Recent Research: Could Sudden Doubling of Federal Physical Science Research Funding Undermine U.S. Competitiveness Goals?

Last year, Congress authorized $5.9 billion in new spending on research, education and entrepreneurship as part of the Bush Administration’s decade-long $50 billion American Competitiveness Initiative (ACI). Though Congress did not appropriate a significant amount of new funding to match this authorization, many remain committed to the goals of the ACI. This initiative would double federal funding over 10 years for research within agencies such as the National Science Foundation (NSF), the Department of Energy (DOE) Office of Science and the National Institute of Standards and Technology. A recent study, however, finds the speed proposed for reaching these goals, while well-intentioned, may be a mistake.
 
In their paper, What if Congress Doubled R&D Spending on the Physical Sciences?, Richard Freeman and John Van Reenen argue that rapid increases in funding for R&D are substantially less effective in improving U.S. competitiveness than steady long-term increases. They examine the consequences of the federal government's effort between 1998 and 2003 to improve the U.S. standing in the life sciences by doubling the research budget of the National Institutes of Health (NIH). Freeman and Reenen find that much of the benefit of this increased funding was undermined by the cost of instituting many new programs and expenditures at once and then readjusting after NIH spending plateaued, which it has done since 2005.
 
In 1998, a bipartisan coalition in Congress pushed for a large increase in federal research spending, particularly at NIH. With the Clinton Administration's support, annual NIH spending grew from about $14 billion in 1998 to $27 billion in 2003. When this rapid acceleration of spending ended, the rate of increase fell to 3 percent in 2004, and continued to fall to 0 percent in 2007. Measured in constant dollars against the Biomedical R&D Price Index, the 0 percent increase in 2007 was equivalent to a 10.9 percent drop in spending. The result is fewer grants being made and decreased overall spending on research.
 
There were several consequences that arose in the wake of this sudden rise and fall in spending. First, as the number of grants rose, researchers submitted a larger number of applications. After the funding surge ended, however, the number of grants awarded decreased, but the applications did not. The result was a large drop in the rate of successful applications. When researchers found that more applications were being denied, many began submitting multiple applications for a single research project. This put a much greater burden on grant evaluators, who were swamped with applications. For researchers, who depend on new and continuing grants to sustain their work, the change led to widespread cuts among research staffs.
 
Second, the consequences disproportionately impacted younger researchers who had entered the field during the years of increased spending. During the years of increasing spending, principal investigators hired a larger number of postdoctoral research and graduate students. Once the number of grants fell, these younger researchers were often the first to be eliminated. Others, who had finished their training during the period of increased spending, found limited opportunities in the job market. Given the current concerns about retiring older researchers and the lack of younger workers to take their place, sudden changes in funding levels can be highly detrimental to competitiveness.
 
Freeman and Van Reenen argue that the NIH initiative would have been more successful if funding had been slowly increased over 30 years, even if the total spending during that time would be less than it was with a sudden increase over a few years. Steady increases allow agencies to initiate projects one-at-a-time and provide a more dependable stream of research funding. The authors acknowledge that long-term steady increases over time can be more difficult to implement politically. They suggest that one option to help support grant-dependent research through times of inconsistent funding would be to include extra overhead for stabilization within research grants. These funds could provide bridge support once for researchers when R&D spending slows down.
 
What if Congress Doubled R&D Spending on the Physical Sciences? was presented by John Van Reenen and Richard Freeman at the National Bureau of Economic Research's (NBER) Conference on Innovation Policy and the Economy on April 15, 2008, in Washington, D.C. The paper will be published as part of the Conference Report, which will be available in the future on the NBER website: http://www.nber.org/