Nationalizing Science

It seems as if you can’t turn anywhere without hearing that industry is destroying science these days. Former editors of the New England Journal of Medicine allege that pharmaceutical companies are perverting health science. The National Institutes of Health have instituted strict new ethics rules that forbid researchers dirtying their hands by collaborating with industry. And at the end of July the American Journal of Public Health devoted an entire supplement to essays alleging that industry was somehow distorting the legal and regulatory processes through a series of laws and judgments.

All these allegations have a similar purpose: to delegitimize industry’s involvement in the scientific process. The result, if they are heeded, will be a de facto nationalization of basic science. As I show in a new paper for the Competitive Enterprise Institute, that would be philosophically unjustified, would ignore financial reality and would ultimately lead to disaster for American science.

The distinction between basic science and applied science (and its development) is, at heart, an elitist and artificial one. It is based on a misunderstanding of the scientific dynamic that was set in policy stone by Vannevar Bush, President Franklin D. Roosevelt’s top science adviser, during World War II. It relies on what is known as the “Linear Model” of science, which states that “basic research” develops a pool of knowledge from which “applied research” draws practical benefits, which are then developed into economic goods. The entire process is called “Research and Development,” or R&D. In diagrammatic form, the model can be depicted as:

Academic Science → Technology → Wealth

The model recognizes that industry is likely to fund applied research and development, but maintains that basic research, being academic in character and generally producing results that are not immediately translatable into profit, is unlikely to be funded by the private sector. Therefore government funding is justified on the grounds that it will be returned in economic wealth at the end of the R&D process.

Yet the model fails to recognize that the process is not linear. As Terence Kealey shows in The Economic Laws of Scientific Research, much of new technology derives from advances in old technology rather than from basic research (and the same goes for new research); academic research can be inspired by technological development, as was the case with the development of radioastronomy and even the Big Bang theory, and there is often cross-fertilization between the two, as is the case with solid state physics.

Kealey therefore proposes a refinement of the linear model to:

Basic science    ↔ Technology   → Wealth

                        ↑                      ↑

Old science       Old technology

Science’s role in the economy, it appears, is mainly dependent on the technological portion — the applied research and development — of the model. Basic research contributes far less than the linear model suggests. Thus, approaching science policy by looking at basic or academic science as a starting point and extrapolating from it is misguided.

Moreover, before Vannevar Bush “basic research” was more often referred to as “pure science,” an elitist approach that suggested that “pure science” was somehow nobler than applied science. Where echoes of this approach survive, the usual problems with elitism surface. For instance, when one of the developers of penicillin suggested to the British Medical Research Council, which had funded his research, that it should patent the discovery, he was told that only tradesmen try to make money from science. The MRC made exactly the same mistake with monoclonal antibodies in the 1970s. As a result, the British taxpayer has paid millions in royalties to the foreign firms that patented the knowledge. When it comes to science policy, the utility of the science is what is important.

This becomes even more apparent when one considers that industry is in fact the dominant player in American scientific research and development. It currently funds over two-thirds of American investment in the field. On the other hand, government is not even a particularly dominant force in the funding of basic science. In 2000, just over 50% of basic science funding came from the federal government, with the rest coming from industry and other private sources such as foundations or private universities. In fact, federal funding of basic research amounts to just 9% of American investment in science.

This is why rules meant to “protect” basic science from the supposedly pernicious effects of industry would be disastrous for American basic science. Basic science cannot be placed in a box. It is part and parcel of a much larger endeavor, where industry plays an important and valuable role. Denying industry science a role in basic science or forbidding government-funded scientists from dealing with industry would make basic science into a ghetto. Like all ghettoes, movement of good scientists would be only one-way: into industry and out of government science.

Moreover, allegations of misconduct and inaccuracy in industry research ignore the self-correcting nature of the scientific process, spurred by commercial realities.

A new report in the Journal of the American Medical Association has revealed that 16% of highly-cited medical studies were contradicted by subsequent ones, and another 16% were shown by later trials to have overstated results. In short, almost a third of medical studies – ones that were cited over 1,000 times in subsequent studies and articles – were found by subsequent research to be flawed. But contrary to initial impressions, this shows that in medical science the scientific process is working.

It is quite plausible that in other areas it is working less well. This may be the case particularly in government-funded basic research, because funding bodies are more likely to fund original research rather than replication studies. This is, of course, because the justification for government funding stated in the Linear Model is to produce original research.

Approaching science policy from the viewpoint of the narrow base of basic research is foolish. We need to accept that industry is driving science and the American economy, and we should welcome and celebrate that fact as a triumph for the American entrepreneurial spirit.

The attitude that private sector interests are corrupting science and that anyone who has any association with a private sector interest is somehow tainted presents a greater danger to American science than the alleged problem of conflict of interest. In fact, the real problem facing science is not corporate influence, but too much government influence.

Based on the statements of those opposed to business-funded research, it becomes clear that the real danger for science is not the influence of private industry, but de facto nationalization.