National Academies Press: OpenBook

The Positive Sum Strategy: Harnessing Technology for Economic Growth (1986)

Chapter: National Science Policy and Technoligical Innovation

« Previous: Technology and Its Role in Modern Society
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 119
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 120
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 121
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 122
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 123
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 124
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 125
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 126
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 127
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 128
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 129
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 130
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 131
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 132
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 133
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 134
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 135
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 136
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 137
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 138
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 139
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 140
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 141
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 142
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 143
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 144
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 145
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 146
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 147
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 148
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 149
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 150
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 151
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 152
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 153
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 154
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 155
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 156
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 157
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 158
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 159
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 160
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 161
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 162
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 163
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 164
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 165
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 166
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 167
Suggested Citation:"National Science Policy and Technoligical Innovation." National Research Council. 1986. The Positive Sum Strategy: Harnessing Technology for Economic Growth. Washington, DC: The National Academies Press. doi: 10.17226/612.
×
Page 168

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

National Science Policy and Technological Innovation HARVEY BROOKS There is lime debate about Be necessity of a federal role in technological innovation when the government is the ultimate user and the goods or services produced are widely acknowledged to be "public goods.'' There is general agreement, too, that where the costs of R&D can be entirely recovered from the future revenue streams generated by products, services, or information, there is lime justification for a government role. In many cases, however, the proper relative roles of the public aM private sector are highly controversial. An important factor in each of these cases is who makes the choices and strategic judgments as the R&D evolves. Here, whether the judgments to be made relate primarily to science or technology considerations or primarily to market considerations is ofterz key. HISTORIC ROLES OF GOVERNMENT IN SCIENCE AND TECHNOLOGICAL INNOVATION Although industry is the dominant source of commercially significant tech- nology in the United States, government has been a much more important and direct influence on the direction and rate of technological innovation than much of our national ideology and public rhetoric would lead us to suppose. Government in particular has been a source of much Generic technology," as well as fundamental science, which has then served as a substrate for technological innovation by the private industrial sector. Government has supported the generation of new knowledge and tech- niques directly, for example, through sponsoring the exploration of the largely unknown American continent in the early nineteenth century, or through the creation of such government agencies as the Agricultural Research Service, the National Bureau of Standards, the Geological Survey, and the National Advisory Committee for Aeronautics (NACA) in He late nineteenth and early twentieth centunes. It has also subsidized the expansion of certain basic 119

120 HARVEY BROOKS industries: the canal system in the early nineteenth century, the westward extension of the railroads in the mid-nineteenth century, the creation of a national highway system to undergird a growing auto industry, the devel- opment of an infrastructure of airports and air traffic control as well as air mail subsidies to sustain the grow of a commercial air transport system, and special tax benefits to stimulate the development of the domestic petro- leum industry to name just a few examples of indirect government involve- ment. These indirect subsidies had the effect of creating a "demand pull" for new technologies, not only within the industries immediately affected, but also in collateral industries that supplied or serviced the subsidized in- dustries. For example, the demand for durable steel rails for the railroads was a major factor driving technological innovation in the burgeoning steel industry (Morison, 1974:72-861. Tax benefits for the petroleum industry not only resulted in cheaper fuel, which stimulated demand for automobiles, but also fostered innovation in oil exploration and drilling technology, in which We United States still leads We world. The subsidy for highways indirectly stimulated innovation in highway construction and planning techniques, but it also influenced the direction of innovation in the automobile industry toward large and powerful cars with increased driving amenities, a stimulus that was reinforced by tax benefits to the oil industry, which effectively lowered gasoline prices. Thus, in hundreds of ways, government throughout American history has influenced the priorities of entrepreneurs and innovators in We private sector. This influence has been no less when it was inadvertent or incidental to some over government purpose, such as national defense, than when it was explicit and intentional, as in We case of U.S. agricultural programs or water development in the West. Throughout American history, also, the military has often been a direct or indirect source of technological innovation. Sometimes security consid- eraiions have been used as an important justification to command a wider political consensus, as was the case win federal sponsorship of the Interstate Highway System in 1956 (Rose, 1979), the financing of aeronautical research after World War ~ through the National Advisory Committee for Aeronautics (Mowery and Rosenberg, 1982; Nelson, 1977:111; Nelson, 1984:51-52), and the creation of He U.S. Naval Observatory in the 1840s. A. Hunter Dupree (1957:62) has observed that "the Naval Observatory is the classic example of the surreptitious creation of a scientific institution by underlings in the executive branch of the government in He very shadow of Congres- sional disapproval." Introduced in He guise of a "Depot of Charts and Instruments," ostensibly to standardize chronometers on naval ships for more accurate navigation, the observatory quickly grew into a major center for studies in hydrography, astronomy, magnetism, and meteorology, so that even today it is the leading world center for astrometric observations and the source of time and star-posit~on standards for practically the entire world.

NATIONAL SCIENCE POLICY AlID TECHNOLOGICAL INNOVATION 121 In the early nineteenth century the military loaned its officers to help survey for the railroads and generally to assist them in solving civil engineering problems. In those days military expeditions and surveys were better staffed and supported, and closer to the best elements of American science, than were any civilian projects (Dupree, 1957:651. Also in the mid-nineteenth century the government arsenals at Harper's Ferry, Virginia, and Springfield, Massachusetts, pioneered in the development and introduction of milling machines and other machine tools, and in proving out the principles of mass production and interchangeable parts (Rosenberg, 1976:20~. Indeed, in the whole evolution of the American scientific establishment to this day one can discern a consistent pattern in which technical sophisti- cation has diffused outward from military science and technology into the civilian economy and eventually into the whole political and social structure. This has even been true for Me introduction of new technologies less ob- viously related to military applications. Medical developments such as an- tibiotics, techniques of blood preservation, and the use of chemical pesticides to control disease vectors were initially introduced in connection with Me military. Much of modern psychology had its origins in techniques of psy- chological testing first used on a large scale in World War I. Frequently the institutional structures created in wartime to push military applications of science have become permanent in Me subsequent period of peace and have been redirected toward the generation of a new level of government support for fundamental science and for advanced scientific and engineering edu- cation, as well as new siding and credibility for the scientific community in its influence on national policy (Brooks, 19701. The Growing Role of Government The role of government in science and technology has been increasing, in all the industrialized countries, but it has probably changed fastest in the United States, especially during and since World War II. Many of the new technologies that have been at the forefront of U.S. economic growth during the postwar period had Heir origins either in World War II or in the subsequent period of the cold war: commercial transport aircraft; semiconductors, solid- state electronic devices, and integrated circuits; computers; nuclear power; satellite communications; microwave telecommunications and radar appli- cations, such as air traffic control; antibiotics; pesticides; new materials, such as high-strength steel alloys, titanium, high-temperature ceramics, fiber-rein- forced plastics, and composites; and new methods of metal fabrication and processing, such as numerical-controlled machine tools or powder metal- lurgy. Much of this has been derivative from military and space activities, although in many cases, once the basic technology was transferred to He private sector, it tended to take off on its own, with rapid proliferation of

122 HARVEY BROOKS incremental improvements, cost reductions, quality enhancements, and an- cillary technologies necessary for wide commercial acceptance. Much commercially significant innovation has also been an indirect de- rivative of the eno~ous public investment in biomedical research. Although innovation in pharmaceuticals and medical devices has been largely generated in the private sector by private research and investment, it is doubtful whether much of this would have taken place without the base of knowledge resulting from govemment-sponsored programs. Much modem medical instrumenta- tion and diagnostics derive from basic advances in the physical sciences, including laboratory instrumentation, which occurred as a result of broad- based government sponsorship of fundamental physics, chemistry, and bi- ology (Handler, 1970:25~257; Grabowski and Vemon, 1982~. It is important to recognize, however, that several other equally innovative industrial areas owe less to government initiative or science sponsorship: industrial chemicals, synthetic fibers, heavy machinery (including construc- tion equipment), electric power generation (other than nuclear steam supply), and telecommunications are specific examples. Moreover, even where gov- emment has been an important influence, the civilian applications, market penetration, and broad economic benefits would not have been realized with- out the strongly complementary initiatives and technical ingenuity of private entrepreneurs. Govemment-generated science and technology were only the starting point and not Me basic driving force. This is nowhere better illustrated than in the semiconductor industry, where government started as almost the sole customer for the early transistors, whereas today military and space end uses account for only about 10 percent of the market for semiconductor devices (Levin, 1982:19, Table 2.11. Government and Basic Science The government role in stimulating the broad development of science for it own sake, rawer than for well-defined special social purposes, is a relative latecomer to the U.S. science policy scene, especially when compared win many other industrial counties. Although the Founding Fathers showed some concern with the development of a national science policy, and even proposed the creation of a national university (Dupree, 1957:1~15, 40), this interest largely lapsed win the rise of a more pragmatic, populist political orientation following the election of Andrew Jackson in 1828. When the British indus- ~ialist Smithson left a bequest to the U.S. government for the founding of a national institution devoted to He cultivation of science in its own terms, Congress debated for 10 years before deciding to accept the bequest, ques- iioning whether the support of science was an appropriate federal function except for specific practical public purposes (Dupree, 1957:7~791. Through- out the nineteenth century American scientists, considering themselves a

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 123 beleaguered minority, continually bemoaned the countries exclusive concern with applied science and its neglect of pure science. They looked with envy at European governments and their tradition of public patronage of pure science (along with the arts). Well into the twentieth century the support of science was not viewed as a government responsibility, and until World War II, the development of American science depended mainly on private pa- tronage, particularly the great private foundations. A few far-sighted indi- viduals were beginning to point out the dependence of the continued advance of the U.S. economy on a broad-based science, and Herbert Hoover in the 1920s actually proposed a government-industry coalition to provide funds for the support of science in the ultimate interest of industrial innovation (Dupree, 1957:340-343; Layton, 1971:Ch. 81. The Watershed of World War 11 World War II marked a true watershed in the development of Amencan science policy. It was the first war in history in which fundamental scientific or engineering developments originated during wartime came to fruition and were used in battle during the same war; hitherto wars had only stimulated technology which became significant in a subsequent war. Though the seeds of the wartime science policy had been sown in the activities of private philanthropy, in selected government activities such as NACA and in advocacy by a few leaders of the scientific community (Dupree, 1957:358-361), a new relationship between government and science was triggered by the crisis of the war. In contrast with World War I, when scientists were brought into the service of the war effort primarily as military officers under the direction of military commanders, He scientific war effort in World War II was organized as an independent civilian enterprise under the Office of Scientific Research and Development (OSRD), directed by Vannevar Bush, and managed by industrial and academic scientists in equal partnership win the military rather than subordinate to it. Although He work was fully funded by government, it operated outside the Civil Service with scientists remaining within their familiar institutional settings. Military re- search and development was conducted under contract to private institutions on an unprecedented scale, with the government bearing all the costs, in- cluding those of administration and infrastructure ("overhead"), on a reim- bursable basis, generally with no profit, and no financial gain or loss to either individuals or institutions. The research contract win fully reimburs- able overhead was a distinctly U.S. invention, which proved to be an ex- traordinanly flexible instrument in the subsequent partnership between government and private institutions that evolved in basic research, hardware development, and even policy analysis and system management during He postwar period.

124 HARVEY BROOKS It was probably the accident of the cold war and the accelerating military- technological rivalry between the United States and the Soviet Union that prevented the system of research contracting that evolved during the war from being dismantled in the postwar period. The political climate after World War II stood in shard contrast with that following World War I, when much of the wartime science apparatus was dismantled and military contractors were widely viewed by the public as "merchants of death," the root cause of war itself rawer than a source of national security. Instead of government and civilian science turning their backs on each over, the institutional "swords" built to fight the scientific World War II were at least partially forged into the "ploughshares" of a postwar policy for the broad development of science in the interests of society (Bush et al., 1960), even though the military influence on overall scientific priorities remained substantial—e.g., the large emphasis on the physical sciences. The consensus in support of even this much of a civilian science policy was to a considerable extent maintained by the threat of the Soviet Union; the support of even the purest science was justified in terms of its possible ultimate value in the rivalry between the superpowers (England, 1983:212, 219, 2801. Nevertheless, it gradually evolved into a full-fledged civilian science policy, increasingly divorced from its national security parentage. THE POSTWAR ERA AND THE NEW SOCIAL CONTRACT BETWEEN SCIENCE AND SOCIETY ''Science the Endless Frontier" The public debate on the postwar organization of science was opened in November 1944 by a letter from President Franklin D. Roosevelt to Vannevar Bush (actually drafted by Bush) asking him to set up a committee to study how the lessons learned in OSRD could be applied in peacetime "for the improvement of national health, the creation of new enterprises bringing new jobs, and the betterment of the national standard of living." The resulting report, Science the Endless Frontier (Plush et al., 1960), became the fundamental charter for American postwar science policy, and its general philosophy, though not its specific organizational recommen- dations, continues to guide government support of science and technology in the United States to this day (Brooks and Schmitt, 19851. It recom- mended the use of public funds to support basic research in colleges and universities and to "foster the development of scientific talent in our youth. " Research was to be supported largely through contracts and grants with universities and research institutes, as well as private firms, leaving `' infernal control of policy, personnel, and the method and scope of re- search to the institutions themselves. " It also proposed that the governance

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 125 of the federal agencies sponsoring research in private institutions be left in the hands of "persons of broad interest in and understanding of the peculiarities of scientific research and education." Thus science was to be accorded a high degree of self-governance and intellectual autonomy, in return for which its benefits would be widely diffused through society and the economy. This diffusion was to be further fostered by the extensive use of contracts with private industry in preference to civil service labo- ratories for developmental activities. As a consequence, industry has ac- counted for between 70 and 75 percent of total public and private expenditures for R&D and between 50 and 55 percent of all federal R&D expenditures (National Science Foundation, 1984c:3, 111. In one important respect the postwar scientific system did not follow the pattern envisioned by the Bush committee. The committee had sug- gested that a single agency be responsible for all extramural research sponsored by the government, to be known as the National Research Foundation. It was to support not only basic research but also long-range applied research that would contribute to various federal missions, in- cluding the military and public health. Instead of a single R&D agency with mission-oriented functional divisions, there evolved a pluralistic sup- port system, with several cabinet-level federal agencies having partially sheltered divisions responsible for the support of long-range research re- lated to their missions, e.g., the National Institutes of Health (NIH) in the case of public health; the Office of Naval Research, the corresponding offices for the Army and the Air Force, and the Advanced Research Projects Agency (now the Defense Advanced Research Projects Agency) in the case of the military; and the National Science Foundation (NSF), responsible only for basic research and science education not tied to any particular federal mission (Brooks, 1973a). Both NSE and the newly cre- ated Atomic Energy Commission (AEC) were prohibited from setting up their own civil service laboratories, but were encouraged or required to '~contract out" the actual conduct of research to private organizations, sometimes created especially for that purpose under independent boards of private citizens. The "contracting out" idea was also adopted by the Air Force, in some measure by the Army, and least by the Navy; it also became the norm when, in the aftermath of sputnik, the largely civil service NACA was converted into the National Aeronautics and Space Admin- istration (NASA) by the Space Act of 1958. Within a few years the trans- for~ed agency was converted from one which was 98 percent "in-house" in its conduct of research, to one which contracted more than 80 percent of its R&D, especially the development part, to the private sector (Bok, 1966) The other major heritage from OSRD was the principle of awarding re- search and development contracts to the most qualified organization, irre-

126 HARVEY BROOKS spective of geographical or other nonscientific considerations. This was a sharp break with the tradition that had been established prior to World War IT, especially in agricultural research, where the policy had been to distribute federal research facilities and support very widely in each state. This proved to be the most controversial of the Bush committee's recommendations, and one on which the creation of the new science agency, NSF, nearly foundered (England, 1983:5-61. Today, nearly 40 years after the publication of the Bush report, the "social contract" between science and society that it advocated remains remarkably intact, despite numerous alarums and excursions whose rhet- oric has generally outrun their practical effect (Brooks and Schmitt, 19851. In the context of the American political system, this is a rather remarkable political phenomenon, and many would assert it has been responsible for American world leadership in pure science and in most fields of advanced technology (Bush, 1970:651. It may also be said, however, that the Bush social contract is probably under more fundamental challenge today than at any time in its postwar history, largely as a result of the erosion of U.S. international competitiveness in the increasingly interdependent world economy (Brooks and Schmitt, 19851. Trends in R&D Expen':litures The course of both federally sponsored and privately supported R&D since the end of World War II can be divided into three distinct penods. The first extended from the beginning of the cold war in the late 1940s to about 1967. This period was characterized by more or less steady growth in R&D ex- penditures, averaging up to 15 percent per year in real terms, with the life sciences considerably exceeding that rate after 1957 following the "take- off'' of the budget of He National Institutes of Health at that time. The bunk of federal R&D expenditures was devoted to space, defense, and militaTy- onented nuclear programs, which reached more than 90 percent of all federal R&D in the early 1960s (Brooks, 19631. The second period started in about 1967, when an abrupt leveling off in the volume of government-sponsored R&D began. This was associated with a severe budgeter, crunch resulting from the attempt of the Johnson admin- istration to maintain a "guns and butter" budget during He Vietnam buildup. However, the period of stagnation was prolonged until about 1977. MilitaIy R&D and space expenditures declined and He basic physical sciences also experienced a fall-off in support to about 14 percent below their 1967 peak when measured in constant dollars (using the GNP deflator). The life sci- ences, riding on the political popularity of biomedical research and backed by an effective political coalition in Congress, maintained continuing, though reduced, grown, considerably assisted by He "War on Cancer" Hat was

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 40 30 J o 1— ~ 20 LL o _e . . - O / ~ / m 10 o 1960 1 965 1 970 1 975 127 . . . . . . . Total .- - . __ . _ .— . \ Federal Private - . - . . . . 1980 1983 1 984 FIGURE 1 Feeder, pnvate, and total R&D expenditures, 196~1984 (billions of constant 1972 dollars). SOURCE: NahonaI Science Foundation, National Patterns of Science and Technology Resources (Washington, D.C.: 1984), Table 5. initiated win bipartisan support in the Nixon administration (U.S. Con- gress,l971; Strickland, 1972; Berger, 1980:62-631. Ouring this period pn- vately financed industrial research continued some growth, but with decreased emphasis on fundamental and longer-range research. The general trend is illustrated in Figure 1 (President's Con~niission, 1985:981. It was dig this second period Mat defense/space R&D dropped nearly to 60 percent of gov- ernment-sponsored R&D, partly owing to spectacular expansion of energy research, development, and demonstration programs, but also partly due to

128 HARVEY BROOKS TABLE 1 Trends in Federal Funding of Research and Development (billions of constant FY 1972 dollars) - FY 1986 FY 1967 FY 1972 FY 1984 (est.) l Defense $12.4 $9.2 $12.1 $16.1 Space 6.6 2.7 0.8 1.1 Heal 1.8 2.0 2.2 2.2 Energy 0.8 0.6 1.2 0.9 General science 0.7 0.7 0.8 0.9 Other 1.7 1.9 1.7 1.6 SOURCE: American Association for Me Advancement of Science, ALAS Report X: Research and Development, FY 1986. Intersociety Wowing Group (Washington, D.C.: 1985). research in support of the Great Society programs and of environmental protection (see Table 2 below for details). The third period began in 1977 with an acceleration in the growth of self- financed industnal research, a gradual restoration of government-supported research in the physical sciences, and a rapid acceleration of defense R&D, which increased even more rapidly in Me l980s after the advent of the Reagan administration (American Association for the Advancement of Science, 1985:27~. After 1980 civilian R&D shrank, particularly in the Department of Energy (DOE) and in the social sciences, and the proportion of space/ defense programs in government-sponsored R&D climbed to 72 percent. By FY 1985 defense R&D had exceeded its FY 1967 peak in real terms and was scheduled to exceed its FY 1967 peak by 30 percent in the FY 1986 budget. However, space and defense together were still 22 percent below their FY 1967 level in FY 1984 and would be 9 percent below their FY 1967 level in FY 1986, according to the President's budget. Total federal R&D was 22 percent lower in FY 1984 and would still be S percent lower in FY 1986 compared win its FY 1967 level. These results are summarized in Table 1. The changing Rends in federal funding of R&D reflect changes in overall national pnonties, which affected science policy. Those changes are analyzed in greater detail below. THREE EPOCHS IN POSTWAR SCIENCE POLICY The postwar period can be divided into three distinct epochs: the cold war period 1945-1965; He period dominated by social priorities 1965-1978; and the period dominated by industrial competitiveness 1978 to the present. In reality these periods overlap, and He onset of each new epoch was fore- shadowed by strenuous policy debates in Washington. The Tree epochs also

NATIONAL SCIENCE POLICY AND TECHNOLOGIC~ INNOVATION 129 coincide fairly closely with the three periods that mark the changes in funding patterns for federal R&D described above. The Cold War Period: 1945-1965 Until 1965, when the first indications of public revulsion against the Viet- nam War began, and when the environmental movement began to be polit- ically effective, the support of even the purest science and of graduate education had been justified to the Congress largely in teas of the military/techno- logical race with the Soviets (England, 1983:154, 218-2211. This science race was enormously stimulated by Soviet space achievements, beginning with the launching of the first sputnik in 1957. A simultaneous buildup of military and space investments followed, generating unprecedented new de- mands for highly trained scientific and engineering manpower. This in turn helped to fuel an expansion of higher education, especially graduate edu- cation, which also coincided with the "baby-boom" generation's coming of age. Accommodating the rising public demand for advanced education grad- ually became a goal in its own right, partially replacing the anticipated manpower demands of federally sponsored programs as a justification for federal support of science and higher education (Brooks, 19651. Military R&D and procurement undertaken in the l950s and early 1960s laid the groundwork for American domination of world markets in com- mercial jet aircraft, semiconductors, and (temporarily) nuclear power (Nel- son, 1982~. At the same time the so-called GI Bill of Rights introduced at the end of World War II laid the foundation for the U.S. postwar lead in the training of technical manpower, and this helped to staff the explosive grown of government technological programs in the 1960s. In the early 1960s, however, Were developed an intense debate among economists and students of science policy as to the net effect of these large government technological programs on the performance of the U.S. civilian economy, a debate that is being revisited today in only slightly revised form. A majority asserted that the civilian "spin-off'' from government programs would stimulate technical progress within the civilian economy, but a significant and increasingly vocal minority argued that the insatiable demands of federal programs would drain scarce talent away from the civilian sector by bidding up salaries and by providing more challenging and interesting technical opportunities for sci- entists and engineers, free of the normal disciplines and economic consents of He commercial marketplace (Hollomon and Harger, 1971; Brooks, 19721. The Social Priorities Period: 1965-1978 The growing technical successes of the space program and of some of He military systems programs such as Polaris created heightened public and

130 HARVEY BROOKS political expectations about what technology, properly mobilized, could ac- complish. If we could organize science and technology to put men on the moon, people said, why could we not organize them to solve problems on earn? If we could accomplish such wonders by pumping money into applied physical science, why could we not do the same by pouring funds into applied biological science or applied social science (Nelson, 1977)? In 1962 President Kennedy expressed a view that later formed one of the underlying assump- tions of the Great Society programs. He suggested Hat "most of the problems, or at least many of them, that we now face are technical problems, are administrative problems" (Schlesinger, 1965:6441. The old debates of He New Deal were seen as increasingly irrelevant to the complex technical decisions of modern society. This was the political impetus that first gave rise to the trend toward `'civilianization" of federal R&D away from its nearly exclusively military/space emphasis of the early 1960s. However, He epoch that began with euphoria about the capacity of science to solve social problems soon gave way to disillusionment and ended In what looked almost like a revolt against science, or at least against big science, technology overreaching itself, and excessive claims of rationality. This was in no way better symbolized Han by cancellation of funds for He cons~uc~aon of Me prototype commercial supersonic transport aircraft by He U.S. Senate late in 1971 (Mowery and Rosenberg, 1982:111 1451. The period also saw the meteoric rise of He Great Society, subseqllen~dy dissolved in He urban riots of the late 1960s, He civil rights movement, He student revolt, and He fiasco of He Vietnam War. The Great Society idea was predicated in many minds on what was seen as die new capacity of He social sciences to serve as He basis for "engineenng" social change. Very soon, however, science and rationality gradually began to be viewed as the source of He problem rawer Man He basis for its solution, and social problems came increasingly to be talked about as He secondary effects of progress in science and technology (Bauer et al., 19691. Then came die "limits to Down" debate, He environmental movement, He energy crisis, Shannon, and He commodity price explosion, quenching He nanon's optimism and its sense of control over its destiny and its environment, which had come to its cuhninabon in about 1963. Yet in other ways the fain in science persisted beneath the surface. The proliferation of technically oriented regulatory agencies, such as the Envi- ronmental Protection Agency (EPA), the conversion of the Atomic Energy Commission fist to the Energy Research and Development Administration and then to the Department of Energy (with a greatly expanded role in energy- related R&D beyond nuclear power), the War on Cancer of the early 1970s all these developments reflected an underlying fain in He capacity of science to cure social ills, including those created by science-based technology itself. In fact, much of He environmental legislation of this period embodied social expectations far exceeding He existing capacity of science to meet them, but

NATIONAL SCIENCE POLICY ANl) TECHNOLOGICAL INNOVATION TABLE 2 Percentage Distribution of Federal R&D Expenditures, FY 1969- FY 1984 - 131 1969 1971 1976 1981 1984 Defense 53 4% 52.0% 50.2~c 55.3% 70.2% Health 7.2 8.3 11.3 11.6 9.6 Energy 2.1 3.6 7.9 10.5 4.9 Space 23.9 19.6 15.1 8.1 4.2 General sciences 3.3 3.3 4.1 4.0 3.8 Transport 2.9 4.7 3.0 2.6 2.4 Natural resources and environment 3.3 2.7 3.3 3.2 1.7 Agnculn~re 1.4 1.7 1.8 2.0 1.6 Education and social services 1.6 1.4 1.2 0.9 0.5 Other 1.5 2.6 2.0 1.8 1.2 SOURCE: National Science Board, Science Indicators 1982 (Washington, D.C.: National Science Foundation, 1983), Table A2-12. the feeling persisted that if society could hold the feet of technical people to the fire strongly enough it could force them to fulfill its high expectations (Brooks; 1982a). Thus the metaphor of landing men on Me moon drove the regulation of technology as much as its promotion. Table 2 shows We percentage distribution of federal R&D expenditures among major budgetary functions during the period from FY 1969 to FY 1984. It indicates the shift from mil~ta~y/space orientation to social orientation and back again dunug this period (National Science Board, 1983:244, Table A2- 121. The socially oriented period also saw the beginnings of concern about U.S. competitiveness in world markets. Following the demise of the U.S. supersonic transport project at the hands of the Senate in 1971, the Nixon administration decided to take an initiative to use govemment-generated technology to revive We competitiveness of the U.S. economy, whose lagging performance it attributed to a decline in technological innovation. Admin- istration spokesmen testified to Congress on the reverse "technology gap" said to be opening up between the United States and Europe, particularly West Germany (Brooks, 19721. At that time Europe was in the midst of a sharp expansion of its support for research and graduate education, while the United States was in the midst of a ''research recession," alleged by some in the scientific community to be the source of its competitive lag (National Science Board, 1983:~. The grandiose federal initiative mostly petered out, leaving only small programs in the National Science Foundation and the National Bureau of Standards (NBS). The NBS Experimental Tech- nology Incentives Prog~n (ETIP) was a rather innovative effort to use government procurement and regulatory programs in a precisely targeted way to induce innovation in the private economy; in its own teens it was rather

132 HARVEY BROOKS successful, but on too small a scale to be of any economic significance (Lewis, 1975; 1976; National Research Council, 19761. The ETIP was in- tended as a pilot program, but was never followed up. Indeed, efforts to use government indirectly to induce innovation for civilian technology had orig- inally been started in the early part of the Kennedy administration, but had run into political roadblocks In Me Congress as a result of opposition from some of the potentially affected industries (NeLkin, 1971; Katz, 19821. The Period of Emphasis on Innovation Policy During most of the 1970s, concern about the declining competitiveness of We U S. economy was mounting gradually, but it had to compete for political attention with energy policy and the public anxieties created by the 1973 and 1979 energy crises. The national civilian R&D investment went largely into energy technology, with We Department of Energy being the most rapidly growing of the federal science agencies. In 1978 the Carter administration took a major initiative to study the impact of federal policies on U.S. competitiveness and to make recommendations for changes in federal policy that would improve private incentives for tech- nolog~cal innovation and industrial investment in R&D. The study, caIried out jointly by the Office of Science and Technology Policy (OSTP) and the Office of the Assistant Secretary of Commerce for Science and Technology (Jordan J. Baruch), involved wide consultation with industry and with science policy experts in We private sector. In contrast with the abortive initiative of We Nixon administration, it Reemphasized direct federal support of in- dustrially oriented R&D and looked to indirect measures, such as changes in patent and antitrust policy, regulatory procedures, and government tech- nical assistance to small business. Many business leaders expressed disap- poinunent in the final recommendations because Hey tended to steer clear of tampering with tax policies as Hey affected new investment and R&D by cocoons or by would-be high-tech entrepreneurs. Nevertheless, this study, which came to be labeled the White House study on innovation (U.S. De- parunent of Commerce, 1979) sewed to raise the subject to near tile top of tile political agenda, and it has since become perhaps the most important topic of national science policy for the 1980s. The recommendations did eventually result in modifications in patent policy (Public Laws, 1980, 1984), a clarification of He Justice Deparenent's interpretation of antitrust legislation to facilitate R&D cooperation among firms (U.S. Department of Justice, 1980; Baxter, 1983), and He setting up of a Regulatory Review group in the Office of Management and Budget (OMB) to evaluate the economic impact of all proposed new federal regulations (Executive Order, 19781. The incoming Reagan administration picked up on the Carter initiatives and made innovation and new entrepreneurship one of the centerpieces of

NATIONAL SCIENCE POLICY AND TECHNOLOGlC~ INNOVATION 133 its economic strategy and science policy, especially after 1982. American competitiveness in the world economy has become the highest-pnority item of public discussion, and almost every government policy is being assessed for its impact on the rate and quality of industrial innovation and competitive performance. Except for increased support of basic research, however, the Reagan administration has systematically eschewed direct federal support of R&D whose main purpose is to lead directly to new products that will be sold in private markets. It has repeatedly asserted that the government's role should "be focused on relatively less costly, high risk, longer tenn high payoff activities that the private sector traditionally has been less willing to undertake" (Office of Management and Budget, 19811. The most dramatic impact of this policy came in the Department of Energy, where the admin- istration rejected "costly near-term activities, such as construction and op- eration of pilot plants and the operation of demonstration plants using company- specific processes" (Office of Management and Budget, 19811. Research for which a specific commercial product was clearly in view as an outcome should be left to the private sector, whose assessment of the potential market for the product was likely to be more accurate than any judgment by a government official or even an industry committee with no financial stake in the outcome. The new policy, however, left considerable room for debate on We ap- propriate dividing line between government and private responsibility. For example, the administration continued to support funding for the Clinch River breeder reactor, which was clearly a demonstration program of the sort it deplored and which the Budget director, David Stockman, had strongly opposed as a congressman, using the same rationale he used later to kill other DOE demonstration programs (Stockman, 19771. On the over hand, it initially proposed to cut back on NASA's financing of aeronautical research, a decision that was later reversed as the result of an evaluation by an OSTP panel (Office of Science and Technology Policy, 19821. It also proposed to "pnvatize" meteorological and remote sensing satellites, leaving their furler development and operation to Me private sector (Gregory, W., 1982; Wald- rop, 1982) This most recent epoch has been characterized by strong industrial R&D spending, which continued to grow even through a severe recession, and a shift from previous emphasis on relatively short-term, product-improvement research to longer-term projects (National Science Foundation, 1983a). The pressure of looming competition from the Japanese has forced continuing emphasis on innovation over near-tenn cost savings from cutting back on R&D expenditures. Much of industry took to heart the lesson learned from the Japanese semiconductor industry, which, unlike its U.S. counterpart, did not cut back on R&D or investment in new plant in 1974 and 1975 and was Hereford in a better position to meet resurgent market demand in the 197

134 HARVEY BROOKS 1979 economic recovery, when its American competitors were unable to meet even domestic demand (Imai and Sakuma, 19831. The turn around in industrial research has been particularly evident in the automobile industry, which, despite a worsening financial condition in the late 1970s and early l980s, continued to increase its R&D investments (Eck- stein et al., 1984:1571. Indeed, one could argue that the U.S. automobile industry has made at least a partial transformation from a typical "mature" industry to a quasi-high-tech industry, with competitive performance much more dependent on technological innovation than in recent history. As a result of We industry's R&D investments in product improvement, prospects for Me introduction of new performance features have greatly expanded (Altshuler et al., 1984:Ch. 4~. As just one example, as fuel-eff~ciency per- fonnance has improved, the possibility of much greater improvements not hitherto considered likely has become apparent (Altshuler et al., 1984: 91-95). More broadly, the emphasis on productivity as an element of competi- tiveness throughout the mass production industries has led to dramatic in- creases in the use of engineers in manufacturing, and sciendf~c and technological employment grown has outpaced labor force grown by a factor of about three (National Science Foundation, 1983b). The new emphasis on industrial innovation in Me present period has been complicated by Me resurgence of defense spending, and particularly defense R&D and procurement. This has helped fuel a dramatic resurgence of demand for technical manpower, particularly that with graduate training, after a long period of slack demand for technical people in Me period of social pnorii~es described above. As a result, undergraduate enro~nents in engineenug schools have doubled since their low point in the mid-1970s, and there has been a strong shift from science to engineenug at bow Me undergraduate and grad- uate levels, as well as a more modest shift from nontechnical fields into science and engineering (National Science Foundation, 1984b:23~. Equally striking is the high percentage of undergraduates in all disciplines who are now declaring an intention to enter business careers. Engineering education has become a priority item of public policy discussion, and more than two- ~irds of all the states have developed programs to stimulate technical edu- caiion, universi~-indus~y cooperation, or Me establishment of high tech- nology industry within the state (Pear, 19831. The rise in technically oriented defense spending has also revived the 1960s' debate on the economic impact of defense on the civilian economy. Resurgence of defense spending in Me 1980s has occurred in an economy with much lower capacity utilization Man existed in Me 1960s, so Mat one might expect less competition for "bottleneck" resources and talents Man existed then (Aspin, 1984~. As we saw earlier, R&D spending for defense and space is still a considerably smaller fraction of total R&D spending

NATIONAL SCIENCE POLICY AND TECHNOLOGIC~ INNOVATION 135 (probably 20-25 percent), especially when industrial spending is also in- cluded, than was the case in the 1960s. On the other hand, today's sophis- ticated weapons systems may have less spin-off benefit for He civilian economy than those of the earlier period. In many component fields, such as semi- conductors and computers, as well as many aspects of avionics and aero- nautical design, Be civilian sector actually leads the military sector in innovative technology, so that there may be much less potential for useful technology transfer from the military to the civilian sector than existed in the late 1950s and early 1960s. In addition, many major items of military hardware, such as ballistic missiles, supersonic aircraft, surface-to-a~r defense missiles, are much less related to possible civilian applications than jet aircraft, microwave radars, tanks, or fire-control computers. Military systems increasingly have to sustain environments that have no relation to what is necessary for civilian equipment. The specific systems aspects of military hardware are relatively more important and, by the same token, less relevant to civilian applications. In a recent study of the impact of R&D and basic research on productivity at the firm level, Griliches (1985) found the most comprehensive evidence yet of the strong correlation between R&D investments and Be economic health and productivity of firms, including a relatively higher impact for basic research than for R&D overall, but he has also shown Rat gove~nment- funded R&D, though making some contribution, contributes much less to productivity growth in a firm than privately funded R&D a result which can probably be interpreted as indicating the small civilian impact of defense/ space-type R&D, among other factors. On the other hand, a recent study by the Congressional Budget Office provides some indication that government R&D, as well as self-f~nanced R&D spending in-an industry, is positively linked to competitive perfo~ance in international trade (Congressional Bud- get Office, 1984~. These two results are not necessarily in conflict. COMPARATIVE INDICATORS OF U.S. PERFORMANCE IN SCIENCE AND TECHNOLOGY A wealth of literature on science indicators attempts to provide quantitative measures for the comparative performance of the United States and over countries in science and technology. These measures are based bow on inputs, such as R&D spending, manpower, and scientific equipment, and on outputs, such as publications, citations, awards, patents, royalty payments, produc- tivity growth, and shares of world markets for high technology products (National Science Board, 1973; Elkana et al., 19781. None of these measures is entirely satisfactory, not only because of the lack of quality factors, but also because of conceptual problems as to what society really ought to expect from its scientific and technological establishment. For example, the more the output measures deal with factors that relate to He interaction between

136 HARVEY BROOKS science and society rather than factors entirely internal to the technical es- tablishment, the more uncertain their significance becomes (Brooks, 1982a: 2-51. Even if we focus on purely economic measures without getting into debates about the quality of life or the distribution of the national product among sectors of society, we have a problem in that technology and inno- vation are only two among many factors that determine economic growth, and the art of differentiating the technical factors from others is an imperfect one. Win this caution in mind, let us look at some of the conventional indices (Brooks, 1985b:33~352~. inputs R&D Expenditures A very common input measure is aggregate expen- ditures on R&D, both private and public, or such expenditures as a fraction of GNP. Even at this level we immediately encounter a conceptual difficulty, for it is quite uncertain whether it is aggregate national R&D expenditures or national R&D expenditures as a proportion of GNP that are of greater economic significance. If Were were no barriers to He transfer of information between the R&D performer and the individual or organization that can make economic use of it, Den one might say that aggregate R&D should be the more significant indicator. By Ads measure the United States completely dominates any other plausible grouping of countries, with the exclusion of We Soviet Union, whose statistics are extremely difficult to interpret because of its wholly different economic system. One such comparison (Brooks, 1985a), shown below, compares the United States; West Ge~any, France, and the United Kingdom as a group; and Japan with respect to total R&D expenditures in 1969 and 1979: 1969 United States $25.6 billion West Germany, France, United Kingdom 8.3 Japan - 3.0 1979 $55.0 billion 39.0 19.3 Although We R&D investment of other counties has grown relative to Mat of the United States, this country still dominates, and in fact has probably increased its lead in the past five years. If we restrict our attention to self- financed industrial R&D expenditures, the situation does not change, as shown below (Brooks, 1985a): 1979 United States $25.3 billion West Gennany, France, United Kingdom 19.3 Japan 11.4 If we assume Eat most privately financed R&D is for commercial purposes

lIATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 137 and that there Is no spin-off at all from military R&D, a very conservative assumption, He United States still dominates the civilian R&D picture. However, using the absolute level of R&D expenditures can be questioned on several grounds. First, the diffusion of the results of R&D is not effortless, so that the institution performing the R&D has an advantage in applying it. This is probably especially true of industrial R&D performed in-house, in close proximity to marketing, production, and general business planning. Second, the diffusion of R&D does not stop at a nation's borders, so that if we assume excellent diffusion within a country we must also assume that the results are nearly equally available to potential industrial innovators out- side the country. Thus, the national aggregate of R&D expenditures is m~s- leading whether we assume perfect or imperfect diffusion of R&D results. If we make the opposite assumption that privately generated research results are entirely proprietary within the organization that generates them, then the average ratio of R&D to sales among all firms becomes the better indicator of innovative effort; each firm is assumed to have to generate most of the knowledge it must use to innovate. With this equally unrealistic as- sumption, the ratio of industrial R&D to GNP becomes a reasonable proxy for innovative effort. By this measure the United States would be behind both Japan and West Germany, as shown below by He ratios of civilian expenditures to GNP for West Germany, Japan, and the United States in 1981 (National Science Board, 1983:197-198, Appendix Tabte 1-61: United States West Germany Japan 1981 1.7% 2.5 2.3 The A&D-GNP ratios above would suggest that Japan and West Germany might be deriving more economic benefit from their R&D expenditures than the United States, but the assumption Hat aggregate R&D is more significant than the ratio of R&D to GNP seems the better of He two approximations. Consequently, not much should be inferred from the A&D-GNP ratio. Despite its implausibility as an indicator, however, Hat ratio is frequently quoted in He literature in support of a presumed lag in U.S. innovative effort. Manpower The United States still leads all major industrial countnes, except He Soviet Union, in He number of R&D scientists and engineers per 10,000 civilian workers, but the margin of superiority over Japan and West Germany narrowed greatly in He 1970s (National Science Board, 1983:8, Fig. 1-51. The number of first-degree scientists and engineers produced an- nually in the United States was still nearly double that in Japan, but 7 percent of He U.S. graduates were engineers compared with 19 percent in Japan, so Hat Japan was graduating more first-degree engineers than this county

138 HARVEY BROOKS (National Science Board, 1983:6, Fig. 1-31. Considerin, the large number of U.S. engineers in defense activities compared with the number in Japan and Europe, the slight U.S. advantage in the number of scientists and en- gineers per 10,000 workers is not much cause for complacency. Of equal importance to technical graduates may be the general level of "technical literacy" of the labor force, and here there is some reason to believe the United States lags its competitors, particularly West Germany and Japan. These countries have stressed scientific and mathematical pro- ficiency in their secondary educational institutions much more so than the United States, and the achievement of U.S. high school students in mathe- matics and science proficiency tests is inferior to that of these major com- petitors (National Science Board, 1983:5; Husen, 1983~. In a world in which changing markets and technology require increasing adaptability and the reaming of new skills on the job, inferior basic skills necessary for such reaming may become a serious competitive disadvantage, which could be compounded when the organization of the workplace discourages the ac- quisition of broader ranges of skill and the capacity for higher responsibility for Nose without a college education (Brooks, 1983~. There is some indi- cation that outmoded management practices and rigid work rules derived from a tradition of adversarial labor relations place many American workers at a disadvantage in adapting to changing conditions, especially compared with their Japanese counterparts (Skinner, 19831. The Scientific Infrastructure The sta Nation of federal R&D support in the 1967-1977 decade may have had its severest impact in the universities in terms of the declining investment in new instrumentation and renovation of physical facilities. In the last five years several studies of the state of instrumentation in university laboratories have documented the fact that uni- versity laboratories have fallen seriously behind government and industrial laboratories In the access they provide to state-of-the-art equipment (National Science Foundation, 1984a; Smith and Karlesky, 19771. Anecdotal evidence indicates that during the 1970s the equipment available to European re- searchers in leading universities and research institutes became superior to that accessible to their U.S. competitors. The problem first became apparent in relation to advanced research equipment, but more recently it has become equally applicable to teaching laboratories, particularly in engineering and computer science. This has been due bow to a lag in investment and to rapid technological change in laboratory instrumentation brought about by the computer revolution. A part of Me ins~umen~tion problem can be ascribed to We impossibility of spreading the most advanced instrumentation in every field among many institutions. Yet there is a tendency for each university department to aspire to comprehensive excellence across a very broad range of fields; the accep-

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 139 tance of specialization and division of labor among institutions and labora- tones tends to lag the growth in complexity, cost, and sophistication of modem instrumentation (Brooks, 19711. This is not a new phenomenon. It began early in the postwar period in such fields as high-energy physics and radioastronomy, and it was partially solved for those fields by the creation of national centers, such as the Brookhaven National Laboratory or the National Radioastronomy Observatory (NRAO), in combination with the funding of "user groups" from universities to take advantage of these national facilities (Brooks, 1978~. The problem today is that the necessity of con- centrating sophisticated resources in a few national centers is extending to more and more fields of research, and now even to several fields of advanced teaching, such as microelectronics and electrooptics. The question is how some measure of cooperative planning can be introduced into this process without eroding the healthy pluralism and competition which have been sources of strength in American science. Indeed, the sharing of expensive equipment and facilities among researchers is already more widespread than is generally realized (National Science Foundation, 1984a:29-331. Unfortunately, one of the consequences of the drying up of federal funding for research infrastructure in universities has been the politicization of the allocation of resources and the increasing use of backdoor channels through the Congress to secure funding for major new facilities that has not been forthcoming through the normal budgetary processes of executive agencies. The result is the distribution of new facilities and equipment by competitive political influence and lobbying instead of by cooperative planning and peer review (Nonnan, 19831. The country may have reached the point, however, that some new initiatives in research infrastructure, however arbitrary and politicized, may be better than none. Outputs Publication Statistics For research, the main quantitative measure of output is publication, and publication colmts have come to be widely accepted as Output indicators of scientific activity, generally bred on a subset of particularly influential and frequently cited journals. Scientists and engineers in the United States (defined as scientists and engineers working in U.S. institutions, not necessarily American citizens or permanent residents) con- sistently accounted for about 37 percent of the world's science and engi- peering literature in the 1970s, considerably higher Man the U.S. share of world GNP, and roughly equal to the U.S. share of world R&D expenditures (National Science Board, 1983:11, Table 1-2~. There are considerable dif- ference among fields, however, ranging from 43 percent in clinical medicine and 42 percent in earn and space sciences to only 21 percent in chemistry

140 HARVEY BROOKS and 30 percent in physics. Only in mathematics did the U.S. share decline significantly between 1973 and 1980, from 48 to 40 percent. In mathematics the absolute U.S. publication count declined by 36 percent in this period compared with only a 23 percent decline for the non-U.S. publication count. For engineering and technology there was a decline of 29 percent in the United States compared with 22 percent outside the United States. Generally speaking, declines In publication counts seem to have followed declines in real research expenditures in various fields (National Science Board, 1983:111. Citations are often used as a rough indicator of the quality of publications. For this purpose a convenient measure is the citation ratio, that is, the ratio of the total proportion of citations in the world's literature to the publications from a given country to the share of all publications produced by that country. A citation ratio of 1.0 for a given country means that there is no preferential citation of the publications emanating, from that country and, hence, that their "quality" is about the world average. The citation ratio for all fields in the United States has remained stable or slightly rising over time: 1.45 in 1978, with a range from 1.88 in chemistry to only 1.09 in general biology. Since the United States accounts for such a large fraction of all publications, there may be a self-citation bias in these measures; the ratio for non-U.S. citations to U.S. papers is in the neighborhood of 1.0 or slightly under 1.0 for most fields, ranging from 0.57 in biology to 1.25 in chemistry. The lower numbers may also reflect a self-citation bias in other countries, so that the Rue citation ratio would lie somewhere between, Gus still indicating some qualitative superiority for U.S. publications (National Science Board, 1983:121. Recent studies of the comparative performance of research institutions made at the Science Policy Research Unit of the University of Sussex, England, also suggest that U.S. institutions obtain more significant results per dollar of expenditure than do comparable European institutions. Most of these studies, however, have been made for fields of "big science," such as particle physics, fields in which much of the U.S. capability has been in place for longer than the European capability and may therefore have had more opportunity to achieve maximum productivity, (Economist, 19841. Other indicators of We comparative scientific performance of the United States come from the award of Nobel and other prizes in science, where the United States enjoys an increasingly predominant position, and from the increasing number of foreign students who come to the United States for advanced training in science and engineering, in the majority of cases wi their own sources of funds. However, each of these indicators may also be criticized. Since Nobel prizes are most frequently awarded for work done some time in the past, they may be unreliable indicators of current perfor- mance. The big influx of foreign graduate and postdoctoral students has been largely from developing countries; in this case the difference in scientific level between the county of origin and the host country is so great that one

NATIONALSCIENCEPOUCY^DTECHNOLOGIC~lNNOVATlON 141 suspects that the choice of the United States as host reflects factors other than a judgment on the quality of science and engineering in the United States compared with other developed countnes. Overall, the studies of publication and citation counts can be interpreted as indicating a modest quality advantage for the United States, confirming other, more impressionistic evidence that U.S. science achieves more sig- nificant results per unit of expenditure than its main competitors. However, the advantage does not appear to be overwhelming, and its indicators tend to lag considerably in time. Patents and International Trade in Intellectual Property While there appears to be little doubt about the continuing excellence of the U.S. per- fo~Tnance in basic science, despite some closing of the gap, especially by Western Europe, the U.S. performance in applied science and in the com- mercialization of new knowledge is much more in question. It is frequently pointed out that on a per capita basis Britain has led the world in pure science for several generations, and yet it has experienced a steadily deteriorating economic performance, apparently owing to a poor capacity for reducing knowledge to commercial practice. Because of this general acknowledgment that superior performance in basic science does not guarantee superior eco- nomic performance' attention has recently turned to the study of patenting statistics and to international trade in intellectual property. The number of patents is believed to be a better indicator of the state of technology in a country. Moreover, the fact that novelty and originality have to be more formally documented for patents than for scientific publications gives certain advantages to the use of patents as indicators. Pavitt and Soete (1979) have shown that it is possible to compare the technological performance of venous non-U.S. countries by studying the share of patents granted to citizens of each country by the U.S. Patent Office. They have also found a remarkably good correlation over time between the relative number of U.S. patents and national R&D expenditures, and, since patent statistics go back much further in time than properly standardized R&D statistics, Pavitt and his colleagues have used patent counts as surrogates for comparative national R&D activity going back much into the nineteenth century. But, by the same token, patent data do not appear to give information which is independent from Nat pro- vided by comparative R&D statistics. A much-quoted statistic indicating a U.S. technological lag is the 38 percent drop between 1970 and 1982 in the number of patents granted to U.S. inventors by the U.S. Patent Office, while in the same interval the number of patents granted to foreign inventors nearly doubled, reaching 26 percent of all patents by 1982 (National Science Board, 1983:13~. In several product areas foreign patents accounted for 50 percent of all patents, while in other areas, such as petroleum refining and natural gas extraction, foreign

142 HARVEY BROOKS patents were only 20 percent of patents (National Science Board, 1983:14, Table 1~. Overall, over 50 percent of foreign patenting in the United States was accounted for by West Germany and Japan, with Japan showing the most dramatic increase in the 1970s. The U.S. share of foreign patents in other countries declined from 65 percent in 1971 to 59 percent in 1981; the number of patents granted to U.S. inventors in other countries dropped by 44 percent in the 1970s. In the last three or four years, however, there appears to have been a resurgence of patenting by U.S. inventors (National Science Board, 19851. The meaning of such statistics can easily be exag- gerated, however, for cross-national patenting may depend as much on judg- ment of potential markets as on innovation per se. In other words, it is not clear to what extent foreign patenting depends on "market pull" as compared with "technology push" (i.e., level of inventive activity). The work of the University of Sussex group, for example, has shown in a study of 40 product groups that for 23 of them the world export share of Britain correlates very well over time with the rate of patenting by British inventors in the United States (Pavitt and Soete, 19791. On the other hand, some studies by U.S. authors suggest that market pull has little influence on the propensity to patent in foreign countries (National Science Board, 1983:141. International payments for Me use of patents, trademarks, copyrights, and proprietary know-how are also frequently used as an indicator of relative innovative capacity. Since the revenue streams considerably lag the time of innovation, this may be a poor indicator when the distribution of innovative activity is changing rapidly. Nevertheless, by this gross measure the United States is doing well. At the end of the 1970s it was earning nine times as much as it was paying out in royalties and fees, and this ratio had hardly declined since 1967 despite a large growth in the absolute balance. It is important to keep in mind, however, that over 80 percent of such receipts are from foreign affiliates of U.S. companies and thus directly related to foreign direct investment (National Science Board, 1983:241. In a way it is rather surprising that, with the growth of R&D expenditures and patenting rates in other countries relative to the United States, there has not been any decline in the ratio of U.S. receipts to payments. If we had been making as successful use of foreign inventions as our competitors had of ours, one would have expected a decline in the payment ratio as, with increasing R&D levels compared with the United States, foreign nations became a larger potential source of commercializable technology. This may be evidence, pointed out by several observers, that the United States may be lagging its competitors in its ability to scan and adopt foreign technology that could contribute to irnprovin=, its economic performance (demon, 19821. Have we been so accustomed to being the leader in all fields that we simply have not learned how to make optimal use of We technology available worldwide and are thus spending too much of our innovative effort on "reinventing the wheel"?

NATIONAL SCIENCE POLICY kD TECHNOLOG1C~ INNOVATION 143 During the same period We ratio of receipts to payments for Japan was showing a rapid approach to balance in the 1970s, going from 0.2 in 1971 to 0.7 in 1981 (Keizai Koho Center, 1983:18, Fig. 4-7~. If one allows for the average age of licenses on which royalties are paid, it seems almost certain that Japan has a positive payments balance on Me recent licenses, which suggest that it is already a net exporter of technology and once again confirms that it is becoming a world center for technological innovation (Gregory, G., 19821. Producavzty So much has been said about the lag in U.S. productivity in the 1970s that one is sometimes surprised to note that the GNP per employed civilian worker in the United States, when properly adjusted for relative purchasing power (rather than currency exchange rates), is still the highest in the world (National Science Board, 1983:17, Fig. 1-101. The problem is Mat from 1960 to 1980 the average annual growth in output per man hour worked in manufacturing has been less in the United States than in any other industrialized country. Moreover, the absolute level of produc- tivity in Japan has overtaken that in the United States in a number of key industries by ~ percent in steel, by 19 percent in electrical machinery, by 11 percent in general machinery, by 24 percent in motor vehicles, and by 34 percent in precision equipment (Lawrence, 19831. There is a close cor- relation between these Japanese productivity gains (despite the fact that over- all Japan's GNP per employed worker was only 75 percent that of the United States) and Japanese success in penetrating the American domestic market. The U.S. productivity lag reflects many factors, of which a lag in in- vestment is undoubtedly one of Me more important. Recent U.S. emphasis on product as compared with process innovation relative to other counties may also be a factor. The fact that Europe and Japan had been putting much more emphasis Man Me United States on materials and energy-saving in- novations in manufacturing may have strengthened the relative competitive position of Nose countries when the era of shortages arrived after 1973 (demon, 1982:15~1561. Changes in the average "quality" of Me labor force may also have been significant, since the U.S. labor force grew more rapidly in the 1960s and 1970s and thus encompassed a higher proportion of relatively inexperienced people. However, there is no real consensus among economists as to Me relative importance of the various suggested causes of the lower rate of productivity growth in the U.S. economy. While Me lower rate may be partly explained by Me process of "catch-up" in the l950s and early 1960s, such an explanation seems less plausible for the 1970s and 1980s. Table 3 provides an illustration of the likely importance of net investment in determining relative productivity growth among, counties over the period 1971-1980. It is doubtful whether these differences in productivity grown can be attributed directly to differences in the level of technology, except to

144 TABLE 3 Relation Between Net Invesunent and Productivity Grown HARVEY BROOKS Grown Rate of Productivity in Net Fixed Invesunent as Manufactunog Course Percentage of GNP 1911-1980 Japan 19.5% 7.4qc France 12.2 4.9 West Germany 11.8 4.9 Italy 10.7 4.9 United Kingdom 8.1 2.9 United States 6.6 2.5 SOURCE: Benjamin N. Fnedman, Saving, investment, and government deficits in the 1980's, p. 400 in Bruce R. Scott and George C. Lodge, eds., U.S. Competitiveness in the World Economy (Boston, Mass.: Harvard Business School Press, 1985), Table 11-2. the extent that a higher net investment rate means that the capital stock would include a larger fraction of the most recent and presumably the most advanced and productive—technology. This is especially true at a time when production technology moves as readily as it does among the advanced industrialized countries, especially through internal transfer within mult~- national companies. In other words, the rate of productivity growth includes a factor reflecting Me rate at which the world's state-of-the-art manufacturing technology is being incorporated into anation's capital stock. In consequence, it is doubtful whether grown in either total factor productivity or labor productivity can be used as an index for We state of a nation's technology . . . . . except In thIS Indirect sense. World Market Shares in High Technology Products The United States has historically had a high concentration of its manufactured exports in products that were it&D-intensive, and also a strong favorable trade balance in such products. However, this position has been slipping, and the balance with Japan has actually turned negative recently (National Science Board, 1983:22, Fig. 1-161. In teIms of world exports, the U.S. share of R&D- intensive manufactured good declined from 31 percent in 1962 to 21 percent in 1977, while Japan's share went from 5 percent to 14 percent. World market shares (including domestic market) of the largest U.S. high technology companies declined from 79 percent in 1959 to 47 percent by 1978 (U.S. Department of Commerce, 1983:411. This sounds dramatic, and it is greater than the drop In We U.S share of world GNP. On the other hand, worldwide economic recovery has been a major goal of U.S. policy since the end of World War II, so that we should not have expected to maintain the kind of overwhelming dominance Nat existed while Europe and Japan were still recovering from the devastation of We war. In addition, We much greater

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 145 mobility of information, ideas, and potential inventors across national bound- aries, partly through multinational enterpnses, has been a powerful force toward equalizing the technological level among the industrial countries (Vernon, 1982:148-154~. Nevertheless, the evidence suggests that the ero- sion of the U.S. position, even in it&D-intensive products, has gone funkier than might have reasonably been expected from world recovery alone. In summary, the aggregate data which I have reviewed suggest that, while there has been some erosion in the overall U.S. comparative position, the erosion has so far been rather slight in pure science, somewhat stronger in patenting and the generation of original technological innovations, but by far the greatest in the rate of adoption of innovations from the rest of the world. The biggest lag in innovation has been in manufacturing technology, both in the creation of such technology and, again, even more in its rate of adoption, conditioned primarily but not solely by a low rate of net addition to manufacturing capital stock. At the same time, the United States still invests slightly more than the total of all its major industrial competitors combined in industrial R&D, even after completely excluding defense and space. The U.S. government overall, however, invests a much smaller frac- tion of its R&D expenditures than over governments in work that is spe- cifically aimed at enhancing the competitive performance of its industries in world markets. However, it is doubtful whether this by itself is an important factor, though it may reflect less political commitment to national compet- itiveness as a social goal in comparison with other countnes. Other Indices of Competitive Erosion If the U.S. position on the aggregate indices of comparative performance in science and technology does not indicate a severe problem, how do we account for the widespread perception of a serious competitive problem? Clearly this arises from the dramatic turnarounds that have occulted in the competitive performance of specific industries, particularly the rapid Japanese penetration of American markets in such sectors as consumer electronics, motor vehicles, steel, and machine tools. Although it has received less public attention the case of machine tools is particularly spiking and disturbing. This industry has been one of the key sectors of American export strength since the last quarter of the nineteenth century, and it has especially broad ramifications in contributing to the com- pentive strength of many other U.S. manufacturing, industries. In the early 1970s U.S. machine tool exports exceeded imports by a factor of two, yet in the past five years the U.S. machine tool industry lost half of its traditional market, and over SO percent of all machine tools purchased in the United States were manufactured abroad, mainly in Japan and to a lesser extent in West Germany. Japan considerably leads the United States in He introduction

146 HARVEY BROOKS of robots into manufacturing. More importantly, while the adoption of pro- grammable manufacturing technologies, including robots, has been concen- trated in He aerospace industry and the auto industry in the United States, it has been much more widely diffused in Japan. In 1980 sales to auto makers in Japan accounted for only 29 percent of robotics sales, compared with 60 percent in the United States. Moreover, in Japan small and medium-sized Grins account for a much higher proportion of programmable automation sales than in the United States, in part because of an explicit government program to encourage adoption of programmable automation widely through- out the economy (Parsons et al., 19841. Equally disturbing is Japanese world market penetration in fields of ma- tenals technology pioneered in the United States. Japan now accounts for half the titanium used in the non-Communist world, mostly exports to the United States, although the basic technology and the industry were fist developed here under defense sponsorship. A similar example is that of carbon-fiber reinforced plastics as a substitute for metals. The United States has pioneered in inventing new ways of using such materials, and it has a spectacularly growing market for them, but Japan accounts for about 65 to 80 percent of the world output, most of which is exported to the American market (Ayres, 1984:1381. A typical Japanese competitive strategy is to target a relatively small market niche for an advanced technology just behind the world technological frontier and then to develop a superior manufacturing technology for that product that yields superior quality and delivery reliability, thereby achieving very rapid market penetration accompanied by experience and scale economies, which eventually produce an impregnable market salient on which continuing product and process improvement and an expanding range of product com- petitiveness can be built (Imai and Sakuma, 19831. Early market success provides He resource base and the infrastructure for an expanding scope of innovation and investment. RELATIVE ROLES OF PUBLIC AND PRIVATE SECTORS ~ GENERATION AND COMMERCIALIZATION OF NEW TECHNOLOGY* The United States has been slow to accept any notion that government has a responsibility for the generation of innovations that will result in goods and services to be sold in private markets. The only exception to this has been agriculture. The conventional view is that market forces alone will be sufficient to direct innovative resources and investment into the areas of highest potential commercial return, and that government attempts to inter- vene in or even influence the process are more likely than riot to be coun- *Brooks (1982b).

NATIONAL SCIENCE POI lcY AND TECHNOLOGICAL INNOVATION 147 terproductive. This view is reinforced by the notion that industrial innovation is driven primarily by "market pull" rather than "technology push." While government scientists and engineers may be rather good at identifying new technical opportunities, they lack the experience and knowledge to assess market potential and user needs, with the result that the typical govemment- driven technological development frequently tends to be a technical success but a commercial failure. The British-French Concorde project is usually cited as the prototypical example of a spectacular technological achievement driven by government initiative which proved to be a commercial disaster (Nelson, 1984:541. In the United States, as noted earlier, such a disaster was probably forestalled by the Senate's killing the prototype supersonic transport program in 1971. However, the United States has seen many other examples of government technical incentives in housing, transportation, and energy that have failed in the market, largely because they were primarily motivated by Me rec- ognition of a technological opportunity rather than a clearly demonstrated market need. Of course, market circumstances can change rapidly, and the dimming of commercial prospects of energy technologies such as the breeder reactor and synthetic fuels owes something to this circumstance. Moreover, as we saw above, many revolutionary new technologies of the twentieth century have had their origins in governmental initiatives, usually undertaken originally for noncommercial purposes. Most of the success stories were cases in which the recognition of a technical opportunity was the crucial factor, as opposed to only the fulfillment of a generally recognized societal need. The preceding generalization, however, is obviously an oversimplification, and does not fit with the generally acknowledged success of federally man- aged agricultural and biomedical research, to cite two examples. The fol- lowing sections attempt to assess the federal role in a more discruninating manner, indicating where there is general consensus that government has a role, and where there is disagreement. A crucial issue in this connection is not only We area and character of the candidate technological development but the locus of the decision-making process wide respect to the strategy and tactics of the development process. Areas of Consensus on Federal Responsibility Government as Customer There is little debate about the necessity of a federal responsibility when the government, acting as agent for Me society as a whole, is the ultimate user and the goods or services produced are widely acknowledged to be "public goods," i.e., goods or services from which everybody benefits whether or not they pay for them. Examples are defense

148 HARVEY BROOKS technologies and the generation of the scientific knowledge necessary to underpin the formulation of environmental, health, and safety regulations. This is not to say that there are not strong political controversies as to how much or what kind of defense we need, or as to how much regulation is in the public interest and what should be regulated. But the principle that government should bear the ultimate responsibility for such public goods is not questioned. Fundamental Research Although the consensus is more recent and not quite as strong as in the case of public goods like defense, there is wide agreement that the federal government has a responsibility to support the generation of knowledge whose potential benefits are widely diffused among many end uses, so that no one user has sufficient stake in those benefits to sponsor the necessary research. Indeed it can be argued that such general- purpose knowledge is a public good. Moreover, in order to be a public good, such knowledge has to be widely shared and freely and rapidly communi- cable. Yet the freer and more open the communication the less the chance that the research sponsor can hope to recover his costs from potential ben- eficiaries. The only hope of recovering the costs of such public knowledge is a compulsory charge to all, namely taxes. The argument for generating this kind of public knowledge in the mostly widely sharable way also involves efficiency: the generation of public knowledge is more rapid and efficient if it can be widely and rapidly shared among all people competent either to build on it for further advances or to use it. In recent years we have come to accept that responsibility for funding basic research does not necessarily entail a responsibility actually to perform it. Since World War II We belief has been increasing that the separation of funding from performance generally contributes to the cost-effectiveness of R&D because it opens up the possibility of drawing on a broader scientific community. This is not universally acknowledged, but it is recognized de facto, as indicated by We fact that only about 25 percent of government- funded R&D today is perfonned in civil service laboratories manned by full- time government employees (National Science Foundation, 1984c:3~. Externalities As mentioned above, where government has a legislated mandate to regulate in matters of environmental protection, heals, and safer, it also has an acknowledged responsibility to generate the necessary scientific knowledge base through government-funded R&D. This is not as definitive a criterion as it sounds, however. For example, in the case of We regulation of prescription drugs, the Food and Drug Administration (FDA) relies pri- marily on the regulated industry itself to generate the scientific data on which bow He safety and efficacy of a new drug is to be evaluated before intro- duction to He market. The FDA has a minimal research program of its own,

NATIONAL SCIENCE POl ICY AND TECHNOLOGICAL INNOVATION 149 though it does retain a scientifically competent staff to review and evaluate the data provided by industry. The Environmental Protection Agency and the Nuclear Regulatory Commission (NRC) also depend heavily on the in- dustries they regulate for their data bases, but in addition they have substantial independent research programs, both in-house and under contact. An issue that arises in these agencies is the degree to which they get involved in research to assess equipment designs or to develop pollution-abatement or accident-prevention technologies. At one time, in its early days, the EPA ran a congressionally mandated research program aimed at actually devel- oping alternatives to the internal combustion engine up to the point of dem- onstration of prototypes (e.g., the "hybrid vehicle" program). The NRC has had large-scale testing programs to evaluate specific engineered safety mea- sures, such as the Emergency Core Cooling System. The line between design research and evaluation research is often a difficult one to draw in practice. Another issue that arises in connection with government responsibility for externalities is the degree to which government should fund research pro- grams designed to develop a knowledge base for the assessment of technol- ogies that are not yet ripe for regulation, and whether such research would be better left to the industries in which detailed expertise on emerging tech- nologies resides. There are arguments Eat industry, left to itself, will un- dennvest in research related to He externalities resulting from its technologies, especially those which are indirect and far in the future—likely to be im- portant only after the technology has been manufactured and marketed on a major scale. The cases of radioactive waste disposal and management of toxic chemical wastes readily come to mind as examples of industry's having probably invested in R&D at less than a socially optimal scale. In the absence of compelling evidence of potential hazard, industry will tend to invest only in that R&D which appears to be necessary to meet existing regulatory requirements and standards, but it will be reluctant to do research intended to anticipate a need for regulation at some time in the future. In this field, in fact, industry has some conflict of interest, in that the discovery of new regulatory needs may appear to increase its costs and reduce its markets. In part the current situation arises from the unusually rapid change in social expectations that has taken place in the last 20 years, particularly as it has been expressed in stricter and stricter interpretations of product liability and managerial negligence. This has led to a considerable change in industry attitudes toward research related to potential but speculative and uncertain negative externalities resulting from its activities (New York Times, 1984:16~. Part of industry's attitude arises from the natural human tendency for developers of a new technology to become advocates and to be slow in accepting the possibility of adverse effects until it Is forced on their attention by evidence generated outside their industry. A part may also be due to the fact that the expertise required to do research on secondary consequences is

150 HARVEY BROOKS likely to be quite different from the expertise required to develop the tech- nology. For example, chemical manufacturers generally lacked expertise in groundwater hydrology and disposed of toxic wastes on their own property in ignorance of the fact that the wastes might eventually migrate into drinking water supplies after a sufficiently long period. One problem is that the time horizon for appearance of externalities tends to be much longer than the time horizon for product or process development. A new development may be completed and the developers dispersed onto new projects before the need to consider waste management becomes apparent. One might formulate the question here by asking whether the government has a responsibility to support a vigorous research pro~,rarn aimed at searching for trouble arising out of industrial technologies, or whether it should wait for others to identify potential troubles before initiating research. The latter has generally been the practice, and still largely is. Even in the case of nuclear power, for which government was responsible for development of much of the generic technology, the relative investment in R&D on radio- active waste disposal was almost certainly less than socially optimal. Often the government does not support "externality" research unless a regulation is already in place whose implementation or enforcement would require such research. The Office of Technology Assessment (OTA) was created by Con- gress in 1972 to carry out what might be described as anticipatory studies to identify the secondary and tertiary consequences of emerging technologies or the expansion in Me scale of use of existing technologies (Public Law, 1972; National Academy of Sciences, 1969; Brooks, 1985c). However, the OTA of necessity restricts itself to synthesis and interpretation of existing research from all possible sources. It lacks the capacity to initiate original laboratory or theoretical research or to leverage such research on the part of other agencies or of industry, except indirectly through the dissemination of its reports or through its channels of communication to the Congress and over government agencies. There is lime to indicate Hat OTA has much influence on He national research agenda, although this is hard to pin down. Simon Ramo (1985:2~27), an industrialist, has proposed a novel scheme for dealing win the government's responsibility for research on externalities. "We should saw," he says, "by assembling from existing regulatory staffs and outside sources, a competent organization to uncover, study, and assess all hazards to safety, health, and the environment." Ramo goes on to say: We would relieve this investigatory unit of all responsibility (or even the slightest ap- pearance of it) for considering positives as well as negatives and attempting balanced decisions.... Ibis group would be equipped with the required experts, tools, facilities, and budget to enable it to track down hazards and potential hazards in existin,, or proposed activities with reasonable depth and thoroughness.... Efficiency, synergism, and or- ganizational flexibility would all be fostered if the specialists and tools were in one strong unit. It would no longer be necessary for Congress, upon its [usually accidental] discovery of a new danger, to launch still another new agency to investigate it.

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 151 Ramo goes on to propose a presidential board to convert all this information into mandatory rules, standards, and enforcement tools, this board to con- st~tute "a pragmatically effective microcosm of the electorate" with the presidential-appointment and Senate-confirmation process guaranteeing that it would be "responsive to citizens' goals and priorities." Although ~ think this proposal is politically naive in that it assumes a separability between knowledge and its application that is unrealistic in a political context, it does pose the issue of government responsibility for broad-based research on externalities in a more direct way than any other proposal I have seen. Areas of Consensus on Inappropriateness of Government Role There is general agreement that where the costs of R&D can be recovered from the future revenue stream generated by sale of the resultant products, services, or information, there is no justification for a government role. However, this prescription is not as simple as it sounds because He appro- pnability of the benefits of research to its sponsors will depend on the degree to which property rights in the resultant knowledge can actually be secured by law to the sponsor. For this reason leaving the initiative to the private sector implies a positive policy on the part of government to protect and enforce intellectual property nghts. The rapid and efficient generation of new knowledge the maximum rate of advance of science demands a wide sharing of knowledge, so that each researcher can build on the advances of others. On the other hand, the assignment of intellectual property rights entails some sacrifice of this public good in order to increase He incentive for private investment in innovation, particularly in the postpatent stage. In other words, the existence of intel- lecn~al property rights is necessary to enhance the total investment in inno- vation and to ensure Hat its later stages are more responsive to the needs of the market Where the optimal public benefit lies as between open and proprietary research is a debatable question on which the balance of public policy has shifted back and form over time. In general, the national policy has been that discoveries—facts of nature- are in the public domain and cannot be vested with property rights, while artifacts human constructs—can be so vested. Even here we have invented He patent system, which creates a property right not previously existent in rem for public disclosure and open sharing of the underlying knowledge. The actual practice of the knowledge is a commodity, but He use of He knowledge to generate new knowledge is not an ingenious distinction which has worked quite well in practice. In the absence of the patent, knowledge would be purely proprietary and hence unavailable to be built on by others. Because of the linkage among many different pieces of knowledge, He vesting of property rights may not always be necessary to secure the benefits of innovation to the innovator. The mere fact of being first in He field may

152 HARVEY BROOKS itself confer a sufficient monopoly to secure the revenue stream to the innovator for long enough for him to recover his investment before imitators can successfully enter the market and erode his monopoly profits. In practice, whether an area of R&D is appropriate for government depends on a highly subjective judgment both as to whether the resultant innovation will provide a social benefit commensurate with its cost and whether, if He government does not undertake the effort, the incentives are sufficient for the private sector to undertake it. This is a difficult balancing act. The more attractive the social benefits, the greater the likelihood that the private sector will see a commercial opportunity, but also the more political pressure there will be for the government to ensure the realization of the benefits in the shortest possible time. Hence, while there is widespread agreement on the theoretical criteria for government intervention or participation in the inno- vation process, it is much harder to get agreement in any concrete case. It is always possible to argue that the social benefits of a prospective innovation exceed the potential private benefits by a sufficient margin to justify public intervention, but because all the benefits are in the future and the costs are uncertain, the possibilities for rationalization of actions desired for nonra- tional reasons are almost infinite. Efforts to compare the social and private returns for particular innovations indicate wide variation among specific cases even though, on the average, social returns appear to exceed private returns by a factor of about two (Mansfield, 19SSb; Griliches, l98S). In practice the benefit of the doubt for private versus public funding tends to shift with the prevailing political climate and ideology. Recently the pendulum has swung heavily in favor of reliance on private incentives and minimizing direct government intervention except in fields directly linked to national security considerations. Even the umbrella of national security, however, can be quite easily stretched, especially in periods of ample resources. Areas of Controversy Following are a number of arguments that have frequently been used as justification for direct federal support of R&D when the relative role of the private and public sectors has been controversial and fluctuating. The li~ni- tations of each argument are also brought out. An important factor in each of these cases is who makes the choices and strategic judgments as the R&D evolves. Here the issue of whether He judgments to be made relate primarily to science or technology considerations or primarily to market considerations is often key. High Risk At times, the technical or market risks are considered so high that it is improbable any profit-seeking entity will undertake the investment. Clear-cut examples include space technology, nuclear power (in its early,

NATIONALSClENCEPOWCY AND TECHNOLOGICALlNNOVAT10N 153 precommercial phase), the breeder reactor, fusion technology, synfuels de- velopment, and some types of exploratory assessment of natural resources. Risk in these instances is compounded of three factors: (1) the magnitude of investment required before commercial success can be predicted with suf- ficiently high confidence, (2) the hazard that government may intervene for public policy reasons to limit the deployment of the resultant technology or the sale of the resultant products or services after a considerable investment has been made (e.g., unforeseeable environmental effects; foreign policy considerations, such as nuclear proliferation; antitrust considerations), and (3) the lack of available expertise to assess the prospects of the technology (e.g., nuclear power and radioisotope applications right after declassification of the Manhattan Project). Such areas are usually candidates for increased cost shanug by the private sector as they progress from basic and generic research toward commercial application. There is always a tendency, however, for government to hang on too long or to distort the commercial judgment of the private sector by overpromotion based too exclusively on technical considerations (e.g., the case of the light water reactor, discussed below). If the project is large, with long lead times before it can be tested in the marketplace, there is also the danger that the circumstances that' made it appear economically attractive In the first place may change without its being recognized (e.g., the fast breeder reactor following the drastic decline in the forecast rate of grown of energy demand after the 1973 embargo and price jump). The cases of the supersonic aircraft, high-speed ground transportation, and prefabricated housing (as in Project Breakthrough) may also be examples of government officials being too much in the position of making market judgments and too much influenced by "technology push" considerations, by the existence of challenging tech- nical opportunities without necessary commercial value. Exceptional Social Returns There are few R&D projects for which the social returns do not exceed the private returns. Mansfield has shown that, on the average, the social returns to industrial R&D exceed the private returns by a factor of two, despite instances in which social returns are less than private returns (Mansfield, 1985b). Thus it is rather easy to argue for gov- ernment participation in industrially oriented R&D projects on the grounds of unusually high social, relative to private, returns, leading to undenn- vestment by private entrepreneurs. This was the argument implicitly or ex- plicitly used to justify Me creation and rapid expansion of the Department of Energy after the 1973 oil crisis. The U.S. dependence on imported oil was seen as an "externality," which made benefits to the county as a whole greater than the sum of cost savings to consumers. In fact, using plausible models of the world oil market and Me impact of l3.S. demand on world oil

154 HARVEY BROOKS prices, one could estimate a marginal cost of an imported barrel of oil that ranged anywhere from a 10 percent to a 100 percent premium over the market price to the private consumer (Energy Modeling Fomm, 19821. In this way one could justify a federal investment ranging up to anything that might be justified by the excess "social premium" of oil imports. These arguments could be used to justify large public investments in new supply technologies as well as in research and "demonstration" of a variety of energy end-use efficiency improvements as long as the total cost, R&D plus capital invest- ment, was less than the product of the effective price premium and the volume of oil imports. Public subsidies for renewable energy technologies were justified not only by the oil import premium but also by the alleged avoidance of negative environmental externalities that might result from the use of these "benign" technologies. Fragmented Industries The principal examples of fragmented industries are medicine and agriculture. In both industries, an important element of the delivery system is individuals or small family enterprises that lack an effective mechanism for joining together for the collective support of research. Such collective support tends to be precluded by the "free nder" problem, the fact that every entity benefits whether or not it contributes to the support of the research. An additional argument is that both food and health care, though sold in part as market goods, are regarded as "meet goods," i.e., private goods to which everybody In society in some sense has an entitlement (Mus- grave, 1974:274-2751. In addition, it may be no accident that in both these examples the research supported with public funds is in Me life sciences. We rely primarily on complementary activities in the physical sciences and engineering canted out by Me private sector to generate the innovations necessary for Me system as a whole. Whether this is an accident of history or can be given a more solid rationale is harder to say. It is certainly true that the benefits of life sciences research are less "appropriable" Man Rose of physical science and engi- neenng research in Me chemical, pharmaceutical, farm machinery, and med- ical devices industries. Nevertheless, there are over fragmented industries, such as housing and construction, in which the public investment in research has been both smaller and less successful. In part this may due to the lack of development of an easily defined natural division of labor between the private and public sectors as there is in biomedicine and food (Brooks, 1982b:3371. Narrow Markets The classic example of a narrow market is "orphan drugs," drugs to deal with life-threatening diseases Mat affect only a small fraction of the population. The private market for such drugs is too small for recovery of development costs, including the extensive animal and clinical

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 155 testing required for FDA approval. Such drugs may not become available at all unless developed at public expense, although a few may be developed for prestige purposes or as a public service by private firms. Clearly this is a case that falls in the category of a "merit good." In the health area, society does not apply benefit-cost analysis to new techniques in the same way as in other fields. It assumes that if a potential capability exists to cure a life-threatening disease there exists a moral obligation to develop that capability. It is a kind of extension of the philosophy underlying the Hippocratic Oath to the development of new technologies (Brooks, 1973b:21~. A society does not necessarily act according to this principle consistently; indeed, if it did, the aggregate cost would be unacceptable. Nevertheless, as a society becomes more affluent, it apparently tends to make larger and larger investments in health care technologies irrespective of the size of benef~t-cost ratios at the margin. Public Policy Many types of goods and services could be sold either on the private market or as public goods. In most societies they are mixed goods and have a tendency to turn gradually into entitlements or rights of citizenship, rather than private goods, as societies grow in wealth. Examples in this category are health care and education, and even to some extent basic ser- vices, such as telephone and electncity. Over services, such as weather forecasts or monitoring of the environment by remote sensing satellites, in principle could be marketed as private goods, but only at the cost of restnct~ng access to a few users who can pay the price. Many fonns of inflation services are also of this character. As a matter of public policy it is decided Mat the social benefit of universal access justifies public development and eventually operation. In this view the "pos- itive externality" resulting from universal, or near-universal, access makes private marketing of anything but "value-added" services, specially pack- aged for users who can justify the premium price, an undesirable policy. It is also often true that the "transaction costs" involved in creating a market or quasi-market, which must include a technique of excluding nonpayers, becomes prohibitive. These are some of the issues bound up with the current administration proposal to "privatize" weather and remote sensing satellite services (Gregory, W., 1982; Waldrop, 19821. Key Industries An even more controversial area is that of maintaining industnes that have become uneconomic or noncompetitive on the grounds that they are essential for some public purpose, usually national security or, possibly, the maintenance of employment levels. A classic example is ship- building and maritime shipping. The United States maintains an uneconomic industry at enormous cost because it might be a vital national resource in wartime. Almost every domestic industry experiencing severe foreign com-

156 HARVEY BROOKS petition uses this argument to obtain government assistance or protection. It is often an argument difficult to prove or disprove, and therefore it tends to be heavily overworked in the political process. It is not always wrong, but it is certainly controversial. Cntics often counter that there are more cost- effective ways of maintaining an industrial capability without requiring con- sumers to foot a large part of the bill (as in the case of import quotas, for example). In general subsidization of R&D or of capital investment em- bodying the latest production technology in order to restore competitiveness is a more cost-effective strategy than~continuing operating or capital subsi- dies. However, it is often resisted by the industries involved because it requires much more organizational adjustment, almost always including the permanent loss of jobs, since overstaffing and undercapitalization are the most frequent sources of loss of competitiveness. There is a tendency for many countries to regard the same industries as "key" either for national security reasons or for reasons of linkage to other elements of the national economy. Information technology, for example, is on everybody's list of key industries. This not only gives rise to political frictions between countnes, but also generates economic inefficiencies in areas in which world- or continental-scale markets are often required to produce the revenues necessary to underwrite innovation and investment. A striking example is the telecommunications equipment industry within Eu- rope, where each of the national Post Telephone and Telegraph (P17) agen- cies procures its equipment from national companies, at least in the larger countries. The key-industry approach is also a prescription for the creation of world overcapacity in certain sectors—overcapacity which eventually re- quires some forte of rationalization and concerted cut-back, and which is also a major source of trade frictions. The petrochemical industry is an especially egregious example, exacerbated, of course, by the dramatic rise of raw material costs and the efforts of OPEC countries with excess natural gas to develop downstream petrochemical industries based on cheap gas (Bower, 1985:2671. Steel and shipbuilding are similar examples. There appears no simple resolution of this problem, which is at root a product of the inherent inconsistency between an increasingly global world economy and He persistence of national sovereignty as a key, and even growing, political force in the world. In general the most successful adjust- ment strategy appears to be more rapid movement into higher value-added sectors of the world market. To the extent that it supports such a strategy, public investment in R&D is therefore a more acceptable policy than over forms of subsidy or market protection. Generic Applied Research There is increasing interest in this county in the possible role of government in the sponsorship of what is called "generic applied research" in areas of industrial interest. This is defined as mission-

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 157 oriented research that is not aimed at the solution of specific product-related problems or at final design of commercial technology. The classic example is aeronautical research as it was conducted by the National Advisory Com- mittee for Aeronautics after World War I, and as it continued under NASA after its conversion from NACA by the Space Act of 1958. NACA did not design airplanes or even operational aircraft engines. Rather it built wind tunnels in which it tested new airfoil shapes and provided test services to aircraft designers. It also had the most advanced programs in the theory and testing of structures. In its reincarnation as NASA, it pioneered the devel- opment of noise-suppression techniques for jet engines and the basic tech- nology of turbofans to improve fuel economy. It worked in close collaboration with designers in bow the engine industry and the airframe industry, but left design and testing to the private sector, albeit providing testing services for a fee. NACA is frequently cited by theorists of the division of labor between the private and public sectors in R&D as the ideal model of synergy between the two sectors (Nelson, 1977:11, 121-122, 12S, 139; Mowery and Rosen- berg, 1982:128-130~. Recently the National Academy of Engineering has put forward proposals for the creation under NSF auspices of industry-university-government co- operative research centers to work on applied problem, of industrial interest, and many universities have set up such cooperative arrangements with in- dustry. The "generic" nature of the research is supposed to be guaranteed by the fact that the results are public and open to all; in fact this could be used almost as Be definition of "generic" (National Academy of Engi- neering, 1984~. There has been more experience with this type of generic research abroad than in this county, and the results have been mixed. The most successful country seems to have been Japan, with its well-publicized programs in very large scale integration (VLSI), "fifth generator computers," and robotics. Industries have received public support for such programs in rennin for agree- ing to share the resulting information widely with each over. The ingredients for success seem to be (1) a substantial financial stake by industry, (2) an equal or dominant voice of industry in the planning of the broad strategy of the research, (3) wide sharing of the research results among all Be paruci- pating organizations, and (4) limitation of development projects to "proof- of-principle'' demonstrations or models, leaving final product development to individual films on the usual proprietary and competitive basis (Bloom, 1984~. International Cooperation Sometimes the possibility of international co- operation can become the justification for an applied research program sup- ported by governments. The most successful current example is probably Be fusion energy program involving the United States, Be European Commu-

158 HARVEY BROOKS nity, the Soviet Union, and Japan, which has aggregate annual expenditures close to $450 million (Thomassen, 19841. In a way the justification for such intemahonal programs can be considered as a simple extension of the ar- guments used for large national government investments in precommercial applied research programs, where commercializable results are in the distant future and the technical risks are high. Such international cooperative pro- grams can be of two kinds. The more usual kind involves an agreed division of labor between national research institutions with wide sharing of results, joint planning of major facilities and experiments, and extensive short-term exchanges of technical personnel (from months up to a year or more). The rarer kind of cooperation involves the setting up of joint laboratories with a more or less permanent multinational staff. An example is the Italian-based laboratory of Euratom at Ispra, generally regarded as less successful than the fusion program. Still another type of international cooperative program is the Super-Phenix fast breeder reactor program, which involves both gov- ernment and industry, and whose objective is a full-scale commercial pro- totype (Nuclear News, 1985~. Although this is a predominantly French project, other European countries have made a significant investment in it in return for sharing in the information and operating experience developed. From the standpoint of participating researchers, such international pro- grams often have the advantage of being less subject to fluctuations in the budgetary priorities or other policies of individual national governments. Budgetary planning gets locked in by the international nature of Me com- mitment and thus tends to provide an environment of greater policy, as well as financial, stability. Other Public Policies for innovation So far the discussion of government intervention in the innovation process has been concerned with direct government sponsorship of R&D or of pro- totype construction and testing. Although this is the most visible and widely debated type of intervention, there are many more indirect policies that may be of equal importance. One of the principal advantages of such indirect policies is that they provide a natural Hearts for leaving decisions about viability in the market to industrial managers who are in the best position to judge what the market needs or is likely to accept. Thus, indirect forms of intervention are most appropriate when market judgments are most significant for success. Tax Benefits to Consumers One way to stimulate innovation is to provide tax benefits to consumers that lower the effective price of innovative products whose consumption the government decides yields public benefits or "ex- temalities" not offered by alternative products During the energy crisis many

NATIONAL SCIENCE POLICY ED TECHNOLOGICAL INNOVATION 159 states as well as the federal government provided tax credits for household investments which improved energy efficiency or resulted in the substitution of renewable for nonrenewable energy sources, e.g., solar hot water heating or passive solar house design. The idea was to accelerate the market pene- tration of new technologies that would result in reduced oil imports, leaving the choice of technologies to the market. One could regard this as compen- sating the consumer for his contribution to the reduction of a "negative externality," since the alternative would be for him to require more imported oil. R&D Tax Credits Almost all He industrialized countries now offer some sort of tax credit or other tax benefit to firms which increase their R&D spending above some base year. This tax benefit could be thought of as compensating the film for He fact that there will always be some spillover effect from its R&D which will benefit other firms, consumers, or the general public, and which it will not be able to recapture in the price of its products. In addition, since there is an apparent correlation between firm grown and R&D spending, it might be argued that there is a generalized benefit to the economy as a whole from stimulating industries with higher growth potential. On the other hand, there is considerable debate as to whether the tax credit actually stimulates private R&D spending or merely provides a reward for spending that would have taken place anyway for competitive reasons or an inducement for firms to redefine existing marginal activities as R&D. In a recent study of the impact of R&D tax credits in the United States, Sweden, and Canada, Mansfield (1985a) has concluded that such credits and other allowances "appear to have had only a modest effect on R&D expenditures," and from this he infers that in their present form R&D tax incentives "are unlikely to have a major impact on a nation's rate of innovation" largely because the price elasticity of industrial demand for R&D is quite low. Technology-Forcing Regulations One way of stimulating industrial in- novation is to use government to set stiff performance standards for industrial products standards that cannot be met without considerable technological innovation and then rely on prospective sanctions to induce private R&D to meet the regulations. This was the strategy followed by the U.S. Congress in respect to the auto industry in three areas: exhaust emissions, fuel eff~- ciency, and vehicle safety. It is also implicit in water pollution regulations, notably in the Clean Water Act of 1977 (Public Law, 1977), which originally required zero discharge into waterways by 1985. The advantage of this approach is Hat it leaves the choice of technology to engineers and managers familiar with the technology of the industry. In the auto industry Here is no doubt that the new regulations stimulated the industry to step up its R&D spending rather dramatically from the early 1970s on, and that many technical

160 HARVEY BROOKS goals that the industry insisted were unrealistic were eventually achieved- most notably the attainment of lower emissions with virtually no sacnf~ce in fuel efficiency. On the other hand, many observers have argued that these technology-forcing regulations hurt the industry seriously at a time when it was just beginning to face severe competition in domestic markets from Japanese imports (Abernathy et al., 1982:83-88; Eckstein et al., 1984:50- 53~. In fact the attainment of the originally specified standards was postponed year after year, and it is at least debatable whether the goals could not have been achieved more efficiently without legislated standards and timetables that had to be continually revised (Goodson, 19771. Voluntary Standards Industrywide standards can be an important factor in encouraging the rapid diffusion and adoption of new technology. On the other hand, standards can sometimes be abused to confer unfair advantages on particular firms. This is another example of the fact that amving at the optimum choice between competition and cooperation (from a societal point of view) is a difficult balancing act. The United States has a unique system of voluntary standard setting through a number of indus~ywide standard- sethng associations, such as the American National Standards Lnstitute (ANSI) or the American Society for Testing Materials (ASTM). Standard setting is camed out and financed by the industry itself under antitrust safeguards that apparently work quite satisfactorily and have been relatively little criticized. It is important Mat the government maintain a legal regime which is supportive of such voluntary standard setting, which has been an important factor in U.S. competitive success in a number of areas. It is one of the instruments for assuring a continental market for new technologies and thus realizing scale economies at a relatively early stage in an emerging technological area. The lag in standardizing designs, for example, may have been an important factor in the faltering performance of the nuclear power industry in Me United States after a promising Stan. Intellectual Property Since the mid-1970s there has been a general trend toward strengthening intellectual property laws so as to improve the appro- pnability of Me benefits of innovation to the innovating organization. The argument for this has been that much of the financial risk involved in intro- ducing a new product or process to the market is incurred after the original invention has been made. Hence, many potentially valuable inventions are not converted into viable innovations because the innovator cannot be con- fident of a temporary monopoly In the market for long enough to recover his postinvention start-up production and marketing costs. Nevertheless, the benefits to commercial competition of stronger intellectual property rights always have to be balanced against He possibility that too much competition in the earlier "generic" phases of new technology development will result

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 161 in wasteful duplication and slower progress due to a lack of cross-fertilization ideas. In some cases it is possible for too much emphasis on patents or proprietary know-how to result in overinvestment in certain areas,' thus re- versing the usual argument that We "positive extemalities" of R&D result in private underinvestment in R&D. There is now a widespread concern that the pendulum has swung too far in favor of the protection of intellectual property rights, particularly in the relations between industry and universities. Simultaneously, there is concern with government moves to regulate the free flow of scientific inflation for national security reasons (Corson et al., 1982; Wallerstein, 19841. There is also a question of the degree to which proprietary research, as well as government regulation of the flow of research information, has the effect of shielding emerging technologies from proper public assessment until after irreversible commitments have been made to final design and deployment (National Academy of Sciences, 1969:32-331. Antitrust Policy In the recent past there has been much criticism of the overly rigid interpretation of antitrust legislation in relation to cooperation among firms in R&D, particularly in the precommercial phases of innovation before the emergence of specific product designs (U.S. Congress, 19841. One of the sources of Japanese success in technological innovation in recent years is believed to be the Japanese govemment's policy of encouraging cooperation among fimns and even agreed division of markets for products in emerging technology areas. The U.S. Department of Justice has now clarified its interpretation of antirust legislation to be more positive toward research cooperation among firms, and the Department of Commerce has been actively promoting We idea of R&D limited partnerships (U.S. De- parunent of Justice, 1980; Me~Tifield, 1982; U.S. Congress, 19841. Planning vs. the Market The current debate over industrial policy in the United States has specific implications for R&D and science policy. The question is whether the overall national pattem of R&D the resultant of government and private R&D decisions should reflect some kind of con- sensual national vision of the future of technology. Even granted the desir- ability of some sort of coherent pattern, there remains a question of We process by which this pattern is arrived at. A decentralized decision-making process does not necessarily imply an incoherent outcome. The pattern does not have to be established deductively from some generalized vision of a future society established by a few "wise men." It can be established in- ductively through a political and market struggle between competing visions. Essential to the successful outcome from such a smuggle, however, is an open process in which ideas and visions can compete ''fairly," with wide- spread public participation. Even We market and the political process can be considered as compenng

162 HARVEY BROOKS processes in which the participants have different relative weights. The prob- lem with market-like processes in both the political and economic spheres is that they may tend to give too little weight to "externalities" and systemic effects. Unfortunately, the same tends to be tme of the political process, especially in pluralistic societies like He United States. Groups try to use the political process to defend or enhance their interests without reference to He "externalities" of their success. An interesting question is whether the highly communicative and consensual process of Japanese decision mak- in~ helps to offset this limitation of the decentralized mechanisms preferred in the United States, with the result that more internally consistent systems of action result while still avoiding the large errors that result from patterns imposed by a small group at the top of a hierarchy. OUTLOOK AND PROSPECT: CAN THE U.S. DECLINE BE REVERSED? Despite the searching self-criticism Hat is going on in the United States, in technological innovation we are still perceived by the rest of the world as No. 1. Nevertheless, our relative position has eroded. Some of this erosion was inevitable, especially given the long-term U.S. political interest in equal- izing weals and technical capacitor among nations in the interest of greater political stability and the strength of the free world consensus against political and military encroachment by He Communist bloc. The world economic and technical dominance by the United States that existed in He 1950s and early 1960s was not sustainable and was essentially incompatible with He legiti- mate aspirations of He rest of He world's peoples. Seven percent of the world's population controlling 50 percent of its GNP was probably not a triable situation for any prolonged period of history. Moreover, He "race for the new frontier" (National Research Council, 1983) does not have to be a zero-sum game internationally any more than it has been nationally among firms or regions of the county. The growing wealth of the rest of He world provides new markets and new opportunities for innovation by U.S. entrepreneurs. In principle I believe the United States still retains the capacity to stay in front of He rest of the industrialized world, but not way in front, if it gives high priority as a society to science, tech- nology, education, and productive investment without sacrificing a reason- able degree of equity among its population. This is not an easy prescription, nor is it an impossible one. REFERENCES Abernathy, William J., et al. 1982. The Competitive Stams of the U.S. Auto Ins smy: A Study of the Influences of Technology in Determining International Industrial Competitive Advantage, Automobile Panel. Committee on Technology and international Economic and Trade Issues, National Research Council. Washington, D.C.: National Academy Press.

NATIONAL SCIENCE POLICY AND TECHNOLOGICAL INNOVATION 163 Altshuler, Alan, Martin Anderson, Daniel Jones, Daniel Roos, and James Womack. 1984. The Future of the automobile. Report of MIT's International Automobile Program. Cambndge, Mass.: ~ Press. American Association for the Advancement of Science. 1985. AMS Report X: Research and De- velopment, PY 1986. Intersociety Worlcing Group. Washington, D.C. Aspin, Congressman Les. }984. Defense spending and He economy. News Release, April. Ayres, Robert U. 1984. The Next Industrial Revolution: Reviving Industry Through Innovation. Cambndge, Mass.: Ballinger. Bauer, Raymond A., with Richard Rosenbloom and Laure Sharp. 1969. Second Order Consequences, A Methodological Essay on the Impact of Technology. Cambndge, Mass.: MIT Press. Baxter, William F. 1983: Transcript of presentation to the National Association of Manufacturers, Prototypists, Inc. Washington, D.C., May 10, 1983. Berger, Edward J., Jr. 1980. Science at the White House, A Political Liability. Baltimore, Md.: Johns Hopkins University Press. Bloom, Justin L. 1984. Japan's Ministry of international Trade and Industry (~l) as a Policy Instrument in the Development of I~orma~on Technology. Program on Infonnation Resources Policy. Harvard University, C~mbudge, Mass. October. Bok, Enid C. 1966. The establishment of NASA. Pp. 161-270 in Sanford A. Lakoff, ea., Knowledge and Power. New York: Free Press. Bower, Joseph L. 1985. Restructuring petrochemicals: A comparative study of business and gov- ernment strategy to deal with a declining sector. Chapter 7, pp. 263-300, in Bruce R. Scott and George C. Lodge, eds., U.S. Cornperinveness in the World Economy. Boston, Mass.: Harvard Business School Press. Brooks, Harvey. 1963. Government support of science. Pp. 11-21 m~cGr¢w-Hill Yearbook Science and Technology. New York. Brooks, Harvey. 1965. Future needs for the support of basic research. Pp. 77-110 in Basic Research and National Goals. Report of the Committee on Science and Public Policy, National Academy of Sciences, to the House Science and Astronautics Committee. Washington, D.C.: U.S. Gov- ermnent Printing Office; also repented as Chapter 6 in Harvey Brooks. 1968. The Government of Science. Cambridge, Mass.: MIT Press. Brooks, Harvey. 1970. Impact of the defense establishment on science and education. Pp. 931-962 in U.S. Congress, House. NationalScience Policy. House Congressional Resolution 666, Heanags Before the Subcommittee on Science, Research, arid Development, Committee on Science and Astronautics, 91st Cong., 2d sess. Brooks, Harvey. 1971. Thoughts on graduate education. The Gr~e Joz~nzal 8(23:319-336. Brooks, Harvey. 1972. What's happening to the U.S. lead in technology?Harvard Business Review, May/June. Brooks, Harvey. 1973a. The physical sciences: Bellwether of science policy, in James Shannon, ea., Science and the Evolution of Public Policy. New York: The Rockefeller University Press. Brooks, Harvey. 1973b. Technology and values: New ethical issues raised by technological progress. ZYGONIJoz~ 1 of Religion and Science 8(1). Brooks, Harvey. 1978. The dynamics of funding, enrollment, cumc~um and employment, in Martin L. Pert, ea., Physics Careers, Employment, and Education. New York: American Institute of Physics. Brooks, Hanrey. 1982a. Science indicators and science priorities. Chapter 1, pp. 1-32, in Marcel C. La Follette, ea., Q~z~ in Science. Cambridge, Mass.: MIT Press. Brooks, Harvey. 1982b. Towards an efficient public technology policy: Cntena and evidence. Pp. 329-380 in Herbert Giersch, ea., Emerging Technologies: Consequences for Economic Growth, Structural Change, and Employment. Symposium 1981, Instinct fiir Weln~schaft an der Univ- ersidit Kiel. (Paul Siebeck). Tubingen: J.C.B. Mohr. Brooks, Harvey. 1983. Technology, competition and employment. Pp. 115-122 in R.J. Miller, ea.,

164 HARVEY BROOKS Robotics: Future Factories, Future Workers. Special issue of The Annals of the American Academy of Political and Social Science. November. Brooks, Harvey. 1985a. Can science and technology rescue the Altering U.S. economy? Materials and Sociery 9(1):1-12. Brooks, Harvey. 1985b. Technology as a factor in U.S. competitiveness. Chapter 9 in Bruce R. Scott and George C. Lodge, eds., U.S. Competitiveness in the World Economy. Boston, Mass.: Harvard Business School Press. Brooks, Harvey. 1985c. Technology assessment and environmental impact assessment. Pp. 105- 122 in U.S.-China Conference on Science Policy, January 9-12, 1983. Washington, D.C.: National Academy Press. Brooks, Harvey, and Roland W. Schmitt. 1985. Current Science and Technology Policy Issues: Two Perspectives. Occasional Paper No. 1, Graduate Program in Science, Technology, and Public Policy. Washington, D.C.: The George Washington University. Bush, Vannevar. 1970. Pieces of the Action. New York: William Morrow. Bush, Vannevar, et al. 1960. Science the Endless Frontier: A Report lo the President on a Program for Postwar Scientific Research. Originally issued July 1945; reissued as part of the Tenth Anniversary Observance of the National Science Foundation as NSF 60-40. Washington, D.C. Corson, Dale R., et al. 1982. Scientific Communication and lVational Security. Report of Panel on Scientific Communication and National Security. Washington, D.C.: National Academy Press. Dupree, A. Hunter. 1957. Science in the Federal Government: A History of Policy and Acnviiies to 1940. Cambndge, Mass.: Harvard University Press. Eckstein, Otto L., Christopher Caton, Roger Bnnner, and Peter Duprey. 1984. The DRI Report on U.S. Manufacturing Industries. Lexington, Mass.: Data Resources. Economist. September 15, 1984. British science policy: Concem over Cern. P. 93. Elkana, Yehuda, et al. 1978. Toward a Metric of Science. New York: John Wiley & Sons. Energy Modeling Forum. 1982. World Oil: Summary Report. EMF Report No. 6, Eric Zausner, Working Group chairman, Stanford University. Especially p. 7, Using the import premium in policymaking, and pp. 67-75, Lee value of reducing oil imports. England, J. Menon. 1983. A Patron of Pure Science. Washington, D.C.: National Science Foun- daiion. Executive Order 12044. March 23, 19?8. Improving Government Regulations. Washington, D.C.: Executive Office of tile President. Fnedman, Benjamin N. 1985. Saving, investment, and government deficits in the 1980's, in Bruce R. Scott and George C. Lodge, eds., U.S. Competitiveness in the World Economy. Boston, Mass.: Haward Business School Press. Goodson, R. Eugene. 1977. Federal Regulation of Motor Vehicles: A Sublunary and Analysis. Report to the U.S. Department of Transportation. Washington, D.C.: U.S. Department of Transportation. Grabowski, Henry G., and John M. Vernon. 1982. The pharmaceutical industry. Chapter 6, pp. 283-360, in Richard R. Nelson, ea., Government and Technical Progress: A Cross-lndusmy Analysis. New York: Pergamon Press. Gregory, Gene. 1982. Japan: New center for innovation, evolving from imitator to inventor. Speaking of Japan 3(18):~-9. Tokyo: Keizai Koho Center, Japan Institute of Social and Economic Affairs. Gregory, William H. 1982. Editorial, Landsat: Public orprivate?Aviation Week & Space Technology, April 5:11. Gnliches, Zvi. 1985. Producav`ry, R&D, and Basic Research at the Firm L - el in the 1970's. Harvard Institute of Economic Research. Cambridge, Mass.: Harvard University Press. Handler, Philip, ed. 1970. The Life Sciences. Committee on Research in the Life Sciences of the Committee on Science and Public Policy. Washington, D.C.: National Academy of Sciences. Hollomon, J. Herbert, and Alan E. Harger. 1971. America's technological dilemma. Technology Review 31 (July/August).

NATIONAL SCIENCE POLICY ACID TECHNOLOGICAL INNOVATION 165 Husen. Torsten. 1983. Are standards in U.S. schools really lagging behind those in other countries? Phi Delta Kappan Journal 64(7):455-461. Imp, K., and A. Sakuma. 1983. An analysis of Japan-U.S. semiconductor friction. Economic Eye, A Quarterly Digest of Views from Japan 4 (June): pp. 13-18. Tokyo: Keith Koho Center, Japan Institute for Social and Economic Affairs, June. Katz, James E. 1982. Planning and legislating technical services: The American experience. Tech- nology in Sociery 4:51-66 Keizai Koho Center. 1983. Japan 1983: An international Comparison. Tokyo: Japan Institute for Social and Economic Affairs. Lawrence, Robert Z. 1983. Changes in U.S. industrial structure: The role of global forces, secular trends, and transitory cycles. Paper prepared for Symposium on Industrial Change and Public Policy, organized by the Federal Reserve Banic of Kansas City, Jackson Hole, Wy. August 25- 26, 1983. Layton, Edwin T., Jr. 1971. The Revolt of the Engineers: Social Responsibility and the American Engineering Profession. Cleveland, Ohio: Case Westem Reserve University Press. Lester, Richard K. 1985. National policy options for advanced nuclear power reactor development. Pp. 4~491 in Richard K. Lester et al., eds., National Strategies for Nuclear Power Reactor Development. MITNPI-PA-002, Program on Nuclear Power Plant Innovation, Department of Nuclear Engineenng. Cambridge, Mass.: Massachusetts Institute of Technology. Lewis. Jordan D. 1975. Incentives for Technological Change, A Progress Report, Experimental Technology Incentives Program. March 26. Memorandum. Lewis, Jordan D. 1976. Director, Expenmental Technology incentives Program, National Bureau of Standards, Statement Before the Subcommittee on Domestic and International Scientific Plan- ning and Analysis, Committee on Science and Technology, U.S. House of Representatives, May 4. Mansfield, Edwin. 1985a. Public policy toward industrial innovation: An international study of R&D tax credits, in Robert H. Hayes, Kim B. Clark, and Christopher Lorenz, eds., The Uneasy Alliance: Managing Ike Producnviry-Tecknology Dilemma, Harvard Business School Press, forthcoming. Mansfield, Edwin. 1985b. Technological change and economic growth. Pp. 1-18 in U.S.-China Conference on Science Policy, January 9-12, 1983. Washington, D.C.: Naiional Academy Press. Me:Tifield, D. Bruce. 1982. Summary of the use of the R&D limited partnership: A means to enhance our international competitive position. Unpublished draft paper. Washington, D.C.: U.S. De- p~ment of Commerce. Monson, Elting. 1974. From Know-How to Nowhere: The Development of American Technology. New York:.Basic Books. Mowery, David C., and Nathan Rosenberg. 1982. The commercial aire~ft industry. Chapter 3, pp. 101-161, in Richard R. Nelson, ea., Government and Technical Progress: A Cross-Industry Analysis. New York: Pergamon Press. Musgrave, Richard A. 1914. On social goods and social bade. Pp. 251-293 in Robin Mams, ea., The Corporate Society. London: Macmillan. National Academy of Engineenng. 1984. Guidelinesfor Engineering Research Centers. Washington, D.C.: National Academy Press. National Academy of Sciences. 1969. Technology: Processes of Assessment and Choice. Report lo the Committee on Science and Astronautics, U.S. House of Representatives. Washington, D.C.: U.S. Government Printing Office. Naiional Research Council. 1976. An Evaluative Report on the E~perimen~l Technology Incentives Program. Evaluative Panel for the National Bureau of Standards, FY 1976. Washington, D.C.: National Academy of Sciences. Naiional Research Council. 1983, 1984. The Race for the New Frontier, International Competition in Advanced Technology—Decisions for America. Panel on Advanced Technology Competition. Washington, D.C.: National Academy Press; New York: Simon & Schuster. .

166 HARVEY BROOKS National Science Board. 1983. Science Indicators 1982. Washington, D.C.: National Science Foun- dation. National Science Board. 1985. Science Indicators 1984. Washington, D.C.: National Science Foun- dation. National Science Foundation. 1983a. Company and federal support produce 17% industrial R&D spending increase In 1981. NSF Science Resources Highlights. NSF 83-313, August 8. Wash- ir~gton, D.C. National Science Foundation. 1983b. Manufactunug employment becomes increasingly technolog- ical. NSF Science Resources Highlights. NSF 83-303, March 10. Washington, D.C. National Science Foundation. 1984a. Academic Research Equipment in the Physical Sciences and Engineering. Prepared for Universities and Non-Profit Institute Studies Group, Division of Science Resources Studies, National Science Foundation. Rock~ille, Md.: Westat. National Science Foundation. 1984b. Nanonal Patterns of Science and Technology Resources 1984. NSF 8~311. Washington, D.C.- National Science Foundation. 1984c. Science and Technology Data Book. NSF 84-331, October. Washington, D.C. Nellcin, Dorothy. 1971. The Politics of Housing Innovation, The Fate of the Civilian Industrial Technology Program. Ithaca: Comell University Press. Nelson, Richard R. 1977. The Moon and the Ghetto: An Essay on Public Policy Analysis. New YorI;: W.W. Norton. Nelson, Richard R. 1982. Government stimulus of technological progress: Lessons from American history. Chapter 9, pp. 451 482, in Richard R. Nelson, ea., Government and Technical Progress: A Cross-I~s~y Analysis. New York: Perga~s~on Press. Nelson, Richard R. 1984. High Technology Policies: A Five-Nation Comparison. Washington, D.C.: American Enterprise Institute for Public Policy Research. New York Times. November 21, 1984. The chemical lobby's "Tumaround." Quotation of Louis Fernandez, Monsanto Company. P. 16. Norman, Colin. 1983. How to win buildings and influence Congress. News and Comment, Science 222 (December 16):1211-1213. Nuclear News. 1984. Report proposes future nuclear strategy for France. 27(9):66 67. Nuclear News. 1985. Vibration problems at ~ix. 28(5):85-86. Office of hlanagement and Budget. 1981. FY 1982 Budget Revisions. Washington, D.C. Office of Science and Technology Policy. 1982. Aeronauncal Researc}: and Technology Policy: Volume 1, Surname Report. Washington, D.C.: U.S. Government Printing Office. Pavitt, Keith, and Luc Soete. 1979. Innovative activities and export shares: Some comparisons between industries and counuies. Pp. 38-66 in Keith Pavitt, ea., Technical Innovation and Boyish Economic Performance. London: Macmillan. Parsons, C., R. Scott, P. Crazier, and B. Guile. 1984. The Development of Programmable Auto- mation Systems in Discrete Parts Manufacturing Ind~stnes: Agricultural Machinery; Auto Parts, and Pumps and Compressors. Draft report to Be Office of Technology Assessment by the Berkeley Roundtable on Intemaional Economy (BRIE). University of California, Berkeley. Pear, Robert. 1983. States fostering high technology. New York Times, August 16:A1, A21 President's Commission on IIIdustnal Competitiveness. 1985. Global Compenizon: The New Reality. Vol. 2. John A. Yotmg, chainnan. Washington, D.C.: U.S. Government Printing Office. Public Law 92~4. 1972. Technology Assessment Act of 1972. October. Public Law 95-217. 1977. Clean Water Act of 1977. December 27. Public Law 9~515. 1980. Patent and Trademark Laws, Amendment. December 12. Public Law 98-622. 1984. The Patent Law Improvement Act. November 8. Ramo, Simon. 1985. The international race for technological supenonty. Bulletin, The American Academy of Arts and Sciences, 38(4) Stated Meeting report.

NATlON'AL SCIENCE POLICY AN'D TECHNOLOGICAL INNOVATION 167 Rose, Mark H. 1979. Interstate: Express Highway Politics, 1941-1956. Lawrence: Regents Press of Kansas. Rosenberg, Nathan. 1976. Perspectives on Technology. New York: Cambridge University Press. Schlesinger, Arthur M., Jr. 1965. A Thousand Days, John F. Kennedy in the White House. Cam- bndge, Mass.: Riverside Press. Skinner, Wickham. 1983. Wanted: Managers for the factory of the future. Pp. 102-114 in R.J. Miller, ed. Robotics: Future Factories, Future Workers. Special issue of The Annals of the American Academy of Political and Social Science. Beverly Hills, Calif.: Sage Publications Smith, Bruce L.R., and Joseph J. Karlesky. 1977. The State of Academic Science: The Universities in the National Research Effort. Vol. I. New York: Change Magazine Press. Stockman, David. 1977. The Market Case Against the Clinch River Breeder Project. Washington, D.C. Memorandum, U.S. Congress, September 17. Stnckland, Stephen P. 1972. Politics, Science, and Dread Disease. Cambridge, Mass.: Harvard University Press. Ibornassen, K. I. 1984. Progress and directions in magnetic fusion energy. Arsenal Review of Energy 9:281-319. Palo Alto, Calif.: Annual Reviews. U.S. Congress, House. 1984. Japanese Technological Advances and Possible United States Re- sponses Using Research Joint Ventures. Hearings before the Subcommittee on Investigations and Oversight and the Subcommittee on Science, Research and Technology of the Commuttee on Science and Technology. 98~ Cong., 1st seas. [No. 45], June 29-30, 1983. Washington, D.C.: U.S. Government Printing Office. U.S. Congress, Senate, Committee on Labor and Public Welfare. 1971. National Program for the Conquest of Cancer. Report of the National Panel of Consultants on the Conquest of Cancer. Senate Document 92-99. April 14. U.S. Department of Commerce. International Trade Administration. 1983. An Assessment of U.S. Cornpentiveness in High Technology Industries. Washington, D.C.: U.S. Goverrunent Printing Office. U.S. Department of Commerce, Office of the Assistant Secretary for Science and Technology. 1979. Domestic Policy Review of Ir~strial Innovation. PB-290403, PB-290404, PB-290407, PB-29~, PB-290413, P13-290415, PB-290417. Springfield, Va.: National Technical Info,~lation Service. Also, We White House. 1979. White House fact sheet: The President's industrial innovation initiatives. October 31. Reprinted, pp. 155-173 in U.S. Congress, House, Committee on Science and Technology. 1980. Analyses of President Carter's Initiatives in Industrial Innovation and Economic Revitalization. U.S.Departmentof3ustice.1980. Antitrust Guide Concerning Research Joint Ventures. Washington, D.C. Vemon, Raymond. 1982. Technology's effects on international trade: A look ahead. Pp. 145-170 in Herbert Giersch, ea., Emerging Technologies: Consequences for Economic Growth, St~uctwal Change, and Employment. Symposium 1981, Insiitut fur Weltwirtschaft an der Universitat Kiel. Tubingen: J.C.B. Mohr (Paul Siebeck). Waldrop, M. Mitchell. 1982. Imaging the earth (If): The politics of LANDSAT. Science 216 (April 2): 4~41. Wallerstein, Mitchel B. 1984. Scientific commurucation and national security in 1984. Science 224 (May 4): 4~466. .

Next: The Role of the Legal System in Technological Innovation and Economic Growth »
The Positive Sum Strategy: Harnessing Technology for Economic Growth Get This Book
×
Buy Paperback | $155.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

This volume provides a state-of-the-art review of the relationship between technology and economic growth. Many of the 42 chapters discuss the political and corporate decisions for what one author calls a "Competitiveness Policy." As contributor John A. Young states, "Technology is our strongest advantage in world competition. Yet we do not capitalize on our preeminent position, and other countries are rapidly closing the gap." This lively volume provides many fresh insights including "two unusually balanced and illuminating discussions of Japan," Science noted.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!