The NBER held its first Annual Conference on Innovation Policy and the Economy in Washington, D.C., on April 11. The conference was organized by NBER Research Associates Adam B. Jaffe of Brandeis University and Joshua Lerner of Harvard University, and NBER Faculty Research Fellow Scott Stern of MIT. The following papers were presented and discussed:
U.S. taxpayers funded $14.8 billion of health-related research last year, four times the amount that was spent in 1970 in real terms. Henderson and Cockburn evaluate the impact of these huge expenditures on the technological performance of the pharmaceutical industry. They conclude that the returns from this investment have been large and may be growing even larger. Public sector science creates new knowledge and tools and produces large numbers of highly trained researchers, all of which are a direct and important input to private sector research. Further, public science sustains an environment in which for-profit firms can conduct their own basic research, which in turn contributes to the global pool of knowledge. Measured quite narrowly in terms of its effect on private sector R and D, the rate of return to public funding of biomedical sciences may be as high as 30 percent per year. And these calculations are likely an underestimate, since they fail to fully capture the wider impact of pharmaceutical innovation on health and well-being. Indeed, the best may be yet to come: the continuing revolution in molecular biology that began in publicly funded laboratories 25 years ago, and continues to be driven by the academic research, promises dramatic advances in the treatment of disease.
Kremer describes the economic rationale for committing in advance to purchase vaccines once they are developed. Historically, governments have stimulated vaccine research by paying for research inputs, for example through grants to researchers. While this approach remains appropriate for basic research, the development of the biotech industry increases the scope for encouraging the later, more applied stages of vaccine research through committing to buy vaccines. This has several advantages over paying for research inputs. First, it gives researchers strong financial incentives to focus on developing a marketable vaccine rather than pursuing other goals, such as publishing academic articles. Second, paying for vaccines, rather than funding research expenditures, provides pharmaceutical firms and scientists with incentives to select only those research projects that have a reasonable chance of leading to a vaccine. With purchase commitments, the public pays only if a vaccine is actually developed. Finally, purchase commitments help to ensure that if vaccines are developed, they will reach those who need them. After outlining these advantages, Kremer explores how commitments to purchase vaccines could be designed.
Shapiro finds that in several key industries, including semiconductors, biotechnology, computer software, and the Internet, the current system is creating a "patent thicket":an overlapping set of patent rights requiring that those seeking to commercialize new technology obtain licenses from multiple patentees. The patent thicket is especially thorny when combined with the risk of "hold-up": the danger that new products will inadvertently infringe on patents issued after they were designed. The need to navigate the patent thicket and hold-up is especially pronounced in industries such as telecommunications and computing in which formal standard-setting is a core part of bringing new technologies to market. Cross-licenses and patent pools are natural and effective ways to cut through the patent thicket, but each involves significant transaction costs. Antitrust law and enforcement, historically hostile toward cooperation among horizontal rivals, can easily add to these transaction costs. Yet a few relatively simple principles, such as the desirability of package licensing for complementary patents but not for substitute patents, can go a long way toward ensuring that antitrust will help to solve the problems caused by the patent thicket and hold-up rather than to exacerbate them.
Why did commercialization of the Internet go so well? Greenstein examines events in the Internet access market and emphasizes three themes. First, commercializing Internet access did not give rise to many of the anticipated technical and operational challenges. Entrepreneurs quickly learned that the Internet access business was commercially feasible. Second, privatization fostered attempts to adapt the technology to new uses, locations, market settings, and applications, and to extend it to other lines of business. This went well beyond what anyone would have forecast by examining the limited noncommercial uses for the technology prior to 1992. Third, the National Science Foundation was lucky. It commercialized the Internet access industry at a propitious moment, simultaneous with the growth of the World Wide Web. The Web thrived under decentralized and independent decisionmaking, fueling further growth.
Mowery and Ziedonis find that the Bayh-Dole Act had only a modest effect on the content of academic research and patenting at Stanford and the University of California. The most significant change in the content of research at these universities was the rise of biomedical research and inventive activity. But the Bayh-Dole Act had little to do with this growth; indeed, the rise in biomedical research and inventions in both of these universities predates the passage of the Bayh-Dole Act. Both University of California and Stanford University administrators intensified their efforts to market faculty inventions in the wake of the Bayh-Dole Act. This enlargement of the pool of marketed inventions appears to have reduced the average "yield" (defined as the share of license contracts yielding positive revenues) of this population at both schools. But Mowery and Ziedonis find no decline in the "importance" or "generality" of the post-1980 patents of these two universities. In contrast, the patents issued to other U.S. universities that entered into patenting and licensing only after the effective date of the Bayh-Dole Act were less important and less general than the patents issued before and after 1980 to U.S. universities with longer experience in patenting. The patents of inexperienced academic patenters appear to have proved less significant (in terms of the rate and breadth of their subsequent citations) than those issued to more experienced university patenters. Therefore, the effects of the Bayh-Dole Act on entry may be as important as its effects on the internal "research culture" of U.S. universities in explaining the widely remarked decline in the importance and generality of U.S. academic patents after 1980.
Romer suggests that innovation policy in the United States has erred by subsidizing the private sector demand for scientists and engineers without asking whether the educational system encourages the supply response necessary for these subsidies to work. He suggests that the existing institutional arrangements in higher education limit this supply response. Romer discusses specific programs that could directly increase the numbers of scientists and engineers available to the private sector.
These papers will be published in an NBER "Annual" by the MIT Press. The availability of the volume will be announced in a future issue of the NBER Reporter. These papers also can be found at Books in Progress.