How Basic is (Patented) University Research? The Case of GM Crops

One of the main reasons for subsidising university research is the widespread belief that it generates proportionally more positive knowledge externalities than corporate research. Over the last two decades, however, this belief has been shaken by the increasingly aggressive patenting of university-based innovation. This perception was supported by Henderson, Jaffe and Trajtenberg (1998) who found both a sharp increase in university patenting and a decrease in the relative 'importance' of university innovation over the later part of their 1965-1992 sample. In this paper, we have compared the knowledge externalities generated by university and corporate patents related to GM crop research. Our main measure of knowledge externalities is the total number of third party cites generated by a patent. Our main result is that patented university research is not associated with greater knowledge externalities than corresponding corporate patents. If anything, corporate patents appear to generate greater numbers of net citations. This basic conclusion survives when we control for a number of variables that could affect citation counts (e.g. patent examiner effects) and when we break our sample into sub-periods. This does not imply that university patents are similar to corporate patents in every respect. We find two main differences. Firstly, there is some evidence that the shape of the distribution of citations is not identical for the two groups of patents as university patents appear to experience a more sluggish start than their corporate brethren. Secondly, even controlling quite narrowly for areas of specialisation, university patents receive a disproportionate number of cites from other university patents. These two results suggest that there are some fundamental differences in the types of knowledge flows generated by university and corporate patents.