The Introduction is divided into three main parts. The first is a background section summarizing the policy context that led to the development of the Advanced Technology Program (ATP) and a description of the program itself. The second part is a summary of the independent papers prepared for this volume. The third is a summary of the highlights of the symposium organized as part of the National Research Council (NRC) review of the ATP. 1
At this writing, the American economy continues to enjoy steady growth, with inflation under tight control, unemployment at historically low levels, and productivity at higher levels than previous decades. 2 It is hard to remember that only 10 years ago many thought the American economy had been surpassed by the combination of patient capital, skilled engineering, and protected domestic markets that characterized the Japanese and other Asian economies. In the 1970s and 1980s, the United States recorded slow economic growth relative to post-war1 This is the second report on the ATP issued as part of the NRC review. For an explanation of the project, its origins, and areas of focus, see the Preface.
2 Productivity increased after 1995. There is some debate about its sources and sustainability. See Dale Jorgenson and Kevin Stiroh, “Raising the Speed Limit: U.S. Economic Growth in the Information Age,” Brookings Papers on Economic Activity, 2000, pp. 125-235. See also Robert Gordon, “Has the ‘New Economy' Rendered the Productivity Slowdown Obsolete?” Manuscript. Northwestern University. June 12, 1999. See also Council of Economic Advisors, Economic Report of the President, Washington, D.C.: U.S. Government Printing Office, January 2001, ch. 1, especially pp. 25-32.
norms, sluggish productivity performance, and a loss of global market share and technological leadership in key U.S. industries, from steel and automobiles to televisions and semiconductors. There was also considerable concern about a rapidly rising trade deficit. 3 The causes of America's sub-par economic performance defied definitive analysis, but dire warnings of U.S. economic decline and the “deindustrialization” of key manufacturing sectors proliferated. 4 U.S. trade competitors, such as Japan, seemed to have developed an effective economic model different in important respects from what Americans believed to be the traditional laissez-faire American approach. 5 A key feature of that model was its emphasis on cooperation between government and industry rather than competition. The ability of different arms of Japanese industry to work with one another, and the close relationship between government and industry in supporting key economic sectors, had created substantial benefits for the Japanese economy. 6
New Cooperative Programs and Policies
A series of public and private initiatives in the 1980s demonstrate the renewed emphasis on cooperation in U.S. public policy, especially a number of major legislative initiatives passed by the Congress. These included the Stevenson-
3 This too has changed. In most quarters, there seems to be little concern about the trade deficit, although it continues to increase. For an analysis of the sustainability of the deficit, see Catherine Mann, Is the U.S. Trade Deficit Sustainable? Washington, D.C.: Institute for International Economics, 1993. The U.S. Trade Deficit Review Commission was established in October 1998 to review the trade deficit and its implications for the economy as a whole. See
4 Questions persist concerning the degree of the U.S. decline in the 1980s, just as questions remain concerning the sustainability of the current recovery. The STEP Board has recently completed a review of the competitive resurgence of the U.S. economy. It includes an assessment of the factors that have contributed to the U.S. recovery with a focus on eleven U.S. manufacturing and service sectors. See National Research Council, U.S. Industry in 2000: Studies in Competitive Performance, D. Mowery, editor, Washington, D.C.: National Academy Press, 1999.
5 For a discussion of the different economic and cultural assumptions underpinning these differences, see Clyde V. Prestowitz, Trading Places, New York: Basic Books, 1988, Chapter 5; and James Fallows, Looking at the Sun: The Rise of the New East Asian Economic and Political System, New York: Pantheon Books, 1994, Chapter 4, pp. 194, 210. For an overview of these issues see National Research Council, Conflict and Cooperation in National Competition for High-Technology Industry, Washington, D.C.: National Academy Press, 1996, pp. 12-40. For a review of the main features of the East Asian economic success story, see World Bank, The East Asian Economic Miracle: Economic Growth and Public Policy, Policy Research Report, New York: Oxford University Press, 1993.
6 For the best early analysis of the Japanese approach, see Chalmers Johnson, MITI and the Japanese Miracle: The Growth of Industrial Policy 1925-1975, Stanford, CA: Stanford University Press, 1982. See also D. T. Okimoto, “The Japanese Challenge in High Technology Industry,” in National Research Council, The Positive Sum Strategy, R. Landau and N. Rosenberg, editors, Washington, D.C.: National Academy Press, 1986; and D. T. Okimoto, MITI and the Market: Japanese Industrial Policy for High Technology Industry, Stanford, CA: Stanford University Press, 1989.
Wydler Technology Innovation Act (1980), the Bayh-Dole University and Small Business Patent Act (1980), the Small Business Innovation Development Act (1982), the National Cooperative Research Act (1984), the Federal Technology Transfer Act (1986), the Omnibus Trade and Competitiveness Act (1988), the National Competitiveness Technology Transfer Act (1989), and the Defense Conversion, Reinvestment, and Transition Assistance Act (1992). These acts are summarized in Box B.
Box B. Principal Federal Legislation Related to CooperativeTechnology Programs 7
7 Drawn, with National Research Council modifications, from Christopher Coburn and Dan Berglund, Partnerships: A Compendium of State and Federal Cooperative Technology Programs, Columbus, OH: Battelle Press, 1995, p. 485.
any R&D collusion was an automatic violation, regardless of a determination of economic damage.
Federal Technology Transfer Act (1986) Amended the Stevenson-Wydler Technology Innovation Act to authorize Cooperative Research and Development Agreements (CRADAs) between federal laboratories and other entities, including state agencies.
Omnibus Trade and Competitiveness Act (1988) In addition to establishing a Competitiveness Policy Council designed to enhance U.S. industrial competitiveness, the Act created several new programs (e.g., the Advanced Technology Program and the Manufacturing Technology Centers) housed in the Department of Commerce's National Institute of Standards and Technology and intended to help accelerate development, commercialization, and application of promising new technologies and improve manufacturing techniques of small and medium-sized manufacturers.
National Competitiveness Technology Transfer Act (1989) Part of the Department of Defense authorization bill, this act amended the Stevenson-Wydler Act to allow government-owned, contractor-operated laboratories to enter into cooperative R&D agreements.
Defense Conversion, Reinvestment, and Transition Assistance Act (1992) Initiated the Technology Reinvestment Program (TRP) to establish cooperative, interagency efforts that address the technology development, deployment, and education and training needs within both the commercial and defense communities.
The Demise of Spin-offs
Another element in the economic and policy landscape at the time was the demise of the “spin-off” paradigm in defense procurement. 8 For years it was believed that investments in sophisticated defense systems had beneficial spillovers into commercial markets. Companies that developed and manufactured high-technology armaments were believed to be building a technological base that would enable them to compete effectively in commercial markets. 9 For a number of reasons—from burdensome government procurement regulations to8 For a discussion of the demise of the spin-off paradigm, see John Alic et al., Beyond Spinoff: Military and Commercial Technologies in a Changing World, Boston: Harvard University Press, 1992.
9 In some cases, such as aircraft and computing in the early years, this assumption did hold. See National Research Council, U.S. Industry in 2000, op. cit., pp. 184-189, and National Research Council, Funding a Revolution: Government Support for Computing Research, Washington, D.C.: National Academy Press, 1999, passim.
accelerating time-to-market demands in commercial markets—this paradigm no longer applied by the 1980s. 10 In fact, it was becoming obvious that firms competitive in commercial markets are often better able than firms with only military customers to provide defense systems with the most advanced capabilities, particularly in rapidly evolving sectors such as semiconductors. 11 Policy makers were therefore searching for ways to improve private-sector commercialization rates, primarily to enhance U.S. competitiveness, but ultimately for national security reasons as well.
One of the strategies adopted by the United States in response to its perceived loss in competitiveness, at least in some sectors, was to encourage greater cooperation among industry and between industry and government. Such collaboration was by no means novel in the U.S. economy. 12 As noted in the Preface, government funds had supported the demonstration and development of the telegraph in the 19th century, and after World War I the federal government fostered an independent radio industry. 13 The federal government also provided active support through a variety of mechanisms for military and civil aviation and10 Some analysts argue that the U.S. defense acquisition system, sometimes accused of providing disguised subsidies for commercial industry, has in fact created disincentives and barriers to the operation of market forces. These include “the unique government oversight requirements, the unique procurement requirements, (and) the unique military specifications” associated with military procurement. See the presentations of Jacques Gansler in National Research Council, International Friction and Cooperation in High-Technology Development and Trade, Charles W. Wessner, editor, Washington, D.C.: National Academy Press, 1997. See also Richard Samuels, Rich Nation, Strong Army: National Security and the Technological Transformation of Japan, Ithaca, NY: Cornell University Press, 1994.
11 In response to changing procurement needs, the Clinton Administration adopted a “dual use” strategy for defense procurement. See National Research Council, Conflict and Cooperation, op. cit., pp. 153-158. See also the presentations of Paul Kaminski and Jacques Gansler in National Research Council, International Friction and Cooperation in High-Technology Development and Trade, op. cit., pp. 130-152.
12 Private cooperation also expanded. For example, in 1983, fourteen companies—mostly computing manufacturers, but also semiconductor, aerospace, and defense firms—banded together to form the Microelectronics and Computer Technology Corporation (MCC). For a review of the origins of MCC and SEMATECH, see John Horrigan, “Cooperating Competitors: A Comparison of MCC and SEMATECH,” Monograph, Washington, D.C.: National Research Council, 1999.
13 Josephus Daniels, Secretary of the Navy during the Wilson Administration, appeared to feel that monopoly was inherent to the wireless industry, and if that were the case, he believed the monopoly should be American rather than British. Britain had dominated pre-war Atlantic wireless traffic as well as the undersea telegraph cable. By pooling patents, providing equity, and encouraging General Electric's participation, the Radio Corporation of America was created. It proved useful in the early 1940s. See Irwin Lebow, Information Highways and Byways: From the Telegraph to the 21 st Century, New York: IEEE Press, 1995, pp. 97-98 and Chapter 12.
the electronics industry. 14 Yet the 1980s and early 1990s saw a conscious effort to expand cooperation, in part by using federal R&D funding more effectively to meet what were seen as unprecedented competitive challenges. As a recent study of federal support for the computer industry observed, most federal support for industry prior to the mid-1980s took the form of research grants or contracts for product development or procurement that often included substantial support for research. 15 In the latter half of the decade, a growing number of programs were established to benefit from partnerships among government, industry, and universities. These included (1) the Semiconductor Research Corporation, which pools industry and limited federal funding to support university research in semiconductors; (2) SEMATECH, which matched substantial federal and industry funding in a consortium of semiconductor manufacturers; 16 (3) NSF Engineering Research Centers that involve industry-university collaboration on engineering problems; (4) expanded Cooperative Research and Development Agreements (CRADAs), particularly at the Department of Energy; and (5) extramural programs at the National Institute of Standards and Technology (NIST).
Some of the other major federal partnerships of this period were the Department of Defense's Manufacturing Technology (MANTECH) Program; the Department of Transportation's Intelligent Vehicle Highway Systems (IVHS) Program and National Magnetic Levitation Initiative (MAGLEV); the National Science Foundation's Research Centers Program (which includes the Engineering Research Centers, the Industry/University Cooperative Research Centers, the Materials Research Science and Engineering Centers, and the Science and Technology Centers); and the Small Business Technology Transfer (STTR) Program. 17
14 D. Mowery and N. Rosenberg, Technology and the Pursuit of Economic Growth, New York: Cambridge University Press, 1989, Chapter 7, especially pp. 181-194. The authors note that the commercial aircraft industry is unique among manufacturing industries in that a federal research organization, the National Advisory Committee on Aeronautics (NACA—founded in 1915 and absorbed by NASA in 1958), conducted and funded research on airframe and propulsion technologies. Before World War II, NACA operated primarily as a test center for civilian and military users. NACA made a series of remarkable contributions with regard to engine nacelle locations and the NACA cowl for radial air-cooled engines. These innovations, together with improvements in engine fillets based on discoveries at Caltech and the development of monocoque construction, had a revolutionary effect on commercial and military aviation. These inventions made the long-range bomber possible, forced the development of high-speed fighter aircraft, and vastly increased the appeal of commercial aviation. Lebow, Information Highways and Byways, op. cit.; and Alexander Flax, National Academy of Engineering, personal communication, September 1999.
15 National Research Council, Funding a Revolution, op. cit., pp. 32-33.
16 In 1996, SEMATECH became a completely private-sector consortium. In 2000 it became International SEMATECH, a consortium that includes Asian and European companies.
17 Coburn and Berglund, Partnerships, op. cit., p. 488. Other examples are listed in the
More Technology Opportunities, More Partnerships
In parallel with the competitive challenges facing the United States, rising development costs for new technologies, the dispersal of technological expertise across firms, and the growing importance of regulatory and environmental issues provided additional incentives for public-private cooperation in many high-technology industries. Reflecting these convergent trends, collaborative programs have expanded substantially, with perhaps as many as 70 federal cooperative technology programs currently under way. 18 This trend has continued across several administrations. It was under the Reagan Administration that, after much debate, SEMATECH was established in 1988 to address the competitive challenge facing the semiconductor industry. 19 As noted earlier, the first Bush administration oversaw the initial implementation of the Advanced Technology Program in the National Institute of Standards and Technology, and that administration recommended substantial increases for the program in its FY1993 budget. 20
. . . and More Opposition
If the notion of additional government-industry collaboration was well developed by the late 1980s, it was by no means universally accepted. One of the earliest calls for increased collaboration, the President's Commission on Industrial Competitiveness, was released in 1983, but few of its recommendations were adopted. 21 The recommendations for expanded cooperation made by a subsequent study effort, the National Advisory Committee on Semiconductors (NACS), also met with limited success, although it served to highlight the importance of this strategic sector to the U.S. economy. 22 Even a broadly supported initiative such as SEMATECH, whose government funding in hindsight may have always
Ibid., p. 481.
19 The Semiconductor Industry Association formally proposed the SEMATECH consortium in May 1987. In 1996 federal funding ceased. In early 2000 the consortium became International SEMATECH, with a global membership roster. For an overview of SEMATECH's early operation and contribution see National Research Council, Conflict and Cooperation, op. cit., pp. 141-151. For one of the most comprehensive assessments of SEMATECH, see John B. Horrigan, “Cooperating Competitors: A Comparison of MCC and SEMATECH,” op. cit.
20 See the budget bar graph illustrating the evolution of program appropriations in
Chapter III below.
21 See President's Commission on Industrial Competitiveness, Global Competition: The New Reality, Washington, D.C.: U.S. Government Printing Office, 1985, 2 vols.
22 See National Advisory Committee on Semiconductors, Semiconductors: A Strategic Industry at Risk, A Report to the President and the Congress, Washington, D.C.: Semiconductor Industry Association, 1989. See also National Advisory Committee on Semiconductors, Toward a National Strategy for Semiconductors, Washington, D.C.: Semiconductor Industry Association, 1991.
seemed secure, encountered serious opposition at its inception and again at its renewal. 23
The Clinton Administration came to office with an emphasis on civilian technology programs, seeking to realign military and civilian R&D expenditures, and it encouraged the use of partnerships between government and industry to restore U.S. competitiveness. 24 As part of this effort, the Clinton Administration substantially expanded the ATP, created the Technology Reinvestment Program (TRP) to facilitate adjustment to the end of the Cold War, and established the Program for the Next Generation Vehicle (PNGV). 25 The rapid expansion of these programs generated significant opposition and may have contributed to the rekindling of the national debate on the appropriate role of the government in fostering new technologies. 26
The Need for Analysis and Assessment
Despite the growth in government-industry collaboration, there has in fact been remarkably little objective analysis of these collaborative efforts. For example, even longstanding programs such as the Small Business Innovation Research (SBIR) Program, created in 1982, have seen very limited external evaluation. 27 At one level, more frequent analysis and review of the partnership
23 The 1992 renewal was undertaken in the era in which the President's economic advisors purportedly saw little difference between silicon chips (semiconductors) and potato chips. Today there is broader agreement that the composition of the economy matters and that high-technology industry has special benefits such as more rapid growth rates, greater R&D expenditure, and, often, higher wages as compared with more traditional industries.
24 See William J. Clinton and Albert Gore, Jr., Science in the National Interest, Washington, D.C.: Executive Office of the President, 1994.
25 For an overview of ATP see Christopher T. Hill, “The Advanced Technology Program: Opportunities for Enhancement,” in Lewis Branscomb and James Keller, editors, Investing in Innovation: Creating a Research and Innovation Policy, Cambridge, MA: MIT Press, 1998, pp. 143-173. For an excellent analysis of the TRP, see Jay Stowsky, “Politics and Policy: The Technology Reinvestment Program and the Dilemmas of Dual Use,” Mimeo, University of California, 1996. See also Linda R. Cohen, “Dual-use and the Technology Reinvestment Project,” in Branscomb and Keller, Investing in Innovation, op. cit., pp. 174-193.
26 As David Hart notes, opposition was both internal and external to the administration. Those members of the administration whose priority was to reduce the budget deficit scaled back support for public-private R&D. At the same time, some Members of Congress professed to believe that President Clinton and his party might reap political benefit from the ATP awards. Following the 1994 election, the new congressional majority opposed the Clinton technology policy initiatives and also called for the abolition of the Departments of Energy and Commerce. David M. Hart, Forged Consensus: Science, Technology, and Economic Policy in the United States, 1921-1953, Princeton: Princeton University Press, 1998, p. 230.
27 The operations of the SBIR at the Department of Defense were recently reviewed by the Partnerships Program, and the PNGV has been subject to a regular review, also by the NRC. See National Research Council, The Small Business Innovation Research Program: An Assessment of the Department of Defense Fast Track Initiative, Charles W. Wessner, ed., Washington, D.C.: National Academy Press, 2000; and National Research Council, Review of the Research Program of the Partnership for a New Generation of Vehicles: Sixth Report, Washington, D.C.: National Academy Press, 2000.
programs under way might contribute to a better appreciation of the role of collaboration between government and industry in the development of the U.S. economy. Writing 20 years ago, one well-known American economist, Columbia University's Richard Nelson, observed that Americans are still remarkably uninformed about the long history of governmental policies aimed at stimulating innovation. 28 A recent comprehensive report on government support for computing research opens by stating that “It is difficult to recall and acknowledge” the federal government's major role in launching and sustaining the computer revolution, both in terms of innovation and infrastructure. 29 Today, many Americans seem to appreciate the contribution of new technologies to the robust economic growth of the past decade. Yet there is little evidence that Americans are aware of the contributions of federal support for technological innovation, from radio to computers to the Internet. 30
In addition to a better understanding of how the U.S. economy developed and the role government has played in fostering that development, careful assessment of support for new technologies is necessary, because government intervention in the market can be fraught with risk. For example, as Roger Noll and Linda Cohen document, in the 1970s the federal government supported a number of large commercial demonstration projects, such as the Synthetic Fuels Corporation and the Clinch River Breeder Reactor. The government also provided substantial funds for the development of the Supersonic Transport. Several of these large-scale technology commercialization programs were sources of major frustration. 31
Noll and Cohen describe the “political capture” of several of the large, commercial demonstration programs, which resulted in sustained funding even in the face of declining technical feasibility. Their analysis stresses the need for financial stability on one hand (a problem for the ATP, see below) and flexible man
28 Otis L. Graham, Losing Time: The Industrial Policy Debate, Cambridge, MA: Harvard University Press, 1992, p. 250. Graham cites Richard Nelson's observations at the end of the Carter Administration; the situation may not have improved. Writing in 1994, James Fallows makes a similar observation. See Looking into the Sun, op. cit., p. 196. See also Thomas McCraw's “Mercantilism and the Market: Antecedents of American Industrial Policy,” in Claude E. Barfield and William A. Schambra, editors, The Politics of Industrial Policy, Washington, D.C.: American Enterprise Institute for Public Policy Research, 1986, pp. 33-62.
29 National Research Council, Funding A Revolution, op. cit., p. 1.
Ibid., pp. 85-135 and pp. 169-183. The report provides a thorough review of government support for computers, the Internet, and related technologies and infrastructure.
31 See Linda R. Cohen and Roger G. Noll, The Technology Pork Barrel, Washington, D.C.: The Brookings Institution, 1991, pp. 97-148, 217-320. The volume is a valuable review of large-scale technology development programs drawn from the 1970s that is still relevant to current policy and commitments, such as fusion reactors and clean coal (p. 36). The analysis is more balanced than its title. In addition to major failures, the authors identify technically successful R&D projects, such as the photovoltaic electricity program (p. 363). The focus of the authors' analysis is not directed to cost-shared programs such as the ATP.
agement on the other hand, willing to accept project failures as a necessary condition for investments in new high-risk technologies. The authors do recognize the contributions of federal policy to the development of new technologies and, more generally, endorse the notion that if research and development are important to attaining national goals, then the adequacy of the R&D effort in the private economy is a matter of concern. 32
Box C. R&D Programs: The Challenge for Policymakers
Seeking to ascertain the factors that lead to successful programs, Cohen and Noll suggest that “The core challenge to public officials is not only to find ways to organize and manage R&D programs so that they are free from large swings in annual expenditures, but also to be sensitive to the fact that because R&D projects are risky, some are likely to fail and should be cancelled or redirected in midstream.”
—The Technology Pork Barrel, op. cit., p. vii.
There are also cases of major success resulting from federal support. 33 A significant example, of course, is the computer and semiconductor industries, in which the Department of Defense served as a source of R&D funding and a reliable early buyer of products. 34 DARPA later partnered with the semiconductor producers to address a competitive and qualitative challenge with the SEMATECH consortium. 35 Sustained NIH support for new drugs and new medical devices has contributed to major improvements in health, and the mapping of the genome has transformed the prospects for drug development.
Because the development of new technologies is inherently risky, regular assessment is vital to ensure continued technical viability, with cost-sharing requirements acting as an effective safeguard. Assessment can also help avoid
Ibid., pp. 3 and 17.
Ibid. The authors identify defining technological successes such as computers, the Internet, hybrid seeds, and aircraft. See also
Box G, discussing “Picking Winner and Losers,” p. 51.
34 Graham, Losing Time, op. cit., p. 2.
35 Market-opening measures, such as the Semiconductor Trade Agreement (SCTA), also played an important role in preventing dumping and improving market access for U.S. firms. For a review of the SCTA see National Research Council, Conflict and Cooperation, op. cit., p. 81, note 211, and pp. 131-141. See also Kenneth Flamm, Mismanaged Trade? Strategic Policy and the Semiconductor Industry, Washington, D.C.: The Brookings Institution, 1996, Chapters 3, 4, and 5, especially pp. 279-293.
“political capture” of projects, especially large commercial demonstration efforts. 36 Even successful collaborations face the challenge of adapting programs to rapidly changing technologies. 37 Assessment thus becomes a means of keeping programs relevant. Assessment can also have the virtue of reminding policy makers of the need for humility before the “black box” of innovation. As one observer notes, “experience argues for hedged commitments, constant reappraisal, maintenance of options, and pluralism of advice and decision makers.” 38
Comparisons in a Global Economy
From an international perspective, understanding the benefits and challenges of programs to support industry is also important insofar as they have been, and remain, a central element in the national development strategies of both industrial and industrializing countries. Governments have shown a great deal of imagination in their choices of mechanisms designed to support high technology industries. They employ a wide range of policies from trade regulations designed to protect domestic products from foreign competition to tax rebates intended to stimulate the export of selected domestic products. Many provide government R&D funding for enterprises of particular interest and, as the chart below shows, a number of countries have substantially increased their expenditures on R&D. Major financial support is sometimes overtly provided through direct grants, loans, and equity investments; more opaque support can also be provided through mechanisms such as tax deferral. 39 Data collected by the Paris-based Organization for Economic Cooperation and Development (OECD) suggest that worldwide government expenditures on support for high-technology industries involve significant resources and are increasingly focused on what policy makers consider to be strategic industries. 40
In the past decade, countries ranging from Taiwan to Finland to Japan have launched accelerated cooperative programs to restore or gain national competitive
36 Noll and Cohen stress that political capture by distributive congressional politics and industrial interests is one of the principal risks for large-scale, government-supported commercialization projects. In cases such as the Clinch River project, they extensively document the disconnect between declining technical feasibility and increasing political support. See Linda R. Cohen and Roger G. Noll, The Technology Pork Barrel, op. cit., p. vii and pp. 242-257.
37 One of the strengths of SEMATECH was its ability to redefine goals in the face of changing conditions. See Grindley et al., “SEMATECH and Collaborative Research: Lessons in the Design of High-Technology Consortia,” Journal of Policy Analysis and Management, 13(4):724, 1994. See also National Research Council, Conflict and Cooperation, op. cit., p. 148.
38 Graham, Losing Time, op. cit., p. 251. Graham is referring to work by Richard R. Nelson in Government and Technological Progress. New York: Pergamon Press, 1982, pp. 454-455.
39 For an overview of the policy goals and instruments, see National Research Council, Conflict and Cooperation, op. cit.,
Box B, pp. 39-40. See also Martin Brown, Impacts of National Technology Programs, Paris: OECD, 1995, especially Chapter 2.
40 See Martin Brown, Impacts of National Technology Programs, op. cit.
ness in key industries. Semiconductors and communications are frequent targets. The semiconductor industry, seen as an enabling or strategic sector, has long benefited from government support in this country and abroad. 41 In Japan and elsewhere, the SEMATECH consortium is seen as a major contribution to the resurgence of the American semiconductor industry. 42 Japan has launched an innovative series of public-private partnerships based in part on the SEMATECH model with the objective of restoring the competitive position of its semiconductor industry. 43 These cooperative activities are by no means confined to traditional competitors for high-technology industry. Finland has a more general program of technology development called Tekes, which brings together key elements of Finnish technology strategy under a single directorate. Parts of Finland's program have substantial similarities with the ATP. Reflecting the Finnish commitment to investments in new technologies, the Tekes program is funded at a similar level to the ATP. In 2000, Finland, a country of 5.1 million people, funded 2,297 research and development projects through Tekes for a total value of EUR 370 million as compared with the approximately $143 million appropriated to the ATP in FY2000. 44
The United States is an active, if unavowed, participant in this global competition to support national industries, through both state and federal governments. Indeed, the United States has a remarkably wide range of public-private partnerships in high-technology sectors. 45 In addition to the well-known cases mentioned above, there are public-private consortia of many types. They can be classified in a number of ways, such as by the economic objective of the partnership—for example, leveraging the social benefits associated with federal R&D activity, enhancing the position of a national industry, or deploying industrial R&D to meet military or other government missions. 46 Partnerships can also be
41 “The semiconductor industry has never been free of the visible hand of the government intervention. . . . [T]he semiconductor industry, wherever it has developed, has been an explicit target of industrial policy—whether in the guise at military policy in the United States or in the guise at commercial policy elsewhere in the world” (original emphasis). Laura D'Andrea Tyson, Who's Bashing Whom? Trade Conflict on High Technology Industries, Washington, D.C.: Institute for International Economics, 1992, p. 85.
42 “A major factor contributing to the U.S. semiconductor industry's recovery from this perilous situation [in the 1980s] was a U.S. national policy based around cooperation between industry, government, and academia.” Hajime Susaki, Chairman of NEC Corporation, “Japanese Semiconductor Industry's Competitiveness: LSI Industry in Jeopardy,” Nikkei Microdevices, December 2000.
43 For a description of these and other national programs in Taiwan, Korea, and Europe, see Thomas Howell, “An Overview of Government Policy Measures in Microelectronics,” Monograph, Washington, D.C.: National Research Council, 2001.
44 See the Tekes website,
(March 2001). Around EUR 230 million of this funding was in the form of grants and loans aimed at company research and development projects and about EUR 140 million at university and research institution projects.
45 See Coburn and Berglund, Partnerships, op. cit.
46 See A. Link, “Public/Private Partnerships as a Tool to Support Industrial R&D: Experiences in the United States,” paper prepared for the working group on Innovation and Technology Policy of the OECD Committee for Science and Technology Policy, Paris, 1998, p. 20.
differentiated by the nature of public support. Some partnerships involve a direct transfer of funds to an industry consortium. Others focus on the shared use of infrastructure, such as laboratory facilities. A partial list would include partnerships in sectors such as electronic storage, flat panel displays, turbine technologies, new textile manufacturing techniques, new materials, magnetic storage, next-generation vehicles, batteries, biotechnology, optoelectronics, and ship construction.
The scope of federal cooperative activity includes programs such as the national manufacturing initiative, the National Science Foundation's (NSF) Engineering Research Centers, NSF's Science and Technology Centers, NIST's Manufacturing Extension Partnership Program, and the multi-agency Small Business Innovation Research Program, among others. The SBIR is a substantial program, six times the size of the ATP, which is designed to draw on the innovative capacity of small firms and to further the ability of U.S. industry to capture the benefits of U.S. research. 47 University-industry cooperation is also on the up-swing, with a significant percentage of university R&D now provided by industry. There are also innovative cooperative efforts such as the Semiconductor Industry Association's MARCO program. 48
In addition, as noted above, the national laboratories now have extensive cooperative agreements with industrial firms in the form of Cooperative Research and Development Agreements. The Stevenson-Wydler Act of 1980 and the Technology Transfer Act of 1986 created this new mechanism for R&D collaboration. Amended in 1989 to allow industry-operated federal labs to participate, these laws (and high-level political interest) stimulated hundreds of CRADAs. Between 1989 and 1995, the Department of Energy alone signed more than 1,000 CRADAs. 49 The requirements of these agreements pose daunting challenges for small businesses, and a reduction of funding for this form of cooperation has resulted in substantial reduction in activity. 50 Nonetheless, this form of coopera
47 The 1992 increase in the SBIR set-aside to 2.5 percent of participating agencies' R&D budgets significantly augmented the program's resources, which are now in excess of $1.2 billion per year. That increase was accompanied by greater emphasis on commercialization.
48 Industry support of university research has been growing, from $1.45 billion in 1994 to $2.16 billion in 1999, an annual increase of nearly 10 percent. See Charles F. Larson, “The Boom in Industry Research,” Issues in Science and Technology, 16(4):27.
49 David Mowery, “Collaborative R&D: How Effective is It?” Issues in Science and Technology, 15(1):38. Cooperative research under these agreements expanded rapidly. At the same time, these programs face many criticisms: the agency focus on intellectual property (IP) rights can obstruct timely completion of negotiations—a genuine problem insofar as many CRADAs are concerned with near-term results.
50 CRADAs also require a high-level of technical sophistication, so that partner firms must make significant investments to support inward transfer and application of results. The last condition suggests that small firms may not be able to participate without financial and business assistance. Ibid., p. 43. Funding, and therefore the number of CRADAs, declined sharply in the late nineties. See Greg Linden, David Mowery, and Rosemarie Ziedonis, “National Technology Policy in Global Markets,”
tion continues to evolve, particularly with large firms, and is making significant technological contributions. 51
Origins of the ATP
The Advanced Technology Program was conceived as one means of addressing heightened concern about U.S. competitiveness and a desire to ensure that the U.S. economy benefited from federal R&D investments through partnerships. The legislation establishing the ATP was part of the Omnibus Trade Act of 1988, a complex bill whose main objective was to provide policy instruments to address the rapidly growing U.S. trade deficit. The sponsor of the Advanced Technology Program initiative was Senator Ernest Hollings of South Carolina, and the initial appropriation was small, only $10 million for 1990.
Initiated as a means of funding high-risk R&D with broad commercial and societal benefits that would not be undertaken by a single company, either because the risk was too high or because a large enough share of the benefits of success would not accrue to the company for it to make the investment, the program lacked the straightforward national security rationale that had underpinned many post-war U.S. technology programs. It did, however, reflect a general trend away from purely mission-oriented research and development towards facilitating more broadly based technological advances. The program's goal is to facilitate the development and application of new, enabling technologies that individual firms would not or could not pursue on their own and thereby encourage the economic growth that comes from the commercialization and use of new technologies in the private sector.
in Albert Link and Maryann Feldman, eds., Innovation Policy in the Knowledge-based Economy, Boston: Kluwer Academic Publishers, 2001, p. 312. Funding of CRADAs declined in the late nineties, from $346 million in 1995 to an estimated $94 million in 1999, and there was a corresponding drop from 1,700 CRADAs in FY1996 to a still significant 700 in FY1999. Ibid.
51 A major advance in lithography was announced by Lawrence Livermore National Laboratory in April 2001 that will allow chip manufacturers to create circuits as small as 10 nanometers wide using extreme ultraviolet (EUV) lithography. This advance was made under a CRADA in which the government laboratories provided facilities and expertise while the private companies such as Intel, AMD, Motorola, and Infineon provided funding and their own expertise. Interestingly, this success drew, in part, on an early ATP project with Lucent Technologies and Bell Laboratories, which significantly improved the accuracy of precision reflective optics, critical for EUV lithography. Beyond the state of the art when the award was made, this high-risk project aimed to find out whether the mirrors constituted a technical barrier that could not be surmounted. This project illustrates that a lack of immediate commercialization does not mean that the new technology will not eventually yield large benefits. William F. Long, Advanced Technology Program: Performance of Completed Projects—Status Report Number 1, NIST Special Publication 950-1, March 1999, pp. 75-77.
Box D. What is the Advanced Technology Program?
A Cost-shared, Industry-driven Technology Development Program
The Advanced Technology Program describes itself as “bridging the gap between the research lab and the marketplace.” 52 Specifically, the ATP provides cost-shared funding to industry intended to accelerate the development and dissemination of high-risk technologies with the potential for broad-based economic benefits for the U.S. economy. The ATP funding is directed to technical research (but not product development). Companies, either singly or jointly, conceive, propose, and execute all projects, often in collaboration with universities and federal laboratories. The ATP shares the cost for projects that are selected for a limited time. Single-company awardees can receive up to $2 million for R&D activities for up to three years. Larger companies must contribute at least sixty percent of the total project cost. Joint ventures can receive funds for R&D activities for up to five years.
What Does the Program Do?
It Shares Costs to Develop New Technologies with High Risk and High Potential.
Based on peer reviewed competitions, the ATP supports the development of a wide variety of new technologies. These have included adaptive learning systems, component based software, digital data storage, information infrastructure for health care, microelectronics manufacturing infrastructure, manufacturing technology for photonics, motor vehicles and printed wiring boards, new tissue engineering technologies, biopolymer repairs, and tools for DNA diagnostics. 53 These technologies are promising but risky. This means that a significant portion of the ATP-funded projects may fail. Recent research, however, suggests that a significant portion are succeeding. 54 Some projects, such as ATP's early Extreme Ultraviolet Lithography (EUV) research, follow an indirect path requiring some time and substantial additional development to eventually be commercialized. 55
What Does it Cost?
For the 41 competitions held from 1990-2000, the ATP made 522 awards for approximately $1.640 billion. These awards went to 1,162 participating organizations and an approximately equal number of subcontractors. Universities and non-profit independent research organizations play a significant role as participants in ATP projects. Universities have participated in over half of the projects, involving more than 176 individual universities. 56
52 See the ATP website.
53 See the paper by Alan P. Balutis and Barbara Lambis, “ The ATP Competition Structure” in this volume. See also the paper by Rosalie Ruegg “ Taking a Step Back: An Early Results Overview of Fifty ATP Awards” in this volume.
The program has a number of characteristics that collectively distinguish it from other publicly supported technology programs. These include its combination of industry initiative in the identification of technology areas and industry leadership in the planning and implementation of cooperative projects. The selection process is based on both technical and economic merit and is designed to encourage collaboration among companies and with universities and laboratories. To succeed in the selection process, proposals have to have well-articulated technical and economic goals and demonstrate not simply the merit of the work but the specific need for ATP funding. ATP awards are of fixed duration, and involve limited funding. They focus on the development of enabling technologies that have the potential for economic benefits beyond those that can be captured by the innovating firm(s).
Box E. Critical Characteristics of the Advanced Technology Program
Independent researchers have summarized ATP's “critical characteristics” that differentiate it from other government R&D programs as follows:
— Maryann Feldman, Johns Hopkins University
From its modest first-year funding of $10 million, the program grew with the support of a Democratic Congress to more than $60 million in the final years of the Bush Administration. As noted above, the Clinton Administration proposed and initially won substantial increases in ATP funding, but this high-profile sup
54 See the paper by Rosalie Ruegg. Ibid.
55 William F. Long, Advanced Technology Program, op.cit, pp. 75-77. See also Glenn R. Simpson “Why Chip Firms want U.S. SVG to go to Dutch,” The Wall Street Journal, New York: Dow Jones and Company, April 26, 2001, B1.
56 See the paper by Alan P. Balutis and Barbara Lambis, “
The ATP Competition Structure” in Chapter V of this volume.
port also generated significant opposition to the program. 57 The political debate about the need for the program, the proper role of government, and the need for government departments such as Commerce and Energy, 58 combined with the pressure for budget deficit reductions, to create what appeared to be a remarkably unstable budget environment. However, as the chart below shows, there was more budgetary stability than the debates of the mid-1990s and the legislative maneuvering in subsequent years might suggest. Essentially, after an initial ramp up in the early years of the Clinton Administration to $340 million, the program stabilized at around $200 million annually with a reduction in FY2000 due in part to the program's inability to fully allocate its funds. Partly as a result of administrative miscalculation, the ATP budget for FY2000 was $143 million, down sharply from $197 million in the previous year. The FY2001 budget saw a return to $191 million (including carryovers from previous years). 59
57 As one observer put it, “the irony of White House leadership on this issue has been that what was once a nonpartisan issue in the Congress acquired a partisan undertone. Legislative objectives that received broad support in both parties as recently as the early 1990s can now be cause, at times, for heated debate on the role of the government.” Coburn and Berglund, Partnerships, op. cit., p. 484.
Box G on page 51, on “Picking Winners and Losers.”
59 The total funds available for FY2001were $190.7 million: $145.7 million in appropriated funds and $45 million in carryovers from deobligations. It is important to note that while the appropriation for FY2001 is smaller in total than for FY2000, more funds are available for new awards in FY2001 than in FY2000 ($60.7 million vs. $50.7 million). For FY2002 the initial Administration position was to make no new commitments, but continue to meet existing commitments, pending a re-evaluation of the program.
The difficulty is that this relative stability is most apparent retrospectively. The policy debates and political maneuvering that have characterized the program's annual authorization have been a source of substantial uncertainty. For a research and development program that relies on the formulation of proposals by private firms, often organized in joint ventures, the uncertainty about the availability of ATP funding, either for new programs or for existing commitments, has been a major drawback. 60
The controversy with regard to the program during the mid-1990s was related to a broader debate over the role of government in society. In a sense, the debates about the program and the frequent assessments that have resulted may be disproportionate to the scale of the program, particularly in comparison with the SBIR or the mission-specific programs of the Departments of Defense, Energy, and Health and Human Services. 61 To the extent that the debate has generated rigorous assessment of the ATP—as it seems to have—the debate can be seen as a positive addition to the policy dialogue.
To put the policy debate in perspective, two subsections addressing different aspects of the debate are included in this report. To provide background on some of the policy questions examined by the U.S. General Accounting Office (GAO) at the request of congressional committees, a summary of the main points of several of the GAO reports with comments by NIST appears in a shaded box below ( Box F). The ATP has also been subject to attack as a program that involves the government picking “winners and losers.” This statement contains a number of implicit assumptions that are articulated in Box G below and are examined in relation to the ATP. This information is provided to give an overview of the questions raised about the program. Some of the issues raised have been debated since the very origins of the Republic, 62 a debate that extends well beyond the charge of the authoring committee. As noted above, the Committee's task was to examine the operations and performance of the program in light of its legislated objectives, not to make a fundamental determination about the intrinsic desirability of government-industry-university collaboration.
Some critics have focused their attention on the goals and operations of the ATP itself. One analysis of the program addressed the difficulty of identifying projects with substantial social benefits in excess of private returns. 63 The study
60 See the observations by Linda Cohen and Roger Noll cited above and in The Technology Pork Barrel, op. cit., p. vii.
61 See the review of the U.S. General Accounting Office reports in Box F.
62 See the Preface.
63 See the paper prepared by Loren Yager and Rachel Schmidt, The Advanced Technology Program: A Case Study in Federal Technology Policy, Washington, D.C.: American Enterprise Institute, 1997.
Page 44accepts that the economic rationales for the ATP are closely related to traditional arguments for federal support for R&D and notes that the ATP awards might expand the amount of R&D devoted to generic, high-risk technologies. The authors caution, however, that the effectiveness of the ATP depends on whether the program can be successfully implemented. They believe that the ATP is an “interesting experiment in whether the government can promote economic growth through an explicit technology policy.” 64 Combining both political and economic perspectives, they argue that “only projects with certain characteristics will enable ATP to achieve its economic goals, and those may not be the same projects that help it to achieve political support.” 65 Based on the information available to them in 1995-96, the authors concluded that the program at that time had “had only limited success.” 66 In particular, the authors stress the difficulty in identifying and measuring the social returns of ATP projects. 67
The challenge of identifying and tracking the impact of social benefits has been of interest to economists at least since Alfred Marshall emphasized it in his discussion of the evolution of the modern firm. 68 An analysis of spillovers produced by Adam Jaffe in 1996 distinguished several different mechanisms by which R&D generates spillovers. 69 He identifies “knowledge spillovers,” “market spillovers,” and “network spillovers” and observes that the three tend to inter-
64 For further information on Yager's views, in addition to the GAO reports, see his presentation in National Research Council, The Advanced Technology Program: Challenges and Opportunities, op. cit., pp. 42-44.
65 Yager and Schmidt, The Advanced Technology Program, op. cit., 1997, p. 42.
67 This assessment of the ATP program is based on the authors' review of the ATP project selection criteria as set forth in its legislation, regulations, and administrative notices; no independent economic analysis of the impact of awards was carried out. However, the review is useful insofar as it emphasizes the challenges of assessing, in advance of the award, the prospects for commercial success for a given technology and the potential social benefits. The trajectory followed by a technology is, of course, often unpredictable. The authors themselves note that predicting the likely profits of R&D efforts is enormously difficult, yet they observe that it is much easier than predicting social benefits. Ibid., p. 42. These analytical concerns do represent challenges for the program, and its assessment from an economic perspective. As noted below, the difficulty in assessing counterfactuals necessarily implies some uncertainty about the program's impact. The paper by Maryann Feldman and Maryellen Kelley in this volume does provide recent evidence that the majority of ATP applicants do not undertake their proposed projects in the absence of ATP funding, and, among those that do, the project is normally smaller in scope. This study also suggests that the ATP projects generate benefits beyond those accruing to the individual firm. In addition, ATP awards can impact decisions on private R&D investments. The decision process for R&D investments, particularly in large firms, is often complex; profit maximization is sometimes a guide to firm decision-making in the same sense that the national interest is a guide to decision-making for government agencies.
68 See Alfred Marshall, Principles of Economics, London: Macmillan, 1920.
69 See Adam B. Jaffe, Economic Analysis of Research Spillovers: Implications for the Advanced Technology Program, NIST GCR 97-708, December 1996.
act in a way that increases their combined effect. 70 His analysis highlights the importance for the ATP of seeking projects that have a large spillover gap and expected private returns sufficient to make commercialization likely. 71 The ATP selection process seeks to capitalize on this approach. It focuses on multi-use technologies, key components applicable to multiple systems, path-breaking technologies which offer significant economic benefits, and projects which seem likely to generate useful knowledge even if the project itself entails substantial risk. While recognizing the inherent complexity of this type of assessment, Jaffe observes that “the empirical evidence suggests that the average research project generates spillovers.” 72 He adds that to the extent the ATP targets projects with better-than-average spillover potential, its awards can generate social returns greater than would otherwise have been achieved. 73
Box F. GAO Reviews of the ATP
The GAO has reviewed various aspects of the Advanced Technology Program, including issues regarding the award process and elements of its assessment program, at different times. Assessing an R&D program is an inherently complex task, perhaps made more complicated by the program's focus on high-risk technologies with potentially high payoffs. The corollary of high risk is a substantial failure rate. Even venture capital firms, which in principle are focused on opportunities farther down the innovation path, encounter many disappointments in their investments. 74
Ibid., p. 4. Knowledge spillovers occur when projects generate knowledge that can be used by other parties without compensation, or compensation less than the value of the knowledge. Similarly, market spillovers result when the market for a new product or process causes benefits to flow to parties other than the innovating firm. Network spillovers result when the value of a new technology depends on the development of related technologies (e.g., software applications for use on a new operating system) and widespread use of related technologies.
Ibid. In the absence of such returns, projects are unlikely to be commercialized and, hence, less likely to produce significant spillover benefits, although it is also true that some technologies generate spillovers through a more circuitous, indirect path (this process is described by Rosalie Ruegg in her symposium presentation in this volume).
Ibid., p. 21.
74 It is not always appreciated by the general public that even successful venture capital firms normally have many more investment failures than successes. Apparently, a success rate of two out of ten projects can prove acceptable in that one or two substantial successes make good the losses and disappointments of the other eight. This high rate of failure reflects the uncertainty associated with early-stage investing. As Paul Gompers and Josh Lerner point out, venture investors “typically concentrate in industries with a great deal of uncertainty, where the information gaps among entrepreneurs and investors are commonplace.” They add that new firms often have substantial intangible
The venture capitalist's goal is to make good the inevitable losses with the substantial benefits derived from major successes. The ATP's focus on innovative technologies appears to be establishing a similar track record. 75 The challenge for the GAO was to review, usually in a short time frame, the operations of a program involving complex technologies amidst considerable uncertainty as to both technological and market outcomes. An additional challenge is that some of the GAO reports were commissioned relatively early in the program when there were limited results to review because of the normal lead times required for the development of new technologies and processes.
Identification of Similar Research
One of the most recent studies undertaken by the GAO, released in 2000, focused on factors in the ATP selection process that could limit its ability to identify similar research elsewhere. 76 The former chairman of the House Science Committee expressed concern that the program “may have funded research that was similar to research already being funded by the private sector.” 77 The GAO reviewed the NIST status report on the first thirty-eight completed ATP projects. The GAO report states in a letter to Representative Sensenbrenner that “as agreed with your offices we determined (1) whether, in the past, ATP had funded projects with research goals that were similar to projects funded by the private sector and (2) if such cases were identified, whether ATP's current award selection process ensures that such research would not be funded in the future.” 78 The agency chose three of the thirty-eight completed projects undertaken in 1990 and 1992. 79 Each project represented a different technology sector—biotechnology; electronics; and information, computers and communications. The specific projects focused on areas such as handwriting recognition, regenerating tissues and organs, and capacity expansion of fiber-optic cables. 80
The GAO found that patents had been issued to others in these general areas and concluded that the three ATP projects addressed similar
assets that are difficult to value and may be impossible to resell. Moreover, the market conditions in many of these industries are often highly variable. Paul Gompers and Josh Lerner, The Venture Capital Cycle, Cambridge, MA: The MIT Press, 2000, p. 3.
75 See Rosalie Ruegg, “
Taking a Step Back: An Early Results Overview of Fifty ATP Awards,” in this volume, and the earlier assessment by William Long, Advanced Technology Program, op. cit.
76 U.S. General Accounting Office, Advanced Technology Program: Inherent Factors in Selection Process Could Limit Identification of Similar Research, Washington, D.C.: U.S. General Accounting Office, GAO/RCED-00-114, 2000.
Ibid, p. 3.
Ibid, p. 4.
Ibid, p. 7.
research goals to projects in the private sector. The GAO also concluded that the ATP's award process is unlikely to avoid funding similar research, insofar as limits on access to proprietary information and ATP conflict of interest requirements limit the program's ability to identify similar research. 81 The GAO did not, however, recommend any changes to the award selection process. Based on their review of the three projects, the agency stated that “we believe that it may not be possible for the program to ensure that it is consistently not funding existing or planned research that would be conducted in the same time period in the absence of ATP financing assistance.” 82
The Counterfactual Dilemma
This conclusion raises the classic “counterfactual” dilemma of knowing “what would have happened if the award had not been made.” Analytically, this is a difficult question to resolve and one which is normally not raised with the same frequency regarding other aspects of federal activity. In its response, NIST disagreed with both the methodology used and the conclusions reached in the GAO assessment. In addition to noting the positive judgments concerning ATP's competitive peer review process, the NIST management argued that this GAO report failed to address a crucial distinction, that is, between projects with similar research goals versus projects with unique objectives and technical approaches in similar fields. NIST noted further that the GAO report contained no reference to the ATP focus on projects with high technical risk, the competitive value of different technical approaches within broad research fields, nor the national benefits that might result from the success of specific projects. 83
Examining the Award Process
A 1997 GAO review of the ATP involved the provision of information on the ATP award selection process. 84 The GAO describes the ATP as “a competitive, cost-sharing program designed for the federal govern-
Ibid, p. 4.
Ibid, p. 15.
83 The NIST director at that time, Ray Kammer, objected to the GAO report, arguing that “the implied argument is that the federal government should not fund research that shares the same overall goal as research funded outside of the government.” Kammer argued that by that criterion federal research would be discontinued on cures for diseases such as cancer and AIDS, as well as on wireless communications, computing technologies, and manufacturing. See New Technology Week, Monday, May 1, 2000, p. 3.
84 U.S. General Accounting Office, Federal Research: Information on the Advanced Technology Program's 1997 Award Selection: Statement of Susan D. Kladiva before the Subcommittee on Technology, Committee on Science, U.S. House of Representatives, Washington, D.C.: U.S. General Accounting Office, GAO/T-RCED-98-92, 1998.
ment to work in partnership with industry to foster the development and broad dissemination of challenging, high risk technologies that offer the potential for significant, broad-based economic benefits for the nation.” The GAO's inquiry focused on whether ATP applicants could obtain funding elsewhere, and whether not funding a project “would create a serious national economic concern.” The report states that the ATP rejects project proposals when it concluded that (1) the applicants could probably find funding elsewhere, or (2) a delay in project progress would not be a serious national economic concern. The GAO's analysis focused on the information that the ATP used to make these determinations.
The GAO noted that the ATP selects projects for funding based on five selection criteria, which are used by peer reviewers in the assessment of proposals. 85 These criteria are
With respect to the availability of private funds, the ATP review process includes requests for information concerning efforts to obtain funding and also provides opportunities to question the applicants on this point during the oral review phase. In addition, the GAO reports that the ATP now requests additional information on company efforts to obtain outside funding as part of the overall decision on the applicant's proposal. With respect to the second question regarding topics of national economic concern, reviewers were asked to evaluate the proposed project in terms of its potential to improve U.S. economic growth and productivity, timeliness, the degree to which ATP funding is necessary, and cost-effectiveness.
Progress without ATP?
A 1996 GAO assessment sought to determine whether research projects would have been funded by the private sector if they had not received funds from ATP—again the “counterfactual” question. 86 The GAO
85 The GAO noted that the 1997 focused program competitions addressed motor vehicle manufacturing technology, information infrastructure for health care, component-based software, and tissue engineering.
86 U.S. General Accounting Office, Measuring Performance: The Advanced Technology Program and Private-Sector Funding, Washington, D.C.: U.S. General Accounting Office, GAO/RCED-96-47, 1996.
found that half of the near-winners continued their projects without relying on ATP funding, albeit at reduced levels of activity, while the other half discontinued their projects. 87 The GAO also examined the ATP's impact on other goals such as aiding the formation of joint ventures. The survey found that more than three-fourths of the joint venture applicants had come together to pursue an ATP project, thus satisfying ATP's goal of serving as a catalyst for the formation of joint ventures 88 —such arrangements tend to encourage the diffusion of research results, a related ATP goal. 89 The report suggests there may be some overlap with private funding. 90 At the same time, the GAO report also concludes that ATP awards encourage the formation of joint ventures and help companies achieve research milestones faster.
In response, the ATP argued that the GAO findings support the view that ATP is accelerating technology development insofar as the nearwinners normally do not proceed at the same pace and scale of activity in the absence of the ATP award. 91
A 1995 GAO assessment focused on NIST's efforts to evaluate the Advanced Technology Program. At the outset, the GAO report notes that “Evaluating ATP poses many challenges. For example, ATP research projects are intended not only to be technical successes but also to have commercial results. The linkage between technical work and commercial results may not always be direct, and may be subject to interpretation. Also, several years can elapse between the end of technical work and the realization of commercial results.” A key GAO conclusion was that it was too early (in 1995) to determine the ATP's long-term economic impact.
In describing NIST's extensive assessment efforts, the report was critical of the program's initial effort to assess the short-term results of its
87 NIST argues that companies may reduce the scope, scale, and nature of the research goals, with the result that under the same project description, a much smaller-sized effort is under way at a slower pace. Ibid., p. 2.
89 This view was subsequently reinforced by the findings outlined in the Feldman and Kelley paper, “
Leveraging Research and Development: The Impact of the Advanced Technology Program,” in this volume. The authors found that firms selected by the ATP were more likely to share the research findings, thus contributing to their dissemination.
90 Feldman and Kelley find that the receipt of an ATP award has a “halo effect” in that it “signifi cantly increases the firm's success in attracting additional funds from other sources for R&D activities,” Ibid., p. 12.
91 The more recent Feldman and Kelley study finds that 62 percent of the non-winners had not proceeded with any aspect of the R&D proposed to the ATP. More than a third did undertake the proposed work, but in more than three-quarters of these cases, the project was ursued on a smaller scale. Ibid.
awards, describing them as overstated. 92 The same report noted NIST's efforts to extend its evaluation by engaging the advice and services of the nation's leading economists in impact assessment and evaluation, conducting microeconomic case studies, supporting the use and development of economic models, and establishing an extensive data collection system. 93
These necessarily abbreviated reviews of the almost annual GAO reports demonstrate the range of questions that can be asked about programs to support the development of new technologies as well as the difficulty of a definitive response. 94 As noted above, the GAO analysts themselves underscore the challenge of evaluation and note that the ATP assessment program is itself seeking to answer similar questions. The GAO reports provide a valuable perspective on various aspects of the ATP. At the same time, it should be kept in mind that the GAO reports are necessarily constrained by the questions they are asked to address and are normally limited in the scope of their analysis. They provide an additional outside source of assessment that has encouraged and is complementary to the assessment process developed by the ATP.
As the survey of the GAO reports on the ATP suggests, assessment of government R&D programs is as desirable as it is difficult. However, some observers and participants object to a role for the federal government in the support of technology development beyond basic research on the grounds that it is unable to do so effectively. Others would accept a federal role for technology development that is limited to specific national missions. Because these views are often raised with respect to the Advanced Technology Program, the box below discusses this perspective in the context of current federal technology policy and particularly in relationship to the ATP.
92 U.S. General Accounting Office, Performance Measurement: Efforts to Evaluate the Advanced Technology Program, Washington, D.C.: U.S. General Accounting Office, GAO/RCED-95-68, 1995. In its response, NIST objected to the GAO findings criticizing the GAO for misrepresenting NIST statements with regard to the short-term impact of ATP and its then-nascent evaluation program. NIST defended its evaluation plan as appropriate, informative, and well-founded. See pp. 14-27.
Ibid., p. 8. NIST contracted with the National Bureau of Economic Research for part of its assessment activity.
94 The difficulty of arriving at a definitive response is not confined to the GAO reports nor the internal ATP assessments. An excellent, comprehensive survey by Paul David, Bronwyn Hall, and Andrew Toole notes, “the paucity of systematic statistical evidence documenting a direct contribution from public R&D” while also noting that “there is a significantly positive and relatively high rate of return to R&D investments at both the private and social level.” Is Public R&D a Complement or a Substitute for Private R&D? A Review of the Econometric Evidence, NBER Working Paper No. 7373 (October 1999), p. 2.
BoxG. “Picking Winners and Losers”and the Advanced Technology Program
In the United States, discussions of best practice concerning partnerships between the government, industry, and universities often include statements to the effect that the government cannot—or should not—“pick winners or losers.” For example, a recent GAO study described two views of the ATP, as follows: “Thus, ATP is seen by some as a means of addressing market failure in research areas that would otherwise not be funded, thereby facilitating the economic growth that comes from the commercialization and use of new technologies in the private sector. Advocates of the program believe that the government should serve as a catalyst for companies to cooperate and undertake important new work that would not have been possible in the same time period without federal participation. Critics of the program view ATP as industrial policy, or the means by which government rather than the marketplace picks winners and losers.” 95 By this expression “picking winners and losers,” it is generally meant that
These arguments are obviously interrelated, and their appeal is grounded in the popular perception of an American economy regularly transformed by individual investors and entrepreneurs. This view is, of course, well founded, both in terms of American economic history and today's economy. Yet it is equally true that the federal government has long played a nurturing role, with the result that in many cases the process of innovation in the U.S. economy is the result of a complex
95 U.S. General Accounting Office, Advanced Technology Program: Inherent Factors in Selection Process Could Limit Identification of Similar Research, op. cit., p. 5.
96 See, for example, the testimony by Dr. Edward L. Hudgins, before the Senate Committee on Commerce, Science and Transportation, 1 August 1995. In addition to recommending the abolishment of the Department of Commerce and NIST, Hudgins argues that “In the area of advanced commercial technologies, that is, the high-tech revolution of the past 15 years, the private sector already does a world-class job in developing new products and technologies. Thus, ATP is unnecessary. The way a market system—as opposed to a corporatist or socialist system—works is that if there is a prospect for a profit, entrepreneurs will risk investing in order to reap profits. For example, the cost of bringing a new pharmaceutical product to market is now on average $390 million. Yet drug companies make such investments. If there is a profit to be made, entrepreneurs will act with or without government handouts.”
interaction of public and private initiative. Arguments that do not take this interaction into account ignore important parts of the history of technology development in the United States. They also do not reflect recent and current practice with respect to the development of new technologies. 97 This is especially true with respect to platform technologies such as the Internet, or contributions to enabling technologies such as semiconductors, or the sustained research and development support of the National Institutes of Health to the pharmaceutical, medical device, and biomedical industries. 98
The government has demonstrated a capacity to make judgments with respect to new technologies. It has been instrumental in developing major new industries through a variety of means, including awards for demonstration projects, provision of long-term R&D support (e.g., the National Advisory Committee for Aeronautics [NACA]), 99 support of a regulatory framework, and provision of early assured markets through government procurement. Major government initiatives for exploration and defense (e.g., the Apollo and Minuteman programs) provided generous “cost-plus” procurement contracts that made a major contribution to the growth of Silicon Valley. 100 One recent analysis argues that “In effect, the U.S. defense policy after WW II was an innovation policy, which greatly benefited Silicon Valley's high technology firms by creating price-insensitive lead customers and by funding pre-commercial research, supporting universities, and training engineers and scientists.” 101
97 See the
Preface and the
Introduction for a brief overview of the historical and policy context in which the ATP was developed.
98 Public grants to non-profits and private companies constitute a significant portion of NIH funding. In FY2000, $1.0 billion, or nearly 7 percent, of NIH funding for research grants and R&D contracts went to for-profit organizations. An additional $1.4 billion, or 10 percent, went to non-profit institutes, some of which are reported to be closely associated with for-profit firms. For NIH funding by performer in FY2000, see: http://silk.nih.gov/public/cbz2zoz.@www.trends00.fy9100.per.htm. For increasing relationships between nonprofit research institutions and for-profit firms, see Chris Adams, “Laboratory Hybrids: How Adroit Scientists Aid Biotech Companies with Taxpayer Money—NIH Grants Go to Nonprofits Tied to For-profit Firms Set up by Researchers,” Wall Street Journal, New York: Dow Jones and Company, January 30, 2001, p. A1.
99 Founded in 1915, NACA made major contributions to aeronautics, as noted above, until it was incorporated into NASA in 1958.
100 A recent analysis of the factors contributing to the development of Silicon Valley emphasized the role of defense spending in the Valley's development. See the articles by Timothy J. Sturgeon, “How Silicon Valley Came to Be,” and by Stuart A. Leslie, “The Biggest ‘Angel' of Them All: The Military and the Making of Silicon Valley,” in Martin Kenney, ed., Understanding Silicon Valley: The Anatomy of an Entrepreneurial Region, Stanford, CA: Stanford University Press, 2000.
101 Martin Kenney, Understanding Silicon Valley, op. cit., p. 5.
Academic critics of some government-supported technology programs record both successes and failures in government efforts to support new technologies through these multiple mechanisms. One well-known assessment of the major commercialization demonstration projects of the 1970s observed that there are frequent failures as well as successes among federal R&D programs and counted among the successes telegraphy, hybrid seeds, aircraft, radio, radar, computers, semiconductors, and communication satellites. 102 In short, amidst both failure and success, government support for new technologies has laid much of the foundation for the modern economy. As a recent National Academies study noted, “The development of the Internet demonstrates that federal support for research, applied at the right place and the right time, can be extremely effective.” 103 Of course, a great advantage of the American system is that, after government actions have contributed to a technological advance, the development of the technology and its commercialization are left in the hands of multiple players in the private sector. 104 The ATP's broad mandate to support risky but promising technologies that may contribute to new products and processes, enhance U.S. competitiveness, or accelerate welfare-enhancing technologies (e.g., improved mammography) is more controversial than, say, DOD funding of computing and networking development, because the program is not tied to a specific mission or constituency.
One of the great strengths of the U.S. economy is that the government sees its central role as an arbiter of economic competition among private actors. To a remarkable degree, this is accurate, yet the fact remains that the government does intervene in the market in many ways, be it through the provision of R&D support, development of a favorable regulatory framework, or procurement decisions for technologies for gov-
102 See Linda R. Cohen and Roger G. Noll, The Technology Pork Barrel, op. cit., p. 3. More recently, the Office of Science and Technology Policy took a similar view, noting that federal funding of research and development has led to such advances as atomic energy, the Internet, the Global Positioning System, lasers, solar-electric cells, storm windows, Teflon, communications satellites, jet aircraft, microwave ovens, genetic medicine, and a wide array of advanced materials and composites. Office of Science and Technology Policy, Fact Sheet on How Federal R&D Investments Drive the U.S. Economy, Executive Office of the President, June 15, 2000,
103 See National Research Council, Funding a Revolution, op. cit., p. 181. The federal government has played a critical role in supporting the research that underlies computer-based products and services. Federal funding rose from less than $10 million in 1960 to almost $1 billion in 1995. The vast majority of the funding has been awarded to industry and university researchers. Ibid., pp. 2-3. See also Michael Hauben and Ronda Hauben, Netizens: On the History and Impact of Usenet and the Internet, 1998,
104 Irwin Lebow, Information Highways and Byways, op. cit., pp. 9-12.
ernment missions ranging from defense, space exploration, and health. The government role, of course, is not confined to investment incentives. Its role in infrastructure building, support for research—both early and applied—and for training are all integral parts of the government's support for economic growth. Although not without controversy, the exercise of government's regulatory responsibilities has played a key role in the computing and telecommunications industries. For example, anti-trust actions in the 1950s were intended to facilitate the entry of other companies and more rapid innovation in the computer industry. 105 In recent years, since the Telecommunications Act of 1996, government and industry have been closely involved in an ongoing debate concerning the optimal regulatory regime. 106 In short, the government has demonstrated the capability to make judgments concerning new technologies but also must make such decisions to carry out its various responsibilities. 107
Federal support for advanced technologies has involved many mechanisms. As noted above, until the mid-1980s, federal support normally took the form of research grants or contracts for product development or procurement programs that often had a significant research component. After 1985, a number of programs were established that involved partnerships among government, industry, and universities. These programs included the Semiconductor Research Corporation, the NSF Engineering Research Centers, SEMATECH, Cooperative Research and Development Agreements (CRADAs), the Manufacturing Extension Partnership Program, and the Advanced Technology Program. 108 Unlike many
105 National Research Council, Funding a Revolution, op. cit., p. 33.
106 Since the Telecommunications Act of 1996, P.L. No. 104-104, 110 Stat. 56 (1996), the stakes for firms with incumbent positions and start-ups with different technologies and business plans have been enormous. Some are seeking new spectrum allocation (or re-allocation), others regulatory support through active enforcement of the act. The government's role is a critical component in the competitive position of many firms.
107 As with the private sector, the government's judgment and capability do not insulate it from failure. Both the private sector and the government face the same uncertainties. Each must place bets, albeit for different reasons. Each cannot avoid the certainty of losers. Each can cover enough points to be assured of some winners. “In short, winners and losers are an inevitable by-product of the process of innovation. Picking winners and losers is the wrong metaphor to characterize the socially useful and necessary activity of government in supporting that process. Government is actually placing bets on our collective future, and from the public standpoint, the magnitude of the potential social gains are sufficiently large to provide a comfortable margin for error in choosing among technologies to back.” See the testimony of Professor Michael Borrus, University of California at Berkeley, before the House of Representatives Committee on Science, Subcommittee on Technology, April 10, 1997. http://www.house.gov/science/borrus 4-10.html.
108 National Research Council, Funding a Revolution, op. cit., pp. 32-33.
cooperative programs, the ATP relies on the private sector to identify promising areas of technology, such as biomedical devices, information technologies and materials, and communications (e.g., optoelectronics) and to design, lead, and manage the projects. The ATP's reliance on peer review by both industry business experts and government technical experts ensures, to the extent possible, the technical quality of proposals and a potential for economic impact. 109 The requirement of a plan for commercialization encourages, but cannot ensure, a pathway to commercial development. Rather, the commercialization plan requirement is designed to ensure that projects that cannot offer at least a potential pathway to development are not supported with public funds. 110
To avoid open-ended commitments of public funds to uncertain technologies, the ATP has incorporated features that serve as “reality checks,” most notably, the matching expenditure of private funds by forprofit firms. This cost-share requirement, the limited financial commitment of the government through one-time awards, and the limited duration of the awards protect ATP against the frequently justified criticism of open-ended government-led technology commercialization programs that characterized the major government initiatives of the 1970s and early 1980s. 111 Compared with these 1970s programs, ATP is a much smaller, more limited, and more focused effort with different mechanisms (e.g., one-time, competitive awards) focused on R&D more likely to diffuse across the economy. 112 Its encouragement of company-universitylaboratory collaboration and coordination with other public and private efforts is another distinguishing characteristic. 113
Perhaps most important is the ATP's support for innovation that is expected to generate significant spillovers yielding broad national economic
109 Some critics of the program argue that the government cannot pick winners, that is, substitute the judgment of bureaucrats for decisions properly made by the market. Other critics argue that the ATP's selection of projects based on input from industry is equally suspect, leading “to the conclusion—certainly the suspicion—that the real reason for individual sector selection was to reap political support from successive groups of corporate leaders. . .” who would provide pressure against future program cuts. Claude E. Barfield, testimony before the Subcommittee on Technology of the House Science Committee, April 10, 1997.
110 For an overview of the ATP selection process, see the analysis by Alan P. Balutis and Barbara Lambis, “
The ATP Competition Structure,” in this volume. For an assessment of the impact of the program, see the analysis by Maryann P. Feldman and Maryellen R. Kelley, “
Leveraging Research and Development: The Impact of the Advanced Technology Program,” in this volume.
111 See Linda R. Cohen and Roger G. Noll, The Technology Pork Barrel, op. cit.
112 Feldman and Kelley, “
Leveraging Research and Development,” in this volume find that projects and firms selected by the ATP are more willing to share their research findings and tend to be firms that open new paths of innovation by drawing on multiple technical areas through R&D partnerships.
Ibid. Universities play a significant role in many ATP projects.
benefits. This approach constitutes an important source of risk for the program. 114 The ATP's interest in enabling technologies with high spillover potential that are also subject to substantial technical challenge means it is also a source of substantial potential benefit for the economy. By definition, the high-risk, high-payoff strategy means that many ATP projects will not achieve success. 115 The program deliberately seeks projects needing the catalytic effect of a government award to bring together the university-industry partners to achieve technological advances. 116
The Policy Context
Lastly, the program has to be considered in context. The ATP is one element of a national innovation system that employs a variety of policies and instruments to encourage the discovery, development, and exploitation of new technologies. 117 It is not a panacea for the challenges facing the U.S. economy. It is a program that the research outlined here suggests is achieving its goals with a degree of success commensurate with the technical and commercial difficulties associated with the program's objectives. It holds the potential of advancing commercially and perhaps socially valuable technologies. It is a program that might be improved, and this report recommends some ways to do so. Whatever improvements might be made in the program, the dialogue about such programs certainly can be improved, not least by careful research, regular assessment, and attention to the initiatives under way around the world. A constructive dialogue about measures to capitalize on the substantial and growing U.S. R&D investment in areas of great promise needs to be advanced both to avoid misallocation of public funds and to capture the substantial benefits of new technologies for the American economy.
The Need forAssessment
As reflected in the reviews and debates summarized above, one consequence of the debates about government-industry partnerships in general, and the ATP in particular, has been a desire for objective analysis of the goals, operation, and
114 Lewis Branscomb and Philip Auerswald, Taking Technical Risks: How Innovators, Managers, and Investors Manage Risk in High-Tech Innovation, Cambridge, MA: MIT Press, p. 145.
115 ATP successes, insofar as they can be judged, are about 16 percent of the program—a rate comparable to venture capital programs which normally operate later in the development cycle. For an overview of the ATP record, see the analysis by Rosalie Ruegg, “
Taking a Step Back: An Early Results Overview of Fifty ATP Awards,” in this volume.
116 See the paper by Jeffrey Dyer and Benjamin Powell, “
Perspectives on the Determinants of Success in ATP-sponsored R&D Joint Ventures: The Views of Participants,” in this volume and the summary of their findings in
Section B on page 60.
117 See Richard R. Nelson, editor, National Innovation System: A Comparative Study, New York: Oxford University Press, 1993.
results of partnership programs. This pragmatic approach with regard to assessment was captured in a 1995 RAND study, which observed that:
The federal government has undergone a sea change the past few years in its approach to the private sector. The broad awareness of and support for these activities in Congress and their spread throughout the $80 billion federal R&D system ensure that they will continue well into the next Administration and beyond. The debate should address not whether these programs will endure, but whether they are shaped properly—at the program and aggregate levels—to achieve the desired benefits. 118
Reflecting this perception, the Senate requested that NIST commission an outside review of the operations of the Advanced Technology Program. 119
The National Research Council Analysis: Report Structure
To carry out this review, the National Academies undertook an analysis of the ATP under the terms of reference of its study of Government-Industry Partnerships for the Development of New Technologies. As described in the Preface, this project is being carried out by a distinguished Steering Committee under the National Research Council's Board on Science, Technology, and Economic Policy. The study is to contribute to improved understanding of partnerships through commissioned research, conferences, and workshops bringing together policy makers, program managers, academic experts, technologists, and representatives of industry.
As noted above, this volume is the second report in this series on the Advanced Technology Program 120 and is designed to further our understanding of the operations and impact of this program. This report has five main parts. The Preface is intended to give an overview of the overall project and fix the ATP assessment within the context of this broader analysis. The Introduction gives an overview of federal technology policy, fixing the ATP in the context of U.S. efforts to develop new technologies through various forms of government incen-
118 Coburn and Berglund, Partnerships, op. cit., p. 487.
119 In Senate Report 105-235, the Advanced Technology Program was directed to arrange for a well-regarded organization with significant business and economic experience to conduct a comprehensive assessment of the ATP, analyzing how well the program has performed against the goals established in its authorizing statute, the Omnibus Trade and Competitiveness Act of 1988.
120 See the Preface. Six reports on various aspects of partnerships have been published; two others are in preparation. The first report was National Research Council, The Advanced Technology Program: Challenges and Opportunities, Charles W. Wessner, editor, Washington, D.C.: National Academy Press, 1999.
tives, support, and cooperation. To facilitate the reader's task, the Introduction also includes a summary of the seven papers prepared for the volume and of the symposium deliberations.
Chapter III provides the Committee Findings and Recommendations for assessing the operation of the program and making recommendations for its improvement. Chapter IV provides a detailed summary of the presentations of the one-day symposium, entitled “Assessing the Advanced Technology Program: Issues and Outcomes,” held April 25, 2000. The symposium brought together policy analysts, economists, Department of Commerce officials responsible for the ATP, and representatives of private industry to review the ATP's objectives, describe its selection process, and review case studies from the assessment program, as well as discuss the ATP's rationale, program strengths and weaknesses, and consider potential areas for improvement. The summary of the deliberations in Chapter IV are complemented by seven papers that describe the evolution and nature of the ATP competition process and assess, from a variety of perspectives, the impact of the ATP awards. The symposium summary and the related case studies represent the second phase of an analysis requested by NIST in response to a congressional mandate to provide an independent review of ATP program operations.
Section B below provides an overview of the papers collected for this volume. They illustrate the scope, quality, and challenges of the assessment effort while providing insights into the operation of the program. The following section provides a summary of the main points of the workshop proceedings.