March 16, 2007: Robin Hanson on “Why Grants Won Over Prizes in Science”

Professor Robin Hanson is an economist, teaching at George Mason University.  Before becoming an economist, he worked as a physicist for NASA. On March 16 at KEI, Hanson presented his research on the history of grants and prizes as funding sources for scientific research.  The presentation was based on Hanson’s 1998 working paper, “Patterns of Patronage: Why Grants Won Over Prizes in Science.” (paper, presentation ).

According to Hanson, prizes have been historically favored by scientists outside of the establishment who sought a more objective process for allocating funds, while grants have been preferred by scientists with established reputations who could use their influence to secure funding.  Hanson cites the example of “chemical physicians” in the 1600’s who were excluded from teaching in British universities because of their unorthodox ideas about medicine.  In 1651, the chemical physicians proposed to the medical establishment that a cash prize of 300 florens be awarded to whichever group was able to save the most patients in a controlled test.  The proposal was rejected by the university physicians who could rely on using their influence to secure funding and stood to gain little from an objective criterion.

Hanson highlighted three general funding strategies to stimulate scientific innovation: funding results, funding effort, and funding for overhead costs.  According to Hanson, funding for results, i.e. prizes, should be the best strategy when would be innovators have access to capital and funders have a particular goal in mind.  But despite widespread historical use of prizes, today funding for effort (grants) is a far more prevalent strategy.  Hanson explained that historians of science have traditionally argued that the shift reflects a recognition that grants are a superior strategy for getting innovative scientific results.  But Hanson hypothesizes that the shift from prizes to grants can be explained by a shift in the types of patrons who fund science.
Specifically, Hanson suggested that more centralized and democratic governments tend to prefer grants, perhaps because they are more susceptible to pressure from establishment scientists and scientific societies which stand to benefit from more discretionary spending (much in the same way that such governments are more susceptible to pressures leading to “pork” spending).  Using 135 18th century scientific societies, Hanson conducted a regression analysis to determine how well grant-like vs. prize like funding was predicted by the type of patron.  He found that patron type was significantly predictive of the balance of grants vs. prizes, and that without appealing to any other factors, the much expanded role of more centralized and democratic governments in science funding today is predictive of the current dominance of grants as a form of patronage.  In Robin’s words, “Grants may have won not, as their advocates claimed, because they were a superior institution, but instead because non-local and non-autocratic governments tended to prefer them.”
Discussion following the presentation focused on medical innovation prize fund proposals, with Robin stressing that the key question in his mind would be how to set the size of the prize fund and evaluate the benefits of drugs.  Robin, who is best known for his work on prediction markets, suggested using prediction markets to evaluate the likely benefits of various drugs.  For instance, if the government were considering buying out a certain drug patent to place it in the public domain, speculators could wager on the health impact if the government went through with the buyout and on the health impact if the patent remained private, and the government could base its decision on the market’s predictions.