Notes from the National Bureau of Economic Research Conference on April 12

By Susan Finston, Board Chair, BayhDole25, Inc.:

On April 12, 2007, I attended the National Bureau of Economic Research (NBER) conference, Innovation Policy and the Economy, organized by Adam Jaffe, Joshua Lerner, and Scott Stern.

The NBER half-day conference took on a central question facing U.S. policymakers in the coming months: does the U.S. patent regime and related technology transfer system harm innovation and growth by favoring right-holders to the detriment of overall growth? Despite the repeated insistence by Robert Litan (Kauffman Foundation, Brookings) that "the system is broken," other speakers, most notably Edward Lazear, Chairman of the Council of Economic Advisors (CEA), provided compelling data that the U.S. economy continues to promote innovative technologies and to create jobs.

While the entire morning was useful and interesting for those with an interest in innovation and economic policy, the first session on university research and commercialization, provided the most fireworks, to the extent possible at a gathering of generally collegial academics, government officials and consultants. This blog focuses on that discussion.

The conference program is available online at wiki.cepr.org/sing/index.php?title=The_NBER's_Innovation_Policy_and_the_Economy_Conference.

As many are aware, U.S. universities have long served as incubators for technology companies, both in terms of IT and biotech. Over time, this has created in the U.S. an effective system for job creation, leading to substantially lower unemployment and higher productivity growth over time as compared to Europe or Japan. This is undeniable. However, persistent criticisms remain that increased patenting in universities is stifling research and innovation.

The first paper, presented by John Walsh (Georgia Tech) and co-authored by Wesley Cohen (Duke and NBER, not present), provided persuasive evidence that new academic research is affected either way by increasing prevalence of academic patenting. This very interesting and accessible paper points out that it is in fact the non-patent aspects of research are empirically much harder to gain access to than the patented information: Materials (cell-lines, GMOs, etc.) and Data which the researchers would have to actively agree to share and then to ship to other interested researchers.

In fact, the data presented by Walsh and Cohen show that only a small percent of researchers pay attention to patenting when initiating new research, demonstrating that it has little if any impact on scientific research in universities. Only 8% of researchers knew about patents in their field, and even fewer, 5% regularly check patent status. Moreover, this had little negative impact on their research projects in practical terms.

The authors also point out that the so-called "private costs of excluding" other researchers from patented areas may be very high and have little impact, whereas it is much easier to exclude other researchers on the basis of materials and data which may actually impose costs on the original researcher to share, although NIH grant terms may require some level of cooperation. Overall, the conclusion is that actual sharing of data and materials is actually increasing, though in the study 19% did not receive requested academic (non-patent) inputs over the time period studied. This is for various reasons —some academic researchers receive literally hundreds of requests and the volume imposes too high a burden. However, it is clear that the issue of patent protection was not significant.

In direct contrast, Robert Litan (Kauffman, Brookings), Lesa Mitchell (Kaufman) and E.J. Reedy (Kauffman) painted a much darker picture, based largely on anecdote, of a system that Robert Litan described stifling of innovation and commercialization of science. Unfortunately, the Kauffman paper, in contrast to others presented on the 12th, included little direct data to back up the claims made by the authors. To read it and judge for yourself, see: www.ssrn.com/abstract=976005 (Large companies like IBM are on record as favoring policies of "work for hire" vs. university patent licensing. While this would clearly be more profitable for well-established MNCs, it is not clear why universities and start-ups should be disadvantaged by lack of patent rights for the benefit of MNCs. )

Based on his anti-trust expertise, Robert LItan described Technology Transfer Offices (TTOs) as "monopolies," by which he appeared to mean that they should be broken up and alternate pathways provided to commercialization. In fact, there is a great deal of research by Tony Heher, formerly of University of Cape Town, South Africa) and others, indicating that TTOs need greater financial and staff resources and substantive support. And there are organizations, notably MIHR and PIPRA, doing good work in this area. However, the Kauffman paper does not reflect an awareness of these efforts. Participants at the NBER conference, understandably expressed skepticism as to both paper's depth of analysis and conclusions. That isn't to say that TTOs don't need more support and resources, they clearly do, and Kauffman may be doing some important work in that area as well.

The bottom line: it is good to base policy on empirical data, to the greatest extent possible, to avoid unintended consequences of policy changes. Understanding that "a swallow is not a summer," it appears from the data presented that university patents do not pose any significant barrier to ongoing university research. In the area of TTO changes suggested in the second paper, though, data is lacking to demonstrate need for the changes proposed by Kauffman, and how they would affect the technology transfer regime in the U.S., for ill or for good.