Estimating the Effects of Federal Research Funding on Universities using Alumni Representation on Congressional Appropriations Committees

This paper estimates the effects of federal research funding on 71 research universities. We focus on the number of and citations to articles published, patents issued, and faculty salaries. Using a panel data set that spans from 1972 to 1994, we control for potential endogeneity and omitted variables bias in our regressions by using a data set that links U.S. Congress members on the appropriations committee with their undergraduate alma mater. Alumni representation lowers the shadow price of federal funding. Using our preferred instrumental variables specification, we find an increase of $1 million in federal research funding (1993$) to a university results, on average, in 12 more articles, .34 more patents, and $152,015 more in total faculty salaries. Citations to articles publish falls by .08 citations. So when the shadow price of federal research funding falls, as a first approximation, universities buy more federal research funding and produce more but not necessarily higher quality research output.