Should NIH Be in the Drug Development Business?

In an excellent review blasting the false dichotomy of more versus less regulation (for additional commentary, see Amanda and Ezra Klein), economist Dean Baker proposes that the government get into the drug development business directly:

...the government could pay for the research upfront and make all research findings and patents fully public. It already spends $30 billion a year financing biomedical research through the National Institutes of Health, an amount almost as high as the pharmaceutical industry claims to spend on its research. NIH research is highly respected, with almost all observers agreeing that the money is, on the whole, extremely well spent. While the NIH focuses on basic research (it also does some later-stage drug research, including clinical testing), there is no obvious reason why the government could not simply double its commitment to biomedical research in order to replace the research and development currently supported by grants of patent monopolies.

But the government may wish to use a different mechanism to encourage drug development. It may choose to establish a small number of master contractors, who would then contract out the awarding of research funds so as to minimize the potential for political interference. Regardless of the structure a particular program would take, expansion of direct funding is clearly feasible.

There would also be large public benefits in addition to lowering the price of drugs to their marginal cost. Eliminating huge monopoly rents associated with drug patents would take away the incentive for drug companies to push drugs in cases where they may not be especially beneficial, or even potentially harmful. Nor would there be incentive to conceal research findings that indicate a drug's weak performance. Furthermore, by placing all research findings in the public domain, so that scientists can quickly benefit from the research done by others, the process of drug innovation would likely accelerate.

Mind you, NIH doesn't have to be the agency to do this, although they would be the first option (in that NIH already exists). I don't think the "master contractors" idea is too good--I think this could stifle innovation (what if they're biased towards or against certain approaches) as well as potentially become corrupted, but NIH should be much more heavily involved in drug development, including phase II and phase III trials (which are very expensive). If nothing else, this might make it easier to find and track willing subjects.

Discuss.