Abstract
The aim of this study is to investigate the invention performance (direct effect) and the organizational influx effects (indirect effect) of collaborative research between academia and pharmaceutical industry in Japan by conducting quantitative study with the dataset of patent applications and the directory service of researchers in Japanese universities. In the first part of this study, we evaluate the quality of patents (direct effect) arising from the collaborative research projects by using Japanese patent application data from 1976 to 2005. Since the pharmaceutical invention takes very long period to become a product, this study introduces four variables as phase-wise performance indicators of patent quality: (A) whether the patent covers a product or not, (B) whether the patent was registered or not, (C) the number of countries where the patent was filed, and (D) the number of forward citations by examiners. The result of the analysis shows no advantage of the IU invention (industry-university co-invention). What is more, the patent quality of IU inventions is significantly lower than that of the internal inventions in terms of (C) and (D). However, in some disease areas, IU inventions get better in quality than internal inventions. In the second part of this study, in order to understand the other aspect of IU inventions, we focused on the effect of information influx (indirect effect) from the collaborative researches by analyzing annual sums of forward citations from the internal inventions and those from IU inventions. The result shows the quality of internal inventions rapidly and transiently soar after the increase in quality of IU inventions. This suggests information and technology influx from universities encourage internal researches. These findings indicate that collaborative researches, although their direct effects seem to be low, play an important role in accelerating internal research.
