WORKSHOP ON BASIC RESEARCH: POLICY RELEVANT DEFINITIONS AND MEASUREMENT ISSUES PAPER Holmenkollen Park Hotel, Oslo, Norway 29-30 October 2001 Background 1. In their conclusions to the CSTP (Committee for Scientific and Technological Policy) meeting held at Ministerial level in June 1999, Ministers recommended that the Committee pursue its work on trends in the financing of basic research in universities and public institutions. In May 2001, the Council of Ministers in the "Ministerial Paper on Growth" concluded: "Governments should give greater priority to basic research; future innovation will be jeopardised without it." 2. This shows that at the political level scientific research is widely recognised as a major driving force behind any modern knowledge-based economy. Therefore, for most OECD countries, science policy is an increasingly important area of government responsibility. Science policy makers, in turn, need up-to-date tools, instruments and concepts for setting priorities, allocating funds, making long-term plans, refining the organisation of the science administration system and, in general, optimising the governance of the science system, including a better accountability. The notions of "basic research" and "applied research" have, for many years, been standard elements of the policy maker s toolkit. However, one trend that has attracted considerable attention is the blurring of the boundaries between basic and applied research, and the impact of this phenomenon on priority-setting and funding decisions in the public and private sectors. 3. The above trend is best shown in the two research areas that presently get the most attention: life sciences and ICT development. In the life sciences some research areas being regarded as "basic" - genomics and post-genomics - are leading to application very quickly; in ICT development a research area being regarded as "applied" - quantum computing - deals with some very basic knowledge questions. 4. Realising that a discussion on these questions might be useful, and acting on a proposal from the Delegation of Norway, Delegates to the 74 th Session of the CSTP agreed to convene a Workshop to examine the relevant concepts and issues.
5. Although the term basic research is often used in science policy discussions, it is difficult to define its operational meaning with precision, and to quantify the resources devoted to basic research by governments and public and private institutions. To date, the statistical analyses have been based mainly on the definition contained in the Frascati Manual (FM). 1 This definition does not really refer to the content of scientific investigations, but assigns a primary role to the intentions of the researcher (or the sponsor). Whether this is an adequate definition is a matter for discussion. 6. Apart from the Frascati Manual, there is no clear explanation of what "basic research" means or comprises. In searching for an improved policy-relevant definition, it may be worthwhile to enumerate some of the ways in which the term basic research is colloquially used: Curiosity-driven research (research aiming at uncovering general laws and principles). Theoretical analysis vs. experimental work. High-risk or very speculative research. Long-term research (insofar as it requires more time, but also as results only become applicable at a later stage). Open vs. proprietary research. Main issues for discussion Session 1: Is the distinction between basic and applied research still relevant? 7. Any effort to clarify the definition of basic research can only be justified if it is first found that the notion of basic research continues to be relevant and useful for policy makers, programme managers, and scientists. If relevance can be established, it will almost certainly be the case that a quantifiable definition is needed for policy analysis purposes, for example, for tracking basic research trends in time or between different scientific disciplines. 8. A proper definition for basic research may above all influence the following: achieving a desirable balance between basic and applied research and between different scientific areas; help deciding whether research should be funded from private or public sources; establishing whether support for basic research is really declining (as many stakeholders claim); measuring economic, social and cultural benefits of scientific research; deciding about investments in education and research infrastructures; deciding about changes in the structure of science systems. 1 The FM distinguishes three types of activities : basic research, applied research, and experimental development. It defines basic research" as follows: Basic research is experimental or theoretical work undertaken primarily to acquire new knowledge of the underlying foundations of phenomena and observable facts without any particular application or use in view. The FM then distinguishing two types of basic research: "- Pure basic research which is carried out for the advancement of knowledge without working for long term economic or social benefits and with no positive efforts being made to apply the results to practical problems or to transfer the results to sectors responsible for its application. - Oriented basic research which is carried out with the expectation that it will produce a broad base of knowledge likely to form the background to the solution of recognised or expected current or future problems or possibilities.
How important is the concept of "basic research" for science policy decisions in OECD Member countries? Is it even more relevant than it used to be for priority setting and funding decisions? Should "basic research" be defined in a broader sense as the "science base", comprising all elements enabling the science system to carry out all types of research (equipment, infrastructure, permanent research and technical staff)? What are the criteria for governments to fund research which does not yield short-term or foreseeable results? Session 2: The mission of public research institutions: basic research vs. the demand to contribute to solving societal problems and to economic growth 9. Traditionally "basic research" meant that researchers - above all those in universities - had a considerable autonomy in the choice of topics and in carrying out research. Basic research was done in a bottom-up approach - the influence of researchers being strong in formulating the objectives of such research and in evaluating its execution and outcome through peer review procedures. This led to research mostly carried out with regard to a well-defined subject and along the lines of a specific scientific discipline. 10. However, basic research does not merely advance the frontiers of knowledge. Among its benefits are: education and training of highly-skilled personnel, establishment of high-technology spin-off enterprises, insight into issues of major social concern (such as rational and science-based approaches to health, safety, energy and environmental issues), creation of new technologies and instrumentation, and enrichment of the overall culture. Governments put increasing emphasis on these broader benefits of research and ask research institutions to take account of these demands. Therefore funding agencies are under an increasing pressure to fund mission-oriented research, which - in the case that overall funding does not increase - means less resources for funding "basic research" in the traditional sense. 11. Another trend influencing research in research institutions which, in the past, have mostly concentrated on basic research is the increasing demand for stronger ties with industry and for a commercialisation of research results. There is a strong feeling in parts of the scientific community that this approach endangers basic research, in particular in areas more remote from industrial applications, that it is putting their scientific and intellectual independence in question, and that research results might not be publicly accessible anymore. Is there still sufficient support for public research institutions to address basic questions? Have changes occurred? Is there the right portfolio of public funding with regard to research fields and research institutions? Or do some research fields have an undue advantage over others? How do research institutions cope to find a balance between the demand for advancing scientific frontiers and the demand for closer links with industry?
Session 3: The role of industry with regard to basic research 12. The share of business in the funding of academic research has increased in most countries over the past ten years. Though it generally remains below 10% and government funding remains predominant, the share has more than doubled in some countries (from 4 to 8% and 10%). Several reasons are given for this increase. One is that some firms have closed down their own research labs or parts of such labs and outsourced research to public research institutions. Others argue that business, above all in the biotech and ICT area, has realised that intervals from basic research to application have become much shorter and therefore want to secure early privileged access to the results by funding such research. Still another argument is that this increase in funding by business is coming mostly from university-based spin-off companies which naturally would invest in close ties with university research. 13. In this connection the question of a division of labour between the public and the private sector arises. Changes in the funding structure for basic research with regard to funding sources might lead to questions relating to public access of research results, including IPR regimes. What is the role of business in basic research? What types of research are firms interested in? What should be the division of labour with regard to funding between the public and the private sector? Should governments rather invest in areas where intervals from basic research to application are still long and where experimentation costs are still high, and leave the funding of fields promising easier commercialisation to business? How can researchers valorise their basic research results in co-operation with industry? Session 4: Problems relating to data availability and data collection 14. Though the Frascati Manual provides the framework (see footnote 1), statistical data on basic research is relatively poor. It seems that only two countries (Australia and the United Kingdom) collect separate data on the two categories "pure" and "oriented" basic research. As to overall basic research, 15 OECD Member countries reported data on basic research in 1996 or later; whereas 18 did so in 1990 (or closest years). This is quite a big decline, given the fact that five new Members entered the OECD. Eight Member countries reporting data on basic research in the early 1990s stopped doing so between 1991 and 1994; and in 1996 14 out of 29 Member countries did not provide such data. It should also be noted that some countries collect data but do not submit them to the OECD Secretariat. 15. Reasons for countries not to collect some kind of data may vary. However, a number of OECD countries have indicated that the definition used in the Frascati Manual does not sufficiently comply with the current situation in their countries. What have been the reasons for countries to stop reporting basic research data since 1990? Do they still collect such information, or not? What problems do statisticians encounter when collecting and processing such data? Is the definition of the Frascati Manual adequate for policy making purposes, or can a more precise formulation be found - one that is also useful for statistical analyses?
Possible results of the Workshop 16. As outlined above, it would be the primary interest of the Workshop to discuss and elaborate on today s meaning of "basic research" in the context of science policy as well as of research statistics. In detail, the Workshop would aim to achieve the following: Contribute to a better understanding of the contribution of various kinds of research to the accumulation of knowledge and economic performance. Arrive at a better understanding of a "division of funding" between the public and the private sector. Suggest changes in the definition and/or characterisation of basic research which correspond better to the use of this concept in the present science policy context. Suggest a set of indicators, e.g. criteria, which facilitate the measurement of basic research for statistical purposes. Provide possible recommendations for a revised definition of "basic research" which may help policy design, and may feed into a revised Frascati Manual.