A science funding contrarian
The Economic Laws of Scientific Research, by Terence Kealey. New York: St. Martin's Press, 1997, 388 pp.
Richard R. Nelson
The premise of Terence Kealey's book-that scientific research would do better without government support-has naturally attracted a lot of attention and generated a lot of emotion. Kealey is an impassioned advocate of market capitalism and laissez-faire. He believes that economies and societies do a fine job of self-organizing if left alone and that governments almost always are incompetent and often venal. His attitudes toward science policy strongly reflect these opinions.
He is fascinated by economic, technological, and scientific history; has read widely (if selectively) in these areas; and projects lessons he draws from history into the present. Yet Kealey, who is trained in biochemistry and medicine and teaches in the Department of Clinical Biochemistry at the University of Cambridge, presents a strangely limited view of the way modern science actually works and of the complex relationships between science and technology.
Kealey's reading of ancient and modern history leads him to articulate two major propositions about science and technology and their interaction. First, science that is nurtured by a state or a society and is isolated from the world of practice is fruitless. This almost surely is true. Second, free commerce virtually automatically generates technological innovation and economic growth. This leads, without government involvement, to the development of whatever science is necessary to support technology. Kealey argues that this natural state of affairs broke down after World War II, as government intervention increasingly stifled scientific, technological, and economic progress.
"Economic laws" unveiled
In the course of making this controversial argument, Kealey puts forth several "economic laws" of scientific research. The first law is that, in the modern world, the ratio of R&D to gross national product (GNP) tends to rise as GNP per capita increases. Kealey emphatically denies that a rise in R&D as a fraction of GNP results in an increase in GNP per capita. Rather, he holds that as per capita income grows, nations spend more on R&D partly because they can afford to do so and partly because their more complex economies draw more intensively from formal science and technology (S&T). Although Kealey's statement of this argument is somewhat crude, many economists would broadly agree.
Kealey's second economic law is that, given the first relationship, greater public funding of civilian R&D results in less private support. Further, according to Kealey's third law, the net result of public spending on R&D is negative: The efficacy of the R&D effort is reduced, as is the overall level of spending. To support his case, Kealey compares Japan, where there is little public support of civilian R&D, with Britain, where there is a lot. He notes that not only is Japan's high-tech industry more competitive on world markets, but its spending on civilian R&D as a fraction of GNP is higher.
In the last part of the book, Kealey lambastes government S&T policies, particularly in the United States and Great Britain. He has little sympathy for government efforts to bring particular civilian technologies into existence or to support certain high-tech industries-efforts that he finds inevitably clumsy and inefficient, and usually fruitless.
I have considerable sympathy for Kealey's overall position, but his arguments here are a bit simplistic. First, some important technologies resulting from government programs have yielded high civilian payoffs. Jet aircraft and the computer are good examples. Kealey might respond that the key government-funded projects in these cases were aimed at developing military technologies, not commercial ones, and that government programs in these areas for the civilian market have virtually all been expensive failures. Although the development of the Airbus jet aircraft consortium in Europe could be cited to rebut Kealey, his argument is basically correct.
On the other hand, government-funded applied research programs have produced enormous payoffs in several civilian technologies. U.S. agriculture is a striking example. In many areas, publicly funded research targeted at providing basic knowledge and tools to solve practical problems has enormously increased the power of privately funded applied R&D. Kealey's own field of biomedical research illustrates this clearly. Virtually all modern pharmaceutical development takes place at for-profit firms funded by private money, but it relies greatly on knowledge gained through publicly funded research. It is odd that Kealey does not recognize this.
Going to extremes
Kealey's view that government should just get out of the business of supporting science is particularly wrongheaded and dangerous. Government support, he says, is wasteful, poorly directed and managed, basically harmful to the workings of the scientific community, and simply not needed. Industry and private philanthropy can be counted on to support directly or indirectly the research that society needs and will do so in a way that is more efficient and conducive to good scientific research.
Kealey further argues that government support of science is rationalized by the linear model, and because this model simply is false, there is no case for government support. It is true that such influential manifestos as Vannevar Bush's Science: The Endless Frontier put forth the linear model to support government funding of basic research, with control of allocation left principally to scientists. It is also true that in most areas of technological change, the linear model is not a good characterization of the relationship between science and technology. But one does not need to believe in the linear model to be a strong advocate of public support of science. And Kealey's own beliefs about the relationships between science and technology seem as inadequate a general characterization as is the linear model. Let me pick up this latter matter first.
Contemporary scholars who study the relationships between science and technology recognize that these are complex and differ greatly from field to field. In many fields, technological advances hardly tap recent scientific developments. Kealey stresses this. Also, in may cases, technological development leads to the initiation of scientific research and even to whole new scientific fields, rather than the other way around. The field of metallurgy arose because steel became an important economic commodity. Solid-state physics became an important field of research after the birth of the transistor. Computer science obviously came after computers. Kealey points to cases like these.
But although these latter cases do not fit the linear model, they clearly involve areas where scientific research done at universities contributes to technological advance in industry. Indeed, that is the intent of research in fields such as computer science and pathology. A strong case can be made that a rapidly advancing technology almost always has a closely affiliated science.
Not all useful science is intentionally oriented toward areas of technology. There are many striking examples where the simple linear model looks right. That is, scientific research undertaken with only the broadest notions of the practical payoffs lays the basis for revolutionary advances in technology. Again, biotechnology is a good example.
One need not believe in the linear model to support government funding of university research. The fact that a field of science contributes to technical advance by intent certainly does not mean that public support of that science is not warranted. The principal arguments for public support of science are that knowledge won through fundamental research is nonrivalrous in use, and that in many cases it is difficult for a person or an organization to keep that knowledge out of the hands of others or to force all who use it to pay a fee.
The first of these arguments is persuasive by itself. Even when it is possible to make basic knowledge won through research private and proprietary, society pays a cost, perhaps a very large one. There are very sound economic reasons for keeping knowledge-a nonrivalrous good-in the public domain.
Keeping knowledge public
We badly need strong and effective arguments for keeping fundamental scientific knowledge public. With the policy discussion in the United States leading the rest of the world, the fashion increasingly is to argue that there is great value in making new science proprietary. The Bayh-Dole Act is predicated on exactly that largely dubious and in many cases quite pernicious idea. Patenting gene fragments is a clear contemporary manifestation of this belief. But access to fundamental knowledge should not be rationed, even if it can be. There are real economic costs associated with privatizing basic science.
Of course, in many areas it is quite difficult to prevent basic knowledge from trickling into the public domain. This surely is one of the reasons why, during the past 80 years, large electronics companies have abandoned or significantly decreased support of basic research. Previously, their powerful market position meant that even if the results of their research trickled out, few competitors could benefit. With the new global competition, that no longer is true. Kealey interprets the fact that private companies occasionally fund basic research as an indication that the knowledge so won is not a public good. But he additional fact that firms other than the funders benefit from the research indicates that it is. Thus, business support of basic research is limited and fragile.
Industry leaders such as William Spencer of Sematech are deeply concerned about the implications of this fact for long-term progress in microelectronics. They struggle to find alternative funding for fundamental research in that field. Frankly, Kealey's proposal that corporations and philanthropy will fund all the valuable fundamental research is bizarre. National governments also seem to be moving away from support of basic research. Like the companies guarding against competition, nations are shifting support toward more applied research, whose results can more easily be captured nationally.
It is not clear how much influence Kealey's ideas will have on policy. My suspicion is that those who already believe his argument about privatizing basic research support will pick up and trumpet it. Those who understand the very powerful case for public support of science with the objective of keeping science public will ignore or explicitly reject it. Now is the time to rearticulate the need for and the payoffs from publicly funded fundamental research. Perhaps the blatant extreme of Kealey's position will serve the useful purpose of focusing the arguments of those who believe in public science.
Richard R. Nelson is the George Blumenthal Professor of International and Public Affairs, Business, and Law at Columbia University's School of International and Public Affairs and is the author of The Sources of Economic Growth (Harvard University Press, 1996.)