Using patent data, this work provides an initial empirical investigation into university production of agricultural biotechnology patents. We develop a consistent theoretical methodology for understanding the university patent production process and then estimate econometric models of university-owned agricultural biotechnology patents on a series of explanatory variables. Among the issues addressed by the econometric models are (i) Do universities with more overall resources for research and development produce more ag-biotech patents? (ii) Do universities with more technology transfer employees produce more ag-biotech patents? (iii) What is the relationship between the emphasis of the university on agriculture and ag-biotech patent production? After investigating the patent production process, the analysis proceeds to consider spillover effects from university patenting in the form of licenses and the potential importance of "star" scientists in patent production.
The empirical investigations demonstrate the importance of the land grant system in ag-biotech innovation. Ag-biotech patent production does seem to respond to the existence of an infrastructure of agricultural colleges. We also demonstrate the importance of having a technology transfer office to the production of ag-biotech patents. We do however, find that there are decreasing returns to scale in the technology transfer bureaucracy. The spillover effects of ag-biotech innovations as a promoter of research in the local area, however, have not been demonstrated to be large at least in this data. Along with agricultural college infrastructure, own institutional support and the presence of "star" scientists proved to be potentially important. Since at many state universities much of this own funding comes from patent revenues, there is the potential for a virtuous cycle with better patent producing universities better able to produce more patents.