Has Licensing Changed Academic Research? Issues of Productivity, Faculty Incentives, and Public Policy

National Bureau Of Economic Research Inc, Cambridge MA

Investigators

Abstract

This research examines the role of faculty in the increasing trend for U.S. universities to engage in patent licensing. The focus is on the characteristics of individual inventors and the role of licensing in the conduct and dissemination of their research. Since 1980, the Bayh-Dole Act has allowed U.S. universities to own and license results from federally funded research, and license-related activity has increased dramatically. Universities cite increased licensing activity as evidence of their increasing contribution to the economy while skeptics question the impact. Critics claim that potential financial returns from licensing have diverted faculty from more basic to applied research. This research is based on a panel database of the research and licensing activity of six to seven thousand faculty members in major U.S. research universities, including Cal Tech, Carnegie Mellon, Cornell, Harvard, MIT, Pennsylvania, Penn State, Purdue, Stanford, Wisconsin, and Texas A&M. These universities have been selected because of their status as major research universities and their commitment to provide data. Most data cover at least the period 1988-1999. With these data, one can examine the evolution of faculty behavior over time and the extent to which it is related to individual characteristics or to environmental factors. Differences in patterns across universities and disciplines can be examined. Three major issues are addressed: i. Does the increase in university licensing over the last decade reflect a shift away from fundamental research and/or an increased willingness of faculty to participate in licensing as well as publish the results of their research? ii. How is the propensity of faculty to disclose inventions (i.e., file a report with the university when they believe a research result has commercial/license potential) re-lated to individual characteristics (e.g., publication record, research sponsorship) and other characteristics (e.g., academic discipline, university and/or department effects)? iii. What is the life-cycle behavior of inventors with respect to research, publication, and license-related activity? The basis of the research is a life-cycle model in which faculty can choose to engage in basic or applied research and where licensing activity as well as publication is possible. This theoretical model suggests three empirical models of research productivity. One deals with publication activity, while the other two deal with licensing activity and basic relative to applied research over the life cycle. For each model empirical results are based on annual data for individual scientists and engineers. The method of analysis is Tobit and logit regression analysis. Preliminary results based on 602 Purdue faculty over the period 1983-96 tend to support the theoretical model. Further, there appear to be changes over time in licensing activity, but little or no changes over time in publication activity or the basic/applied nature of faculty research. Results will be reported in publications and relevant professional meetings.