Skip to main content



Patent systems are supposed to spur incentives to innovate 1) by granting temporary monopoly rights and 2) by facilitating follow-up inventions by making information about technical inventions available to the public. In other words, patents make information about inventions available to society at the cost of a “privatization” of this knowledge. This project makes use of a natural experiment to investigate the effects of “knowledge privatization” for academia and the private sector.

In general, it is hard to examine causal effects of patent rights since patent systems exist in all industrialized economies so that a counterfactual situation, i.e. how an economy would look like without patent rights, is missing as a benchmark. This project contributes to the previous literature on the effects of patents - that mainly reports correlations - by exploiting the introduction of software patents in the U.S. in the mid-1990s as a unique research setting that allows identifying causal effects.

The first part of the project investigates the extent to which patent rights impact scientific progress. Using a unique database that was created for this study, difference-in-difference analysis was performed that compares U.S. university scientists in the field of computer science to a benchmark group of European university scientists in the field of computer science that were not affected by a similar law change. The results show that software patents led to a smaller quantity of higher quality scientific publications in the U.S. Based on these results we can reject the concern that the introduction of patent rights had a negative impact on university science.

The second part of the project investigates whether patents rights lead to a brain drain from academia to the private sector. Preliminary results from a difference-in-difference analysis comparing U.S. university researchers to European peers in the field of computer science shows that overall there is no evidence for a brain drain from academia to industry.

The third part investigates the effects of software patents on corporate R&D investment. Results from fixed effects regressions show that - despite the concerns about the social and economic value of software patents - investment in R&D increased in the software industry relative to other U.S. industries after patent rights have been introduced. Capital investment in the relevant sector, in contrast, has not changed relative to other industry sectors in the same time frame. The finding is in line with the notion that intellectual property rights strengthen incentives to invest in R&D.

The last part of the project investigates the effect of patent rights on the concentration of R&D activities in an industry. On the one hand, patents should encourage R&D activities by granting exclusivity rights hence leading to less concentrated R&D activities in an industry. On the other hand, patents might lead to increased industry concentration if firms with a dominant market position use patents to establish or strengthen market entry barriers. Empirical results show that patent rights do not lead to a long-run monopolization of R&D in an industry. The spread of R&D activities is accompanied by a decline in the sales concentration suggesting that firms are able to transform their investment in R&D in sales increases.

The results of the project are of interest for policy makers being interested in science policy and technology transfer. The first part of the project illustrates that patent rights seem to not have negative effects on science production as many scholars fear. The research setting of the project allows to provide advice on Bayh-Dole Act-like policies, i.e. those policies that aim at strengthening universities’ intellectual property rights in order to foster technology transfer, because the introduction of patent rights operates the same mechanism as these policies that were introduced in many countries in the recent past. The advantage of the current research setting is that the introduction of patent rights was exogenous to universities’ previous technology transfer activities (rather than a response to unsatisfactory technology transfer activities) so that the effects that could be identified here are most informative. The results of the second part of the project are of interest for policy makers aiming to foster innovation in the private sector since they show that patent rights stimulate R&D activities and also reduce their concentration within an industry.