Real Effects of Academic Research : Comment

Compares Jaffe's work on the use of patents as a measure of the spillover of university research with the work of Acs and Audretsch in which innovation activity is measured by number of innovations. Jaffe's work, which modified the knowledge production function proposed by Griliches, showed a positive relationship between corporate patent activity and commercial spillovers from university research. This research approach was criticized by many. In 1987, Acs and Audretsch proposed measuring innovative activity by the number of innovations recorded in 1982 by the U.S. Small Business Administration. It was believed that using number of innovations, using those provided a more direct measure than Jaffe's work because inventions that were not patented but were introduced into the market were counted and inventions that were patented but never introduced were not counted. This analysis seeks to compare the two works. Jaffe used a pool of data that spanned an eight-year period while Acs and Audretsch considered a single year, 1982. It is shown that using a single year sample in Jaffe's model does not greatly alter the results, which means that both private corporate expenditures on R&D and university expenditures on research both positively and significantly influence patent activity. The impact of university spillovers is greater on innovations than patents using Jaffe's model. By directly substituting the innovation measure for the patent measure, this research approach shows further support for Jaffe's findings and arguments. (SRD)