Science Funding

This book review of Science Funding, by Joseph P. Martino, (c) 1992 (Transaction Publishers, ISBN 1-56000-033-3, 392 pages) was written by Barry Fagin and posted on an Internet mailing list.

A couple of months ago, I thought it might be interesting to look at applying public choice theory to science funding. I thought that by bringing the principles of public choice theory to bear, the "public goods" argument could be satisfactorily refuted and scientists would better understand why federal support for R&D is neither necessary nor desirable.

Last month, however, I learned that a research scientist has spent a considerable amount of time over the past few years doing just this. His name is Joseph Martino, and the results of his research have just been published in a book titled Science Funding: Politics and Porkbarrel. Completed with the assistance of the Reason Foundation and other groups, Martino's book weaves history, economics, and science together to show the problems associated with governmental support of research.

The book is extensively researched, with a large amount of data on federal support of R&D. Martino attempts to address every argument advanced in favor of governmental support of research, and is able to marshall considerable evidence against it. Briefly, the classic argument in favor of federal support of R&D was first given by Kenneth Arrow in 1962. Arrow states that a firm will increase spending on research only so long as the marginal return on it exceeds the marginal return from other investment opportunities. Problems arise when the firm's return is less than the return to society as a whole (i.e. when it can't "appropriate" all the benefits of the research). Under these conditions, society would be better off if the firm spent more money on research, although the firm would lose money. Under these conditions, society will be better off if government funds the additional research, equating marginal social cost with marginal social benefit.

Unfortunately, the flaws in this argument are not obvious to most scientists, and in fact as far as I can tell they have never been refuted satisfactorily in the scientific press. Martino notes that Arrow's argument has three problems, any one of which is sufficient to render it invalid:

  1. Empirical observations show that under competitive conditions firms can actually overinvest in R&D, in the sense that private and social benefit is less than social cost.

  2. Even if firms do actually underinvest in R&D, there is no way for government to determine, even in theory, the level of funding necessary to equate marginal social costs with marginal social benefits. The social return from foregone investments if government funding of research is increased simply cannot be calculated. Given what we now know about centralized planning and distributed decision making, it is clear that the federal government cannot and never will be able to determine the "right" amount of money to spend on research. The answer according to scientists will always be "more".

  3. Arrow's argument implicitly assumes that government is a neutral agent, weighing costs and benefits in an attempt to achieve optimal outcomes. But we know from public choice theory that this is simply not so. Agents of government are human beings who, like the rest of us, respond to incentives. The incentive structure created by government virtually guarantees that even if the right amount of research funding could be calculated, it would not be spent on nonappropriable research. Instead, it would be spent on politically useful research, rent seeking, and other non-productive activities.

The book is also enlightening in its survey of scientists attitudes, many of whom believe steady funding to be an entitlement. Numerous scientists who responded to a survey wanted "automatic funding if continuous publication is present", "longer grant periods with fewer strings on renewal", and so forth. As a study in sociology, Martino's book is a depressing reminder of how pervasive the entitlement mentality has become, even among the intellectual and professional class.

Martino's book is available from Transaction Publishers, Rutgers -- The State University, New Brunswick, NJ, 08903.