Friday, September 15, 2006

Our Broken System of Federal Research Funding

By Richard Vedder

Tens of billions of dollars are spent annually on various forms of federal research support for universities, through agencies such as the National Institutes of Health (NIH), National Science Foundation (NSF), the Department of Energy, and dozens of other funding sources. In addition, a growing percentage of funds are distributed directly by politicians in a non-merit-based way through congressional earmarks. There are lots of problems with the system from an efficiency/productivity standpoint.


We have been yelling a lot about earmarks (Congressional directed spending for specific research projects). During our visit to George Mason Law School yesterday, someone suggested that we try to deal with the one arguably defensible argument for earmarks in a better way. The semi-legitimate argument is that an old boy system of evaluators channels research funds to friends of Establishment researchers in a way that is only partly merit-based. A solution to this problem, if it is one, is to ban earmarks (that go mainly to smaller less prestigious universities), but devote part of the freed up funds for an Emerging University Research Program, providing money for schools not in the top 50 grant recipients, with awards made on the basis of strict merit, not political clout. Getting this approved may be harder than getting the sun to rise in the west instead of the east, but it is a decent idea.


Each institution currently negotiates an overhead rate for research grants -- often 50 or 60 percent of basis research costs --with the feds. This perpetuates inefficiency, has high administrative costs, and even sometimes leads to some almost corrupt practices. New approaches are needed. One is adopting a national uniform overhead rate around the rate prevailing in low overhead schools. Call it the "best practices" approach. Another is forcing NSF, NIH, etc. to use overhead costs as a criterion in making awards --maybe 10-15 percent of the total evaluation. High overhead, lower overall rating scores on research proposals.


Now when Professor X gets her award from NIH, it is for a very specific narrow topic on which only she has applied --she is the monopoly applicant for that topic. Maybe more funds should be awarded for topics chosen by a panel of national experts in various disciplines, with a request for proposals being widely circulated amongst the research community, and cost considerations counting a significant amount in the determination of the final award. This instills a bit of competiton and cost consciousness into the process. There are limits to this approach -- committees often fail to recognize the truly original, often slighty wacky ideas that lead to important research discoveries. But greater experimentation with this approach might be wise.

This merely touches the surface. Another huge issue is institutional support for research through low standard teaching loads, a form of research support that increasingly is totally unjustified on any rational cost-benefit calcuation.


superhiker said...

This is getting more and more idiotic. Overhead rates at universities are typically around 50%. In industry and national labs, they are far, far higher, often a multiple of 100%. The universities are giving the government a bargain.

Butter Cup said...

Superhiker - I'm not sure what your definition of overhead is. By my definition, if I owned a company and my overhead costs were 100%; Why would I be in business?