Ocde 1999b Mgt Science Systems

OECD

STI

SCIENCE

TECHNOLOGY

INDUSTRY

THE MANAGEMENT OF SCIENCE SYSTEMS

THE MANAGEMENT OF SCIENCE SYSTEMS

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT

TABLE OF CONTENTS

Introduction .................................................................................................5 Closer links between science and innovation ..............................................5 Scientific production in OECD countries ....................................................6 University research at the heart of science systems.....................................7 Science funding: stable but increasingly subject to conditions .................10 Setting up regulatory structures and frameworks: partnerships and entrepreneurship ............................................................15 The science workforce: adjusting to change..............................................16 Evaluation of institutions and researchers .................................................17 Science and information technologies: meeting the new challenges.........18 Problems specific to the social sciences ....................................................20 Preventing the widening of the gap between science and the general public...............................................................................20 Conclusion: some policy issues .................................................................24 REFERENCES..............................................................................................27 ANNEX.........................................................................................................29 Bibliometric data .......................................................................................29 Financing of university research................................................................33 Management of the science base ...............................................................36

3

THE MANAGEMENT OF SCIENCE SYSTEMS

Introduction For some years now, science systems have been subjected to significant pressures. These have created new tensions but also opened up new opportunities. In today’s increasingly globalised economies, science systems are expected to contribute still more to innovation, at a time when governments, confronted with budgetary problems, are reallocating their support. At the same time, science systems are having to contend with specific problems associated with their own evolution: entry into new fields, ageing and renewal of research personnel, adjustment to the new information and telecommunication technologies, societal concerns in the face of certain aspects of scientific “progress”, and so on. The Committee for Scientific and Technological Policy (CSTP), and more specifically its Group on the Science System (GSS), has conducted a series of studies and conferences to gain a better understanding of the changes taking place and related problems. This brochure presents the main observations and conclusions of this work. Publications resulting from these activities are included in the references. Topics requiring in-depth study in order to provide decision makers with further policy insights are also identified. Closer links between science and innovation The aim of science is to understand the laws of nature (and the characteristics of societies); that of innovation is to develop and market new products and processes. These two activities are therefore intrinsically different (OECD, 1998a; 1997a). Innovation involves certain activities that have little to do with science, such as the relation to the market, technical development, creation of a firm. Moreover, research results seldom lead directly from science to innovation. In-depth studies on how innovations are developed show that

5

scientific research contributes to innovation indirectly rather than directly by providing solutions to economic and societal problems, by transferring instruments developed for research purposes to industry, by training researchers who are later employed by industry (Martin and Salter, 1996). Moreover, when asked about what led them to innovate, most firms ranked feedback from clients, suppliers, etc., far ahead of basic research. Moreover, the knowledge required for innovation differs from that produced by science. In the latter, knowledge is structured and produced in a fragmented way with little connection between disciplines and sub-disciplines, through a process of deepening and accumulation. Between scientific advances and innovations in the form of products or processes, knowledge is organised around technology areas of a generic or multi-application nature. Here, progress is based on a process of integrating separate elements. This is followed by dissemination, and/or further integration, of technology into different applications at the more detailed level of product development. Yet, although their relationship is rather complex, innovation appears increasingly to depend on scientific progress. While innovation often preceded science in the distant past – the steam engine was invented well before the discovery of the principles of thermodynamics – scientific advances increasingly determine technological progress, as evidenced by developments in electronics and, more recently, biotechnologies, where science and technology are tightly interwoven. Bibliometric studies, carried out in the United States in particular, show that patents now rely more on academic scientific publications than they did in the past (Narin et al., 1997). This phenomenon is undoubtedly encouraged by the possibilities for automated access to scientific databases. It reflects a trend, however, which is particularly visible in new sectors: information technologies, health (pharmaceutical products and biotechnologies) and new materials (Albert, 1998). Consequently, the world is witnessing the rapid and varied development of a body of knowledge at the science/innovation interface. This knowledge is becoming increasingly interdisciplinary in nature, informal in character, and practical in its applications and focus. It is sometimes referred to as “mode 2” knowledge to differentiate it from classic scientific knowledge (Gibbons, 1994).

6

Scientific production in OECD countries Bibliometric data, which measure scientific production in the form of publications, provide information on the amount and orientation of scientific activity in OECD Member countries (see Annex Tables A1-A3). It appears that: ♦ The United States is the source of more than 30% of the articles published in mainstream scientific journals. Well behind, with between 9% and 5%, come Japan, the United Kingdom, Germany and France. ♦ In terms of scientific productivity (number of publications relative to GDP), the Northern European countries stand first, led by the Englishspeaking countries (the United Kingdom, Ireland). ♦ There are notable differences in terms of disciplinary specialisations. English-speaking countries present a rather well-balanced profile, while the Asian countries are more oriented towards engineering and technology and the Nordic countries towards medicine and clinical research. ♦ There are also important differences among countries in the degree of internationalisation, as measured by the share of articles with co-authors from other countries. European countries, particularly the “small” ones, show a high degree of internationalisation, while the United States and Japan are much less internationally oriented. However, the degree of internationalisation has increased substantially overall over the last decades. ♦ Finally, in terms of links between science and innovation, countries show significant differences. Links are strong in the English-speaking countries, but weaker in the Nordic countries (Denmark is an exception), as well as in German-speaking and especially Asian ones. University research at the heart of science systems The boundaries of science systems are not easy to define. There is a certain blurring, particularly in terms of statistical measurement. What is the weight of science systems in research systems, the boundaries of which are clearly defined thanks to the Frascati Manual (OECD, 1994)?

7

On a very narrow definition, science systems can be equated with basic research, which represents at most 15% of the total R&D effort. If academic research is used as a (very imperfect) proxy, science systems account for between 15% and 35% of the R&D effort (Figure 1) and 15% to 60% of the related workforce (Figure 2). The share is greater in small developed countries where public research is generally moderate and/or in less developed economies (where industrial research is limited). Figure 1. Percentage of GERD performed in the higher education sector

35

30 Italy

Canada

Iceland

Netherlands

Sweden

Norway

30

25 United Kingdom

25

20 20

Japan

Denmark

15

15 Germany

United States

France

10

10

1985 86 87 88 89 90 91 92 93 94 95 961997

1985 86 87 88 89 90 91 92 93 94 95 961997 75

45 Portugal

40

Spain

65 Turkey

55

Belgium

35

Austria

45 30

Finland Switzerland

New Zealand Mexico

Greece

Poland

35 25 Ireland

20

25 Korea

Australia

15

15

Hungary

5 1985 86 87 88 89 90 91 92 93 94 95 961997

1985 86 87 88 89 90 91 92 93 94 95 961997

Source: OECD, S&T databases, March 1999.

Countries vary widely in terms of the position of university research in the science system and, more especially, relative to public research. Several profiles can be identified, which are related to a country’s socio-cultural context and its economic structure (OECD, 1998b; 1998c): ♦ In English-speaking countries, the university is the principal setting of fundamental research, but there is also significant public research in sectors of national interest, such as defence, energy, agriculture, medicine. 8

Public research bodies may undertake fundamental research, if the need arises, but generally focus on applied research. ♦ In the larger countries of continental Europe, university research coexists with a large public sector closely involved in fundamental research in its own laboratories (e.g. the Max Planck institutes in Germany, the CNRS in France, the CNR in Italy). Universities also carry out applied research, either to provide R&D infrastructure (as in Germany) or for mission-oriented activities (as in France and Italy). ♦ In other countries of continental Europe, public research tends to focus essentially on applied research, while the bulk of fundamental research is carried out in the universities. Here again, countries differ widely: in some, like Norway, the public sector is large; in others, like Sweden and Switzerland, it is very small. ♦ In the countries of eastern Europe influenced by the Soviet model, university research was generally very limited (with some exceptions, such as Poland). Academy of Sciences institutes conducted basic research. Reforms undertaken since the beginning of the 1990s aim at increasing research activities in universities while reducing, or even eliminating, research in the academies. The R&D activities of the branch institutes, which carried out industrial research, have declined considerably owing to budgetary cuts and the adoption of a market economy. ♦ Finally, in the East Asian countries (Japan, Korea), where research has generally focused on technical applications, the scale of university research has been modest until recently, owing to lack of funding, excessive regulation and constraints related to teaching commitments. The situation is changing rapidly and these countries, especially Japan, are now vigorously encouraging their own fundamental research. Similarly, there are wide differences in the functioning of university research and the behaviour of teachers/researchers in different university systems. In the English-speaking countries, and particularly in the United States, academic research, while adhering strictly to the usual criteria of academic excellence, is well aware of the market principles that govern the economy as a whole and thus evolves in a strongly competitive environment. Researchers are very concerned with publication of their findings and are under constant scrutiny by their peers. They are highly mobile, moving easily from one university to another on the basis of offers received. They are strongly encouraged to obtain contracts with industry, government agencies and local authorities in order to finance their research. They often spend part of their career in the private sector 9

and have a greater propensity than researchers in other countries to set up their own business. This pattern is very different from that in countries where researchers are under less pressure, more privileged, less motivated to publish, and less mobile. They also have fewer opportunities to diversify their research commitments and careers. Figure 2. Higher education researchers (or university graduates) as a percentage of national total 50

50

Switzerland

Italy

45

Canada

Sweden

Finland

45

Netherlands

40 40

35 30

Germany

35 France

25

United States

30 Japan

20 United Kingdom

25

15 10

Denmark Norway

Iceland

20

1985 86 87 88 89 90 91 92 93 94 95 961997

1985 86 87 88 89 90 91 92 93 94 95 961997 80

70 Spain Greece

65 60

70

Portugal

60

55

Turkey Austria

Mexico

New Zealand

50

Poland

50 Australia

40

45 Belgium

35

Korea

30

40 Ireland

30

20

Hungary

10

1985 86 87 88 89 90 91 92 93 94 95 961997

1985 86 87 88 89 90 91 92 93 94 95 961997

Source: OECD, S&T databases, March 1999.

Many countries are now concerned to revitalise their university and public sector research systems and have introduced reforms to this end: reduction of tenured posts, more competitive resource allocation, aids to researcher mobility, and so forth. These reforms are encountering resistance from the unions and research bodies concerned.

10

Science funding: stable but increasingly subject to conditions Interestingly, in view of the increasing budget restrictions that have marked the 1990s, the governments of most Member countries have tried not to reduce their support for scientific research. Various indicators confirm this: public funding of university research, general funds to universities, and government R&D spending on non-targeted research have remained stable or even risen against other public R&D expenditures in relative terms, with the notable exceptions of health and the environment (Figure 3). 1

Figure 3. Trends in the objectives of governments’ R&D budgets, 1989-96 Number of countries reporting increase or decline Environment General University Funds Social Health Research Infrastructure Industry Space Agriculture Earth and Atmosphere Down

Defence

Up

Energy 15 -15

10 -10

5 -5

0

5

10

15

20

1. Government budget appropriations or outlays for R&D (GBAORD) at 1990 GDP prices. Source: OECD, S&T databases, January 1998.

This indicates that the public authorities attach a good deal of importance to basic research and consider it a public good justifying public support. The continuing support for university research contrasts with the gradual withdrawal of support to government laboratories engaged in research which is generally of a more applied nature. Nevertheless, the relative share of public sector financing of university research has diminished substantially in all countries (Annex Table B1 compares funding levels between the mid-1980s and mid-1990s), while that of the business sector has risen slightly. In fact, the main increases in relative value have come from the increase of universities’ own funds (student fees) and contributions from non-profit institutions.

11

Some important changes have taken place in the way in which government support is allocated (OECD, 1998b; 1998c): ♦ In most countries, core (institutional) funding allocated to universities has declined relative to project funding (Figure 4 and Annex Table B2): allocation criteria for these funds have been severely tightened in the case of funds allocated by research councils for basic research or made conditional upon supplementary financing by industry for research of a more technological nature. ♦ In most countries, there has been a steep increase in the numbers of students enrolled in university, but in many, research funds, especially the share from institutional financing, have not increased proportionately (Annex Table B3). Thus, teachers are having to restrict their research activities in order to meet their teaching commitments. ♦ University systems have diversified, particularly with the development of technology institutes, which have started to undertake research activities. This diversification is useful, but it implies more competition and greater selectivity if an inefficient dispersion of funds is to be avoided. Too much selectivity, on the other hand, may be detrimental, in that funds generally go for the most part to the well-established universities. ♦ In general, research areas and costs have been growing at a faster rate than the means at the disposal of the public authorities, who therefore face difficult choices: should they concentrate their resources on priority areas on the basis of known strengths and risk excessive specialisation or should they maintain a balance among disciplines and risk the danger of insufficient critical mass? ♦ The creation of centres of excellence has become widespread. For governments, they represent a way to give concrete form to their priorities, encourage interdisciplinary approaches and capture industry interest, while ensuring the involvement of competent research groups. In some cases, the centres are virtual in nature, drawing on the possibilities offered by information and communication technology (ICT). As a rule, support is given for a limited length of time (a few years) and renewed only after evaluation. At the same time, there is evidence of slow growth in funds from the private sector, in part as a response to government initiatives. These remain modest, however, not exceeding 5% of university research funding except in a minority of Member countries (notably Canada, Finland, Germany, Switzerland, the 12

United Kingdom, and the United States). In no country does in-house basic research in industry account for more than 10% of industrial research. Also on the increase is funding by globalised firms seeking poles of competence worldwide that meet their needs, while retaining the bulk of their own research structures in their home country.

13

1

Figure 4. Changes in the pattern of government financing of science research, 1989-95 Number of countries reporting increase or decline

HE direct

Government basic research

HE GUF

-10 10

-55

Decline

0

5

Little growth

10

Growth

15

20

Major growth

1. At 1990 GDP prices. Source: OECD, S&T databases, November 1997.

All these developments are causing increasing concern in some circles, which fear that the basic foundations of science in our societies will ultimately be undermined by the “short-termism” that seems to characterise combined support from the public and private sectors. Further in-depth studies will be necessary to determine whether these fears are justified. Supranational funding has become more and more widespread, particularly in the European Union, owing to the strengthening of the Framework Programmes and the development of structural funds allocated to R&D infrastructures in the less developed countries. International funding can represent 10% or more of the funds allocated to research. The balancing of national and international funding has become ever more complex; nonetheless, there is agreement about programmes that justify joint effort: very large facilities (CERN, for example), large public infrastructure projects (transport, health), global issues (climate change). At the same time, sub-national levels (local and regional authorities) are becoming more involved in the financing of infrastructure and projects. Some countries have a long tradition in this area (for example, Germany’s Länder cover basic university funding); in other countries, regional efforts are the result of laws in favour of decentralisation (France, for example). Everywhere,

14

however, institutions of higher education and associated research activities are considered essential for making local economies more dynamic. Setting up regulatory structures and frameworks: partnerships and entrepreneurship In response to the shifts in public funding policy, universities in several countries have had to become more like businesses, if not to make a profit at least to manage and develop their competencies so as to respond to “paying” demand from firms, local authorities, international programmes, etc. The ability to exploit these contacts depends, of course, on their location and the disciplines in which they have expertise. Those specialising in technology and management can easily form profitable partnerships with industry and the private sector; the same applies to universities working with hospitals in the field of medical and clinical research. In fact, outside funding may finance most of the research effort of such universities. This is not the case for more fundamental research. Nonetheless, the question of a change in universities’ status to facilitate their integration into the economy is increasingly pressing and calls for consideration of the steps to be taken to encourage the necessary changes. Privatisation of public laboratories has also been on the agenda in a number of countries, even for fundamental research far removed from immediate applications (as in astronomy). Evaluations are needed to gauge the effects of these initiatives, particularly on the nature of the research undertaken. In any event, to stimulate their contribution to innovation, universities and public laboratories have been encouraged to develop their relations with industry through various types of partnerships and structures. The setting up of enterprise incubators on campuses, the introduction of offices for technology transfer and licensing, and so on, has caused changes in the internal culture of many universities. It should be noted, however, that the success of these operations depends largely on a dynamic atmosphere and the possibility of attaining critical mass. The most active technology licence offices at universities tend to be highly concentrated – in the United States, seven universities account for about 50% of royalties from licensing agreements – and earnings are far less than the resources gained from research contracts with industry (20% of industry contracts in the case of the United States). There is also a tendency to give universities priority over the teachers/researchers they employ for the rights to intellectual property. Some fear that these trends will inhibit individual creativity. More detailed studies are needed to determine their possibly harmful effects.

15

Among the most powerful sources of innovation is researcher mobility to industry, which can be particularly rewarding if it leads to enterprise creation. However, the phenomenon is only significant in a few, principally Englishspeaking, countries. This is partly due to the presence of factors unrelated to the research environment, such as buoyant venture capital markets, and partly to the fact that researchers in those countries have strong incentives and encounter few obstacles. Adjustments need to be made in other countries as regards conditions of reemployment, transfer of pension rights, return on capital investment in nascent enterprises, etc. (OECD, 1998a). The science workforce: adjusting to change Differing disciplinary trends are causing major changes in demand for scientific personnel. There is currently less recruitment in physics, chemistry and other traditional disciplines than in biology, computer science, etc. Many countries face adjustment problems, as there is a dearth of skilled personnel in some areas and a surplus in others. These mismatches are more pronounced in countries where education systems are inflexible. In several Member countries, moreover, unemployment among scientists is on the increase, as they are also affected by the slowdown in economic activity and by structural change – to a lesser extent than the rest of the population, no doubt, but nonetheless significantly. The problem is particularly marked in countries where young academics fail to find employment or are obliged to settle for temporary post-doctoral positions. The ageing of the research workforce is another reason for concern in a number of countries, particularly in continental Europe. The researchers of the babyboom generation, recruited in large numbers during the 1960s, are now approaching retirement age, and in some countries more than one-third of scientific personnel will retire over the next decade. This will create a void impossible to fill with the numbers now being trained. The situation is aggravated by the fact that, in some countries, young people show a lack of interest in science (OECD, 1997b). In-depth analysis is needed to evaluate the scale of the problems. It should take into account recent geopolitical, economic and technological changes that are indirectly but dramatically transforming the world’s scientific labour market. The opening up of the eastern European countries, Russia and China has greatly increased the supply of high-level researchers, with varying degrees of negative effects on recruitment of home-country researchers in firms and universities in several countries. The globalisation of industry has also had an effect, as firms now “shop around” for research to obtain best value for money. Finally, the development of the Internet, by facilitating communication among researchers,

16

is likely to affect their mobility, without necessarily reducing it. It is necessary, therefore, to grasp the implications of the various trends at work. Evaluation of institutions and researchers Governments, concerned to make public spending as cost-effective as possible and to optimise resource allocation, have been placing great emphasis on ex ante but, above all, ex post evaluation. Evaluation exercises have increased in number in the agencies and ministries that finance research as well as in university or public research establishments, on the basis of government guidelines. Here, again, the influence of national cultures and traditions is clear (OECD, 1997c). In some countries, evaluations, carried out according to very strict rules, are closely linked to resource allocation. In other countries, evaluation is more often seen as a way to facilitate and introduce reforms (sometimes very directly when it has been necessary, as for the former East Germany, to adapt structures drastically). In still other countries, evaluation focuses on institutions, more to upgrade management efficiency than to stimulate reform. Finally, certain countries, such as the Nordic countries, have a long-standing tradition of evaluation and make extensive use of internationalised procedures to ensure that their research meets international standards of excellence. An examination of evaluation practices nevertheless reveals some general tendencies (OECD, 1997c): designing evaluations as exercises for improving understanding and behaviour by closely associating all the actors concerned; going beyond purely quantitative assessments such as bibliometric performances (OECD, 1997d) to introduce qualitative judgements (notably through peer review); introducing criteria that take specifically into account not only the production of research results but also individual efforts to demonstrate their value and transfer them. The general view is that much remains to be done in most countries to make these concerns part of the evaluation process. In fact, it would be useful to develop procedures and criteria for evaluating individuals with respect to the type of research activities they engage in. Those who undertake theoretical and speculative research should be evaluated differently from those who are in direct contact with industry and involved in applications. “Entrepreneurial” researchers, those team leaders who seek out financing of various sorts, cannot be evaluated on the basis of the criteria used for researchers working in more traditional structures who rely essentially on institutional financing.

17

Science and information technologies: meeting the new challenges Initially invented for research and defence purposes, the Internet spread rapidly to scientific circles. Surveys conducted in the mid-1990s showed that more than half of US researchers used the Internet to communicate with one another. Internet use has now become widespread in the developed countries and is increasingly reaching the less developed parts of the world. The “Global Research Village” offers both challenges and promise (OECD, 1998d; 1999a). The ways in which research practices are being affected by the development of telecommunications, and more especially by the Internet, remain unclear. Communication has certainly intensified, facilitating the development of collaborative work in particular, but it is not clear whether overall creativity is increasing. “Peripheral” research teams certainly also now have easier access, but it is unclear whether this is significantly altering the conditions of competition between teams and their ranking. On the other hand, the implications for the publication of results are considerable. The raw data on which studies are based can be published with the relevant articles. Peer-review procedures are being modified, if not adversely affected, by the fact that work can be made known rapidly to large numbers of people at a preliminary stage. “Soft-copy” journals can and are being developed. The attribution of intellectual property rights is likely to be profoundly affected, and new rules and procedures need to be devised. Little real progress has been made on this front in recent years, despite early recognition of the problems by researchers and publishers alike. The development of information and telecommunication facilities holds immense potential for the renewal of the research infrastructure: very largescale international databases, digital libraries with almost infinite storage capacities, virtual laboratories that can draw together research teams from all over the world, etc. Countries are exploiting these opportunities with varying degrees of enthusiasm and boldness. It is important to examine the implications for science systems of the different initiatives being taken here and there, at more or less experimental stages. Very important developments are also taking place in the telecommunications infrastructure, which, thanks to advances in digitalisation, make possible massive increases in data transmission and storage capacities, while lowering costs. In this regard, there are significant gaps in planning for equipment between North America and the rest of the world, including Europe and Asia (Table 1). This is a matter of concern in some circles and may in fact create a regional imbalance that could persist and cause problems.

18

19

Table 1. Infrastructure and information technology networks* Europe

United States

’90 IXI (64 Kbps)

’69 ARPANET (DoD)

’92 Europa NET (64 Kbps)

’86 NSFNet (56 Kbps) ’88 NSFNet (1.54 Mbps)

’97 TEN-34 (34 Mbps)

Japan/Asia ’97 NACSIS (150 Mbps)

98 2003 APAN/IMNET (1 Gbps)

’91 NSFNet (44 Mbps) ‘98/99 TEN-155 (155 Mbps)

’95 vBNS (155 Mbps) ’99 Abilene (2.4 Gbps)

*Flows in Kilobits per second (Kbps), Megabits per second (Mbps) et Gigabits per second (Gbps). Source: OECD.

Problems specific to the social sciences The social sciences represent an important segment of science systems, with between 10% and 40% of R&D personnel, depending on the country (Figure 5). They also attract a considerable share of students, a share which exceeds significantly their share of national R&D expenditures. However, the social sciences and the humanities are faced with a number of problems (OECD, 1999b). Compared with the natural sciences, the social sciences and humanities suffer from a problem of image and status. Social science researchers are partly responsible: occasional laxity of approach, insufficient quality control or infighting between “schools of thought” do not help to improve the image of the social sciences. But this does not explain everything. The social sciences are the “victim” of something more fundamental: the difficulty of self-analysis inherent in any society. Self-analysis naturally encounters great resistance; furthermore, it is inevitably biased by ideology and subjectivity. Institutional rigidities are particularly acute in the social sciences and take different forms: the questionable but persistent stratification of disciplines, difficulty of full-fledged integration, proliferation of hybrid disciplines that serve as niches which isolate and protect the various actors. Governments take various approaches to remedying this state of affairs: they develop interdisciplinary programmes or centres, change criteria for awarding chairs and

20

professorships, etc. However, in most countries, these measures have as yet had little impact on structures. Collaboration between the social sciences and natural sciences is also being encouraged where the value is clear, as in matters relating to the environment, urban life or education. Here as well, much remains to be done. Figure 5. Share of social sciences and humanities in domestic expenditure on R&D and R&D personnel Percentages 50 45 GERD

40

Personnel

35 30 25 20 15 10

Sweden 1995

Portugal 1995

Poland 1996

Norway 1995

Mexico 1995

Japan 1995

Iceland 1992

Ireland 1994

Hungary 1995

Spain 1995

Denmark 1993

Germany 1993

Czech Rep.1996

Canada 1993

Austria 1993

0

Australia1994

5

Source: OECD, based on the S&T databases.

The conditions of social science research could well be transformed by advances in information technology that make it possible to collect, process, store and disseminate huge quantities of data (OECD, 1999a; 1999b). It is becoming possible, given the necessary technical and legal measures, to interconnect countries’ existing databases in many fields and undertake large-scale integrated and comparative analyses. It is becoming possible to launch vast surveys on all kinds of subjects via the Internet and to track, in real time, developments in the perceptions and behaviour of very diverse populations. It is becoming possible to set up networks of virtual laboratories that bring together teams of social science researchers from all over the world. Besides their impact on the conditions for observing and analysing human and

21

societal realities, these developments should also be useful in encouraging the breakdown of barriers between disciplines and the diffusion of information to decision makers. This could be the focus of extensive national and international programmes. Preventing the widening of the gap between science and the general public If science policy is to be conducted effectively, it needs the backing of public opinion for raising funds, choosing priorities, etc. Moreover, for a satisfactory renewal and enlargement of the population of researchers, science needs to have an attractive image throughout society as a whole. Yet, as surveys show (see the example of the United States, Figure 6), public understanding and perception of science have not improved over the last decade, despite extensive media developments (television especially) and efforts made in science museography and other areas. Significant progress could be made by drawing on display and outreach techniques that have proved successful the world over (OECD, 1997b; 1997e). The development of international special-interest TV channels devoted to science and technology, combining the best of individual countries’ achievements, could also be envisaged. Stimulating the interest of young people in scientific studies calls for measures at the levels of primary and secondary education (OECD, 1996). The experience of the few countries which have tackled this issue successfully indicates that the way in which scientific subjects, including mathematics, are taught needs to be improved. Courses need to be grounded in everyday life, and selection and grading methods should not unduly or prematurely discourage young people. This can produce significant positive results, even in very underprivileged environments (OECD, 1997b). Finally, society’s perception of science in the years ahead will depend crucially on how the ethical issues posed by the evolution and uses of science are addressed. These will particularly affect the progress of biotechnology and genetics (human cloning, gene therapy, etc.), which raise legitimate concerns (see the comparison with other technologies, Table 2). These are areas in which countries diverge greatly because of their cultural traditions. Given the ethical problems raised as well as inequalities in the conditions under which research can be performed, international agreements on certain limits seem desirable.

22

Figure 6. Public understanding of various questions in the United States Percentages considering themselves very well informed

50

E n v iro n m e n ta l q u e s tio n s

40

E n e rg y

E c o n o m ic p o lic y

(in c lu d in g n u c le a r)

M e d ic a l d is c o v e rie s

30

F o re ig n p o lic y

20

10

S c ie n tific d is c o v e rie s

N e w te c h n o lo g ie s 0 1979

80

81

82

83

84

85

86

87

S p a c e e x p lo ra tio n 88

89

90

91

92

93

94

1995

Source: National Science Foundation, Science and Engineering Indicators – 1996.

Furthermore, scientists are increasingly required to speak out on sensitive and politically delicate issues in many areas (mad cow disease, climate change, contaminated blood, nuclear waste). Their opinions should be sought in a timely way by governments, so that informed decisions may be taken in the best interests of democracy. Table 2. Perception of the effects of new technologies on the quality of life Responses from 15 EU countries plus Norway and Switzerland Solar

Informatics

Biotechnologies

Telecommunications

New materials

Space

Improve

74

76

44

80

65

49

No effect

14

8

9

10

12

28

Worsen

4

9

22

4

6

8

Don’t know

8

7

25

6

17

15

Source: D. Boy (1999), “Les biotechnologies et l’opinion publique européenne”, Futuribles, No. 236, Paris.

23

Conclusion: some policy issues Science systems are undergoing changes which, although gradual, are nonetheless profound. The science systems that are emerging at the eve of the 21st century are quite different from those of even a decade ago. The latter still fit quite comfortably into the frameworks set in place after World War II. They were then just beginning to be affected by transformations resulting from the new geopolitical situation, by the reduction of government budgets, by the globalisation of industry, and by the new forms of interaction between science and innovation. They were not yet faced with the problems of ill-adapted structures and the need to renew their personnel. Neither the possibilities offered by ICTs nor the ethical questions raised by advances in genetics had yet been recognised. The preceding observations suggest a number of general orientations for government policy as it accompanies these changes (Annex Table C1 provides some example of best practices drawn from the experience of Member countries). ♦ Governments should integrate science policy into appropriate coordination mechanisms at the highest executive level, so that it is a part of an overall understanding of and strategy for developing the country’s innovation capacity. ♦ Adequate support must be found for scientific research, especially for university research which largely depends on public funding. Resources should be sufficient to ensure the continuation of long-term research efforts and related training activities. Institutional funding has decreased in many countries, while the expansion of university education has led to increased teaching commitments. A good balance has to be maintained between core and contract-based resources to ensure productive synergy between the scientific community and its environment, and between mission-oriented and curiosity-driven research. ♦ Research structures and regulatory frameworks require constant adjustment to facilitate adaptation and stimulate new efforts. As financing conditions change, universities need to be able to act flexibly in order to sound out “research markets” and sell their skills. While the development of “university entrepreneurship” is desirable in many respects, it must not be allowed to undermine the foundations of long-term research. It is also necessary to remove obstacles to researcher mobility and, in particular, to reduce the deterrents to entrepreneurship by researchers (e.g. pension rights, conditions of secondment and reinstatement, stake in equity). 24

Evaluation of researchers, programmes and institutions should be strengthened, preserving the current criteria of scientific excellence, but also taking technological relevance into account. ♦ Adjustments are needed in scientific and technical education in order to maintain an adequate balance of demand for and supply of qualified personnel and to remedy the mismatches in certain disciplines in many countries. Some countries should also react vigorously to looming shortages in researchers as the large numbers recruited two to three decades ago near retirement. It is important to act not only at the level of higher education but also at the level of primary and secondary schools so as to encourage young people to opt for scientific studies and careers. This implies that the conditions of science teaching should be re-examined. ♦ The revolutionary changes in information technologies seem to be strongly influencing research conditions, opening up huge possibilities for communication among researchers around the globe, altering conditions for publishing research results, and making possible the development of new tools (digital libraries, virtual laboratories). It is important to adjust regulatory frameworks and ground rules so that these new opportunities may benefit the greatest possible number. It is also necessary to take care that the gaps between different regions of the OECD area in the development of new technology facilities do not widen. ♦ Science/society interfaces need to be strengthened. It is important to improve scientific and technical culture and, in order to do so, to make better use of advances in the media and museography. The ethical issues arising from recent advances in science, notably in medicine and biology, require timely responses and mobilisation of all the actors in appropriate frameworks. ♦ Greater use should be made of the social sciences to tackle various societal problems. They need to break down barriers and become better integrated in both research and training. Use should be made of the advances in information technology that will transform the ways in which databases will be developed and exploited, thereby significantly changing social science research. ♦ Governments must be prepared to deal with the globalisation processes that increasingly affect the scientific enterprise. The effects of multinational corporate strategies on national scientific potential need to be precisely tracked. The process of integration in the major OECD

25

regions (Europe, North America, Asia-Pacific) is also leading to readjustments of research funding conditions and linkages between the national and supranational levels. Lastly, new forms of international co-operation need to be deployed boldly and imaginatively to meet the challenges presented by world-scale problems, such as climate change.

26

REFERENCES

Albert, Michael B. (1998), The New Innovators : Global Patenting Trends in Five Sectors, US Department of Commerce, Office of Technology Policy, Washington, DC. Gibbons, M. (1994), The New Production of Knowledge, Sage, London. Martin, B. et B. Salter (1996), The Relationship between Publicly-Funded Research and Economic Performance : A SPRU Review, Science Policy Research Unit, Brighton. Narin, F., Hamilton K.S. and D. Olivastro (1997), “The Increasing Linkage between US Technology and Public Science”, Research Policy 26, pp. 317-330. OECD (1994), Proposed Standard Practice for Surveys of Research and Experimental Development: Frascati Manual 1993, OECD, Paris. OECD (1996), Changing the Subject: Innovation in Science, Mathematics, and Technological Education, OECD, Paris/Routledge, London. OECD (1997a), Oslo Manual. Proposed Guidelines for Collecting and Interpreting Technological Innovation Data, 2nd edn., OECD/Eurostat, Paris. OECD (1997b), Science and Technology in the Public Eye, OECD, Paris. OECD (1997c), “The Evaluation of Scientific Research: Selected Experiences”, OCDE/GD(97)194, Paris. OECD (1997d), “Bibliometric Indicators and Analysis of Research Systems: Methods and Examples”, STI Working Papers 1997/1, OECD, Paris. OECD (1997e), Promoting Public Understanding of Science and Technology, OECD, Paris. OECD(1998a), “Managing the Science Base”, in Technology, Productivity and Job Creation – Best Policy Practices, Chap. 6, OECD, Paris.

27

OECD (1998b), University Research in Transition, OECD, Paris. OECD (1998c), “University Research in Transition: Country Notes”, DSTI/STP/SUR(98)5/FINAL, OECD, Paris. OECD (1998d), The Global Research Village: How Information and Communication Technologies Affect the Science System, OECD, Paris. OECD (1999a, forthcoming), “The Global Research Village”, STI Review, No.24, OECD, Paris. OECD (1999b), The Social Sciences at a Turning Point?, OECD, Paris.

28

ANNEX

Bibliometric data Table A1. Scientific effort and performance in OECD countries, 1995 Table A2. Patterns of international collaboration in science and engineering research, 1991-95 Tableau A3. Specialisation patterns in science by selected scientific fields, 1991-95 Financing of university research Table B1. Share of government-financed domestic R&D expenditure performed in the higher education sector Table B2. Share of direct and institutional financing in university research Table B3. HERD per student (full-time, ISIC levels 6 and 7) Management of the science base Table C1. Management of the science base

29

Table A1. Scientific effort and performance in OECD countries, 1995¹ GDP per inhabitant as a % of the OECD average, 1996

Gross domestic expenditure on R&D as a % of GDP

Researchers per 10 000 workers

Scientific and technical articles per unit of GDP2

United States

138

2.6

74

20

Norway

129

1.7

73

21

Switzerland

123

2.7

46

37

Iceland

117

1.5

72

15

Japan

116

2.8

83

23

Denmark

112

1.8

57

31

Belgium

110

1.6

53

25

Canada

109

1.7

53

20

Austria

109

1.5

34

18

France

106

2.3

60

24

Australia

105

1.6

64

21

Germany

105

2.3

58

31

Netherlands

103

2.0

46

20

Italy

100

1.1

33

13

United Kingdom

98

2.1

52

41

Sweden

97

3.6

68

29

Ireland

96

1.4

59

35

Finland

94

2.3

61

16

New Zealand

87

1.0

35

29

Spain

77

0.9

30

16

Korea

69

2.7

48

5

Portugal

65

0.6

24

7

Greece

63

0.5

20

16

Czech Republic

46

1.2

23

15

Mexico

39

0.3

6

20

Poland

34

0.7

29

2

Hungary

34

0.8

26

17

Turkey

31

0.4

7

4

1. Or latest available year. 2. Number of articles per billion USD – see National Science Foundation, Science and Engineering Indicators – 1998. Source: OECD calculations on the basis of the MSTI database; CHI Research, National Science Foundation, 1998; and OECD, Science, Technology and Industry Outlook, 1998.

31

Table A2. Patterns of international collaboration in science and engineering research, 1991-95 Number of scientific articles and shares in percentage

Source Country/region

Total (thousands)

Share of multiauthored (%)

International co-authored (%)

Share of all articles

Share of total international coauthored

United States

773.7

56

16

31.7

21.3

United Kingdom

186.2

52

26

7.6

8.1

1.06

Germany

174.6

50

30

7.2

8.9

1.24

France

132.2

61

32

5.4

7.1

1.31

Italy

76.9

70

33

3.2

4.3

1.35

Other southern Europe

74.0

56

32

3.0

4.0

1.33

Nordic countries

96.3

65

36

3.9

5.8

1.47

Other western Europe

136.2

61

38

5.6

8.8

1.57

Japan

200.6

50

13

8.2

4.3

0.52

Canada

103.9

57

28

4.3

5.0

1.17

Former USSR

134.6

29

17

5.5

3.9

0.70

Degree of Internationalisation¹ 0.67

Eastern Europe

58.1

56

41

2.4

4.0

1.68

Israel

25.6

66

37

1.1

1.6

1.51

Near East/North Africa

17.6

55

37

0.7

1.1

1.52

Other Africa

21.3

60

39

0.9

1.4

1.60

Australia, New Zealand

62.6

51

25

2.6

2.6

1.03

India

43.8

33

13

1.8

1.0

0.55

Central America

9.6

65

46

0.4

0.7

1.89

South America

32.7

62

40

China

30.8

52

29

1.3

1.5

1.18

Asia NEIs²

37.5

53

24

1.5

1.5

1.00

9.2

70

57

0.4

0.9

2.32

2 438.0

54

24

100.0

100.0

1.00

Other AsiaPacific Total

2.2

1. Share of internationally co-authored articles divided by the country’s share of total articles. 2. NIE = newly industrialised economies. Source: National Science Foundation, Science and Engineering Indicators – 1998, OECD calculations.

32

Table A3. Specialisation patterns in science by selected scientific fields, 1991-95 1981 Biology Biomedical research Chemistry Clinical medicine Earth and space sciences Engineering and technology Mathematics Physics 1995 Biology Biomedical research Chemistry Clinical medicine Earth and space sciences Engineering and technology Mathematics Physics Differences between periods Biology Biomedical research Chemistry Clinical medicine Earth and space sciences Engineering and technology Mathematics Physics Correlation between years

USA 104 108 65 105 121 111 106 89 USA 97 113 73 106 121 108 109 83

JAP 89 89 187 68 35 133 63 135 JAP 82 89 137 87 45 115 37 140

GER 85 73 141 92 71 106 156 122 GER 76 76 148 87 72 92 101 143

FRA 68 102 137 95 92 65 111 130 FRA 67 129 116 80 88 80 182 118

ITA 54 93 149 101 99 65 63 133 ITA 59 83 111 106 86 89 112 128

UK 106 100 93 107 104 99 73 85 UK 101 100 88 115 104 96 77 79

CAN 158 102 92 80 136 106 129 84 CAN 172 98 84 88 158 125 114 70

USA -6.1 4.6 7.8 0.9 -0.1 -3.0 3.4 -6.6 USA 0.954

JAP -6.7 -0.6 -50.3 19.0 10.5 -17.3 -26.0 4.7 JAP 0.898

GER -9.3 3.3 6.7 -4.8 1.0 -14.7 -55.4 20.0 GER 0.734

FRA -1.2 27.5 -20.7 -15.1 -4.3 15.3 71.2 -11.8 FRA 0.594

ITA 5.0 -10.2 -30.8 4.9 -12.5 24.3 49.5 -4.9 ITA 0.637

UK -5.4 0.0 -5.4 8.2 0.0 -3.7 4.8 -6.2 UK 0.913

CAN 14.3 -3.9 -7.5 7.5 21.9 18.3 -15.5 -13.9 CAN 0.929

33

Table A3. Specialisation patterns in science by selected scientific fields, 1991-95 (continued) 1981

AUS

AUT

DEN

FIN

NET

NOR

SWE

SWI

Biology Biomedical research Chemistry Clinical medicine Earth and space sciences Engineering and technology Mathematics Physics 1995 Biology Biomedical research Chemistry Clinical medicine Earth and space sciences Engineering and technology Mathematics Physics Difference between periods Biology Biomedical research Chemistry Clinical medicine Earth and space sciences Engineering and technology Mathematics Physics Correlation between years

206

58

75

61

95

121

59

46

83 90 91

54 106 142

96 55 153

93 61 154

121 110 92

101 72 138

115 68 153

92 108 116

144

50

61

59

80

98

45

65

67 87 69 AUS 231

73 131 100 AUT 65

35 82 85 DEN 127

63 93 70 FIN 111

73 84 121 NET 119

43 61 47 NOR 158

61 49 56 SWE 99

88 53 139 SWI 65

83 74 102

70 100 135

98 65 128

79 67 144

100 77 120

78 73 129

97 72 133

101 104 106

122

66

106

87

92

160

74

71

90 79 66

66 101 102

46 90 77

78 51 68

70 77 82

69 86 46

68 66 82

60 74 133

AUS

AUT

DEN

FIN

NET

NOR

SWE

SWI

25.6

6.3

51.6

49.8

24.5

37.3

40.3

19.4

-0.8 -16.4 10.5

15.8 -5.9 -6.2

2.4 9.8 -25.2

-13.9 5.5 -9.8

-20.4 -33.6 28.0

-22.9 1.6 -8.8

-17.8 4.4 -20.0

8.5 -4.2 -10.7

-21.7

16.2

45.3

28.0

12.4

62.1

28.5

6.1

23.6 -8.1 -2.4

-6.9 -30.7 2.2

11.0 7.8 -7.9

14.8 -41.8 -1.4

-3.1 -7.3 -38.9

25.7 25.5 -0.8

6.2 16.4 25.9

-28.1 21.1 -5.8

AUS 0.948

AUT 0.929

DEN 0.700

FIN 0.598

NET 0.123

NOR 0.768

SWE 0.863

SWI 0.866

Note: For a given country (region) and field, this indicator is defined as the share of publications in that scientific field in relation to the total number of publications by that country (region), divided by the share of that field in total world publications X 100. Values greater than 100 indicate relative specialisation. Source: National Science Foundation, Science & Engineering Indicators – 1998, OECD calculations.

34

Table B1. Share of government-financed domestic R&D expenditure performed in the higher education sector 1986 State 93.7 91.6 97.61 86.1 79.4

1991 Firms 6.3 2.1 1.7 8.6 4.1

State 93.0 91.55 97.44 76.2 71.7

1996

Firms Germany 8.0 Australia 5.2 Austria 2.07 Belgium 10.68 Canada 11.8 Korea 50.5 Denmark 92.7 1.1 89.6 1.6 1.9 Spain 98.3 1.5 86.8 10.0 7.5 United States 87.1 4.2 74.1 5.3 5.7 Finland 89.62 3.8 91.2 3.6 5.78 France 96.2 2.0 93.1 4.2 3.2 Greece 97.5 4.8 80.0 6.1 5.68 Hungary 83.5 14.4 2.9 Ireland 75.7 6.8 66.0 8.6 6.98 1 Iceland 75.3 0.6 90.8 4.9 4.3 Italy 98.4 1.1 94.4 4.0 4.7 Japan 52.7 1.7 49.5 2.4 2.4 Luxembourg Mexico 76.77 3.4 78.48 1.48 2 8 Norway 91.0 4.5 90.4 4.7 89.4 5.38 N. Zealand 63.8 4.6 54.68 9.48 Netherlands 95.6 1.2 96.3 1.2 86.5 3.8 Poland 80.9 11.3 Portugal 95.9 0.9 94.65 0.7 87.78 0.88 Czech Rep. 65.7 0.4 United Kingdom 80.3 5.7 72.0 7.8 66.5 6.7 Sweden 88.12 5.9 84.3 5.2 83.68 4.68 Switzerland 96.7 3.3 91.66 1.8 88.5 6.2 Turkey 87.8 10.4 74.7 18.0 Note: The sum of the percentages for each year and each country generally does not attain 100%. The differences corresponds to funds from non-profit instutitions and universities’ own funds. 1. 1985 2. 1987 3. 1988 4. 1989 5. 1990 6. 1992 7. 1993 8. 1995

Source: OECD, based on the S&T databases.

35

Firms 7.0 2.2 1.8 15.4 9.1

State 91.0 90.3 97.27 72.98 65.8 44.0 87.7 73.6 73.6 89.08 90.0 72.48 85.0 62.08 89.1 93.3 49.1

Table B2. Share of direct and institutional financing in university research Percentages 1986

1991

1996

Australia Austria

20.7 9.4

11.3

Belgium Canada

51.1 52.4

35.0 59.6

52.1 61.0

4

27.6 14.9

6 7

Czech Republic Denmark

15.4

23.8

100.0 29.0

Finland France

39.8 48.6

3

30.6 50.6

36.1 50.3

7

Germany Greece

23.6 6.5

3

23.2 8.2

22.3 18.3

7

Hungary Iceland

39.8

2

94.3

89.9

Ireland Italy

13.9

37.3

32.3

Japan Korea

23.2

19.2

14.7

Luxembourg Mexico

16.5

Netherlands New Zealand

7.8

Norway Poland

21.0

6

36.5

4.9 23.6

7.7 35.7

3

22.1

21.8 100.0

31.3

17.5

3

35.4 18.8

32.0 17.2

Portugal Spain

26.4

Sweden Switzerland

38.0 21.8

Turkey United Kingdom

30.4

31.4 34.7

63.6 44.6

76.2

100.0

100.0

1

United States

5

1. From 1990, general university funds, which were included in the sub-total of government financing of higher education, are included in own funds of the higher education sector. 2. 1985; 3. 1987; 4. 1989; 5. 1992; 6 1993; 7. 1995. Source: OECD, based on the S&T databases.

36

7

7

7

7 7

7

Table B3. HERD per student (full-time, ISCED levels 6 and 7) Thousands of constant USD (1990 prices and PPP) 1986 4.65 2.88 3.26 2.83

1991 3.02 2.73 4.75 3.13 0.20 2.40 2.63 2.74 3.53

6

1995 3.85 3.40 4.60 3.00 0.69 2.89 2.22 2.41 3.07 0.85 1.37 3.16 2.24 1.47 6.02 0.83

Australia 2 4 7 Austria Belgium Canada Czech Republic Denmark 2.63 Finland 2.34 8 France 2.73 Germany 2.79 7 Greece Hungary 4.12 Iceland Ireland 1.61 2.01 5 Italy 1.59 1.86 Japan 5.46 5.76 Korea Luxembourg 7 Mexico 0.51 0.56 Netherlands 6.18 8.45 3.94 2 2.32 2.09 New Zealand 2.20 3 Norway 5.79 4.15 3.68 8 Poland 0.61 5 Portugal 0.84 1.14 Spain 0.47 0.75 0.89 3 11.63 6.57 Sweden 12.96 6 8 Switzerland 6.52 11.04 10.61 Turkey 1.47 0.92 United Kingdom 5.39 4.90 3.82 United States 3.60 3.69 3.97 1. HERD was deflated using implicit GDP deflators (rate of price increase/decrease) and converted into USD using 1990 PPPs. 2. 1985. 3. 1987. 4. 1989. 5. 1990. 6. 1992. 7. 1993. 8. 1994. ISCED: International standard classification of education. PPP: purchasing power parities. Level 6: Programmes leading to a first university degree or equivalent. Level 7: Programmes leading to a postgraduate university degree or equivalent. Source: OECD, based on the S&T databases.

37

Table C1. Management of the science base Policy areas

General policy principles

Cases of best policy practices

Science policy and government structures

Incorporate science policy into central government decision making and overall economic development strategy by appropriate mechanisms.

Structure of the R&D effort (performing organisations)

Establish and maintain an appropriate structure in the R&D effort, with an adequate balance between industry, government and university.

Finland with the Science and Technology Policy Council, Japan with the S&T Policy Council and the long-term plans, Canada with the coordinating role played by Industry Canada (Federal S&T and Industry Ministry). Germany, the Netherlands, Switzerland, the United Kingdom, the United States.

General organisation

Funding of the science base Overall funding

Maintain or increase overall government support to university and public research with a longterm view.

Denmark, Finland, Iceland, Japan.

Funding of university research

Maintain and establish an adequate ratio between sure and precarious resources for university research at the overall level (around 70/30%); at the institution level, maintain a minimum percentage of 50/50 between core and contract-based funding on average.

Policies pursued at the national level by the Netherlands and Finland; at the institution level, see examples provided by wellperforming universities in a number of countries, including Germany, Switzerland, the United Kingdom, the United States. Countries maintaining a strong network of government laboratories performing strategic research of industrial interest include Finland, Japan, Korea, Norway; core funding provided by government can exceed 50% of laboratories’ budget.

Funding of government Maintain a minimal level of laboratories government research of collective interest and establish funding mechanisms accordingly.

38

Funding of the science base (continued) Management of funding Separate criteria for funding of schemes basic research (excellence) and applied/technical research (relevance).

Most countries now follow such principles.

Financing of basic research in industry

To date, none of the OECD countries seem to have come up with incentives to prevent the drying up of inhouse basic research in industry.

Maintain a minimal level of effort by appropriate subsidies and tax incentives for in-house research.

Science/industry interfaces General framework

A climate favourable to academic/industry collaboration is characterised by: the absence of regulatory obstacles (regarding financial earnings, pension schemes, etc.); flexibility regarding teaching obligations; and autonomy in the development of new faculty structures (interdisciplinary).

Australia, Canada, Ireland, the United Kingdom and the United States present favourable climates with few obstacles. Climates favourable to institutional experiments can be found in Nordic countries. Switzerland and Germany used to present excellent interactions in specific sectors, but these need to be reinvigorated. Ad hoc centres Centres of excellence (for basic UK and Canadian research) and co-operative R&D schemes for centres of centres (for more applied research), excellence, and if properly funded and focused, have Australian and US both proved to be efficient schemes for co-operative mechanisms for joint research work. R&D. See also examples provided by Finland, Japan, Korea and Sweden. Research programmes If well-designed and generously Significant programmes funded (notably at the level of of the first type can be individual projects), such found in the United programmes can have a critical Kingdom (e.g. LINK) and impact on S&T field concerned; if in Japan (on specific moderately funded, they can be technologies). There are instrumental in developing many examples of the science/industry networks. second type of programmes (see the European Union for complex, multi-country schemes).

39

Science/industry interfaces (continued) Placement of scientists Placements can be promoted on in industry an ad hoc basis with specific linkages with a given institution or professor, or through more general incentives.

Source: OECD (1998a).

40

The UK TCS and the Canadian Industrial Research Fellowship for the first type of programmes; and the French, Dutch and German incentives (paying part of the cost of employment of researchers by SMEs) for the second type of programme.