Francis Collins, director of the National Institutes of Health (NIH), recently testified before a Senate appropriations subcommittee about what he characterized as a funding crisis facing the NIH. Collins highlighted two core problems: a decline in the purchasing power of the NIH budget and the dangerously low success rate of research grant proposals. These problems are quite real but they have deeper roots than the federal government’s recent budget tightening. The more fundamental problem facing the NIH is a lack of stable, predictable growth and the perverse dynamics of the biomedical research workforce.
To a large extent, Collins’s testimony focused on the effects of the sequester, the across-the-board budget cuts implemented in 2013. However, these recent burdens need to be placed in a broader context. The chart below shows the growth in the NIH budget from 1960 to 2014 in both current dollars and in real dollars adjusted for inflation. The inflation-adjusted NIH budget grew at an average rate of almost 9 percent a year from 1960 to 1998. Beginning in 1999, congressional enthusiasm for biomedical research investments resulted in very high annual growth rates for the NIH budget ranging from 14 to 16 percent through 2003 known as the NIH budget doubling. But after 2003, the inflation-adjusted value of the NIH’s budget declined by over 20 percent.
The wild fluctuations in the NIH budget over the past two decades are a problem in their own right, independent of the burdens imposed by sequestration. Sound research and development planning in universities and in industry requires predictability, like with any long-term investment.
In addition to the funding crisis, Collins noted that the NIH is experiencing a research-grant proposal crisis. NIH’s research-grant proposal success rate is the share of reviewed grant applications that receive funding. In 2014, the success rate was 18 percent, down 27 percentage points from its 1974 peak of 45 percent. Success rates this low are demoralizing to early-career and established biomedical researchers alike and make it extremely difficult to provide stable funding for laboratories. Collins naturally tied these low rates to the recent troubles of the NIH budget. However, the decline in proposal success rates is a decades-old phenomenon, long pre-dating the sequester.
The problems facing the NIH are not fundamentally rooted in the congressional appropriations process; they are a consequence of the structure of the biomedical research workforce itself. Since graduate students make up a substantial share of the research workforce, any increase in funding (such as the 1999-2003 doubling) brings with it a large increase in the number of graduate students. These students enter their research careers several years later (as funding levels off) and are often unable to successfully compete for their own grants, exacerbating the original crisis.
Responding to Collins’s very real funding concerns with another dramatic investment in the NIH is likely to replicate the same dynamic of overproducing graduate students to staff existing labs. As long as the production of new graduate students is tightly coupled with biomedical research, even stable growth in research funding will result in a declining grant acceptance rate over time. A permanent resolution of the NIH’s troubles would require a stable budget environment, but even more important, it would require fundamental reforms in the way the scientific enterprise is structured in the United States. The production of new graduate students should reflect the career prospects of graduates rather than the staffing requirements of research laboratories.