Direct government funding of R&D empirically spurs innovation successfully

There is widespread agreement over the need for some amount of direct centralised financing for scientific research, although there are major disagreements over precisely the role that such funding should play. The consensus is that government needs to finance basic research that is not likely to yield commercial benefits in the immediate future. Few firms are willing to undertake basic research where the prospective return is not only uncertain, but often not even well-defined. For example, while the commercial benefits from the Internet are now quite evident, this was hardly the case in the 1960s and 1970s when the United States Department of Defense was financing the research that led to its creation. Similarly, much of the basic biomedical research financed by the National Institutes of Health has no immediate commercial use, even if follow on innovations may prove to be quite profitable if subject to patent protection. If these types of basic research were not supported through some mechanism of direct financing, the research would mostly not be done. The risks would be far too great to support large investments through a prize mechanism.18 Moreover, as Mazzucato (2014) has shown, many privately successful ventures have been dependent on government funded innovation for their successes.

While the merits of direct financing through a centralised system for basic research are widely accepted, this is much less the case for research focused on developing end products such as prescription drugs or medical equipment. The argument against this sort of financing is that a centralised mechanism is likely to be overly bureaucratic, slow-moving in response to changes in technology or changes in needs, and subject to political interference. This raises the risk that government financing could end up being largely wasted, with favoured companies or individuals being perpetually funded for work that ends up being of little value. The same issues exist with funding for basic research; however, in the case of basic research there is no feasible alternative. Therefore, these problems, insofar as they arise, are an inevitable cost. In practice, the peer review system used by, for instance, the National Science Foundation in the US, has proven itself to be free of these problems. Similarly, DARPA (the Defense Advanced Research Projects Agency), widely credited with creating the Internet, has proven itself as effective sponsor of advanced research.19