1
Background: The Postwar U.S. Technology Enterprise

The United States' collective capacity to create, develop, and deploy new technology constitutes its national technology enterprise. The nation's human, physical, and financial capital, and the publicand private-sector institutions (firms, universities, government agencies, nonprofit research laboratories, financial and regulatory systems, etc.) that organize and direct these resources in service of the interests of U.S. citizens, are all elements of the U.S. technology enterprise. Since World War II, the U.S. technology enterprise and the private and public strategies that have sustained it have been profoundly shaped by the nation's unique economic and geopolitical position in the decades immediately following the war (Nelson, 1990). This period of unchallenged U.S. economic and commercial technological preeminence was a time when the most urgent scientific and technological challenges to the nation were defined by the Cold War, the space race, a domestic war on cancer, and the quest for world leadership in virtually all areas of scientific research.

For most of the past 40 years, the U.S. technology enterprise and supporting public- and private-sector technology strategies have served the interests of U.S. citizens—their security, their economic welfare, their global influence, and their many other needs and wants—quite effectively. Recent changes in the global political and economic environment, however, have raised serious doubts about the adequacy of public- and private-sector institutions, actions, and assumptions that have characterized the U.S. approach to technology development and deployment for the last four decades. The Cold War is over. Meanwhile, other industrialized nations, led by Japan and Germany, have caught up with the United States, first in manufacturing performance, and more recently in some pivotal areas of product technology.



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



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 7
1 Background: The Postwar U.S. Technology Enterprise The United States' collective capacity to create, develop, and deploy new technology constitutes its national technology enterprise. The nation's human, physical, and financial capital, and the publicand private-sector institutions (firms, universities, government agencies, nonprofit research laboratories, financial and regulatory systems, etc.) that organize and direct these resources in service of the interests of U.S. citizens, are all elements of the U.S. technology enterprise. Since World War II, the U.S. technology enterprise and the private and public strategies that have sustained it have been profoundly shaped by the nation's unique economic and geopolitical position in the decades immediately following the war (Nelson, 1990). This period of unchallenged U.S. economic and commercial technological preeminence was a time when the most urgent scientific and technological challenges to the nation were defined by the Cold War, the space race, a domestic war on cancer, and the quest for world leadership in virtually all areas of scientific research. For most of the past 40 years, the U.S. technology enterprise and supporting public- and private-sector technology strategies have served the interests of U.S. citizens—their security, their economic welfare, their global influence, and their many other needs and wants—quite effectively. Recent changes in the global political and economic environment, however, have raised serious doubts about the adequacy of public- and private-sector institutions, actions, and assumptions that have characterized the U.S. approach to technology development and deployment for the last four decades. The Cold War is over. Meanwhile, other industrialized nations, led by Japan and Germany, have caught up with the United States, first in manufacturing performance, and more recently in some pivotal areas of product technology.

OCR for page 7
To better understand the significance of these and other challenges currently facing the U.S. technology enterprise, it is useful first to examine the distinguishing characteristics of that enterprise as they have evolved since the Second World War. FOUR DISTINGUISHING CHARACTERISTICS OF THE U.S. TECHNOLOGY ENTERPRISE As in all market economies, a vast majority of the resources and operational intelligence of the U.S. technology enterprise has resided in private companies and has been organized and driven by the logic of markets. At the same time, the structure, goals, and performance of the U.S. technology enterprise as well as its foreign counterparts have been heavily influenced by the contributions of other publicand private-sector institutions, such as government laboratories, universities, and not-for-profit research institutes. Beyond these broad commonalities, however, the U.S. technology enterprise has been most distinguished from that of other industrial countries by the following four characteristics since World War II: The federal government has focused on mobilizing technical resources to further specific national missions. These missions, undertaken by federal agencies, have included national security, the cure of disease, space exploration, food production, and world leadership in basic science. National economic development and international competitiveness have rarely been explicit objectives of federal technology policies and investments.1 Technology strategies of federal government agencies and some of the most rapidly expanding segments of U.S. industry have focused on R&D-driven breakthroughs in product technology as the key to sustained technological and economic leadership. At the same time, process-related R&D and the organization of technical activities downstream from R&D that drive continuous improvement of existing products and processes have received considerably less public- and private-sector attention (Ergas, 1987; Florida and Kenney, 1990). The federal government and the private sector have maintained a division of roles with regard to the funding of research versus the funding of development and deployment of technology for most sectors of the nation's economy.2 Basic research and the development and application of technology relevant to accepted federal agency missions (though conducted principally by private-sector actors) have been regarded as legitimate activities for funding by the public sector. The identification, development, and adoption of technology for commercial products and services not directly associated with public missions has been seen as the legitimate preserve of the private sector.

OCR for page 7
Responsibility for making and implementing science and technology policies has been dispersed among a diverse collection of federal agencies, state and local governments, and private-sector participants; U.S. science and technology policy is explicitly pluralist, only loosely coordinated, and, at the federal level, largely disconnected from economic policymaking as well as highly influenced by constituency politics (Cohen and Noll, 1991). These fundamental characteristics of the U.S. technology enterprise—each of which is discussed in more detail below—form an important background for understanding the policy challenges facing the United States in the 1990s. The Primacy of Public Missions For more than 40 years, federal government support of the U.S. technology enterprise has had a persistent focus on mobilizing technical resources for national security, space exploration, the cure of disease, the exploitation of nuclear energy, world leadership in basic science, and the other primarily public missions.3 With few exceptions it has been assumed that national economic development need not be specifically addressed by federal science and technology policies. Federally funded basic research would provide a rich feedstock of new science for industry to exploit. "Spillover" technologies—those technologies discovered or developed in the pursuit of public missions and subsequently picked up and applied by private companies driven by market incentives alone—would do the rest. The extent to which national security and other federal agency missions have set R&D priorities for the nation's technology enterprise is well documented (see Table 1.1). Although 80 to 90 percent of all research and development performed in the United States over the past four decades has been performed by private-sector entities (companies, universities and colleges and nonprofit institutions), the federal government has directly funded more than half of the nation's total R&D for most of this period. Since 1960, on an annual basis federal mission agencies have accounted for 60 to 70 percent of the nation's total investment in basic scientific and engineering research, 35 to 56 percent of the nation's total applied research investment, and 40 to 68 percent of the total U.S. investment in technology development. During the past three decades, the federal government's share of the nation's total research and development investment (basic research, applied research and development) has declined significantly—from 65 percent in 1960 to 43 percent in 1992. However, at 43 percent it remains proportionally more

OCR for page 7
TABLE 1.1 Federal Government Role in the U.S. R&D Enterprise, Shares in Percent   1955 1960 1970 1980 1992 (est.) Federal R&D spending as share of total R&D spending 57 65 57 47 43 Federal share total U.S. spending: basic research * 60 70 70 61 applied research * 56 54 45 37 development * 68 55 43 41 Federal defense-related R&D as share of total R&D 48 52 33 24 26 Federal health-related R&D as share of total R&D 2 3 4 6 6 Federal space-related R&D as share of total R&D 1 3 10 5 5 Federal energy-related R&D as share of total R&D * 3 2 6 2 Federal R&D funding as share of total R&D performed by U.S. industry 47 59 43 32 28 Federal R&D funding as share of total U.S. academic R&D 54 63 71 68 57 * Data not available. SOURCES: National Science Foundation (1990a, pp. 55; 1992, pp. 46–48, 52, 56, 60, 62, 69). than twice as large as that of the Japanese government, which at present accounts for only 20 percent of all Japanese R&D spending, and a fifth again as large as that of the German government, which funds 36 percent of all German R&D (National Science Board, 1991; National Science Foundation, 1992). The most direct involvement of the federal government in the nation's R&D enterprise is through the national system of federal laboratories established to serve federal agency missions. Federal agencies currently support more than 700 federal laboratories with a combined budget for FY 1991 of $20.9 billion. These laboratories employ roughly 120,000 R&D scientists and engineers nationwide.4 The reach of federal agencies, however, extends well beyond the federal laboratories to large segments of U.S. industry and U.S. universities. During the past 40 years, publicly supported R&D in the service of federal missions and federal procurement of technologically advanced products, systems, subsystems, and components have contributed significantly to the development of some of the most successful and rapidly growing commercial

OCR for page 7
industries in the United States. These include aerospace, communications, and biomedical and pharmaceutical industries.5 It is estimated that in 1992 over $31 billion of federally funded mission-oriented R&D was performed by U.S. private industry, which, in turn, represented more than 30 percent of all R&D performed by U.S. industry that year (National Science Foundation, 1992a). Throughout the postwar period, federal mission agencies, in particular, the National Institutes of Health (NIH), the National Science Foundation (NSF), and the Department of Defense (DOD), have provided the vast majority of funding for U.S. university-based research6 (see Table 1.2). Although federal agencies' share of total university-based research has declined in recent years, as of 1991 these agencies collectively funded nearly $10 billion, or 56 percent, of all research on American campuses (National Science Board, 1991). Through their funding of university-based research and their pull on labor markets for advanced-degree scientists and engineers, federal mission agencies also contributed significantly to the expansion of the nation's science and engineering work force during the 1950s and 1960s.7 Most notably, rapid growth of the U.S. defense-related R&D effort during this period helped to create and sustain a much larger population of R&D scientists and engineers than in any other Western country. In 1965 the ratio of R&D scientists and engineers to total work force in the United States was nearly three times that of its major industrial competitors (see Figure 1.1). Not until the late 1980s did Germany, Japan, and other industrialized nations achieve ratios approaching those of the United States.8 Driven by the imperatives of the Cold War, national security has long received the highest priority for federal R&D funds. In 1992, national TABLE 1.2 Support for U.S. Academic R&D, Percent Shares by Sector: 1960–1991   1960 1970 1980 1991(est.) Federal government 62.7 70.5 67.5 56.1 State and local government 13.2 9.4 8.2 9.0 Industry 6.2 2.6 3.9 7.3 Academic institutions 9.9 10.4 13.8 19.7 All other sources 8.0 7.1 6.6 7.8 TOTAL 100.0 100.0 100.0 100.0 NOTE: Percentages may not sum to 100 because of rounding. SOURCE: National Science Board (1991, p. 349).

OCR for page 7
FIGURE 1.1 Scientists and engineers engaged in R&D per 10,000 labor force, by country: 1968 and 1989 NOTE: Latest available U.K. data is from 1988. SOURCE: National Science Foundation (1992, p. 67) security accounted for 59 percent of federal R&D spending and 26 percent of total national (public and private) R&D expenditures (see Table 1.3). Roughly 90 percent of defense-related R&D spending has been for "development, testing and evaluation" of weapons and other systems having no markets other than military.9 To a large extent, the demands of the national security mission have determined the structure and objectives of the government's system of federal laboratories, particularly in the physical sciences and engineering research. In 1991 DOD laboratories accounted for nearly half of all federal laboratory obligated expenditures as well as 50 percent of all federal laboratory research scientists and engineers.10 National security has also defined the focus of government support of much industrial and university-based engineering research and development. Most of the federally funded R&D performed by U.S. industry has been concentrated in a few industrial sectors such as aerospace and electronics that have both civilian and national defense components. In 1990, 63 percent of all federal funds for industrial R&D went to the aerospace sector,

OCR for page 7
and 18 percent went to the electrical machinery and communications sector, accounting for 76 percent and 38 percent, respectively, of these industries' total R&D spending in that year (see Table 1.3). Although some of the industrial R&D supported by defense monies has yielded "dual-use" technology (having both civilian and military applications), the vast majority of defense-related R&D performed in these sectors has been for "development" of weapons and other systems having no markets other than military. Although the share of total academic research supported by federal defense agencies has declined significantly during the last 30 years (from 60 percent of all federally supported academic R&D in 1954 to roughly 8 percent in the mid-1980s), the Department of Defense remains a major funder of university-based engineering research. During the late 1980s, DOD provided 32 percent of all funds for academic engineering research: TABLE 1.3 National Security's Contribution to the U.S. R&D Portfolio, Shares in Percent   1955 1960 1970 1980 1992(est.) Defense R&D as share of federal R&D 85 80 58 51 59 Defense R&D as share of total U.S. R&D 48 52 33 24 26 Defense share of total federal support of academic engineering research * * 45a 55 46 Defense share of all government-funded R&D in U.S. industry b * 81 68 63 68 Federal R&D funding as share of total R&D performed by U.S. industry 47 59 43 32 28 Federal share of total R&D funds in aerospace industry 88c 89 77 72 76d Federal share of total R&D funds in electrical machinery and communications 66c 65 52 41 38d NOTES: * Data not available a 1971 data b Department of Defense only, data for 1962, 1970, 1981 and 1989 c 1957 data d 1990 data SOURCES: National Science Foundation (1990a, p. 55; 1991a; 1992, pp. 46–48, 62, 69 and unpublished data, 1993).

OCR for page 7
50 percent for electronics and electrical engineering research, 42 percent for aerospace engineering research, 20 percent for mechanical, 6 percent for civil and 4 percent for chemical engineering research (National Science Foundation, 1990a, 1992).11 Finally, national security has been a significant claimant on the nation's technological work force during the past three decades. Because of the high engineering intensity of defense-related economic activity, it is estimated that the national security mission currently commands roughly 18 percent of the total U.S. engineering work force. 12 In summary, federal involvement in the technology enterprise through pursuit of agency missions has been extensive. Federal agencies, through their procurement of goods and services and their investment in R&D, have had a profound influence on the growth and direction of U.S. science and engineering research and education, on the growth and deployment of the nation's science and engineering work force, on the pace and direction of technical change in major sectors of the U.S. economy, and on allocation of technological and complementary investment resources throughout the economy (Casagrande, 1992). The Priority of Research and Development in Postwar Policy Throughout the postwar period, the U.S. public sector and some of the most rapidly expanding segments of U.S. industry have tended to focus on research and development aimed at technological breakthroughs as the key to sustained technological and economic leadership. In general, public and private technology strategies have been based on the (sometimes not fully acknowledged) premise that the supply of new technological ideas and concepts, rather than market demand, is the pacing factor in economic progress. Actual or potential market demand for products and services has been assumed to be sufficiently large and well-organized to pull substantial fractions of scientific and technological discoveries into use as fast as they emerged from the laboratory. A driving force behind this post-World War II orientation toward R&D-driven new technology creation was the recognition that despite its industrial supremacy, the United States lacked a sufficiently broad-based institutional capacity for scientific research and development (Bush, 1945). For more than half a century before World War II, large sectors of U.S. industry had risen to global preeminence by drawing extensively on the results of foreign research and development. The spectacular productive performance of the United States in World War II, confirmed U.S. superiority in technical areas downstream from research, such as design, development, engineering and production. However, despite major gains in domestic R&D capability

OCR for page 7
during the war, the U.S. government and many sectors of U.S. industry and academe considered the nation's science and engineering research base inadequate to the nation's new economic and geopolitical role. During the 1950s and 1960s, public- and private-sector leaders assumed that unparalleled capacity to generate new science and technology was not only a necessary but also a sufficient condition for the maintenance of U.S. technological and economic leadership in newly emerging industrial sectors such as computers and semiconductors, or more established science-based industries such as pharmaceuticals or chemicals. The U.S. domestic market was by far the largest, wealthiest and most technologically sophisticated in the world. U.S. leadership in mass production and distribution, then the world's leading system of industrial production, was unchallenged. Foreign competition was minimal or nonexistent in most of the new high-tech sectors. These factors, when combined with an exceptionally large U.S. population of scientists and engineers, made it possible for U.S. firms in these high-growth industries to convert new technological ideas into commercially viable products more rapidly than their competitors as well as to translate that advantage into global market dominance in many R&D-intensive industries.13 Meanwhile, in less R&D-intensive manufacturing and service industries such as automobiles, steel, machine tools, construction and financial services, firm technology strategies, to the extent they were articulated at all, tended to be focused on incremental improvements in existing products with an emphasis on product design and marketing. Relatively immune to foreign competition as a result of the large economies of scale in production and distribution afforded by the U.S. domestic market, most firms in these industries devoted little attention and fewer resources to process innovation or the pursuit of product technology breakthroughs. By the late 1970s, many of the deeply entrenched organizational and managerial practices of U.S. firms with regard to product development, design, production and marketing—practices that served these firms so effectively during the 1950s and 1960s—were beginning to handicap U.S. companies in several high-tech and non-high-tech sectors. Many U.S. companies were slow to pick up on major improvements in the pace and efficiency of product development and in the flexibility, efficiency and quality of production systems made by major Japanese firms and would later regret their relative inattention to process technology and the integration of R&D with its complementary downstream technical activities.14 Nevertheless, in the unique circumstances pertaining for much of the postwar period, U.S. citizens and institutions enjoyed an unprecedentedly high but, as it proved, temporary probability of capturing virtually all the potential economic returns on the nation's large public and private investments in research and development.

OCR for page 7
The Division of Roles Between the Public and Private Sectors The postwar U.S. technology enterprise has been characterized by a relatively sharp division of roles between the federal government and private-sector participants with regard to the funding of research and development and the application of commercial technology for most sectors of the nation's economy. Since the late 1940s it has been widely accepted that the federal government should play a central role in supporting the nation's basic research enterprise and its system of advanced scientific and engineering education (Bush, 1945). Like technology procured by the federal government to advance particular federal agency missions, basic research and advanced technical education have been viewed as essential public goods—goods that benefit society at large. Because of their public goods character, these areas of technological activity have not attracted sufficient investment from the private sector alone to meet societal needs, and have, therefore, been deemed appropriate areas for government intervention and support. In contrast, nearly all technology not directly procured for use by the government has been perceived to be a private good that—like equipment or real estate—could be treated by a company as an asset (all or most returns on investments in technology were assumed to accrue to the investor). Likewise, technical activities beyond basic research, or not directly associated with the specific public missions, have been viewed as the exclusive responsibility of private-sector participants operating within competitive markets; the identification, development and adoption of commercially useful technology has been left to private companies.15 The federal government has, of course, set the climate for these private-sector technical activities through pursuit of growth-oriented macroeconomic policies, the regulation of markets, the guarantee of intellectual property rights, and other critical market-sustaining policy actions. A similar public-private division of roles has been assumed for ''spin-off" benefits from public science and technology missions, such as technologies emerging from federally funded space, defense and biomedical research programs. Here again, it was generally accepted that market incentives alone were sufficient to motivate the private sector to pick up and adapt these developments for commercial use.16 The notion of a clean dichotomy of public-sector responsibility for basic research and private-sector responsibility for commercial development and use of technology has been an important signpost in U.S. debates over the role of government in civilian technology. Economic theory and historical experience argue that many areas of commercially relevant technological activity beyond basic research yield, or promise to yield, high returns to society as a whole yet pose risks too high or offer private returns too low to

OCR for page 7
attract sufficient private-sector investment (Brooks, 1986; Nelson, 1989; Nelson et al., 1967). These activities include Research and development related to "pathbreaking" technologies that might give rise to a major new industry or transform existing industries. Pathbreaking technologies are characterized by high technical risk and by uncertain and possibly long-delayed economic payoffs, which may discourage private-sector investment (Alic et al., 1992). Examples of past and present pathbreaking technologies include nuclear medicine, biotechnology, semiconductors, aircraft engines and communications satellites. Research, development and institutional and technical support related to "infrastructural," or "generic," technologies—generally low technical risk, relatively low-cost technologies that enhance the performance of a broad spectrum of firms in the near to midterm, but whose benefits cannot be predominantly captured by any one firm. 17 Infrastructural technologies include the development of engineering methods; compilation and validation of technical data; development and characterization of materials, measurement tools and instrumentation; and refinement of manufacturing processes. Investments in research, development, institutional and technical support and complementary assets that facilitate the timely identification, adoption and diffusion of new scientific and technological knowledge throughout the national economy. These include investments in work force training, travel budgets for resident research and advanced production personnel, development or use of information data bases and networks and provision of technical or industrial extension services. Despite the compelling logic for public-sector intervention to compensate for these market failures—articulated by many scholars and political and industrial leaders over the last 40 years—efforts to develop an explicit federal role in this area have rarely taken hold.18 As a result, federal support for commercially relevant technological activities has not been broadly institutionalized. That is not to say that the federal government has not contributed, in some cases significantly, to the development and diffusion of pathbreaking and infrastructural technologies through its advancement of federal agency missions. Clearly, it has.19 In general the federal government has not considered the development of commercially relevant technologies or commercial technology diffusion a legitimate part of its technology investment portfolio. Notable exceptions are the relatively small-scale industrial technical support (standards, testing, and evaluation) provided historically by the National Bureau of Standards (renamed the National Institute of Standards and Technology in 1989), and more recent, limited initiatives such as the Advanced Technology Program and the Small Business Innovation Research (SBIR) program, or the cultivation of industry-federal laboratory cooperative research and development agreements (CRADAs).20

OCR for page 7
The reluctance of the federal government to explicitly breach the division of responsibilities between public research and private technology in the commercial sphere since World War II can be explained by a number of factors. Throughout the 1950s and 1960s, the relatively impressive performance of the U.S. commercial technology enterprise had not suggested any obvious gaps in the nation's commercial technology portfolio. U.S. leadership in new high-technology industries and in the development of new products and services was taken for granted. Certainly, several major U.S. high-technology industries benefited greatly from federal mission-related R&D and procurement during their rise to commercial dominance. Recognition of this fact in the absence of clear threats to the nation's commercial technology base, however, did not translate into a persuasive call for a larger, more explicit federal role in support of commercial technology development and diffusion. Yet, in recent decades, even as the existence of important gaps in the nation's commercial technology portfolio has gained wider credence, the analytical and political impediments to an expanded federal role in this area have remained formidable. The strong ideological commitment of American government to the power of free markets and limited government intervention in the nation's economy has for the most part contained congressional attempts to expand the government's role in civilian technology. At the same time, many of those who acknowledge the need to redress gaps in the commercial technology portfolio in principle have been reluctant to take concrete policy actions. This is, in part, because the theoretical and empirical bases for identifying, setting priorities for, and deciding at what level to fund worthy areas of infrastructural or pathbreaking technology are not well established. An even greater impediment to policy action, however, is the fact that the benefits of public-sector investments in these areas of commercial technology are hard to measure, slow to diffuse and slow to mature. In short, the need for elected representatives and government officials to demonstrate concrete, short-term results to their constituencies may discourage them from investing much political capital in such diffuse, long-term yield initiatives. A corollary to this political imperative is the fear that direct industry funding might be yet another breeding ground for "pork."21 The "Nonsystem" of U.S. Science and Technology Policymaking A final distinguishing feature of the postwar U.S. technology enterprise has been the pluralist, decentralized, loosely coordinated structure of U.S. science and technology policymaking. This structure has been characterized by the lack of an explicit commitment at the federal level to supporting technology development and deployment for economic development, and

OCR for page 7
by the corresponding divorce of science and technology considerations from the making of federal economic policy. In addition, there has been little effort to coordinate or consolidate the autonomous yet often overlapping science and technology policies of the diverse federal mission agencies that collectively define U.S. federal technology policy overall. For example, it is only in the last three to four years that the federal government has begun to take stock of (let alone, begin to coordinate) the investments of various mission agencies in technology areas of mutual interest, such as manufacturing and advanced materials.22 The fragmented nature of science and technology policymaking at the federal level, its disconnectedness from economic policymaking, and the federal structure of the U.S. political system have contributed to even greater decentralization and fragmentation of U.S. science and technology policymaking at the subfederal level. With the explicit objective of advancing economic development within their jurisdictions, many state, regional, and local entities have pursued science and technology policies of their own (Carnegie Commission, 1992a; Clarke and Dobson, 1991; Feller 1992a,b; Plosila, 1987; Shapira et al., 1992). Today at least 46 states and countless municipalities and counties pursue a range of policies aimed at promoting the creation, dissemination, and application of commercial technology within their jurisdictions. 23 Funding for these efforts involves much more modest resources than those invested by the federal government in mission-oriented R&D; state governments' spending for research and development and R&D plant totaled a mere $1.2 billion in fiscal year 1988, compared with a federal R&D investment that year of approximately $58 billion.24 Historically, there has been little coordination (formal or informal) among these subfederal programs or between them and federal agency efforts, though this is beginning to change.25 In many respects, the pluralist nature of U.S. technology policymaking has both reflected and reinforced the pluralist structure of the technology enterprise proper. To an extent far greater than in other industrialized nations, operational responsibility for research, technology development, and technology application in the United States is distributed among a large, highly diverse population of public- and private-sector participants. These include private companies, trade and industry associations, universities, private research institutes, community colleges, professional associations, private or private/public consortia, and local, state, and federal government agencies. Moreover, the fact that the nation's science and technology capabilities are dispersed over a large number of regions and political constituencies has greatly increased the importance of constituency politics in federal technology policymaking and implementation. On the one hand, the decentralized nature of U.S. science and technology policymaking has allowed diverse and locally adaptive policy responses

OCR for page 7
to technology-related challenges at the federal and the subfederal levels. On the other hand, this same highly distributed, highly fragmented quality of U.S. science and technology policymaking, in combination with the near total divorce from economic policymaking at the federal level, have greatly impeded collective action on issues and problems that cut across political jurisdictions, and inhibited cooperation in the setting and implementation of national priorities. THE POSTWAR PERFORMANCE OF THE U.S. TECHNOLOGY ENTERPRISE IN PERSPECTIVE Throughout much of the past 40 years, the distinguishing features of the U.S. technology enterprise have served the nation's multiple interests effectively. By and large the most important goals of mission-oriented publicly funded research and development have been achieved. The United States has achieved and sustained preeminence in defense-related technologies, which it has used effectively to strengthen U.S. national security and U.S. influence throughout the world. The United States has been on the forefront of biomedical research, leading the world in the ability to treat and control many diseases.26 As a result of heavy, sustained federal support, the U.S. basic research enterprise and U.S. advanced science and engineering education, after World War II, quickly achieved and continue to enjoy world leadership status. Furthermore, the direct and indirect contributions of defense and other public missions to U.S. civilian technology development have been substantial. Federal agency R&D and large-scale federal procurement of advanced technology products generated important spin-offs. Some of those spin-offs were seminal to the growth and development of industries that have been major engines of U.S. and world economic growth, such as the aerospace, microelectronics, telecommunications, and computer industries.27 Heavy federal mission agency funding of university science and engineering research departments helped to provide the intellectual underpinnings and the highly skilled human capital base for many newer, high-growth, science-based industries (computer science, pharmaceuticals, chemicals, and technical advances in many other areas). Likewise, throughout the 1950s and 1960s, the performance of U.S. companies at home and abroad tended to confirm belief in the effectiveness of the nation's division of responsibility between the public and the private sectors with regard to research, development, and the commercial application of technology as well as its collective focus on research and development as the key to technological leadership. As of the early 1970s, U.S. productivity (gross domestic product per capita) was one and a half times that of Germany and Japan, and U.S. industry accounted for half of world

OCR for page 7
high-tech production, more than a quarter of global high-tech exports, and nearly half of the total world stock of foreign direct investment (Maddison, 1989; National Science Board, 1989; U.S. Department of Commerce, 1989b) (see Figure 1.2).28 The U.S. domestic market appeared to be largely immune to foreign competition, and the nation's ability to spawn new products, services, and industries was unrivaled.29 The 1970s and 1980s witnessed growing concern in the United States about the health and performance of the nation's commercial technology enterprise. As major U.S. manufacturing industries were outdone by foreign competition, both U.S. industry and the federal government sought to understand better the changing nature of international competition and its implications for U.S. competitiveness. This, in turn, led to a series of reevaluations of the private-public division of labor with regard to civilian technology development.30 FIGURE 1.2 National shares of world high-tech production and trade, by country: 1970. NOTE: Based on data valued in current U.S. dollars; uses OECD definition of "high intensity technology products," see Chapter 1, note 28. SOURCE: National Science Board (1989, pp. 371, 377).

OCR for page 7
Nevertheless, even as the nation's vision of the federal role in the commercial development and application of technology began to be questioned, the focus of both public- and private-sector technology strategies has remained on research, early development, and acceleration of the rate of generation of new technology as the principal technological response to early signs of the nation's declining competitiveness. The concentration on invention and new technology development also resonated with a deeply ingrained "product-cycle" view of national industrial evolution. As U.S. companies lost market share to foreign competitors in what were considered "technologically mature" or lower valued-added industries or segments of industries, it was assumed that U.S. economic preeminence would be continually renewed (Kodama, 1991; Thurow, 1980; Vernon, 1966). This renewal would be achieved chiefly by exploiting the perpetual ability of the nation's technology base to create new, technologically dynamic, high-growth industries, such as computers, telecommunications, commercial aircraft, and pharmaceuticals. In short, U.S. national interests were thought to be best served by technology strategies that focused on maintaining U.S. leadership in the creation of new ideas and new technologies. These, in turn, could be counted on to seed new industries and new, higher value-added markets to compensate for ''old" industries and markets lost to foreign competition. During the past 10 to 15 years, however, the global political and economic environment has undergone a profound change. As a result, the strategies and tactics that have guided U.S. investment in science and technology over most of the second half of the twentieth century have become less effective and are likely to become even less so during the next 50 years. The next chapter discusses changes in the global context that challenge the adequacy of established U.S. public-and private-sector approaches to technology development and deployment. Notes 1.   Henry Ergas (1987) has classified national technology strategies as "mission-oriented," "diffusion-oriented," and "hybrid." The United States, Great Britain, and France are cited as examples of "mission-oriented" strategies, Germany, Sweden, and Switzerland as examples of "diffusion-oriented" strategies, and Japan as somewhere in between. 2.   This sharp division has frequently been stricter in theory and rhetoric than in practice since World War II. See Brooks (1986), Cohen and Noll (1991), Kash (1989), Mowery and Rosenberg (1989), and Nelson (1989). 3.   During this period, federal support for mission-oriented research and development has been diminishing in defense, steadily rising in public health, and highly volatile in other public mission areas such as space, energy, environment, and housing (National Science Foundation, 1990a). 4.   This total includes intramural agency laboratories as well as federally funded research and development centers (FFRDCs). FFRDCs and many intramural agency laboratories are government-owned, contractor-operated laboratories managed by universities (Los Alamos, Lincoln Laboratory), university consortia (Brookhaven, Fermilab), industrial contractors on a not-for-profit basis (Oak Ridge National Laboratory administered by Martin Marietta,) and independent nonprofits (MITRE Corporation, Draper Laboratory, RAND). Other intramural agency laboratories are government-owned and government-operated (National Institutes of Health, NASA's space flight and space science laboratories, National Institute of Standards and Technology, Naval Research Laboratory, Naval Surface Weapons Center). In addition to these differences in management structure, federal laboratories are very diverse in size, character, and purpose. Most are single-office facilities employing a small number of researchers, whereas others are large organizations that employ thousands of scientists and engineers (Committee on Science, Engineering, and Public Policy, 1992, pp. 67–79).

OCR for page 7
5.   Several recent studies have noted that the "spillovers" from defense and other federal agency missions were greatest during the early postwar period, when many of the technologies that later became new and highly competitive commercial industries were in the early, more "fluid" stage of their technology life cycle. In most cases, these mission-related spillovers have declined in importance as these industries and their relevant technologies have matured. However, there are notable exceptions. Aerospace technology generally, and aircraft engine technology in particular, remain as "dual-use" as ever. The massive federal investment in health-related research clearly continues to yield significant spillovers to the biomedical and pharmaceutical industries. See Chapter 2, pp. 53–54 below for further discussion. See also Alic et al. (1992); Mowery (1987); Committee on Science, Engineering, and Public Policy (1992); Utterback (1987). 6.   The National Institutes of Health and the National Science Foundation came to account for the largest shares of federal support for academic research during the 1960s and 1970s, paralleling growth in academic life sciences research both absolutely and as a share of total academic research. In 1989 NIH accounted for 47.9 percent of total federal obligations for academic research and development, NSF for 14.5 percent, and DOD for 13.7 percent. In 1989 life sciences research accounted for 54 percent of academic science and engineering research expenditures (National Science Board, 1991, pp. 355, 360, appendix tables 5-6, 5-8; Government-University-Industry Research Roundtable, 1989, p. 2–23). 7.   The Servicemen's Readjustment Act of 1944 (the G.I. Bill of Rights), which provided funds for World War II veterans to continue their education, contributed significantly to the early growth of the nation's advanced technological work force. The G.I. Bill enabled more than 2.2 million ex-servicemen to attend colleges and universities; see Ginzberg (1986). 8.   As of 1989, the United States had 76 scientists and engineers engaged in R&D per 10,000 labor force, compared with Japan's 74, Germany's 59, and France's 50 per 10,000 labor force (National Science Foundation, 1992, p. 67). 9.   It is worth noting the large contrast in the distribution of effort between publicly funded defense and nondefense research and development. Ninety percent of public funding for defense-related R&D is for development, testing, and evaluation, with applied research, basic research, and R&D plant accounting for the remaining 10 percent. In contrast, public nondefense R&D spending is divided more evenly among the 3 major categories with 30 percent for development, 30 percent for applied research, and 30 percent basis research, with the remaining 10 percent for R&D plant (National Science Board, 1991, pp. 94–95). 10.   If Department of Energy (DOE) laboratories that focus primarily on nuclear weapons research are added to those of DOD, the national security mission laboratories account for roughly 55 percent of total federal laboratory expenditures and 60 to 70 percent of total laboratory researchers. At present slightly less than half of all DOE laboratory resources are dedicated to weapons research (Committee on Science, Engineering, and Public Policy, 1992, pp. 68, 74, tables 2-1, 2-3). 11.   Alic et al. (1992) note that most of the university-based engineering research sponsored by DOD, DOE, the Atomic Energy Commission (AEC), and the National Aeronautics and Space Administration (NASA) "was 'engineering science'—i.e., investigations of natural phenomena underlying engineering practice—rather that engineering design, manufacturing operations, or the construction and testing of prototype equipment."

OCR for page 7
12.   The U.S. Congress, Office of Technology Assessment (1992a) estimates that 342,000 engineers were engaged in defense work in 1990 out of total U.S. engineering work force of 1.86 million. Henry and Oliver (1987) in their survey of the U.S. defense build-up from 1977 to 1985 and its claims on the U.S. labor force estimated that 15 percent of the nation's "technical professionals" were employed in defense-related economic activity. 13.   Ergas (1987), Florida and Kenney (1990), and Nelson (1992) all note the mutually reinforcing character of U.S. technology strategies (public and private) and the postwar evolution and development of a broader set of U.S. institutions—research universities, venture capital markets, regulatory agencies, company law, etc.—with regard to the U.S. focus on technological breakthroughs and resulting comparative advantage in new science-related technologies and industries. 14.   The preoccupation with R&D has been reflected in the technical "caste" systems and reward systems of companies in these industries under which conceptualizers and analysts involved in marketing, invention, and research have enjoyed considerably greater status and financial rewards than their colleagues in design, development, and manufacturing who transform broad concepts into working systems. 15.   There are three recent exceptions to this characterization. First, following on legislation of the early 1980s that made technology transfer from federal agency laboratories to the private sector an explicit objective of federal policy, the 1986 Federal Technology Transfer Act authorized the establishment of cooperative research and development agreements (CRADAs) between government-operated laboratories and industry. Second, the National Institute of Standards and Technology, NIST, (formerly the National Bureau of Standards, NBS) has recently had its mission expanded to include support of "generic" advanced technologies important to certain sectors of the civilian economy with the establishment of the Advanced Technology Program. And third, the Small Business Innovation Research (SBIR) program was created in 1982 to direct a small share (not less than 1.25 percent) of each major mission agency's total annual R&D budget to fund R&D at small and medium-sized firms and to stimulate the commercialization of new products and services. The SBIR program was significantly expanded by Congress in the fall of 1992. For further discussion of these initiatives see Committee on Science, Engineering, and Public Policy (1992) and U.S. General Accounting Office (1992c). See also the Small Business Research and Development Enhancement Act of 1992 (P.L. 102-564). The SBIR program, CRADAs, and the expansion of NIST's mission beyond the much narrower standards-oriented mission of NBS are in an infant stage. The NIST/NBS budget has been virtually level for most of the last 10–15 years, reflecting a long-standing low priority given to insfrastructural research and technology. However, significant increases in NIST funding in the fiscal 1992 and 1993 federal budgets reflect a growing recognition in Congress and the administration of the increased importance of NIST's charge. 16.   Among industries that serve accepted public missions as well as commercial markets, the division of responsibility among private firms and government with regard to research, technology development, and technology deployment has varied considerably from industry to industry. Unlike defense, which has been a purely public mission monopolized by public funding, a natural division of labor developed between the government and industry in other fields such as agricultural and health. In these two fields, government provided funding primarily for the life science aspects where it was to develop appropriable knowledge, while the private sector funded the physical sciences and engineering, where the knowledge has tended to be more appropriable. Thus, in biomedicine and agriculture there developed private sectors whose R&D expenditures approximated in magnitude federal expenditures but were complementary. A somewhat different division of labor developed in aeronautics. Here the government provided generic knowledge and testing facilities such as wind tunnels, which became a source of public knowledge available to all competitors, while industry carried out the design and development work and took responsibility for commercialization of actual aircraft. Similarly, the markets for these industries have been "mixed" (public and private) to varying degrees. The aircraft industry's market has consisted largely of regulated air carriers and the military, where government had control of the market ground rules. For space technologies, the market has been virtually synonymous with the federal government. Even in satellite communications, where the products are sold by private companies, the market was a regulated monopoly throughout most of its postwar development. In the pharmaceutical and agricultural sectors, the market ground rules have been heavily regulated by government—with price supports in agriculture, and safety and efficacy regulation in pharmaceuticals and medical devices. In the latter two industries, there has been a substantial government market from Medicare, Medicaid, and the Veterans Administration (probably a good deal more than 30 percent of the total market, with much of the rest determined by third-party payers, which were subject to price regulation by the states). Somewhat similar considerations would apply to nuclear power where the government provided a great deal of R&D funding and some infrastructure, while the heavily regulated electric utilities financed the development of, and capital investment in, actual power reactors. Here again is a case of mixed economy, ostensibly private, but with heavy government regulation and complementarity between the public and private roles. For further discussion of the diverse mix of public and private roles in these industries, see Brooks (1982), Committee on Science, Engineering, and Public Policy (1992), and Kash (1989).

OCR for page 7
17.   The concepts of "pathbreaking" and "infrastructural" technologies were taken from Alic et al. (1992, chapter 10). Fundamental technical problems can and do arise at any stage of the technological life cycle, not just in the conceptual or exploratory phase; the lack of critical knowledge, empirical or theoretical, can appear as a barrier to incremental improvement even in a nominally mature technology. The research necessary to turn a new technological concept into a commercial product may or may not be a private good. Such research creates useful technical knowledge, which contributes to the function of the product or service, or contributes to reducing the cost or otherwise improving the efficiency of producing a product or service. If the results are widely applicable to many products or even several different industries, the payout may not be sufficient for any single company to support the work. For example, if the knowledge is in the form of data such as characteristics of materials or aerodynamic performance of wing shapes, or flow characteristics of various shapes of orifices, it is simply a fact of nature and can seldom be held proprietary and hence economically appropriable to the creator of the knowledge. In such cases, there may be sufficient mutual advantage among many product lines and industries to justify a collective or shared investment (perhaps with additional government support) in acquiring the relevant knowledge even if the demand for that knowledge is generated by immediate commercial concerns. See Brooks (1991). 18.   Congress refused to support the Kennedy and Johnson administrations' attempts to develop modest research support programs within the Department of Commerce for textile, building, and machine tool industries during the early 1960s (Economic Report of the President, January 1963; Katz, 1982; Nelkin, 1971). The Nixon administration's grandiose federal initiative to use government-generated technology to bolster the competitiveness of the U.S. economy during the early 1970s produced only a few small pilot programs in the National Science Foundation and the National Bureau of Standards (including the limited, yet rather successful NBS Experimental Technology Incentives Program) that were never followed up (Lewis, 1975, 1976; National Research Council, 1976). See also the Carter administration's 1978 initiative to study the impact of federal policies on national economic competitiveness and to make recommendations for changes in federal policy to improve incentives for private-sector technological innovation and industrial investment in R&D (U.S. Department of Commerce, 1979).

OCR for page 7
19.   For further discussion, see pages 9-16 and note 16 above. For an overview of the role of the Advanced Research Projects Agency and its successor the Defense Advanced Research Projects Agency in the development of dual-use technology, see also Mowery and Rosenberg (1989). 20.   See Chapter 1, note 15, for further discussion. 21.   It should be noted, however, that all of the uncertainties and risks associated with targeting specific technology areas as worthy of public support have not deterred major trading partners of the United States from doing so, albeit with mixed success. See Keck (1993), U.S. Congress, Office of Technology Assessment (1991b), and U.S. Department of Commerce (1992b). 22.   In 1989 President Bush's new science and technology adviser and Office of Science and Technology Policy (OSTP) director, D. Allan Bromley, reestablished the dormant Federal Coordinating Council on Science, Engineering, and Technology (FCCSET) as a means to get federal agencies to coordinate their R&D programs. Since then FCCSET has launched six assessments of federal agency R&D programs in particular technology areas (technology "crosscuts"), including advanced materials and processing, biotechnology, global change, high-performance computing and communications, math and science education, and advanced manufacturing. 23.   State-level industrial extension and economic development programs are long-standing. It was not until the 1980s, however, that state governments launched major initiatives emphasizing technology development, the search for new products and processes, and the launching of new spin-off firms (Carnegie Commission. 1992a; Clarke and Dobson, 1991; Feller, 1991, 1992; Osborne, 1989; Plosila, 1987; State of Minnesota, 1988). 24.   In its recent report entitled Science, Technology, and the States in America's Third Century, the Carnegie Commission on Science, Technology, and Government (1992a) estimates that the total public and private resources leveraged by state spending (mostly matching investments by private industry) in fiscal 1988 was in excess of $2 billion. See also National Science Foundation (1990b). 25.   Often, matching funds from state or federal resources are required by one or both parties for the financing of broad-based research centers. 26.   Nevertheless, as recent analysis and debate of the U.S. "health care crisis" suggest, it would be wrong to claim that the United States actually leads in bringing to bear this superior capability in delivering health benefits to all its heterogeneous population. 27.   While the spin-off benefits of national security spending have been larger than they would have been had no federal money been spent, they are almost certainly smaller than they would have been had comparable amounts of money been invested directly with an explicit mission of economic development. In other words, the spin-off benefit per dollar of expenditures is probably tiny compared with the potential benefit of a comparable amount of commercial industrial R&D and capital investment. This may constitute a politically unrealistic standard of comparison, because it is hard to construct a counterfactual political scenario in which the U.S. body politic could have been persuaded to devote similar amounts of money to commercially oriented R&D. Nevertheless, it is an important point that has been poorly understood by both sides in the spin-off debate. See Alic et al. (1992), especially pp. 54–81 for further discussion. 28.   The OECD classification of "high intensity technology products" relies on directly applied R&D expenditures in its calculation and includes those products with above-average R&D intensities. Direct R&D expenditures are those made by the firms in the product group. The OECD classifies the following industries as high-tech: drugs and medicines (ISIC 3522); office machinery, computers (ISIC 3825); electrical machinery (ISIC 383 less 3832); electronic components (ISIC 3832); aerospace (ISIC 3845); and scientific instruments (ISIC 385). The Department of Commerce definition of high-technology products (DOC-3 high-technology products) includes products that have significantly higher ratios of direct and indirect R&D expenditures to shipments than do other product groups. Direct R&D expenditures are those made by the firms in the product group. Indirect R&D describes the R&D content of input products. The DOC-3 industries include guided missiles and spacecraft (SIC 376); communication equipment and electronic components (SIC 365–367); aircraft and parts (SIC 372); office, computing, and accounting machines (SIC 357); ordnance and accessories (SIC 348); drugs and medicines (SIC 283); industrial inorganic chemicals (SIC 281); professional and scientific instruments (SIC 38 less 3825); engines, turbines, and parts (SIC 351); and plastic materials and synthetic resins, rubber, and fibers (SIC 282). Comparisons of U.S production data for "high-intensity technology products," as reported to the OECD, with U.S. total shipment data for "high-technology" products—as reported to the Department of Commerce according to DOC-3 definition—show that the OECD data represented 96 percent and 100 percent of the DOC-3 data in 1980 and 1986, respectively (National Science Board, 1989, pp. 149–150).

OCR for page 7
29.   Imports accounted for less than 5 percent of total U.S. domestic consumption of high-tech products in 1970 (National Science Board, 1989, p. 375, table 7-7). 30.   Among the most important and influential studies were the 1979 report entitled Domestic Policy Review of Industrial Innovation from the U.S. Department of Commerce and the Office of Science and Technology Policy; the report of the President's Commission on Industrial Competitiveness (1985); and the formal statement of U.S. technology policy by the Bush administration (Executive Office of the President, 1990). See also Council on Competitiveness (1991) and National Academy of Engineering (1988). For a review of recent reports on U.S. technology policy, see Mogee (1991).