Wednesday, June 18, 2014

Love Surveys University Researchers on Patenting

While there have been extensive debates about patenting federally funded research under the Bayh-Dole Act, there is little evidence about what the researchers themselves think of this regime. Professor Brian Love (Santa Clara Law) has tackled this problem with a survey of electrical engineering and computer science professors, and he describes the results in Do University Patents Pay Off? Evidence from a Survey of University Inventors in Computer Science and Electrical Engineering (forthcoming Yale J.L. & Tech.). Love sent an email survey to all 2,387 tenure-track faculty in the top 20 ECE and CS departments (as ranked by U.S. News), and his 269 respondents were highly representative based on measurable characteristics. Here are some of his findings:
  • 66% of respondents have filed a university patent application (median: 4); 55% have an issued patent (median: 3); 35% have a licensed patent (median: 2).
  • Researchers with licensed patents reported that those licenses brought in a mean of $140k and median of $9k for their universities. 25% of licensors said their licenses had brought in $0. Based on estimates of the amount spent on prosecution, Love concludes that high-tech patenting appears to be a net loss for these 20 universities in the aggregate. However, Love also reports that respondents had little knowledge of royalty-sharing agreements with inventors, so there may have been royalties they didn't know about.
  • Only about 10% of respondents agreed that the prospect of obtaining patents encouraged them to do more or better research than they would have otherwise. No respondents ranked patent rights as their #1 motivation for doing research; the most common responses were (1) enjoyment / desire to advance knowledge; (2) publication; (3) peer recognition; and (4) promotion.
  • Many respondents stated that university patents harm their ability to bring in research funding, to collaborate with professors at other institutions, and to share their discoveries (see Figures 5-7).
  • Evidence was mixed on whether patents help commercialization, which is the most compelling justification for Bayh-Dole patents. Of professors who had founded a startup, 42% said that patents help commercialize university research, while 24% said that patents hurt.
  • 51% of respondents think patenting is taken into account in tenure decisions, and 38% think patents impact raises. And over 1/3 of respondents believe patents enhance their own and their universities' reputations.
Anyone interested in Bayh-Dole policy should download the paper and take a closer look, especially given the dearth of other data in this area. As Love notes, it is hard to draw firm policy conclusions, but these results do raise questions about the societal benefit of universities' current approach to patenting in high-tech fields, especially because one might expect patenting to be most successful at the top schools. The results also suggest that universities could do a better job of working with their researchers to improve the patenting system, such as by being more transparent about royalty sharing programs or more judicious in seeking only those patents that are most likely to be beneficial.