solutions. She emphasized that there were no new disciplines in that list, and she suggested that climate change is perhaps marginally different from other topics of study, but does not necessarily require different disciplines or fundamentally different approaches, merely more sustained efforts at interdisciplinary work.

Nordhaus also noted that more consideration must be given to the complementarities between public and private R&D, both in terms of synergy and also whether or not public R&D may crowd out some private R&D. John Weyant suggested that there may be lessons in looking at other innovation systems, such as the National Institutes of Health (NIH). Weyant also pointed out that there are certain gaps in the innovation chain that are not filled because they fall between basic science research and venture capital opportunities, since venture capitalists tend not to take large technological risks. On the subject of risk, Nakicenovic mentioned that IIASA has used its mathematical MESSAGE model to treat uncertainty explicitly with regard to technology investment risk. An assumption was that investors were willing to pay a risk premium to hedge against some of that risk—as the risk premium approaches 5 percent, the dynamics of the entire system fundamentally change. Investments in the lower-cost options start to happen earlier and there will be more diversity in terms of technologies, and the more costly and risky technologies get introduced as well. Weyant also mentioned that in the field of robotics, most big innovations are based on a series of old patents, with one or two new patents building on these to lead to breakthroughs. He related this notion to the common conception of innovation being one of bold pathbreaking changes, which overlooks the minor breakthroughs that might bridge the gap enough to make new technologies commercially viable.

Bryan Hubbell relayed the anecdote of research funding for land grant universities working on genetically modified crops in the late 1980s and early 1990s—this basic research was designed to maximize spillovers and thus public benefit. However, as public funding tightened and then as public/private cooperatives emerged, the nature of the research changed to one that would minimize spillovers and thus allow private entities to capture the rents. He questioned whether or not this situation could be managed differently with regard to paradigm-changing energy and climate technologies. Adele Morris also noted the important linkage between R&D and international participation, specifically that there are potentials for international spillovers, and this information could inform international negotiations with an eye toward maximizing the global impact of such spillovers.

Skip Laitner mentioned that Moore’s law is not a physical law, but an extrapolation that has become a self-fulfilling prophecy that continues to be driven by business models. Graham Pugh concurred and noted that there may be lessons from the semiconductor industry’s experience with the research collaborative Sematech. Sematech is a pre-competitive R&D consortium, whereby leading semiconductor companies pooled resources and worked collaboratively. From the industry point of view, the costs were too great to be borne by any one company, and this collaborative effort allowed them to move toward the production frontier in a pre-competitive model, driven by the perceived need for constant innovation.

Communicating Results

To conclude the workshop, participants discussed how to take ideas forward and improve communication channels between policymakers and the analytical community. As Francisco de la Chesnaye and others remarked, it is incumbent on analysts to spend more time comparing and synthesizing similar analyses to better communicate their insights. A participant questioned whether reduced-form models that could be operated by congressional staff or other lay people might be useful, but Dick Goettle replied that after 20 seconds, a consumer would begin asking the detailed questions that only the more-complicated models can answer. Computer time is cheap and there are good models out there, and so he advised that the full-form models be run.

Ed Rubin’s simple advice to analysts was to “get the sign right,” a reference to the need to better communicate where and why there are negative cost opportunities to be had. This message seems to get lost when discussing the overall costs to the economy.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement