Association for Public Policy Analysis & Management

U.S. universities spent over $37 billion on life sciences R&D in fiscal year (FY) 2012, of which approximately 61 percent ($22.7 billion) was funded by federal agencies, with the ultimate goal of producing innovations that will lead to better medical treatments and prevention of disease (NSF 2014). Recent studies have demonstrated that increases in universities’ federal life sciences R&D funding yield subsequent increases in investment by non-federal funders, and in universities’ successful applications for drug and medical patents. In addition, some 10% of new drugs approved by the U.S. Food and Drug Administration (FDA) since 1990 were covered by patents acknowledging government funding.

Despite the evident contribution of federal research funding to universities’ R&D non-federal sources, to patenting, and to downstream pharmaceutical R&D, the complex dynamics and potential feedback effects inherent to the so-called triple helix interactions of government, industry, and academe preclude econometric estimation via a single equation, or via multiple independently estimated equations. Several recently developed panel data estimation techniques permit consistent estimation of nonlinear models with correlated outcomes, even in the presence of unobserved and unmeasured heterogeneity or feedback from current year outcomes to future year inputs. I employ these techniques on a newly-created longitudinal dataset combining university R&D funding by source, university-assigned patents, university-industry alliances, and patents associated with all FDA new drug approvals since 1980, to address the following specific questions:

(a)  Is the observed increase in subsequent non-federal life sciences R&D investment due to university patenting of inventions discovered through federally funded research?

(b)  Does university patenting discourage subsequent industry investment in R&D, collaborative agreements, and university-industry alliances?

(c)   When industry R&D funding follows closely behind federal funding, does that increase or decrease the likelihood any resulting invention will be patented by the university?