Against the one-way-street: analyzing knowledge transfer from industry to science

Knowledge flows from industry to science have mostly been neglected by empirical studies. This work aims at contributing to this issue by analyzing differences in the factors that influence the probability of knowledge transfer within industry and from industry to science in the biotechnology sector. In order to model these knowledge flows we conduct a citation analysis on the basis of patent data. We then estimate a weighted bivariate probit model on the citation probability of industry and science on the basis of a combined sample of citing and cited patent pairs and an equal number of control patent pairs. The empirical results suggest that there are considerable differences in the citation probability. Cultural closeness for instance has a positive effect on the citation probability from industry to industry while the citation probability of scientific institutions is not affected by cultural distance.