Empirical Analysis of Industry-University Collaboration and Research Productivity

Abstract

In this study, the empirical analysis of the effect of industry-university collaboration on the research productivity of firms is carried out using Japanese firm-level panel data. In consideration of estimating the model in which the dependent variable is the number of patents and the number of papers, the panel-Poisson model, which is a count data model, was used for estimation. As a result, industry-university collaboration has a positive effect on R&D productivity of firms. Even if the scale of R&D input, the firms' absorptive capacity, and the industry characteristics and time trends are controlled, the number of patents and papers by firms publishing with university are larger than that by firms with no publication by industry-university collaboration. The number of patents and papers by firms collaborating with university tends to be larger than that by firms not collaborating. The same results could be found in the effect of the number of industry-university collaborative papers, the number of industry-university joint patents, and the amount of research expenditures paid to universities on the number of patents and papers by firms.