Licensing and technology transfer

Licensing agreements have the potential to reveal previously unobserved insights into technology transfer. Technology transfer has mainly been a theoretical topic relying on limited empirical data to draw hypotheses or tests. By utilizing a privately-held database of complete full-text technology licensing agreements collected from public U.S. sources, I explore technology transfer with licensing contracts as their proxy. The objective of this thesis is to develop a data-driven foundation for analyzing technology transfer.; First, given the apparent connection linking economic growth in nations with access to trade, I explore hypotheses relating to sectoral technology transfer incidence with theories of international trade. I compare previous technology transfer empirical survey results with my data, and use a probit methodology to examine pair-wise cross-sector technology licensing transactions over the entire U.S. economy. I show that intra-industry international trade theory provides testable technology transfer predictions, suggesting exploration into other aspects of international trade theory may lead to further insights.; Second, because empirical studies rarely define direct mechanisms of technology transmission, I break out technology transfer into its constituent components. I propose a general framework to model technology transfer that hypothesizes technology matching. I use a probit methodology on pair-wise cross-sectoral U.S. manufacturing industry technology licensing transactions. I observe evidence of matching in technology production levels between sectors, and that licensor capital spending intensity and licensee labor spending intensity appear to be positively correlated with the increased incidence of licensing.; Lastly, I build a framework to address how firms bargain over the license of a technology. My framework consists of a model where a licensor and licensee bargain over a two-part tariff royalty payment for the use of a pharmaceutical technology license. The model distinguishes itself in that the party with private information is the licensee who knows the value of the technology being licensed. I test and confirm the propositions derived from the model using a least squares methodology while controlling for sample selection.