The Market for Intellectual Property: Evidence from the Transfer of Patents", Working

This paper studies the economics of intellectual property transfer, using new data on patent transfers. First, it presents the key stylized facts of the transfer of patents from a dataset I have compiled. The main stylized facts are four: (i) Nearly 20 % of all U.S. patents are traded at least once over their life cycle. (ii) Better patents, as defined as those receiving more citations, are more likely to be traded. (iii) The transfer rate, defined as the proportion of traded patents conditional on renewal, monotonically decreases following the date of issue with the exception that it jumps up immediately after a patent is renewed. (iv) Previously traded patents, and in particular those traded recently, are more likely to be retraded and less likely to expire. Second, this paper develops a model with costly technology transfer to interpret the stylized facts. The model has two mechanisms, a selection effect and a horizon effect. The cost of technology transfer creates a selection effect so that better patents are more likely to be traded. This effect accounts for the discontinuous increase in the transfer rate after renewal, and also for why traded patents are more likely to be retraded and less likely to expire. There is a horizon effect that explains why transfer rates decrease as the patent gets closer to its expiration date. A shorter horizon implies less time to amortize the cost of technology transfer. Finally, I structurally estimate the parameters of the model to quantify what are the gains from trade in the market for patents. I find preliminary estimates that show that the existence of this market adds 5.08 % to the average value of a patent. ∗I am especially grateful to Thomas Holmes and Sam Kortum for their advice and patience. I would als