP 6 (2002-2006). Overall, EU research funding increased substantially under FP 4 and has remained at approximately the same level in FP 5 and FP 6. The SME share of this increased funding has also remained constant at around 10%, although the Commission has set a target of 15% for FP 6. Questions have been raised as to whether this target is achievable given that the new instruments introduced under FP6 raise the barriers for participation by SMEs. It is also noteworthy that SME participation, measured by proportion of contracts received, is significantly higher than the proportion of research funds. This can be explained by the fact that approximately half the SME contracts relate to the special stimulation measures for SMEs (TSMEs) described above. TSME contracts have only a fraction of the funding allocated to mainstream research contracts. This section will trace SME participation in successive R&D programmes from FP 3 to FP 6 and will differentiate between participation in mainstream research and participation in the special SME stimulation measures. The issue of the quality of SME participation, and whether it has kept pace with the increase in the number of contracts awarded, and with the overall increase in resources allocated to EU Research, will be examined in a later section. Table 5.1

SME Funding Trend and Participation (contracts) in Framework Programmes (1990–2006)

Total Funding (Euro bn) SME Share (Euro bn) SME % of Funds SME Participation (propn. of contracts)

FP 3 4 0.7 17.5% n.a.

Source: EU reports (various). * This figure excludes about Euro 1.3 bn for Euratom.

FP 4 13 1.4 11% 28%

FP 5 15 1.4 9.4% 33%

FP 6 16* 2.2 (est) 14% (est) n.a

67

5.3.1 SME Participation in FP 3 (1990–1994) and FP 4 (1994–1998) Good data are available for SME participation in these two programmes31. The overall Framework budget increased more than threefold from 4 bn Euro in FP 3 to over13 bn Euro in FP 4. The funding for SMEs doubled from 0.7 bn Euro in FP 3 to 1.4 bn Euro in FP 4. Figure 5.1 illustrates how the funding was distributed between the three categories of project in which SMEs participated and how this distribution changed between the two programmes. 90% of SME funding in FP 3 came from participation in mainstream research projects but this figure fell to 77% in FP 4. Conversely, funding for SME participation in the special TSME measures increased from 10% in FP 3 to 23% in FP 4. Figure 5.2 shows that the number of SME contracts increased nearly threefold from 5,425 in FP 3 to 14,754 in FP 4. This increase was shared equally between mainstream research and Special SME Stimulation Measures. SME participation in mainstream research projects reached nearly 8,000 enterprises under FP 4. Figure 5.1. FP 3 €732m

Funding of SMEs in FP 3 and FP 4 by Project Type Mainstream Research Co-operative Research (CRAFT) Exploratory Awards 9%

FP 4 €1,427m

1%

Mainstream Research Co-operative Research (CRAFT) Exploratory Awards 4%

19 %

77 %

90 %

31

The data in this section are taken from EU 1998, “Framework Programme 6, SME Participation 1994–1998”. FP 4 projects signed up to August 1999 are included.

68 Figure 5.2. FP 3 5,425 contracts

Participation of SMEs in FP 3 & FP 4 by Project Type Mainstream Research Co-operative Research (CRAFT) Exploratory Awards

FP 4 14,754 contracts

Mainstream Research Co-operative Research (CRAFT) Exploratory Awards

18 % 29 %

49 %

53 %

29 %

22 %

5.3.2 SME Participation in FP 5 (1996–2002)32 The overall research budget for FP 5 was increased from 13 bn Euro (in FP 4) to 15 bn Euro and 10% of this budget was earmarked for SMEs. In the event a total of 20,000 SMEs participated and the targeted SME budget of 1.4 bn Euro was met. The Commission Services took additional measures to improve SME participation in FP 5. A completely new strand “encouraging SME participation” was included in the Workprogramme for the Specific Research Programme relating to Innovation & SMEs. This encompassed establishing “a single complementary entry point” and introducing common management tools. A new action line, “Technological and Economic Intelligence”, was initiated in order to “guide the choices and orientation of SMEs towards research”. These measures were aimed at SMEs “with limited or no internal research capability”, in industry groupings dominated by SMEs, and on specific topics of interest to large communities of small and medium sized enterprises. Collective research projects involve at least one European industrial association, or at least two national industrial associations from at least two different Member States. The associations define the research topics that are of interest to their member SMEs and then outsource the required research to RTD performers. Figure 5.3 compares the allocation of funding to SMEs by project type under FP 4 and FP 5. This shows that there was no change in the proportion (77%) 32

A number of FP 5 contracts were still outstanding at the end of 2002 and the statistics for this Framework Programme are provisional.

69 allocated to mainstream projects. In addition, updated figures for the full FP 5 period (1996–2002) show that SMEs received around 1.4bn Euro under both of these FPs. Thus the SME share of an increased budget fell from 11% to 9.4%. Figure 5.4 compares the proportion of contracts going to mainstream projects as opposed to special SME measures. Even though the total number of SME contracts increased from around 15,000 to approximately 20,000 under FP 5 about half the total were for contracts under the special TSME measures (Exploratory Awards and CRAFT), receiving only some 20% of the SME funding allocation. In summary, SME participation under FP 5 registered only a qualified improvement. The number of firms participating increased but the budget they shared did not. As a result the SME share of total budget fell from 11% to 9.4%. Figure 5.3. FP 4 €1,427m

Funding of SMEs in FP 4 & FP 5 by Project Type Mainstream Research Co-operative Research (CRAFT) Exploratory Awards 4%

FP 5 €1,186m (April '99April '00)

Mainstream Research Co-operative Research (CRAFT) Exploratory Awards 2%

19 % 20 %

77 %

78 %

70 Figure 5.4. FP 4 14,754 contracts

Participation of SMES in FP 4 & FP 5 by Project Type Mainstream Research Co-operative Research (CRAFT) Exploratory Awards

18 %

FP 5 9,240 contracts (April '99April '01) 26 %

53 %

Mainstream Research Co-operative Research (CRAFT) Exploratory Awards

50 %

29 % 24 %

5.3.3 SME Participation in FP 6 (2002–2006) The overall research budget for FP 6 was increased to 16 bn Euro (excl Euratom) and 15% of this budget is earmarked for SMEs, who have an allocation of 2.2bn Euro as opposed to 1.4 bn Euro (in FP 5). It is projected that 82% of the SME budget will be awarded through the mainstream research instruments and the balance through TSME measures. This would reverse the historical trend towards a greater proportion of the budget going to SME stimulation measures. Achieving greater SME participation in thematic areas, however, may prove difficult because two new instruments have been introduced for the seven priority research streams, namely “integrated projects” and “networks of excellence”. The integrated projects are designed to achieve “ambitious, clearly defined, scientific and technological objectives.”33 They will require a critical mass of activities and resources, valued up to several tens of millions of Euro. In practice the number of partners is likely to be very large. Networks of excellence are joint programmes of activities “designed to strengthen Europe’s research excellence in a particular research topic”.34 The net effect of these new instruments may be to raise the barriers to SME participation in mainstream EU research. Changes have also been introduced in the SME stimulation measures. The Exploratory Awards have been discontinued and the Collective Research

33 34

EU Commission, DG Research, November 2002. Ibid.

71 instrument, pilot-tested in FP 5, has been introduced. Co-operative Research (CRAFT) has been retained.

5.3.4 Evaluation of SME Participation Participation by SMEs in the FPs remains below general expectations, notwithstanding the sustained efforts of the Commission Services. The first calls from FP 6 have now reached contract negotiation stage and the indications are that the 15% funding target is unlikely to be achieved. While SME participation in selected FP 6 priority themes reached 13% of the available budget, other priority areas, such as “health” and “food quality and safety”, barely reached 8% of budget. The Commission has reacted by outlining “further corrective measures to boost the number of small companies participating in FP 6”.35 Successive Monitoring Panels, reporting on the Innovation and SME Specific Programme, have expressed concern that the quality of SME participation in Community research may also be diluted.36 The Exploratory Awards (discontinued under FP 6) are not for research proper, but to ensure the preparation of better research proposals. In the case of the Co-operative Research projects, the results of the research are owned by the SMEs but the research itself is carried out by external research institutes. No qualitative analysis37 has been carried out to date on SME participation in FPs and, as a result, there is a dearth of evidence on the extent to which these stimulation measures have increased the capability of SMEs for in-house research. The Sixth Report of the SME Observatory (2000) carried out a survey on SME participation in EU programmes. Only 30% of those surveyed were aware of the existence of the R&D programmes and, of those that were aware, only one in three had participated in the course of the previous five years.

5.3.5 Barriers to SME Participation in KNOGG Countries More detailed information is provided in Table 5.2 relating to the barriers to participation that were experienced by KNOGG countries (The survey is for a period preceding participation by the new member countries). This analysis is confined to those SMEs that responded that were aware of the EU R&D programmes. The table also gives the average score for EU 19 countries.

35

Mr Xabier Goenaga, head of unit at the Commission’s Research DG. Cordis Focus. Number 233, page 3, 2003. 36 The 2001 Innovation and SME Workprogramme introduced a number of activities aimed “in particular at those SMEs with limited or no research capability of their own”. (page 3). 37 A report on the qualitative and quantitative aspects of SME participation in Framework R&D programmes is scheduled for the end of 2003.

72 There are some common features but also important country differences with respect to barriers to participation. The only barrier of significance (17% for EU 19) that can be readily redressed by action at Community or national level is probably the information deficit. There is no doubt but that the Commission Services have taken this on board in recent years because a number of remedial actions have taken place, including the setting up National Contact Points and major improvements in management procedures and information dissemination. In advance of these steps, it would appear that countries like Finland (6%) had solved the information problem themselves while Greece (38%) remained very isolated. Perhaps the most important finding from this survey is the extent to which, for a variety of reasons, firms do not feel capable of participating in Community R&D. If one aggregates the barriers caused by lack of in-house resources, EU research not compatible with particular needs of firms, and a general lack of capability (projects too complex) one finds that these three account the majority (54% for EU 19) of the reasons for not participating. It is interesting that counties with the strongest capability, Finland (63%) and the Netherlands (40%), are the most conscious of their own limitations. Greece (32%) and Ireland (33%) were less disposed to accept internal factors as reasons for failure. Table 5.2

Finland Greece Ireland Netherlands EU 19

Barriers to SME Participation in EU Framework Programmes – KNOGG Countries (% of enterprises aware of the support available) Too No Lack of Complicated Appropriate In-house Programme Resources 10% 27% 26% 13% 17% 2% 9% 19% 5% 8% 18% 14% 22% 20% 12%

Failed Lack of Applications Information 0% 16% 14% 1% 6%

5% 38% 18% 11% 17%

Other Reasons 4% 20% 17% 6% 8%

Source: ENSER Enterprise Survey 1999.

5.3.6. Policy Implications The capacity to significantly increase SME participation in EU R&D programmes, under present circumstances, may be approaching saturation. Further participation can still be achieved but at the risk of diluting the quality of that participation. This is a difficult proposition to accept given that only 20,000 SMEs participated in FP 5, out of a total of several hundred thousand high tech

73 SMEs across Member States.38 The SME-R&D participation problem is not confined to small countries but is particularly acute for small countries such as Ireland where indigenous industry is largely populated by SMEs. The R&D Framework Programmes are designed to strengthen Europe’s research excellence and to give it a leading place in major current and emerging research areas. The national research system actors best equipped to further that objective are research institutes, public and private research laboratories, universities and other institutes of higher education and large enterprises with strong research capability. The participation and contribution of high tech SMEs is important, however, because this introduces legitimate tensions and keeps the research system aware of the market imperative. The European Charter for Small Enterprises39 promotes the thesis that small enterprises are the backbone of the European economy. They are acknowledged as a key source of jobs and a breeding ground for business ideas. They are also believed to be “a main driver of innovation, employment, as well as social and local integration in Europe”. There is no question about the Commission’s strength of commitment to improving the research capability of SMEs and enhancing their participation in Framework R&D Programmes. The only issue relates to the effectiveness of the policies that are being pursued. Notwithstanding the determined efforts that have been made to facilitate the participation of SMEs in Community research programmes, the response has been disappointing and the technological capability of EU SMEs remains much weaker than that of its main competitors. The net result of the Commission’s efforts has been to increase the number of contracts awarded to SMEs from about 15, 000 in FP 4 to 20, 000 in FP 5. But it is difficult to refute claims that the quality of SME research participation has been diluted in the process of attracting more applications. The increased participation is, for the most part, in the special measures introduced for SMEs “with little or no internal R&D capability”. Results are awaited, from a qualitative analysis40 of SME participation, currently in progress, to establish if the special stimulation measures have improved the capability of SMEs to undertake in-house research. The prospects for increased SME participation under FP 6 are not encouraging, even though a target of 15% of total funds has been set and further attempts have been made to alleviate the administrative burden on SMEs that participate in EU research. The substantive changes introduced under FP 6, however, are inimical to SME participation. Exploratory Award contracts, the most popular of the 38

The European Observatory for SMEs, in its Sixth Report (2000), gives a figure of 0.75 million high tech SMEs, nearly half in computer and related activities. 39 Give particulars of launch as part of Lisbon goal. 40 A report on the qualitative and quantitative aspects of SME participation in Framework R&D programmes is scheduled for the end of 2003.

74 special stimulation measures, have been discontinued. The two new research instruments, Integrated Projects and Centres of Excellence, are intrinsically less suitable for SMEs than for organisations carrying out research on a large scale. One must conclude with a question: are expectations, with respect to SMEs, from the EU R&D Framework Programmes pitched too high? It may be preferable to place more emphasis on regional and national, rather than international, research and development networks for the vast majority of SMEs. One may also ask if this is the sentiment underlying an EU Regulation, to stimulate SME research at the national level, proposed by the Commission in May 2003. This Regulation will enable Member States to grant aid research carried out by their own SMEs, without notifying the Commission beforehand: national governments may, in future, cover up to 100% of SME expenses for fundamental research, 60% for industrial, and 35% for pre-competitive development R&D. When introducing the measure Research Commissioner Philip Busquin welcomed the timing of the new Regulation. He reaffirmed the message that “research in Europe is facing an uphill struggle: lack of co-ordination, insufficient public support, difficult access to capital, and unfriendly fiscal, legislative and regulatory frameworks hamper European R&D effectiveness.” EU level STI policy may be excessively top down, while it is the market and grass-roots SMEs that may define success. Optimal technology policy may therefore need to follow, and conform to, market trends at the local level.

5.4 Other EU policies and knowledge-based growth In addition to pure research-oriented programs, there also other Community level policy programs which can contribute positively to the member countries’ potential for knowledge-based growth. Incentive funding constitutes only a proportion of the total public R&D budget of each country. It appears that specific Community programs have had varying relative impacts and influences on national policies. Infrastructure financing was mainly channeled through the EC Structural Funds to those countries which are less advanced in research. The interaction between Framework Programme specific actions, Structural Funds and national policies was obviously important in those countries (Ireland, Greece, Portugal and, to a lesser extent, Spain and Italy). Structural funds can be used as a means of creating favourable business infrastructure and environment, and even to support R&D in some regions. European Regional Development Fund and European Social Fund are important instruments in that, but their effectiveness depends eventually on the regional growth potential and administrative efficiency – but not on country size. Community policies generally have had only a slight influence on the structuring of national policy of the most advanced countries, while its impact has often been

75 significant in the ‘less favoured’ ones. In the latter groups, the Community programs have inspired the design and adoption of similar national programs and procedures. Structural funds in the EU are instruments aiming at the improvement of economic structures and are as such addressing a variety of problem areas. Two among them appear in their current form relevant for the enhancement of National Innovation Systems: The European Regional Development Fund (ERDF) and the European Social Fund (ESF). The areas relevant for the knowledge based economy in their context are research, development and technological innovation (RTDI), human capital (HC) and the enhancement of the Information Society (IS): Table 5.3

Structural funds and knowledge-based economy ERDF X

RTDI HC IS

ESF X X

X

Structural policies are mainly designed at the national and regional level. The European Regional Development Fund (ERDF) supports regional development through co-funding of multiannual programmes agreed in partnership between the member state, the region and the European Commission. Eligible regions are those having a GDP, which is lower than 75% of the EU average and known as Objective 1 or “less favoured” regions (LFRs). Some support but of a much lower funding is also given to industrially declining regions, in an effort to facilitate their restructuring. These programmes, known as Community Support Frameworks have a broad scope and cover all aspects of development. Over the 1988–2006 period of programmed regional support there are regions entering and leaving the group of those eligible for support (see table 5.4). Table 5.4

Exit and entry of regions

Category of region LFR during the previous period New entries – political New entries – falling behind Exits Total Permanent LFRs LFRs in two periods LFRs in one period only Source: Tsipouri, L., 2003.

1988–1993

1994–1999 48 7 5

48 (46+2) 42 6

61 (59+2) 42 13 6

2000–2006 60 6 3 12 58 42 7 9

76 The majority of regions have thus kept their LFR status over the whole period. There is no overlap between country size and eligible regions for structural support. Small countries may either be rich (the Netherlands, Belgium, Denmark with no Objective 1 regions and recently Ireland) or poor (Greece and Portugal, having the whole territory broken down into Objective 1 regions), and so can big countries (Spain and Southern Italy being composed of Objective 1 regions). For designing an adequate innovation system and preparing for the knowledgebased society small countries differ mainly from big ones in their ability to generate economies of scale and utilise externalities in a small market; thus they are obliged to open up to the international market more often than large countries. The opening can operate only in a narrow range of competitive sectors. It follows, that specialisation is an imperative to small countries. Because of specialisation, the small countries face the risk that these sectors are ill-selected or demand for their products is declining and restructuring is either delayed or insufficient. As the regional level is the most appropriate for structural intervention (because of agglomeration effects) there seems to be hardly an issue of a separate treatment of small countries in the implementation of the development policies of the EU. Nevertheless, it is important to clarify two issues: • what are the recommendations emerging for small and big countries alike, based on the evaluation of the current experience of the past CSFs • in which areas could structural intervention help eliminate/reduce the high risk/specialisation aspect inherent in the small countries development model in the knowledge-based economy? The results presented hereafter are based on the RTDI and IS thematic evaluations of DG Regio, as well as on the various evaluations of the ESF, and are combined with the basic small country perspective, as perceived in the KNOGG project.

5.4.1 Enhancement of RTDI expenditure by the Structural Funds The technology policy option for regional development, which is being provided by the Funds, induces fundamental long-term changes in the development capabilities of the regions. In addition, the Funds are changing the culture of regional development and providing new tools for regional actors. Innovation and technology development are now on the policy agendas of many regions, for the first time. Many of these regions are still trying to understand the dynamics of these new tools, to find appropriate institutional designs and management systems.

77 Training, exchange of experience and learning opportunities will have to be improved. So too will real time monitoring and evaluation of best practices. Regions may be seen to over-invest, under-invest or mis-invest in technology. Determining the most appropriate strategy for a region is not an easy task. This is mainly because the determinants are not only the general guidelines and national priorities, but also the regional specificities, path dependencies and parameters related to political and administrative strengths and the power relations of the region, as well as the need to adapt strategies to new environments and circumtances. The following aspects need thus to be taken into consideration: (A) Capacity enhancement – RTDI infrastructure The need for improvement of public RTD infrastructure in the Objective 1 and Objective 6 regions continues to be significant, especially in regard to training and human capital enhancement. But, more consideration must in future be given to the demand (and not just public sector demand) for such infrastructure. Demand anticipation is important, especially for the fast growing regions (a minority), but moving too far ahead of demand brings a risk of inefficiencies and under-utilised capacities. While regional requirements and specificities will determine how quickly and in what direction a particular region might go, in upgrading its public RTDI infrastructure, the important consideration is the extent to which utilisation of capacity, driven primarily by the productive sector, may be expected. (B) Innovation and technology transfer While the demand from industry for the various schemes in operation is high, often they are in “overbooking”, it is nonetheless difficult to see yet any significant industrial impacts from the Funds. Spillovers and externalities stemming from the investments made in RTD have been limited. In the majority of cases one has to assume that the major benefit was the effect on training and skills and on human resource enhancement. Particular attention is needed to ways of mobilising and increasing the demand for RTDI from industry in the lagging regions. In the majority of cases this will involve greater attention to innovation, to networking and linkages, to co-operative structures and clusters, to technology transfer, to diffusion of information, technology auditing, the mobility and placement of human resources and less to research and research infrastructures.

78 (C) Management and implementation processes Management inefficiencies and discontinuities are reducing the effectiveness and impacts of the Funds. (D) Policy development and delivery In this area, it is important to distinguish policy development from policy delivery. The problems exist especially in the policy delivery phase. Good schemes and programmes are failing because of weaknesses in management. Part of the problem appears to be that there is limited experience of designing or managing industry oriented programmes. Most regions do not yet have this type of experience. There are some problems with policy formulation, too. The strategic vision is not always apparent at national level. Priorities are not always clear and funding is being dispersed over too many areas. Building critical mass on the basis of established strengths requires a clear focus on priorities. Concentration and specialisation will now be more important concepts for the Funds and for the participating regions. Clearly, the next stage needs to pay more attention to fewer and stronger projects, to clearer definition of priorities and to assisting the establishment of critical mass in well selected areas and technologies. This will place a major emphasis on planning and policy formulation and on monitoring to ensure that these are not just aspirations and empty promises. Performance based systems would help to introduce greater selectivity and focus. (E) Economic impacts and Community added value The available evidence indicates that the direct economic impacts of the Funds for RTDI in the beneficiary regions are so far fairly modest. However, a much needed strengthening of the long term capacity of regions for economic development is occurring, with the assistance of the Funds. The additionality of the Funds is also evident, as is apparent that many, if not most, of the investments supported could not have been undertaken by the region or Member State, without the assistance of the Funds. The availability of the Funds is speeding up the enhancement of RTDI capabilities, and improving both the pace and the scale of development. Small countries may in this context be better off in the sense that policy delivery can be more easily agreed in a smaller political constituency, but as capacity building seems to be the most funded area, special care should be taken to balance it with innovation and technology transfer measures that will enhance productive capabilities and thus reduce risk and promote flexibility. These are issues to be keep in mind when designing the application measures

79 of the RTDI parts of the ERDF rather than specific actions for small countries. 5.4.2. Human capital development The European Social Fund has a particularly difficult assignment in the current circumstances of high and persistent unemployment. Most efforts are addressing the reduction of unemployment and social equity rather than the promotion of the knowledge-based economy. Yet, the very specific case of Life-Long-Learning (LLL) included now in three of the European Employment System (EES) Guidelines is dominant for the support of the knowledge-based-economy. Equally the support and enhancement of entrepreneurship is an important element for the improvement of human capital and leads to easier restructuring when necessary. The improvement of the quality of education and vocational training is important for big and small countries alike and progress is made in both groups as stated by the evaluations. Life-long learning will be promoted and there is a need for comprehensive and coherent strategies. As LLL can only demonstrate its merit in the medium to long term there is some evidence that it is praised in the rhetoric more than it is applied in current measures. Progress here is slow but again there is no distinction between small and big countries. The emerging Common European LLL paradigm is at different stages of implementation (quality, labour market needs matching, basic skills development, no drop outs, e-learning targets) and it is only in the case of matching skills with labour market needs that small countries may have an advantage, as they policy makers may have a better overview in the market. Most importantly small countries appear as better off in two more cases, where the ESF evaluations see major needs for the future: promoting LLL through social partners agreements and the improvement of impact and effectiveness evaluation tools. In both cases the closer links and easier access to information can act as facilitating elements, thus making pilot cases out of small countries. The Employment Strategy also seeks to maximise the job creation potential of economic growth. In this context, the Entrepreneurship Pillar of the Guidelines calls for increasing entrepreneurial awareness, facilitating new start ups and the hiring of staff, encouraging self-employment, exploiting the potential of job creation at local level, in public sector as well as in the private service sector, and for a reduction of the tax burden, especially on labour. From the policies used entrepreneurial education, e-government and e-business are those most relevant for the knowledge-based society. The national evaluation reports draw attention

80 to the fact that evaluation of this pillar is complicated by the lack of quantitative data linking business developments to the policies promoted through the Employment Guidelines. But overall there is evidence of a strong effort of individual member states to promote entrepreneurship by ESF-co-financed actions, and through improvements of the administration. The promotion of self-employment, including through active labour market measures is also considerably developed. Yet it is mainly the big countries, which have taken active measures in this direction: Italy has focused on selfemployment in recent years with several measures which attracted a considerable number of people previously in the black economy. In Spain growth in selfemployment was partly due to changes in work organisation processes, but the hiring of workers by self-employed people was facilitated through tax incentives. Germany and France encouraged unemployed people to start a small business by allowing them to temporarily maintain their social security status or to return to employee status. Entrepreneurial training has also grown rapidly and in several Members States it has expanded to cover all forms of training, from academic curricula to school level courses. The notion that business education should be provided as a part of lifelong learning now tends to be commonly accepted. The EES has inspired a number of initiatives which make it easier for women to become entrepreneurs. Women constitute an increasingly significant proportion of entrepreneurs in the SME sector. The majority of Member States have launched particular measures to encourage women to become entrepreneurs. AT, FI, DE and SW have developed targeted schemes such as the promotion of start-up advice for women. Most of these schemes were initiated to improve the socio-economic position of women. Small countries in this context cannot be seen as disadvantaged. They have to face human capital development as a major opportunity and exploit it for LLL and high quality (partly generic) skills. In addition, because of their size and ability to achieve consensus, they can be used as important pilot cases in areas where the ESF identifies major weaknesses in most European member states. In the context of entrepreneurship and self-employment small countries seem to be lagging behind the big ones and should re-orient into this direction, since it is an important precondition for the flexibility they need to constantly adapt their specialisation patterns. 5.4.3. Support for the enhancement of the Information Society The information society is a broad topic with many facets and, in the context of regional development, there are many layers of complexity. Despite the diversity of regional contexts and the variety of approaches adopted in information society programming, some clear trends and tendencies have emerged: an increasing

81 commitment to information society development and more coherent approaches to information society planning than in the past. The thematic evaluation estimated, on the basis of 156 regional programmes and 3 national programmes for the information society that the Structural Funds can be expected to co-finance a total of just under EUR 16 billion of information society investment in the period 2000–06. This represents an estimated 7.36% of total Structural Funds investment allocated to information society investment and suggests that national and regional decision makers are increasingly committed to information society development. Considerable variations between regions in per capita investment levels exist, of course, from an estimated EUR 358 per capita (or 38% of Total Structural Funds expenditure) in the Border, Midlands & West region (Ireland) to zero (apparent) investment in about half of the regions that we investigated. More and more regions are tending to adopt coherent and strategic approaches to information society planning a higher proportion of Objective 1 regions has adopted a structured and strategic approach to programming the information society in the 3rd CSF, yet still about half of the regions investigated do not appear to prioritise information society development (or had provided no financial information regarding it). National programmes for the information society have provided leadership in some of the new member countries (big and small ones alike). Ongoing dialogue between the national authorities and the regional partnerships is critical and this is easier to achieve in small countries. Wherever possible, national programmes need to be adapted in terms of regional implementation to emphasise region-specific needs and priorities. In terms of regions smaller ones are investing more (per capita) in information society development than larger regions, but there is no evidence that the same applies to small and big countries. It is rather the countries with the more sparsely populated and more peripheral regions, which propose to commit more per capita on information society development. Regional programming priorities for the information society were broadly consistent with the priorities of the e-Europe 2002 Action Plan. Regions vary in their priorities on the information society according to their Member State, which again is size-independent. Yet, overall information society investment tends to be dominated by supply-side measures with too little attention paid to stimulating the demand for ICTs. This makes things particularly difficult in those small countries, where the language is unique to the country: in the small countries, where the language is the same as with bigger member states (Ireland, Austria, Belgium) the problem does not arise, but for countries like Portugal, Greece or Finland the need to enrich national content and attract a higher share of the population is a barrier to ICT literacy. In this case small countries need a special treatment.

82 The evaluation has identified a slowness of the Structural Funds programming, approval and implementation processes in the face of fast-changing technologies and changes in demand. In the context of information society planning, and despite the importance of regional and cohesion policy and the volume of EU funds committed to it, the quality and availability of regional data to support policy analysis, review and formulation is sometimes inadequate. Small countries should have an advantage in that because of their size. While the need to adapt to the knowledge-based society is encountered in all EU policies, especially after the setting of the Lisbon target, structural policies still see the support towards this goal as secondary to their primary objectives, which are regional development and social cohesion. In this context one should note that there is virtually no instrument in this context, which addresses the needs of laggards towards restructuring into a new economic development model. In general small countries need to follow the general recommendations regarding the use of structural funds like the big countries: put more emphasis on RTDI demand and technology transfer, assure a good life-long learning strategy, efficient interventions in the educational system and promote the information society. These are the instruments that will facilitate a quick adaptation to the knowledge economy. The problem that most evaluations identify is that there is still a tendency to apply input proxies as indicators for development planning (implicitly assuming some kind of linearity relative to output proxies). This problem needs to be very seriously taken into account. It is precisely for this reason that the CSFs still under-fund intangibles. Moreover, data and research capacities have improved tremendously over the years for researchers to be able to analyse the effectiveness of inputs in producing the desired output. However, there are some areas, where improvements are easier in small countries, because of the small size of the stakeholder communities. In this case small countries have to be more ambitious than bigger ones and can act as pilot experiments for a better overall structural fund efficiency. Examples include: 1. the involvement of the social partners in the implementation of an active lifelong learning strategy; 2. the adoption of new tools for effectiveness and impact assessment of the ESF-supported educational policies, 3. a better coordination of RTDI policy delivery. Unlike these areas, where the size of the home market can constitute an advantage for a more efficient policy, other areas constitute a real barrier:

83 1. the promotion of entrepreneurship, though very important for the flexibility needs of small countries, seems less developed than in bigger ones and 2. the size of the language market may create a real problem, in the sense that the less educated population will face an additional obstacle in their way to ICT literacy, unless national content is enhanced.

5.5 ERA and small countries As early as 1994 a Commission communication suggested that more coordination would be necessary at a number of levels of research policy. There should be a shared knowledge base which would form a basis for better coordination between national policies and European policy; co-ordination should also be improved between national and European activities supporting research and innovation; and between representatives of member states and the European Commission in international negotiations on global research programs involving non-EU countries. The planners of ERA realised in the mid-1990s that co-ordination had failed to produce any large-scale initiatives. As a compromise, it was hoped that Europe could concentrate on “the pursuit of a limited number of key objectives” (Caracostas and Muldur, 1997). The concept of European research co-ordination was revived in January 2000 when the Commission’s communication formulated the need to avoid fragmentation in a following way: We need to go beyond the current static structure of ’15 + 1’ towards a more dynamic configuration. This has to be based on a more coherent approach involving measures taken at different levels: by the Member States at national level, by the European Union with the framework program and other possible instruments, and by intergovernmental co-operation organisations. A configuration of this kind would make for the essential ‘critical mass’ in the major areas of progress in knowledge, in particular to achieve economies of scale, to allocate resources better overall, and to reduce negative externalities due to insufficient mobility of factors and poor information for operators. The emergence of the European Research Area strategy is the response of policymakers and other actors in the European research and innovation system, to a systemic failure to utilize the widely distributed European resources in a more concerted and efficient way. Thus ERA seeks to set up of new research and information infrastructures and facilities, and to mobilize existing research

84 capacities in virtual research centers and networks, capable of competing and cooperating with their US counterparts. At the moment there is no common European policy on research. National research policies and Union policies overlap without forming a coherent whole, and the national policies dominate the picture. The European market for knowledge and technology still remains largely to be created. For it to develop and function a real European research policy needs to be defined (EC Comm. Jan 2000/ERA).41 At the Lisbon European Council in March 2000, the concept of “open method of co-ordination” was introduced in order to better implement the long-term strategy for a competitive knowledge-based economy with more and better employment and social cohesion. 5.5.1. What kind of ERA would be best for Europe / small countries? Europe needs a dynamic ERA capable of competing with the US. Of course, competing in itself is not important, but establishing a competitive ERA is a necessary prerequisite for achieving faster economic growth. Is simply increasing research funding the best way of doing this? A key issue relating to the debate on the history of European research institutions is the tension between a “static economies of scale approach” (single market, relaxing competition rules, target strategic industries and big companies through R&D programs) and a “diversityconnectivity-networking approach”. The discussions about the ERA strategy of promoting the emergence of European poles of excellence have highlighted the risks of too much concentration. New centres of excellence may partly grow at the cost of marginalizing emerging and atypical research teams, and of transforming pre-project discussion from the scientific to the bureaucratic level. The European Commission – not surprisingly – advocates the pursuit of concentration on the basis of its interpretation of the principles of subsidiarity (European research policy should be decided on Community level) and on the need to reduce fragmentation. Concentration of resources around new poles of excellence means creating the critical mass to compete internationally. On the other hand, member states often defend their own spending priorities and their own institutions. Quite obviously, establishing a successful ERA with a limited number of highperforming platforms is a challenge for small European countries, at least for those located in the periphery and for those lagging behind in economic and scientific development. From the small country point of view, there are tradeoffs,

41

See European Commission, 2002, too.

85 and the question of subsidiarity – to what extent research policy should be done on national or regional level – should be considered carefully.42 The tension between scale and diversity in the European integration process has been analysed by many authors. According to Matthews and McGowan (1992, p. 230), “any tendency to eliminate technological diversity represents a diminution of the technological resource pool, even if, via such processes as gains from economies of scale, short term efficiency is increased by convergence”. Gregersen, Johnson and Kristensen (1994) also assess critically the logic of integration: “On the other hand, (growing integration) might also hamper innovation processes and interactive learning in the long run, if established domestic link-ages are broken”. However, creation of a large internal market may also promote innovation. The growth of multinational corporations through mergers and acquisitions across borders, which is a likely result of European integration, is expected to lead to a better innovation performance since, in Schumpeterian terms, bigger firms in bigger and more specialised markets are better innovators. Also the rewards for successful innovators are higher in large integrated markets. Market-driven development is likely to be efficient in economic terms, but it is far from certain that it will benefit all countries and all regions equally. In order to defend original national institutions and the autonomy of science, some have suggested the establishment of an “EU Science Foundation”, or more recently a “European Research Council” (or “European Agency for Science and Research”, as proposed by Sapir et al., 2003). From the small country point-of-view, there are worries concerning the possible negative impacts of European policies on national policies. While the broad objectives of ERA are beneficial to all, Community institutions may lead to a certain alignment of national policies with European policy which can result in a growing distance of national programs from national priorities. Furthermore, the promotion of European partnerships could lead to an unexpected and marketdriven international sharing of tasks in the major technological fields resulting in agglomeration. Such a development can be expected from the current location of business sector R&D activities (and from the US experiences). Countries with large companies and strong R&D centres would benefit and lead the way by using e.g. small companies of less favoured countries as subcontractors. The logic of market size and the benefits of agglomeration may also lead to a situation where it is the larger countries (together with some successful regions of small countries) which would benefit more from the possible positive

42

An interesting paper on the possible implications of ERA in the European innovation system is Caracostas and Soete, 1997.

86 externalities of the reallocation of research resources, notably in terms of high skilled research jobs.43 It is especially important for small countries and countries far away from the economic core of the EU to create their own research platforms and poles of excellence which could be attractive parts of the new European research landscape. It is clear that it may be hard for small countries to create competitive platforms with sufficient critical mass. This may only be possible in some key sectors. As a consequence, regional and national RTD policies are still needed to maintain regional cohesion and to avoid brain drain from small countries and distant locations to large European centres. To facilitate this, small countries (and regions) should be allowed a certain leeway in policy measures to develop local centres of excellence which could be parts of the ERA. It is very likely that this requires internal reallocation, restructuring and relocation within small countries, as well as higher public RTD spending and measures to support and increase business R&D spending. It is necessary that well-resourced and highly competitive European Centers of Knowledge will be created. Network-based centres of excellence may not be enough. Instead, they should be high level institutions which could compete in their capacity with the US institutions. There should be such advanced and permanent research platforms in all EU countries in order to avoid too much concentration.

43

See also Kuhlmann, 2001.

87

6. Knowledge-based growth and other policy areas The European Union and its member countries have widely accepted common objectives; they wish to promote employment, social cohesion and well-being, economic growth and innovations. Ambitious objectives of knowledge-based growth have been endorsed by the European heads of state and government in Lisbon in 2000 and in Barcelona in 2002. It is understood that the only sustainable way to promote economic growth in the long term is to support the build-up of the knowledge-base of the European economies – i.e., human capital and skills, research and technology, and innovation. It is only natural that the Lisbon targets are connected to the idea of European Research Area (ERA), which aims to strengthen the European science and applied research. However, at the same time it has to be acknowledged that knowledge-based growth does not only depend on narrowly defined STI policies but also on many other things. Efficient and ambitious STI policy is not enough, it is also important to create favourable economic and fiscal environment for private sector R&D and innovation. Growth-oriented economic policies are needed to enable sufficient private sector investments.

6.1 Focus on the firm It is enterprises and entrepreneurs that are the real drivers of innovation – most innovations come from firms, not from government sector or science (Acs and Audretsch, 1990). Although R&D is mostly done by large multinationals and government laboratories, many product innovations are made by small and medium sized enterprises.44 This has also been the case in the advance of ICT in the 1990s. Successful entrepreneurs are also crucial for the growth of regions (and for overall economic growth, too (see e.g. Audretsch and Thurik, 2001). In some regions there is more entrepreneurial activity than in others. There are also differences across countries in this respect. The research on the determinants of innovations on one hand and on entrepreneurship on the other has revived in recent years (see e.g. European Commission, 2003). The both topics are interlinked; innovations usually require entrepreneurs, and it is small firms and many times even start-ups which contribute disproportionally to the output of innovations (see e.g. Kaiser, 2003). It follows that NIS should be tightly connected to enterprises through networks and co-operation; applied research institutes should be responsive enough to the needs of firms. In practise that would require more flexibility in universities.

44

The special characteristics of entrepreneurs may also be important, see Uusitalo, 1999.

88 The creation of innovations requires entrepreneurial spirit, and sufficient population of entrepreneurs. Hence entrepreneurship should be encouraged, and capital markets developed to improve the supply of private venture capital. This has been acknowledged by the Commission but it requires national action. In this respect most EU countries are clearly lagging behind the US. Small and medium sized firms do not have much resources for R&D. However, many of them are innovative. Denmark and many other countries have managed to create successful and innovative SME sectors without considerable R&D investments, and created considerable wealth without technological innovation (see e.g. Lundvall, 2001). In fact, there is a big difference between innovation and R&D. A change of mindset is needed to recognise that R&D performance is not a surrogate for innovation. A cursory inspection of STI statistics will show that, in all Member States, the number of firms that innovate is a multiple of the number that engage in significant R&D. There is a great diversity among firms in their innovation process, innovation inputs and outputs. Many sectors are major users of technology advances generated elsewhere, and make a decisive contribution to economic growth by incorporating these advances in the new products and processes they introduce. It is worth noting that the SME Charter (2000) does not explicitly mention R&D but concentrates on the need to strengthen the technological capability of firms. “We will strengthen existing programmes aimed at promoting technology dissemination towards small enterprises as well as the capacity of small business to identify, select and adapt technologies”.45 The innovation imperative of SMEs is best served by a system-based innovation programme to run in parallel with, and complementing, the R&D Framework programme. This programme would prioritise learning and the diffusion and absorption of the knowledge created in the R&D programme and elsewhere. Its instruments would be networks of regional technology centres, technology transfer services, cluster programmes, demonstration schemes etc. The innovation programme would need comparable resources and comparable commitment to that that currently invested in the Community R&D programme.

45

Action line number 8 in the SME Charter adopted by the General Affairs Council (June 2000).

89 Figure 6.1.

Investment in Venture Capital, in percent of GDP, average for 1995–1999 Investment in venture capital in % of GDP, 1995-1999.

0,25

0,20

0,15 Expansion Early stages 0,10

0,05

0,00 Hungary (1998-99)

Japan (1995-98)

Greece

Korea (1995-98)

EU

Finland

Ireland

Netherlands

OECD-19

United States

Source: OECD

Percent share of high-tech sectors in total venture capital, 1995-1999. United States OECD-19 Ireland Finland Netherlands EU Greece Hungary (1998-99) Korea (1995-98) Japan (1995-98) 0

10

20

30

Communications

40

50

60

Information technology

70

80

90

100

Health/biotechnology

The small and medium sized firms should be stimulated and supported to do more in-house research and product development. In this respect they currently lag well behind their competitors in other economic regions. Such a breakthrough, however, must first be achieved at local and national level. It is

90 reassuring that a regulation introduced by Commissioner Philippe Busquin in May 2003 now allows Member States to grant-aid research carried out by their own SMEs, without notifying the Commission in advance.46

6.2 Role of innovation policy Science budgets and education are important and necessary for knowledge-based growth. However, flourishing universities do not automatically yield innovative business sector. Strong science and education are vital but they are not enough. Strong applied and business-oriented research is vital. The main problem in many countries is that only a small proportion of S&T budget goes directly to firms – that proportion is much higher in the US and Japan, and also Finland. There is a need for external inputs in firms, but direct subsidies and grants are not always effective. There is evidence which suggests that networking and R&D cooperation are more important to firms and innovations than subsidies. Cooperation and networks by government intermediaries (like TEKES in Finland) with business firms, and training and competition-based R&D support are important. The vision and the consistency of a country’s industry-oriented innovation policies are decisive in achieving knowledge-based economic growth. Ultimate responsibility for product and process innovation rests with enterprise. However, the innovation performance of enterprise reflects the effectiveness of industry innovation policies. There are lots of research results, which indicate that entrepreneurial firms are dependent on risk capital, and that they are constrained by credit rationing. The same applies to self-employment (see Bottazzi and Da Rin, 2003 and Lindh and Ohlsson, 1996 and 1998). Entrepreneurship requires certain character, and financial means (Blanchflower & Oswald, 1998; see also Rickne and Jacobsson, 1996 and 1999). It follows, that the availability of risk capital is likely to be central especially to start-ups (Keuschnigg and Nielsen, 2003 and 2004) analyse these questions.). The venture capital market is deeper in the USA than in the EU, as shown in Figure 6.1. Only the Dutch venture capital market is close to American level. Figure 6.2 shows that a very large share of American venture capital has gone to finance new technologies.47

46

When introducing the measure the Commissioner reaffirmed the message that “research in Europe is facing an uphill struggle: lack of co-ordination, insufficient public support, difficult access to capital and unfriendly fiscal, legislative and regulatory frameworks hamper European R&D effectiveness”. 47 On the American lessons, see Henrekson and Rosenberg, 2001.

91

6.3 Better understanding of interfaces with other policy areas It is widely accepted that entrepreneurship, risk-taking and business sector growth should be encouraged in Europe. There is vague understanding that excessive regulation and high taxes are likely to discourage entrepreneurship. It is also commonplace to think that better incentives and different attitudes have contributed to more dynamic business environment in the US (See Rosenberg, 2000 on the US experience). However, it is not very well understood how different institutional and regional factors affect entrepreneurship and supply of risk capital. The role of public policies like taxation, regulation, competition policy, and macroeconomic policies as determinants of private sector growth have not been fully covered by research. Entrepreneurship has not been a subject of extensive research in economics; to the contrary, business firms have in most cases treated like ‘black boxes’ in economic analysis: they have been seen as unknown entities which have used certain inputs to produce output in order to maximise profit. Widespread innovation and entrepreneurship are necessary ingredients of knowledge-intensive productivity growth. However, they may themselves be dependent on business cycle and demand conditions. In a rapidly growing economy, the supply of risk capital and the willingness of potential entrepreneurs to accept risks are likely to be higher than in a stagnant economy with modest prospectives. Traditional instruments of economic policy and regulatory framework are important for favourable business environments and innovative entrepreneurship; such policies include the use of fiscal instruments (most notably taxes), regional policies, subsidies to and taxation of risk capital, intellectual property rights, and in-built incentives of R&D policies (e.g. networking and co-operation) as well as other enterprise and entrepreneurship policies.

92 Figure 6.3.

Tax subsidies to R&D Tax subsidy percent in large firm R&D, 1999

Spain Netherlands United States Ireland Japan United Kingdom (2000) Finland Greece -0,050

0,000

0,050

0,100

0,150

0,200

0,250

0,300

0,350

Source: OECD, STI/EAS Division, May 2001.

There is a large variation of different policies within the EU. Figure 6.3 shows one example, the tax subsidies to R&D, which range from the 35 percent subsidy of total R&D expense to the negative subsidies (i.e., net taxation) in Greece and Finland. There should also be better understanding of factors affecting location of innovative activities.48 For better understanding there should be better databases to facilitate quantitative evaluation of STI and enterprise policies, and there should also be more performance monitoring and policy evaluation.49 There is also a question of subsidiarity. How far should the Community level policy harmonisation and co-ordination go in STI policies, and in other policies having impact on knowledge flows, business formation and location? As suggested above, such policies cover almost all economic and regulatory measures. To what extent should there be room for national differences? It can be argued (on the basis of small country experiences) that especially for small countries and for disadvantaged countries it is important that there remain some room for national initiatives and priorities in STI policy. Earlier European nation states took care of their economic competitiveness by using various measures of economic policies. Today that is not an obvious 48

See Ekholm and Hakkala, 2003, and Fagerberg and Verspagen, 1996. These questions have recently became into the focus of research, see Fujita et al., 1999 and Fujita and Thisse, 2002. 49

93 solution anymore. In Euroland there is no national monetary or exchange rate policy, and the use of fiscal policy and business subsidies is restricted by the rules of Single Market and Growth and Stability Pact.

94

7. STI policy and knowledge-based growth: guidelines The idea of the project KNOGG was to analyze the experiences of some small European countries in order to provide practical guidelines for national and Community level STI policies. That was done by identifying good practices in terms of public policies and national STI institutions and innovation systems. The findings of the project give rise to some tentative policy guidelines, both on national level and on the Community level.

7.1 National level: The role of national STI policies in small countries The challenge of national STI policies is to find ways to improve the absorptive capacities of the economies, and help technology diffusion. Successful absorption is key to diffusion, and much more important than innovation. An additional, and more ambitious target would be to seek high level original research and innovations through national STI policies. The appropriate role of public sector and public policy in any small country innovation system depends on the political ambitions and available resources. In any case, and even with very limited resources, national STI policies should help to strengthen the knowledge base and the main clusters of the country. Without any policies small and peripheral countries may risk losing their growth potential. A small country located close to larger markets may be able to benefit from the location and attract knowledge-intensive firms without active national or regional policies. That is more difficult or perhaps even impossible for a country which is located far from the economic core. Among the cases studied in the project KNOGG such countries were Finland, Greece and Ireland. The other three KNOGG countries, namely the Netherlands, Hungary and Slovenia are located closer to the large Central European economies. Well-functioning NIS should help to establish knowledge-creating and intermediating institutions. The purpose of the former ones is to foster adaption. the latter ones enable the knowledge to flow from knowledge-creating institutions (like research institutes and universities) to business firms and to the use of entrepreneurs and innovators. There is a need for flexibility and clustering, although it is clear that the relation between being a small country and having sufficiently strong industrial clusters is not simple. But in certain clusters there might be room for real innovations and critical mass even in a small country. Some large companies in the Netherlands (e.g. Philips) and in Finland (e.g. Nokia) are examples of that. One of the main findings of the KNOGG project is that a well developed NIS is of central importance in the development of knowledge based society which supports innovations. A well-functioning NIS consists of strong basic research at

95 least in some key areas, extensive education, and effective intermediating institutions between the research community and the corporate sector. Strong research is not itself enough to produce innovations. Co-operation and networking with innovative enterprises are also needed. The public STI institutions and government policies should promote linkages between universities and industries, and commercialisation of public research findings. They should enable knowledge diffusion and help to reduce the barriers to knowledge transfer, promote international co-operation in STI and the national participation in international organisations and programs. Role of national decision-making and ability to make long-term commitments are important in STI policies. The policies to develop NIS should be stable and widely accepted. This requires certain consensus among the political decisionmakers about the importance of knowledge. That is not possible without largescale public support. Hence governments should also assist institutions which are dealing with the social understanding and popularisation of science and technology. Public attitudes are important for the political acceptability of long term STI policies. In addition to focusing on the institutions of NIS, national policies should especially in small countries encourage private enterprise, innovative environments, and TNC presence through FDI. In order to benefit effectively from the imports of knowledge it is important to have the national capacity to adopt new knowledge sufficiently high. This also means that strong NIS is important even in countries which mostly import knowledge and are mainly imitators in the world market. The development of innovation basis, including national innovation system takes a long time. Therefore long term commitment is needed. This concerns e.g. both the development of education system and scientific research basis in a country. The build-up of the STI policy infrastructure is usually strongly path dependent. The studies of KNOGG project strongly point to the fact that it is important to invest to the domestic scientific and research capabilities. And this cannot succeed without the development of adequate education policy. The development of these take a long time and effects of these development actions are felt only gradually. But without sufficient capacity even the adaptation into the new international environment becomes more difficult.

96

7.2 The role of Community level STI policy The Community policies are expected to improve the European competitiveness by creating centres of excellence, and at the same to promote economic convergence and social cohesion between different countries and regions. The Union level decision-making should also create some value added, or something which cannot be achieved by national STI policies. This task is not easy, although it is clear that by combining resources a union level technology policy can make a more decisive impact than the uncoordinated measures of the 15 or 25 member countries. However, Community level research and technology policy by supporting excellence may promote centralization of RTD activities. The logic of regional concentration may undermine the objective of regional convergence. In the EU there is a governance gap with regard to R&D policies because different levels of R&D policies are only loosely linked. Therefore there is a need for more co-ordination of R&D policies at the EU level in order to stimulate knowledge creation and spillovers. New European Research and Innovation Area aiming to excellence is necessary: Large-scale and high-level European research establishments are needed to enable large-scale R&D projects with technological breakthrough potential and externalities, and to create new competitive platforms with critical mass and sufficient resources. However, for small countries such development may be risky: how they will be able to keep talented people and innovative firms which are necessary to knowledge-based growth, if the mobility of firms and their key personnel increases? The small country aspect and the need to improve cohesion should be kept in mind in the formation of the ERA. The flexibility and human capital requirements placed on small countries, particularly those located at the edges of Europe, may in the long run result in a scatter of science and technology cores in Europe. Common EU policy should accommodate accordingly to support these developments as they imply beneficial diversity for innovation, competition and options for European companies and experts to locate in. In the long run, ideas and innovations flourish from diversity and external demands. Heterogeneity on the European spectrum could well prove a competitive advantage on the global scale, provided that the potential is let to emerge and develop. The most recent growth wisdom asserts that growth enhancing policies and institutions tend to be context specific. Policy-makers have substantial room for creatively packaging sound economic principles into institutional designs that are

97 sensitive to local capabilities, constraints and take advantage of local opportunities (Rodrik, 2003). From the perspective of small countries the new European STI policies should help to avoid the new risks which the small countries are facing.

98

8. Conclusions and suggestions for further research 8.1 Conclusions The project KNOGG was about small countries and their possibilities to achieve and sustain knowledge-based economic growth in the era of globalization. The project compared the STI policies of six European countries: Finland, Greece, Hungary, Ireland, the Netherlands and Slovenia. All of these countries are ‘small’ not only in a global context but also when compared to some larger European countries. That is obvious for Slovenia, Ireland and Finland, which have populations of only 2–5 million people, but it holds also for Greece and Hungary (which have a population of 10 million but in economic terms are not much larger than Finland) and even to the Netherlands, which with 16 million people is the 7th largest country in EU. The economies of small countries are small and open. This means that in order to utilise economies of scales they have to engage themselves into international trade and specialise in their production and exports. As a consequence, small are countries are more vulnerable than larger ones to external developments like industry-specific shocks and technological changes. In the context of science and technology policy – which was the main focus of the project KNOGG – smallness is a constraint with respect to availability of resources and scope of R&D. Perhaps the most important specific feature of small countries is that in their R&D and also in the development of their knowledge base they depend more on external sources of knowledge and innovations than larger countries. Most of the applied knowledge in small countries is based on imported knowledge. To be successful the national innovation systems of small countries have to take this feature which into account. Small countries may lack the critical mass needed in domestic R&D, particularly in basic research. Due to the fact that their investment in R&D will represent only a small fraction of global R&D invested in almost any field of technology, small countries are unlikely to win dominance on the global market. Due to the small size of their domestic market, they are also unlikely to be able to keep the positive spillovers of domestically generated knowledge confined to their local economies. Research capabilities in small countries may also be vulnerable to brain drain. This is especially the case if the country belongs to the lower echelon of the per capita income levels or if the country is not able to provide sufficient working conditions for highly skilled people or critical mass for research. One of the main findings of the KNOGG project is that a well developed NIS is of central importance in the development of knowledge based society which

99 supports innovations. A well-functioning NIS consists of strong basic research at least in some key areas, extensive education, and effective intermediate institutions between the research community and the corporate sector. Strong research is not enough to produce innovations. Co-operation and networking with innovative enterprises are also needed. Hence the public STI institutions and government policies should promote linkages between universities and industries, and commercialisation of public research findings. They should enable knowledge diffusion and help to reduce the barriers to knowledge transfer, promote international co-operation in STI and the national participation in international organisations and programs. Governments should also assist institutions which are dealing with the social understanding and popularisation of science and technology. Public attitudes are important for the political acceptability of long term STI policies. The basic role of NIS is that it should improve the adaptability and knowledge base of a country. Thus knowledge creating and intermediate institutions are important in the NIS. Globalization is changing rapidly the innovation environment. New centres of knowledge are emerging in various parts of the world. Transnational corporations create their own research platforms. These changes are connected to the increased mobility of firms. It follows that in this new globalised environment NIS should also become more ‘global’. For small countries this may mean that their innovation systems become part of the global production structure of TNCs. Due to differences in history and institutions, different countries have chosen different strategies in their technology policies. This is true also in the case of the KNOGG countries, which represent a great variety of policies and institutions. Broadly speaking, there is a choice between strategies based either on “innovation” or “imitation”. For a small country it is the most natural and least risky strategy to encourage “imitation”, or technology diffusion through investment and learning. However, in some cases even small countries can be innovators and technological leaders. Among KNOGG countries, Finland is clearly an example of the innovation-based growth strategy. In Finland, national S&T policy has been concentrated on the creation of centres of innovation and platforms for technology firms. The Finnish higher education system has supplied lots of highly skilled labour force. As a result, Finland is the only one of the KNOGG countries which has already achieved the Lisbon target of R&D spending exceeding 3 percent of GDP, and that actively promotes S&T policies that enforce the national innovation system. These policies have made the country an attractive place for R&D intensive inward FDI. The highly successful growth strategy of Ireland has been based on different instruments. In Ireland tax policies have been practiced with a main eye to attract

100 FDI. Ireland has thus relied clearly on technology diffusion. In economic terms the results have been good. There are marked differences in how the KNOGG countries have formulated their policies towards FDI. The way out of the constraints of being a small country is to make an optimal use of the increased internationalisation of technology and globalisation by upgrading the domestic knowledge base through the right policies. A special role in this effort can be played by transnational companies (TNCs) or foreign direct investments (FDI). The small domestic markets of small countries do not enable firms to recoup fully their R&D costs, particularly in a time of shortening product life cycles and increased competition. Consequently, R&D intensive firms from small home countries need export outlets and outward foreign direct investments in order to achieve growth. This internationalisation of technology from domestic firms in small countries cannot be considered as a weakness in the national innovation systems but as a logical consequence of being small. At the same time the existence of large domestic TNCs in small countries (like the big Dutch multinationals or Nokia in Finland) make it difficult for policy planners to control domestic R&D structures because a large share of NIS may actually controlled by one large firm. From the viewpoint of small host countries, TNCs are of extreme importance as a means to improve their science and technology base. As a consequence, it is necessary to develop policies that make small countries attractive for TNCs to integrate them in the international production chains at a high technology part of the production process. In the Netherlands the most important fiscal incentives with R&D aims is a tax relief (so called WBSO) aimed at stimulating R&D. With regard to S&T policies also measures exist to stimulate R&D-co-operation among firms and between firms and S&T institutions. Technological loans for high-risk innovations are provided as well. However, the budget cuts on education in the last twenty years have led to a shortage of technologically skilled workers. This is expected to go on and make the Netherlands a relatively less attractive place for high technology investments in the future. In Greece some fiscal measures exist but these are not part of explicit S&T policies. The labour market of skilled workers is rigid and there is a shortage of engineers. With regard to Hungary tax deduction opportunities for R&D activities exist also for R&D-activities sourced out to other organizations like universities and nonprofit R&D institutes. Hungarian tax policies include tax breaks, free trade zones and industrial parks as well as grants, interest subsidies and loan guarantees for the SME sector. With regard to S&T policies, R&D expenditures are fully tax deductible since 2001.

101 There is also a Supplier Target Program which aims at increasing links between inward FDI and local firms in order to increase spillovers from FDI to Hungarian firms. This fits in the theoretical policy advices suggesting to pay attention to the development of local firms in order to make them more receptive for foreign technology. Yet the results have been mixed. Hungary does not have strong applied R&D institutes, and universities are not an active part of the innovation system due to lack of resources. In Slovenia corporate taxes are not considered an impediment for FDI. There are no special incentives to encourage S&T co-operation and development. Slovenian investments in the education system are insufficient to make the country an attractive place for high technology components of high technology firms. Although Ireland, the Netherlands and Hungary try to attract FDI with linkages to their domestic economy, there often are no linkages to the national innovation system. Greece and Slovenia use financial incentives to attract TNCs but do not have explicit policies that aim at increasing linkages. In order to benefit effectively from the imports of knowledge it is important to have the national capacity to adopt new knowledge sufficiently high. This also means that strong NIS is important even in countries which mostly import there knowledge and are mainly imitators in the world market. The development of innovation basis, including national innovation system takes a long time. Therefore long term commitment is needed. This concerns e.g. both the development of education system and scientific research basis in a country. The build-up of the STI policy infrastructure is usually strongly path dependent. The studies of KNOGG project strongly point to the fact that even in the case of leaning either on the increasing FDI or even if a country follows mainly the imitation strategy, it is important to invest to the domestic scientific and research capabilities. And this cannot succeed without the development of adequate education policy. The development of these take a long time and effects of these development actions are felt only gradually. But without sufficient capacity even the adaptation into the new international environment becomes more difficult. The European Union has become an important medium for promoting internationalisation of STI in Europe and transmitting global processes. Achieving an innovation performance that makes the European Union a world reference for innovation represents an enormous opportunity that can translate into raised living standards over the coming years. Progress towards such a more innovative European economy is however proving tentative and fragile. Enhancing innovation is a cornerstone of the strategy to meet the target agreed by the European Council in Lisbon in March 2000 – otherwise the Union cannot

102 become the most competitive and dynamic knowledge-based economy by the end of the decade. European STI policy can support the national efforts, and the development of the ERA could help to take the needs for long term commitment into account. From the perspective of small countries the new European STI policies should help to avoid the new risks which the small countries are facing. So far the largest part of EU transfers, i.e. structural funds, has not been very effective in improving the knowledge-base. The Framework Programmes of European research have been important especially to countries which have been weak in R&D. However, the framework programmes still face challenges in better integrating SMEs. EU policies should complement the national policies. Market forces are likely to promote further concentration of innovative activities and the European research capacity. From the perspective of small countries that is something which should be avoided. The new European policies should not prevent national policy measures to support national and regional innovation systems. The ERA can support new networking and help to create European centres of excellence. If the Lisbon targets are to be realized Europe needs high level scientific institutions. But achieving excellence may also mean more geographical concentration of the scientific capacity and R&D activities and to the development of strongly concentrated industrial clusters. This kind of development is a problem for small countries if they do not belong to the ‘core’ or central clusters. It is therefore important that in the European policy development the potential risks for small countries are to be taken into account. Otherwise two Union objectives, the search for excellence on one hand and regional cohesion on the other, may to some extent be in conflict. To avoid polarization, the creation and support of regional centres of excellence with strong international participation could and should be a goal of Union level research and technology policies. The EU policies should hence support the establishment of regional RTD centres of excellence also in small member countries. In most cases they should not only be virtual networks of excellence but permanent research centres and innovation platforms. We think that is necessary in order to engage the key partners to longterm work. That of course makes it necessary to member countries to choose their own priorities and redirect their own RTD funding to such centres, too. The creation of the European Research Area may yield a permanent change in the RTD policy formation and implementation in Europe. It can play a crucial role in the more organic integration of the new and disadvantaged member countries into the European R&D structure. Of particular importance here is the capacity of the country to compete successfully for funds and the capacity of the economy and society to absorb and productively utilise new knowledge, science

103 and technology. Small countries may be in unfavourable position in the creation of more abundant and more mobile human resources. Therefore this needs a special attention in the future work of the development of EU policies in this area.

8.2 Suggestions for future research What is left to be explained for future research is a more detailed investigation of types of R&D co-operation between foreign firms on the one hand and domestic firms and S&T institutions on the other hand in a number of small and large EUand accession countries. A quantitative analysis of these forms of R&D cooperation would reveal more systematically and in detail the determinants of foreign firms to get involved in co-operation networks with domestic firms and S&T institutions. In addition it would make clear at what target variables future S&T policies in small countries should aim at and how such policies can be framed at the national and EU level. The knowledge-base of STI policy planning should be strengthened through coordinated joint European efforts. There is a need to develop national statistics and indicators on STI which facilitate the systematic measurement and international comparisons. Benchmarking exercises have become a popular and useful instrument also for STI decision-makers on different levels of activities. The idea behind benchmarking is simple: find out what the best performers are doing and catch up, fast. It is a valuable idea for those who are exposed to international competition and are behind the standards of the best performers. The benchmarking exercise is a very rough evaluation of a country’s performance as a whole and it has been argued that one cannot build a synthesis of the country’s performance on the grounds of performance on individual indicators. Foresight is an other important multipurpose and increasingly popular instrument for policy makers. Governments, corporations, international organisations and academic groups used science and technology forecasting for the support of their long term goals. Foresight studies are aimed at the development of a common vision for the society, they identify future technologies and technological changes, they intend to guide national technology policies and offer “intelligence” for the different firms and institutions. They combine science vision, business vision and social vision on science and technology. Some experts consider it as a vitally important instrument of STI policies. There are certain important conditions which are essential for the proper use of foresight: multidisciplinary approach, the participation of a wide range of knowledge “producers” and users and the appropriate understanding of factors which influence R&D and STI policies. It must be kept away from the influence of

104 lobbies as much as possible. Foresight exercises may be more difficult to small countries where the expertise may be limited compared to large countries; if that is the case, small countries need international co-operation in this field. Most European countries need better data to generate knowledge on their economies and the impacts of policies, in order to carry out policies that are truly knowledge-based. Good data is the prerequisite for informed choices and strategies. The quality and productivity of R&D hinges on it. Policymakers need to know the impacts of R&D subsidies in order to target them effectively. At the European level, the provision of reliable comparable data would benefit research and policy comparison efforts. Further research using more comprehensive European data sets is needed to clarify our view of the state of European research, innovation systems and entrepreneurship. We need better understanding of knowledge flows between universities, research institutes and firms, and better understanding of the role of entrepreneurship and of the factors affecting it.

105

References Acs, Z.J. – Audretsch, D.B. (1990): Innovation and small firms. Cambridge: MIT Press. Andersson, T. (2001): Seizing the opportunities of a new economy: challenges for the European Union, No. 8. OECD Growth Project: Background Papers 1. Audretsch, D.B. – Thurik, R. (2001): Linking entrepreneurship to growth. OECD Working Papers No. 2001/2 Beers, C. van (2003): The Role of Foreign Direct Investments on Small Countries’ Competitive and Technological Position. VATT Research Reports 100. Government Institute for Economic Research, Helsinki. Berghäll, E. – Heikkilä, T. – Hjerppe, R. – Kiander, J. – Kilponen, J. – Lavrac, V. – Stanovnik, P. (2002): The Role of Science and Technology Policy in Small Economies. VATT Research Reports 91. Government Institute for Economic Research, Helsinki. Blanchflower, D.G. – Oswald, A.J. (1998): What makes an entrepreneur? Journal of Labour Economics, 16(1). Blomström, M. – Kokko, A. (2003): The Economics of Foreign Direct Investment Incentives. NBER Working Paper 9489. Bottazzi, L. – Da Rin, M. (2003): Financing entrepreneurial firms in Europe: facts, issues, and research agenda. Cesifo Working Paper No. 958. Caracostas, P. – U. Muldur (1997): Society, the Endless Frontier, Brussels and Luxembourg: European Communities Publication Office. Caracostas, P. – L. Soete (1997): The Building of Cross-border Institutions in Europe: Towards a European System of Innovation? In Edquist, C. (ed.): Systems of Innovation Technologies, Institutions and Organisations. London: Pinter, pp. 395–408. Castells, M. (1996–1998): The Information Age: Economy, Society and Culture. Volumes 1–3, Oxford and Malden. Castells, M. – Himanen, P. (2002): The Information Society and the Welfare State. The Finnish Model. Oxford University press, Oxford, UK. Cogan, D. J. (1992): Innovation Output of Irish Industry in 1991. Based on research for Literature-Based Innovation Output Indicators, a project initiated by the Institute for Prospective Technological Studies of the EU Joint Research Centre in Seville.

106 Cogan, J. – McDevitt, J. (2003): Science, Technology and Innovation Policies in Selected Small European Countries. VATT Research Reports 96. Government Institute for Economic Research, Helsinki. Cogan, J. – J. McDevitt – M. M. Godinho (2001): Internal and External Factors in the Catching up of European Less Favoured Regions: a comparative study of Portuguese and Irish growth experience. Paper in Converge Project (financed by TSER Programme, DGXII). Cohen, W.M. – Levinthal, D.A. (1990): Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly, 35. Cordis Focus. Number 233. 2003. Dicken, P. (1992): Global Shift. The Internationalization of Economic Activity. Paul Chapman Publishing Ltd, London. Dosi, G. (1988): Sources, procedures, and microeconomic effects of innovation. Journal of Economic Literature, 26. Driffield, N. – J. H. Love (2002): Does the Motivation for Foreign Direct Investment Affect Productivity Spillovers to the Domestic Sector. RP0202, Aston Business School Research Papers. March. Dumont, M. – Tsakanikas, A. (2002): The evolution of intra- and inter-sector knowledge spillovers in the EU Framework Programmes. CESITpaper2002/01. EC DG Enterprise (2000): Innovation in a knowledge-driven economy. COM(2000) 567 final. EC DG Enterprise (2001): Innovation Policy in Six Candidate Countries. Edquist, C. (ed.)(1997): Systems of innovation: Technologies, institutions and organizations. London: Pinter Publishers. Ekholm, K. – Hakkala, K. (2003): Location of R&D and high-tech production by vertically-integrated multinationals. CEPR Discussion Paper No. 4078. EMCO/22/060602/EN_REV 1, IMPACT EVALUATION OF THE EES. Background paper. Entrepreneurship. EMCO/24/060602/EN_REV 1, IMPACT EVALUATION OF THE EES. Background paper. Lifelong learning. European Commission (2000): Towards a European Research Area. Brussels, 18 January 2000: European Commission, COM (2000) 6. European Commission (2002): The European Research Area: Providing New Momentum. European Commission Communication. Brussels: European Commission.

107 European Commission (2003): Third European Report on Science & Technology Indicators: Towards a knowledge-based economy. DG Research. European Observatory for SMEs (2000): Sixth Report. European Commission. Fagerberg, J. – Verspagen, B. (1996): Heading for divergence? Regional growth in Europe reconsidered. Journal of Common Market Studies, 31(5). Florida, R. – Kenney, M. (1993): The New Age of Capitalism: InnovationMediated Production. Futures 25 (6): 637-652. Freeman, C. (1987): Technology policy and economic performance: lessons from Japan. London: Pinter Publishers. Fujita, M. – Krugman, P.R. – Venables, A.J. (1999): The spatial economy: cities, regions and international trade. Cambridge: MIT Press. Fujita, M. – Thisse, J. (2002): Economics of Agglomeration. Cambrdige: Cambridge University Press. Gregersen, B. – B. Johnson – A. Kristensen (1994): National Systems of Innovation and European Integration. Paper presented to the EUNETIC conference on Evolutionary economics of technical change: assessment of results and new frontiers, Strasbourg: European Parliament, 68 October 1994. Henrekson, M. – Rosenberg, N. (2001): Designing efficient institutions for science-based entrepreneurship: lesson from the US and Sweden. Journal of Technology Transfer, 26. Hjerppe, R. (2003): Social Capital and Economic Growth Revisited. VATT Discussion Papers 307. Government Institute for Economic Research. IMD (2003): World Competitiveness Yearbook 2002. Geneva. Jacobs, D. (1989): Small Countries’ Opportunities for Participating in ScienceBased Development. In van Tulder, R. (ed.): Small Industrial Countries and Economic and Technological Development. SICRA/NOTA. Kaiser, U. (2003): Product innovation and product innovation marketing: theory and microeconomic evidence. CEBR Discussion Paper No 2003-06. Katzenstein, P.J. (1985): Small states in world markets: industrial policy in Europe. New York: Cornell University Press. Keuschnigg, C. – Nielsen, S.B. (2003): Tax policy, venture capital and entrepreneurship. Journal of Public Economics. Keuschnigg, C. – Nielsen, S.B. (2004): Start-ups, venture capitalists, and the capital gains tax. Forthcoming in Journal of Public Economics. Kristensen, P.H. – J. Levinsen (1983): The Small Country Squeeze. Roskilde: Forlaget for Safundsokonomi.

108 Krugman, P.R. (1991): Increasing returns and economic geography. Quarterly Journal of Economics, 99(3). Kuhlmann, S. (2001): Future governance of innovation policy in Europe: Three scenarios. Research Policy, 30, pp. 953–976. Lindh, T. – Ohlsson, H. (1996): Self-employment and windfall gains: Evidence from the Swedish Lottery. Economic Journal, 106 (439). Lindh, T. – Ohlsson, H. (1998): Self-employment and wealth inequality. The Review of Income and Wealth, 44(1). Lundvall, B.-Å. (1988): Innovation as an interactive process: from user-producer interaction to the national innovation systems. In Dosi, G. – Freeman, C. – Nelson, R.R. – Silverberg, G. – Soete, L- (eds.): Technology and economic theory. London: Pinter Publishers. Lundvall, B.-Å. (1992): National systems of innovation – towards a theory of innovation and interactive learning. London: Pinter. Lundvall, B.-Å. (2001): Innovation, growth and social cohesion. Edward Elgar. Matthews, M. and F. McGowan (1992): Reconciling Diversity and Scale: Some Questions of Method in the Assessment of the Costs and Benefits of European Integration 1992, N° 59 1e tr. in N° Spécial de la Revue d’Economie Industrielle ‘Technological diversity and coherence in Europe’. Moon, H. – A.M. Rugman – A. Verbeke (2001): A generalized double diamond approach to the global competitiveness of Korea and Singapore. In D. van den Bulcke – A. Verbeke (eds.): Globalization and the Small Open Economy. Cheltenham: Edward Elgar. OECD (2001): Science, Technology and Industry Outlook. Drivers of growth: Information technology, innovation and entrepreneurship. Paris: OECD. Palmberg, C. – Leppälahti, A. – Lemola, T. – Toivanen, H. (1999): Towards a better understanding of innovation and industrial renewal in Finland – A new perspective. VTT Group for Technology Studies Working Papers. Rickne, A. – Jacobsson, S. (1996): New technology-based firms – An exploratory study of technology exploitation and industrial renewal. International Journal of Technology Management, 11(3/4). Rickne, A. – Jacobsson, S. (1999): New technology-based firms in Sweden – A study of their direct impact on industrial renewal. Economics of Innovation and New Technology, 8(2). Rodrik, D. (2003): Growth Strategies. CEPR Discussion Papers No. 4100. Romer, P. (1986): Endogenous Technological Change. Journal of Political Economy, 94, 1002-37.

109 Romer, P. (1990): Endogenous Technological Change. Journal of Political Economy, 98, 71-102. Rosenberg, N. (2000): Schumpeter and the endogeneity of technology: Some American perspective. London: Routledge. Sapir, A. – Aghion, P. – Bertola, G. – Hellwig, M. – Pisani-Ferry, J. – Rosati, D. – Viñals, J. – Wallace, H. – Buti, M. – Nava, M. – Smith, P.M. (eds.) (2003): An Agenda for a Growing Europe: The Sapir Report. Oxford University Press, Oxford. Simai, M. (ed.) (2003): Practical Guide for Active National Policy Makers – What Science and Technology Policy Can and Cannot Do? VATT Research Reports 103. Government Institute for Economic Research, Helsinki. Smith, Keith (2002): What is the “knowledge based economy”? UNU-INTECH discussion paper 2002-6. ISSN 1564-8370. Suetens, S. (2001): Cooperative versus competitive R&D behaviour in experimental duopoly markets, CESIT-paper 2001/04. Széchenyi (2000): Széchenyi Terv – Információs társadalom – és gazdaságfejlesztési program (Széchenyi Plan – Information Society and Information Economy). Gazdasági Minisztérium (Ministry of Economics), Budapest. Tsipouri, L. (2003): Can less favoured regions change their destiny? Lessons from Europe. In Fuchs, G. – Shapira, P. (eds.): Rethinking Regional Innovation: Path Dependency or Regional Breakthrough? Kluwer. Tsipouri, L. and M. Xanthakis (1993): Impact of the EC Science and Technology Policy on the Greek S/T Policy, Athens: University of Athens. Uusitalo, R. (1999): Homo Entreprenaurus. VATT Discussion Papers 206. Government Institute for Economic Research, Helsinki. Walsh, V. (1987): Technology, Competitiveness and the Special Problems of Small Countries. A Survey of the Literature. IKE Workshop on Technological Change and the Competitiveness of Small Countries, Denmark. von Zedtwitz – O. Gassmann (2002): Market versus technology drive in R&D internationalisation: four different patterns of managing research and development. Research Policy, 31, 569–588. World Bank (2001): World Development Indicators 2000. World Economic Forum (WEF) (2003): Global Competitiveness Report 2002. www.cordis.lu/trendchart/.

110

Glossary The following abbreviations appear in this publication AKMON

Research Centres Development and Services Supply Projects with User Participation)

BERD

Business Expenditure on R&D

BUNT

Business Development Using New Technology – launched in Norway in 1989

CC

Candidate Country

CRC

Co-operative Research Centre (Hungary)

EES

European Employment System

EIS

European Innovation Scoreboard

EPET III

Community Support Framework 2000–2006 (Greece)

EPO

European Patent Office

ERA

European Research Area

ERDF

European Regional Development Fund

ESF

European Social Fund

ESPRIT

Inter-country research programmes within the EU

EU

European Union

FDI

Foreign Direct Investment

FORFAS

Policy Advisory and Co-ordination Board for Industrial Development and Science and Technology (Ireland)

FP

European Union RTD Framework Programme

GDP

Gross Domestic Production

GERD

Gross intramural Expenditure on Research and Development

GSRT

General Secretariat for Research and Technology Greece

HE

Higher Education

HEI

Higher Education Institution

HC

Human capital

ICSTI

Irish Council for Science, Technology and Innovation (Ireland)

111 ICT

Information and Communications Technologies

IMT

Innovation Management Techniques

IS

Information Society

JPO

Japanese Patent Office

KETA

Centre of Entrepreneurial and Technological Development (Greece)

LFR

Less favoured regions

LLL

Life-Long-Learning

MINT

Managing the Integration of New Technology

MTA

Hungarian Academy of Sciences

NIC

Newly Industrialized Countries

NIS

National Innovation System

NTBF

New Technology-based Firm

OECD

Organisation for Economic Co-operation and Development

OMFB

National Committee for Technological Development (Hungary)

PAPHOS

Initiative encourages benchmarking and the assimilation of best practice and world-class technologies

PAT

Programme in Advanced Technology (Ireland)

PEPER

Promotes the participation of Greek firms in large-scale demonstration and TT programmes

PRIs

Public sector research institutions

R&D

Research and Development

RTD

Research and Technological Development

RTDI

Research, Technological Development and Innovation

SENTER

Executive Research and Innovation Agency (Netherlands)

SME

Small and Medium Enterprises

STI

Science, Technology and Innovation

SYNTENS SME Support Measure - formerly Innovation Centres (Netherlands) TEKES

National Technology Agency (Finland)

TNC

Transnational Companies

TNO

Contract Industrial Research Institute (Netherlands)

112 TRIPS

Agreement on Trade-Related Aspects of Intellectual Property Rights

TT

Technology Transfer

US

United States

USPTO

US Patent & Trademark Office

VTT

Technical Research Centre (Finland)

WP

Work Package (KNOGG Project)

VATT-TUTKIMUKSIA -SARJASSA ILMESTYNEITÄ PUBLISHED VATT RESEARCH REPORTS 74.

Laine Veli – Uusitalo Roope: Kannustinloukku-uudistuksen vaikutukset työvoiman tarjontaan. Helsinki 2001.

75.

Kemppi Heikki – Lehtilä Antti – Perrels Adriaan: Suomen kansallisen ilmastoohjelman taloudelliset vaikutukset. Vaiheen 2 loppuraportti. Helsinki 2001.

76.

Milne David – Niskanen Esko – Verhoef Erik: Legal and Institutional Framework for Marginal Cost Pricing in Urban Transport in Europe. Helsinki 2001.

77.

Ilmakunnas Seija – Romppanen Antti – Tuomala Juha: Työvoimapoliittisten toimenpiteiden vaikuttavuudesta ja ennakoinnista. Helsinki 2001.

78.

Milne David – Niskanen Esko – Verhoef Erik: Acceptability of Fiscal and Financial Measures and Organisational Requirements for Demand Management. Helsinki 2001. (Not yet published).

79.

Venetoklis Takis: Business Subsidies and Bureaucratic Behaviour. Helsinki 2001.

80.

Riihelä Marja – Sullström Risto: Tuloerot ja eriarvoisuus suuralueilla pitkällä aikavälillä 1971-1998 ja erityisesti 1990-luvulla. Helsinki 2001.

81.

Ruuskanen Petri: Sosiaalinen pääoma – käsitteet, suuntaukset ja mekanismit. Helsinki 2001.

82.

Perrels Adriaan – Kemppi Heikki – Lehtilä Antti: Assessment of the Macro-economic Effects of Domestic Climate Policies for Finland. Helsinki 2001. Tulossa.

83.

Venetoklis Takis: Business Subsidies and Bureaucratic Behaviour, A Revised Approach. Helsinki 2001.

84.

Moisio Antti – Kangasharju Aki – Ahtonen Sanna-Mari: Menestyksen mitta? Vaihtoehtoisia mittareita aluetalouden kehitykselle. Helsinki 2001.

85.

Tuomala Juha: Työvoimakoulutuksen vaikutus työttömien työllistymiseen. Helsinki 2002.

86.

Ruotoistenmäki Riikka – Babygina Evgenia: The Actors and the Financial Affairs of the Northern Dimension. Helsinki 2002.

87.

Kyyrä Tomi: Funktionaalinen tulonjako Suomessa. Helsinki 2002.

88.

Räty Tarmo – Luoma Kalevi – Koskinen Ville – Järviö Maija-Liisa: Terveyskeskusten tuottavuus vuosina 1997 ja 1998 sekä tuottavuuseroja selittävät tekijät. Helsinki 2002.

89.

Hakola Tuulia: Economic Incentives and Labour Market Transitions of the Aged Finnish Workforce. Helsinki 2002.

90.

Venetoklis Takis: Public Policy Evaluation: Introduction to Quantitative Methodologies. Helsinki 2002.

91.

Berghäll Elina – Heikkilä Tuomo – Hjerppe Reino – Kiander Jaakko – Kilponen Juha – Lavrac Vladimir – Stanovnik Peter: The Role of Science and Technology Policy in Small Economies. Helsinki 2002.

92.

Räisänen Heikki (toim.): Rakenteellinen työttömyys. Tutkimusinventaari ja politiikkajohtopäätökset. Helsinki 2002.

93.

Moisio Antti: Essays on Finnish Municipal Finance and Intergovernmental Grants. Helsinki 2002.

94.

Parkkinen Pekka: Hoivapalvelut ja eläkemenot vuoteen 2050. Helsinki 2002.

95.

Junka Teuvo: Maailman kilpailukykyisin maa? Tuottavuus ja investoinnit Suomessa 1975-2000. Helsinki 2003.

96.

Cogan Josehp – McDevitt James: Science, Technology and Innovation Policies in Selected small European Countries. Helsinki 2003.

97.

Perrels Adriaan – Kemppi Heikki: Liberalised Electricity Markets – Strengths and Weaknesses in Finland and Nordpool. Helsinki 2003.

98.

Sarvimäki Matti: Euroopan Unionin itälaajentuminen ja maahanmuutto Suomeen. Helsinki 2003.

99.

Räty Tarmo – Luoma Kalevi – Mäkinen Erkki – Vaarama Marja: The Factors Affecting the Use of Elderly Care and the Need for Resources by 2030 in Finland. Helsinki 2003.

100.

van Beers Cees: The Role of Foreign Direct Investments on Small Countries’ Competitive and Technological Position. Helsinki 2003.

101.

Kangasharju Aki: Maksaako asumistuen saaja muita korkeampaa vuokraa? Helsinki 2003.

102.

Honkatukia Juha – Forsström Juha – Tamminen Eero: Energiaverotuksen asema EU:n laajuisen päästökaupan yhteydessä. Loppuraportti. Helsinki 2003.

103.

Simai Mihály (ed.): Practical Guide for Active National Policy Makers – what Science and Technology Policy Can and Cannot Do? Helsinki 2003.

104.

Luoma Arto – Luoto Jani – Siivonen Erkki: Growth, Institutions and Productivity: An empirical analysis using the Bayesian approach. Helsinki 2003.

105.

Montén Seppo – Tuomala Juha: Muuttoliike, työssäkäynti ja työvoimavarat Uudellamaalla. Helsinki 2003.

106.

Venetoklis Takis: An Evaluation of Wage Subsidy Programs to SMEs Utilising Propensity Score Matching. Helsinki 2004.

107.

Räisänen Heikki: Työvoiman hankinta julkisessa työnvälityksessä. Helsinki 2004.

108.

Romppanen Antti: Vakaus- ja kasvusopimuksen ensimmäiset vuodet. Helsinki 2004.

109.

Vaittinen Risto: Trade Policies and Integration – Evaluations with CGE Models. Helsinki 2004.