Research and Development

I. Industrial research and developmentJacob Schmookler

BIBLIOGRAPHY

II. Financing social researchHenry W. Riecken

BIBLIOGRAPHY

I Industrial Research And Development

While, as commonly used, “research” describes the quest for new scientific or engineering knowledge and “development” refers to the creation and the reduction to practice of new products or processes, “research and development” (R & D) is less than the sum of the two, for the phrase ordinarily denotes only research and development occurring in organizations established to perform them. The word order of the phrase may convey the impression that scientific or engineering discovery invariably precedes development, but development often leads to new discovery or occurs without it. Industrial R & D, to which this article is largely confined, is that portion of all R & D performed by or for business. It thus excludes R & D by nonindus-trial institutions for themselves or for each other but includes R & D by such institutions for industry.

Research and development are not only ancient human activities but continue outside as well as inside R & D establishments. Although the importance of R & D has undoubtedly increased greatly, both absolutely and relatively, in recent decades, it is noteworthy that of 61 major inventions in the twentieth century, about half were made by independent inventors. As recently as 1953, less than half of all inventions patented in the United States came from corporate R & D establishments. The scientists, engineers, and supervisors in the operating divisions of industry contributed about half as many inventions to their employers’ patent portfolios as did their R & D counterparts. The greater part of the reduction in production costs over time, even in the largest enterprises, typically results from hosts of minor, often unrecorded, changes in technique, introduced at the site by production engineers and others, not from the larger innovations stemming from the enterprises’ own R & D establishments. Hence, if based solely on statistics of R & D expenditures or employment, estimates of the volume of all resources allocated to advancing science and technology are too low, whereas, because of the greater relative growth of R & D, estimates so based of the rate of growth of all resources are too high.

Thus, the distinctive feature of R & D is not in the ends sought but in the formal organization of the means of attaining them, i.e., in the employment of professional scientists and engineers by lay bodies, private or public, for the specific purpose of advancing science and technology. In modern times the state—the princes and absolute monarchs of the seventeenth and eighteenth centuries—first provided the support for such employment, initially mainly as patrons of the New Learning; later, and increasingly, in the hope that discoveries and inventions of military, industrial or medical value would result. Not until the 1860s did business enterprise begin to consider that science had matured enough to make its systematic cultivation profitable at the level of the individual firm. For various reasons, chief of which, perhaps, was the superiority there of scientific and technological education, that development took place first in Germany, in the dyestuff industry, under the stimulus of the Englishman William H. Perkin’s accidental discovery of synthetic mauve. The rise of industrial R & D under the auspices of individual business enterprises followed in other countries at varying intervals, in each case roughly determined by the extent of the lag in quality and character of the requisite education.

General Electric’s laboratory, established in 1901, was the first in the United States and was followed by du Pont’s in 1902; but not until World War i cut off the United States and the Allies from German pharmaceuticals and dyestuffs were their chemical industries able to develop appreciably in the fields pioneered by the German industry. The roughly one hundred private R & D establishments on hand in the United States at the start of the war tripled, with government aid, during its course. Most of the new ones survived in the postwar years, and by contributing to the stream of product and process innovations they helped drive more and more rivals to establish R & D programs of their own. The experience of World War n and the Korean War duplicated, on a larger scale, that of World War i, except that the number of American firms engaged in R & D has apparently not increased appreciably since 1953, growth since then taking the form primarily of increased R & D per firm engaged.

The strong association of industrial R & D and defense explains why the aircraft and missiles industry alone accounted for one-third and the electrical and communications equipment industry for almost one-fourth of all R & D outlays in private firms in the United States in 1960. However, when only private R & D expenditures are considered for the same year, the chemical and allied products industry ranks first, with about one-fifth of the total, followed closely by electrical and communications equipment.

The pressure of defense requirements on industrial R & D is also reflected in the fact that the four American industries which received the largest amounts of federal funds for R & D in 1960— machinery, instruments, aircraft and missiles, and electrical and communications equipment—were the only industries that year to use at least 80 per cent of the R & D funds at their disposal for development alone. The expenditures in these industries on basic research in that year amounted to only 1 per cent of total R & D expenditures in aircraft and missiles, 3 per cent in electrical and communications equipment, and 2 per cent each in machinery and instruments. By contrast, basic research expenditures amounted to 18 per cent of total R & D expenditures in petroleum refining and extraction, 12 per cent in chemicals and allied products, 7 per cent in primary metals and food, and 4 per cent in industry as a whole. The small ratio of basic research to total R & D in the defense-linked industries is believed to result both from the characteristic complexity of the equipment produced and the arduous testing required and from the pressure for immediate results which usually attends publicly financed R & D expenditures in the United States, particularly in defense fields. However, other, more complex factors may also be involved.

Interindustry variations

The relative importance of R & D differs greatly among industries. In terms of company-financed R & D per employee among companies engaged in R & D , the leading industries in the United States in 1960 were chemicals and allied products, with outlays of $980 per employee, followed by the instrument industry, with $710, and the electrical and communications equipment industry, with $600. The average for all companies performing R & D was $390 per employee. At the bottom were textiles and apparel, with $60, and lumber, wood products, and furniture, with $90. Since industries differ greatly in the proportion of companies engaged in R & D (and in the same direction as R & D outlays per employee), even these data considerably underestimate the true interindustry differences. Similarly, among the companies engaged in R & D in 1957 and having 2,500 or more employees, company outlays on R & D per dollar of capital expenditure varied from highs of $2.83 in scientific and mechanical measuring instruments and $1.30 in drugs and medicines to lows of 37 cents in industrial chemicals and five cents in primary metals, with an average of 41 cents for all industries surveyed.

While such differences presumably reflect corresponding interindustry differences in such factors as market opportunity, technological potential, durability of equipment, and comparative advantage of professional over nonprofessional research and development, the relative weight of these underlying determinants remains to be established. Indeed, while a profit-maximizing firm would, under conditions of certainty, allocate funds between physical plant and R & D so as to equalize the return to the marginal dollar spent on each, the actual relationships are unknown because data needed for calculating the returns to R & D under typical circumstances are nonexistent. Similarly, efforts to explain interindustry or interfirm differences in growth rates in terms of correlated differences in R & D rates have not been wholly successful thus far, because the lines of causality run in both directions.

Organizational considerations

As implied above, in Western nations generally, private firms now do most of the R & D , whether for their own account or the government’s. This tendency is most pronounced in the United States, where in 1962/1963, for example, of an estimated national total of $16,400 million spent on all R & D , 70 per cent was spent in, and 30 per cent was spent by, business firms—compared with 16 per cent spent in, and 67 per cent by, the federal government. Of the R & D expenditures of business enterprise, only 3 per cent was for work done in nonprofit institutions, and most of that was on behalf of the individual spending enterprise. By contrast, in the Commonwealth countries and western Europe relatively more government R & D is done “in house,” and relatively more R & D for industry is performed in industry-operated research centers, financed by industry alone or jointly with government. Its traditionally more confined role has caused the government of the United States to eschew a major direct part in advancing the technical arts, outside of the military and health fields, and such new and expensive fields as nuclear power and space, which are felt to be politically or militarily important, or such old, economically distressed and politically powerful fields as agriculture and coal mining. The minuscule amount of cooperative research in American industry is explained in part by the fact that the patent laws create monopolies that would be illegal under the antitrust laws if created by other means.

Debate on the relative merits of alternative ways of organizing industrial R & D, as well as on such related matters as the efficacy of existing patent laws—and particularly whether the government or the private contractor should hold title to patents on government-financed business-produced inventions—has swelled with the increase in R & D expenditures and with the growing recognition of the importance of technological progress to economic growth.

Thus, some argue that a shift in the site of industrial R & D from the individual firm to an industry center might eliminate wasteful duplication, permit economies of scale in R & D and the pursuit of projects too expensive for a single enterprise, and in principle make immediately available to all what would otherwise be patented or kept secret. There is, however, some danger that under such an arrangement unusual but occasionally great projects would be neglected; that individual firms would not risk innovating, when rivals could freely imitate after most of the risks had been eliminated by the pioneering firm; and that being removed from the point of application and lacking competitive pressure, R & D would be inefficiently conducted. Only about 15 per cent of the inventions patented by the U.S. Department of Agriculture and the Tennessee Valley Authority are used by private industry (although they were developed for that purpose), while between 50 and 60 per cent of inventions patented by private firms are used commercially. Since the R & Dprograms of the two agencies resemble those of industry centers, these differences in utilization rates have obvious pertinence to the question.

Undoubtedly the growth of elaborate, technologically complex, and extremely expensive systems has increased the optimal size of some R & D projects, as in the armaments, space, power, and communications fields. Hence, some individual R & Dprojects cost billions. This fact, also, has seemed to justify greater centralization of industrial R & D to insure efficient articulation of interdependent elements of industrial technique and apparatus. However, the weight of present evidence suggests that decision making in large R & D projects may already be too centralized in the United States when radical advances are sought, because the discoveries and inventions on which such advances must be based cannot be forecast with the precision needed to preplan their interrelations.

Scale considerations

On another plane, some have urged relaxing the antitrust laws because of presumptive economies of scale in R & D. Large firms, it is argued on theoretical grounds, can afford R & D projects that small ones cannot. Large firms, having a longer time-horizon, can wait longer for returns; having better access to capital, they can more readily implement R & D findings; being more diversified, they are more likely to find useful the unexpected results of R & D; and having a more protected market position, they are able to capture more of the gains from new knowledge and so have a greater incentive to finance any given R & D project. These factors, it is argued, explain why, among American manufacturing firms in 1953, for example, only 10 per cent of those with 8 to 499 employees had or financed R & D programs, compared with 42 per cent for firms with 500 to 999 employees, 60 per cent for firms with 1,000 to 4,999 employees, and 94 per cent for firms with 5,000 or more employees; and they also help explain why in 1960 company-financed R & D for American manufacturing firms engaged in R & D amounted to 2 per cent of sales for firms with 5,000 or more employees, compared with 1.4 per cent for firms with 1,000 to 4,999 employees. The greater progressiveness of large firms, which such data and theoretical reasoning appear to imply, it is maintained, cannot be fully exploited if the individual firm is prohibited by law from achieving dominance in its market.

However, small firms appear to have certain offsetting advantages in the domain of R & D . Because the small firm has shorter lines of communication, its R & D scientists and engineers are likely to know the firm’s production problems and the product requirements of its customers more precisely than are similar personnel in a large firm. Perhaps this is one reason why small firms use commercially about three-fourths of the inventions they patent, whereas big firms use only half. The greater contact of the scientists and engineers of the small firm with the problems the firm confronts may result in less lost motion in the R & D process itself, for firms with under 5,000 employees spend half as much on R & D per patented invention as do larger firms. Moreover, research and development on the part of operating scientists, engineers, and supervisors probably occurs more often in small than in large firms, since the division of labor in the latter is characteristically narrower and more formal. Hence, variations in R & D per dollar of sales between firms of different sizes are a misleading index of the corresponding variations in the amount of new knowledge produced within the firms. Finally, the relative frequency of projects so large that only a giant enterprise can mount them does not seem high, for after a certain, not very large size, further increases in the size of the firm are not associated with any increase in R & D as a percentage of sales or with an increase in the proportion of firms engaged in R & D .

The ambiguity of existing evidence suggests that the net effect of different patterns of industrial organization and of the organization of industrial R & D upon the discovery and use of new knowledge may be small, perhaps because a strikingly favorable effect of a given pattern in one direction tends to be more or less offset by adverse effects in other directions. In general, it seems probable that an acceptable analysis of the problem must await a firmer understanding of the processes by which new knowledge is created and disseminated.

Jacob Schmookler

[See alsoInnovationandPatents. Also related areEngineeringand the articles underScience.]

BIBLIOGRAPHY

Barber, Bernard; and Hirsch, Walter (editors) 1962 The Sociology of Science. New York: Free Press.

Carter, Charles F.; and Williams, Bruce R. 1957 Industry and Technical Progress. New York: Oxford Univ. Press.

Carter, Charles F.; and Williams, Bruce R. 1958 Investment in Innovation. New York: Oxford Univ. Press.

Freeman, C 1962 Research and Development: A Comparison Between British and American Industry. National Institute Economic Review, No. 20:21–39.

Griliches, Zvi 1957 Hybrid Corn: An Exploration in the Economics of Technological Change. Econometrica 25:501–522.

Jewkes, John; Sawers, David; and Stillerman, Richard 1958 The Sources of Invention. London: Macmillan; New York: St. Martins.

Kapp, Karl W. (1950) 1964 Social Costs of Business Enterprise. 2d ed., rev. New York: Asia Pub. House.

Mansfield, Edwin 1961 Technical Change and the Rate of Imitation. Econometrica 29:741–766.

Mansfield, Edwin 1963 Intrafirm Rates of Diffusion of an Innovation. Review of Economics and Statistics 45:348–359.

Mansfield, Edwin 1964 Industrial Research and Development Expenditures: Determinants, Prospects, and Relation to Size of Firm and Inventive Output. Journal of Political Economy 72:319–340.

Mansfield, Edwin 1965 Rates of Return From Industrial Research and Development. American Economic Review 55, no. 2:310–322.

Mansfield, Edwin 1967a The Economics of Technological Change. New York: Norton.

Mansfield, Edwin 1967b Industrial Research and Technological Innovation. New York: Norton.

National Advanced-technology Management Conference, Seattle, 1962 1963 Science, Technology, and Management. Edited by Fremont Kast and James Rosenzweig. New York: McGraw-Hill.

Nelson, Richard R. 1959 The Economics of Invention: A Survey of the Literature. Journal of Business 32: 101–127.

Nelson, Richard R. 1961 Uncertainty, Learning, and the Economics of Parallel Research and Development Efforts. Review of Economics and Statistics 43:351–364.

Nelson, Richard R.; Peck, Merton J.; and Kalachek, Edward 1967 Technology, Economic Growth, and Public Policy. Washington: Brookings Institution.

Organization For European Economic Cooperation 1954 The Organisation of Applied Research in Europe, the United States and Canada. 3 vols. Paris: O.E.E.C. → Volume 1: A Comparative Study. Volume 2: Applied Research in Europe. Volume 3: Applied Research in the United States and Canada.

Patent, Trademark, and Copyright Journal of Research and Education. → Published five times a year since 1957.

Schmookler, Jacob 1959 Technological Progress and the Modern American Corporation. Pages 141–165 in Edward S. Mason (editor), The Corporation in Modern Society. Cambridge, Mass.: Harvard Univ. Press.

Schmookler, Jacob 1966 Invention and Economic Growth. Cambridge, Mass.: Harvard Univ. Press.

Schumpeter, Joseph A. (1942) 1950 Capitalism, Socialism, and Democracy. 3d ed. New York: Harper; London: Allen R & D Unwin. → A paperback edition was published by Harper in 1962.

Terleckyj, Nestor E. 1963 Research and Development: Its Growth and Composition. National Industries Conference Board Studies in Business Economics, No. 82. New York: The Board.

U.S. National Science Foundation Funds for Research and Development in Industry: Performance and Financing. → Published annually since 1957. Supersedes U.S. Bureau of Labor Statistics, Science and Engineering in American Industry, 1953/1954–1956.

U.S. National Science FoundationReviews of Data on Research and Development. → Published monthly since 1956.

U.S. Office Of Scientific Research And Development 1945 Science: The Endless Frontier. A report to the President by Vannevar Bush, director of the Office. Washington: Government Printing Office.

Universities-national Bureau Committee For Economic Research 1962 The Rate and Direction of Inventive Activity: Economic and Social Factors. National Bureau of Economic Research, Special Conference Series, No. 13. Princeton Univ. Press.

II Financing Social Research

As the social sciences have evolved from their origin in moral and political philosophy to their present state as empirically grounded observational and experimental sciences, they have not only lost some of their speculative tone but have added the costs associated with the systematic collection of data. In this respect they share some of the financial problems of the physical and biological sciences, even though the social sciences generally do not use such extensive (and expensive) equipment. Instead, the costs of research in the social sciences consist principally of payment for the skilled (and often professional) labor of interviewers, observers, experimenters, coders, data analysts, and those who construct questionnaires and develop tests of various kinds. In recent years outlays for high-speed electronic computers and the professional and technical labor associated with their use have increased rapidly— a trend that seems likely to continue to grow as it becomes possible (and desirable) to work with larger and larger bodies of data. The travel and other expenses associated with cross-cultural or multinational research are another factor in rising costs.

For these reasons, modern social research is too costly to be borne by the individual investigator, and yet, with the exception of some kinds of applied research, it does not produce results that have sufficient private commercial value to encourage business ventures. Hence, like the physical and biological sciences, the social sciences depend upon grants and gifts from private and public sources or on contracts for research services, usually with governmental agencies. The nature of the source of support and the similarity or difference in interests and motives on the part of the grantor and the researcher make for some interesting and difficult problems. These include the possibility of conflict between the “fundamental research” interests of the social researcher and the need which the supporting organization has for “practical” results; the possibility that the sponsor will directly or indirectly try to influence the outcome of the research or the conclusions drawn from it; and the possibility that the sponsor’s interests may make it risky or impossible even to inquire into certain topics. Less sinister but just as important is the fact that the grantor’s interests are likely to be directed toward immediate and very contemporaneous social problems, so that a researcher with a particular topical interest may be overwhelmed with funds for a while and, later, when interest has shifted to another question, find himself without a patron in sight. Finally, there seem to be some problems that are chronically low on a society’s priority list (such as the psychology of aesthetic experience) and others that usually have a high place (the analysis of economic fluctuations, for example). None of these problems is specific to the social sciences, for an instance of each can be found in one or more of the physical and biological sciences. Some are, perhaps, exacerbated by the close connection between social research and human welfare, economic interests, political power, and ethical, moral, and judicial issues. Since the patterns of support in the United States are highly varied, and since they are relatively well documented, discussion in this article will be confined to the United States.

The role of universities

In the United States scientific research of all kinds has traditionally been carried on largely in and through universities and operating government bureaus, rather than through independent institutes or academies of science. Furthermore, until recently at least, the largest amount of scientific research was carried on in the universities, although an increasing share of the cost of research has been borne by the government.

Accordingly, among performers of research the most important institution is the college or university. Private universities in the United States are supported chiefly through endowments, fees from students, and rents or royalties, supplemented by gifts (from nongovernment donors) and by grants or contracts for research or services performed. Public universities obtain their funds chiefly from the legislative bodies of the individual states or cities in which they are located, although they may also have small endowments, rents, royalties, and the like, and they usually receive substantial amounts of grant or contract funds for research and services. In both private and public universities, funds for grants or contracts for social science research may come from the federal government, from industrial corporations, from private foundations, labor unions, and other nonprofit organizations, and from private individuals.

There is considerable variety in the specific arrangements for performing research in universities. Often an individual member of a faculty will, with the help of a graduate-student assistant and a clerical worker, carry out the major share of the collection and analysis of data and the preparation of a technical article on a topic in which he has a special personal interest and which he himself has chosen and defined. At the other extreme is the large, teamwork project, in which several senior investigators and many juniors, assistants, and clerks take part. Often such projects are initiated and contracted by organizations outside the university, and the interest of the faculty member is likely to be tangential, partial, or secondary. Between these extremes one can find a wide variety of intermediate arrangements.

We can distinguish, then, two dimensions: the size of the research project (in money, time, and personnel) and the extent to which the problem originates spontaneously from the scholarly or professional interests of the senior investigator(s). These two dimensions are not precisely parallel, but they are far from orthogonal.

To a considerable extent the sources of funds are correlated with these two dimensions of research. The lone researcher with a single assistant is more often supported through the university’s own research funds, when such are available; the large team working on some problem presented by an outside agency is likely to be supported through a contract with a firm or a government agency. The chief reasons for this situation are historically rooted in both the development of methods of research in social science and the emergence (especially after 1945) of new forms of support. Beginning in the classical tradition of individual scholarship, which placed reliance on documentary evidence and direct but somewhat unsystematic observation, social science research in the universities was originally the act of individuals, working slowly, independently, and on a fairly small scale. Gradually, the introduction of new techniques, especially sample surveys and experiments, together with machine processing of data and statistical techniques, enlarged the scholar’s need for assistance. The private foundation’s grant or fellowship for travel, study, and field work served the individual scholar well during the period when methods were developing, but the costliness of the new techniques taxed these resources and virtually overwhelmed the universities’ own resources. Social science research was severely limited, chiefly by lack of funds, through the 1940s and 1950s. There are still financially imposed limits, and the social sciences do not yet enjoy the extent of support in the United States that the physical and biological sciences do. After World War n, however, the development of government financing for social science research greatly changed the situation and permitted social scientists to take on larger projects and more extensive and complex studies than had previously been possible for any except a few.

The increased interest of government bureaus in social science research is probably attributable to three things: the demonstration, during the war, of the practical contribution of social science research to selection, training, and other manpower problems; the growth of agencies in the government (from 1935 on) that were concerned with administering social and economic legislation, with the lagging realization that such administrative efforts could not succeed without research on the problems that the agency had to handle; and the developing awareness that basic research on society and human behavior should be supported both for its own sake and in order to develop the basis for future applied work. Since the pattern of supporting research through university facilities had already been laid down in the natural sciences; since the trained personnel were available in universities (and it could be argued that sponsored research there provided opportunities for training graduate students); and finally, since there was no existing tradition or machinery for the establishment of independent research institutes or academies, the pattern of federal support through the universities grew up.

The detailed arrangements for financing government-supported projects (which form a substantial portion of all sponsored research) at universities can be exceedingly complicated, and this article will attempt to outline only the major aspects of the most common cases. As a rule, the senior investigator on a project is a regular member of the university faculty, and normally part of his salary is paid out of the university’s own funds. The junior professional members of the project group are usually paid exclusively out of the grant funds, as are the clerical and other assistants. Although this pattern is in flux, it is generally true that research staffs have enlarged more rapidly than regular faculty positions, with the result that there are sometimes, in effect, two groups of equally competent social scientists at a university: one, the regular faculty, tied to the university through teaching and through tenure appointments; the other attached more uncertainly, through the research grant, with few or no teaching opportunities and with an appointment whose duration depends solely on the duration of the grant. Indeed, this development of two groups whose link with the university is different has produced many stresses and is one of the more serious problems arising from short-term financing of large-scale research.

Agencies of the federal government, operating on annual appropriations from the Congress, have tended to make short-term commitments to university researchers. Some agencies fund research on an annual basis, while others commit funds from the annual appropriations to support selected projects for as much as seven years in the future. Research funded on an annual basis suffers from uncertainty, the need to solicit additional funds at frequent intervals, and the difficulty of providing stable support for the members of the project staff.

On the other hand, the presence of outside research funds has often enabled the university to enlarge and enrich the opportunities for instruction and practical training of graduate students, as well as to provide technical consultation and assistance to the regular faculty.

One further aspect of the financial arrangements between the universities and the sponsors of research remains to be commented upon, namely, the support of indirect costs which accompany the establishment of research projects. As long as social science research was the work of individual scholars who used only the space, light, heat, books, journals, and other facilities that the university was accustomed to provide for its staff, there seems to have been no question raised about “overhead” costs. But with the beginning of federal government contracts for research, which often involved expanding space, facilities, staff, and ancillary services, the situation changed. Universities began to notice that sponsored research, while offering increased research opportunities for their faculties and students, also resulted in increases in operating costs for the university. Accordingly, a system of payment for indirect costs was established and has been perpetuated. Private foundation grants do not always include an overhead cost for the university, while federal government grants and contracts always do. This situation arises partly out of the fact that research in the United States has not, on the whole, been conducted through separately established institutes, where the capital costs of research facilities would be reflected in direct outlays for buildings, equipment, supporting services, and maintenance.

The role of the federal government

In terms of funds spent, the federal government plays a distinctly secondary role in the performance of social science research. While the several departments of government maintain laboratories and other research units, the research efforts in these units are smaller in size than those of their counterparts in the universities. Establishments such as the Socio-Environmental Laboratory and the Adult Psychiatry Branch of the National Institute of Mental Health, the Wright Air Development Center of the Air Force, the research sections of the Bureau of the Census, the Personnel Research Branch of the Army, and the Agricultural Research Service and Agricultural Marketing Service of the Department of Agriculture are all budgeted in the departmental appropriation made annually by the Congress, are staffed by civil servants, and in general direct their research toward the practical problems which they encounter in performing their assigned missions. A unique case is the Smithsonian Institution, which is both public and private. Established by private initiative and receiving support, in part, from private funds, it nevertheless has a charter from the Congress and enjoys some measure of federal government subsidy, which is appropriated annually. The Smithsonian’s chief contributions to social science have been in anthropology and archeology.

Most of the research performers within the federal government devote themselves to applied social science rather than to basic research, which receives much less support. This situation arises, no doubt, from the fundamentally practical or administrative orientation of most government bureaus and reflects insufficient recognition of the present state of social science, where a great deal of basic work must be done before the development of practical applications from a scientific base is possible. Basic research usually demands smaller amounts of money than applied work does but requires consistent funding over relatively long periods of time, during which no “useful” results may emerge. It is difficult to resist the demand for “practical” information which administrators and legislators can use. One of the habitual and trying cycles in social science is that which begins with administrative optimism; is followed by an increase in support, together with an increase in demand for immediate results; and then fades into a period of administrative disillusionment, terminated by the cutting off of funds. Instability of support hampers immediate accomplishment and discourages research workers from undertaking substantial, long-range projects that might yield greater results. These considerations, it should be mentioned, have pertinence to government, as both a performer and a source of research funds, but have also afflicted other sectors, such as private foundations, on occasion.

The role of other performers

Universities and federal government research units account for nearly four-fifths of the funds spent for social science research in the United States. Of the remaining one-fifth, only a small amount is spent by the private foundations, which play a very minor role as performers; there are very few social scientists who are active in research on foundation staffs. The Russell Sage Foundation is an exceptional case, and its operations are perhaps typical of the few performers in this area: a small staff in social psychology devotes some share of its time to the pursuit of research that its members organize and that is related both to the scientific interests of the staff and to the objectives of the foundation.

A larger share of funds is disposed of by a variety of nonprofit research institutions such as the Stanford Research Institute (SRI), the National Opinion Research Center at the University of Chicago (NORC), the RAND Corporation, the Systems Development Corporation, the Bureau of Social Science Research, the National Bureau of Economic Research, the Institute for Social Research at the University of Michigan, and the Bureau of Applied Social Research at Columbia University (BASR). These nonprofit, nongovernmental enterprises are a postwar phenomenon in the United States. Some have university connections, but as a rule they are without endowment and without students and tuition fees. These institutions have developed and flourished simply by engaging to conduct both basic and applied research on questions that are important to somebody, usually somebody outside the staff of the institute or center. They have received support from, and conducted research for, industrial corporations, labor unions, agencies of the federal government, private foundations, individuals, and colleges and universities. The pattern of financial arrangements, if it can be called a pattern, is extremely variable, ranging from the “one-client” organizations, such as the Operations Research Office, which depends almost entirely upon federal government funds, to the more variegated institutions (NORC, SRI, BASR), which accept commissions from all of the categories of “clients” enumerated above. The chief financial-administrative problem of such units is to maintain continuity of professional personnel and the scientific integrity of their work, in the absence of any continuous support from a source whose interest in social science research is not restricted to a single problem. In other words, the task of the director of such an institute or corporation is to provide continuing funds for the operation of the whole enterprise: he must not drive his research staff to distraction by frequent and arbitrary shifts in their assignments, yet he must avoid periods of low income during which valuable staff members must be let go because their salaries cannot be paid. The glory of a director of a private nonprofit institution stems from his skill in maintaining a substantial scientific productivity by the agency while at the same time providing the “client” with useful answers to his (not always scientific) questions.

There are a smaller number of independent organizations which operate on a profit basis. Except for this financial difference, they share many of the characteristics and problems of the independent nonprofit institution, although they do not have even a tenuous university affiliation. Such organizations are common in the physical sciences, where the pattern was established, and there are a growing number in psychology, which receive funds from both government and industry sources.

Social Science Research Council

The Social Science Research Council is not itself a performer of research but acts as a source of stimulation for needed research, better training for research, and support for research. A private organization of social scientists, it has developed close relations both with private foundations and with government agencies that share its interests. The governing board of the council consists of social scientists, as does its professional staff. Its committees, which are of two kinds, are composed of social scientists, largely university faculty members who serve without remuneration. Some of these committees administer programs of research-training fellowships and grants for research; others survey the status of research in selected fields, develop plans for needed research, organize conferences, and supervise a variety of programs, for which the council receives grants from private and public sources. Its continuity is assured by the fact that it has a basic endowment from private sources, although the bulk of the work accomplished by council committees and grantees is funded by short-term grants received from various sources. The governance of the council is effectively in the hands of professional social scientists.

Sources of funds

The matter of where the funds for social science research come from can be dealt with more briefly than can the financial complexities of performers’ activities. There are two major sources of support: the federal government (chiefly the Department of Health, Education, and Welfare, the Department of Defense, the Department of Agriculture, and the National Science Foundation) and private foundations (chiefly Carnegie, Ford, Rockefeller, and Russell Sage). The principal flow of funds is from these two sources to the universities, with considerably smaller amounts going to nonprofit institutions. Private industry also supplies funds for social science research, although data on expenditures are not easy to obtain. Private industry is probably more important as a source of funds than as a performer of research, although it is difficult to estimate the amounts spent by industrial performers because of the problem of deciding which activities should be called social science research (as distinguished from routine data collection for the purpose of making, say, a marketing decision).

Further problems

Besides the problems already mentioned—the growth of “secondary” research staffs at universities and the lack of continuity and stability in independent research organizations—the foregoing description of the institutional arrangements for financing social science research may have raised other questions, especially in the minds of those who are not entirely familiar with the operations of the system. For example, it is obvious that the federal government, as a principal source of revenue, could exert a large, perhaps an undue, influence on social science research. Under such conditions, are social scientists able to work on important problems or are they frequently distracted and diverted either to trivial questions or away from significant but controversial ones?

It is important to realize that although university-based social scientists are heavily dependent upon government sources for support of their research, it is equally true that government agencies are heavily dependent upon university-based social scientists for advice and for the professional and technical skills needed to carry out the research which government agencies want. Virtually every government agency that makes grants or contracts for social research uses advisory committees made up of faculty members of university social science departments. Many of the staff members in agencies making grants and contracts are themselves former university professors, whose understanding of the needs, requirements, and values of the social science community is therefore fairly good. To a considerable extent, then, policies governing federal government support of social science research are shaped by the values, and indeed the interests, of those who are most affected by these policies. It is true, of course, that the particular mission of the government agency will affect the sort of social science research it supports and encourages. Some agencies, however, devote all or a substantial part of their funds to the support of basic research in the social sciences, and they presumably are governed almost entirely in their project-granting decisions by the interests and capabilities of available social scientists who apply to them for assistance. At the opposite extreme, there are a very small number of research projects that are virtually ordered by agency representatives, who conceive them, initiate the negotiations for their accomplishment, and specify how they are to be done. Such highly specific task requests play a relatively small role in shaping the given field of inquiry, both because there are comparatively few such commissions and because most university and independent research agencies feel cramped by such specific requests and avoid accepting them whenever possible. Extreme financial need is usually the reason given by the research director of an independent institute which accepts such a commission from a federal agency or, for that matter, from industry or business. Somewhat more influence derives from the extensive support that federal agencies have provided to certain fields: for example, agricultural economics and clinical psychology. Both these fields have been the objects of truly massive support for a number of years—agricultural economics since the 1920s and clinical psychology since the 1940s. The availability of funds in these two fields has undoubtedly influenced the career choices of graduate students who had no clear commitment to a particular subfield. Perhaps the strongest influence wielded by federal government agencies has been a negative one—that is, the government has exerted influence either by deciding not to support a field or by simply failing to take responsibility for the nurturance of a particular field. In this respect, both history and political science have probably experienced greater neglect from federal government agencies than any other of the social sciences. Psychology has been relatively well supported; anthropology and sociology somewhat less well. Perhaps because of a conservatism inherent in the situation, newly developed fields of inquiry, such as the administrative sciences, have often had great difficulty becoming established and securing support through federal agencies. To some extent the same point can be made with regard to particular topics or problems within a given discipline—topics that have been undersupported or wholly ignored. Until relatively recently at least, research on the social and behavioral aspects of birth control has been regarded as taboo or at least very touchy and dangerous for a federal agency. Similarly, inquiry into problems associated with racial segregation or the social changes associated with integration has had relatively little support from federal agencies. It seems probable that research on contemporary national politics has a dim future among federal agencies. This last example illustrates the point that it is not merely bureaucratic prudence that keeps agencies from encouraging research on certain controversial topics. Rather, it is quite evident that it would be difficult to achieve detachment and objectivity in judging research proposals when the research directly or indirectly affected the institution proffering the support. For that reason it may be more appropriate for private foundations to undertake the support of research on national politics, federal government operations, the evaluation of foreign policy, and other questions on which individual federal agencies are not likely to be able to maintain objectivity and detachment.

The implication should not be drawn from the preceding paragraph that government agencies are inherently more conservative in the social science research which they finance than are private foundations. In terms of intellectual and scientific innovation, both federal agencies and private foundations have been sometimes conservative and sometimes innovative. The chief difference arises, however, in the duration of support for a “new idea.” It is characteristic of private foundation funding that after having given initial impetus to some plan, the foundation will withdraw its support and turn its attention to other ventures, whereas the federal agency is typically more likely to continue extended and dependable support of a given venture for which it has taken initial responsibility. For this reason, there may be in federal operations a certain degree of inflexibility, which grows as the number and variety of programs to which the agency is committed increases. As it is difficult to bring any given activity to an end, it becomes increasingly difficult to undertake new activities unless there is a constant increase in the funding ability of the agency. This disease, however, does afflict some private agencies and should not be thought of as something unique to the public sector. Of course, although it means that government-supported social science research is somewhat less flexible in its terms of preference, it may also mean that it is more dependable and continuous.

Finally, it should be clear that the growing support for social science research at universities has had certain consequences for the style of life and pattern of organization at these institutions. Increased support has made it possible to undertake larger-scale projects, often in remote and inaccessible parts of the world, and often concerned with a much larger volume of data than social scientists of fifty years ago would have believed possible to handle. This growth in size and in sophistication has provided better opportunities for training future social scientists during their graduate school years, and it has produced new roles of an administrative, managerial, and entrepreneurial sort for faculty members. Although the day of the lone scholar working in a library or with direct personal observations has far from vanished, nevertheless, the spread of larger research teams has demanded an increase in scholars with managerial abilities. The head of a large research team must be skilled at defining the problem to be studied, at recruiting and employing personnel, and at managing the logistics of the operation. Above all, under the project system of grant and contract making, he must have the entrepreneurial skill necessary to secure, often from more than one source, the required support and to maintain the flow of funds, so that the continuity of the team effort will be upheld.

Furthermore, the existence of team efforts of this sort has increased the pressure on university administrations to establish institutes, interdepartmental committees, and other interstitial organizational units within the traditional academic structure. Such a demand for new organizations does not reflect, of course, simply the presence of additional funding but rather indicates the existence of new strains, new lines of interest, and new intellectual developments which may seem to justify novel organizational arrangements. The role of outside funds in this kind of development is simply to make possible what might have been difficult or impossible within a traditional departmental structure. To repeat, such institutional innovations are not new (the Institute of Human Relations at Yale University was established in the 1920s), but the trend has been accelerated by increased funding in recent years.

Henry W. Riecken

[See alsoFoundations; Universities.]

BIBLIOGRAPHY

Alpert, Harry 1959 The Growth of Social Research in the United States. Pages 73–86 in Daniel Lerner (editor), The Human Meaning of the Social Sciences. New York: Meridian.

U.S. Congress, House, Committee On Government Operations 1967 Use of Social Research in Federal Domestic Programs: Staff Study for the Research and Technical Programs Subcommittee. 4 parts. 90th Congress, 1st Session. Washington: Government Printing Office.

U.S. National Science FoundationFederal Funds for Research, Development, and Other Scientific Activities. → Published since 1950/1951. Volumes 1 to 12 were published as Federal Funds for Science.

Research and Development

Research and development (R&D) is a process intended to create new or improved technology that can provide a competitive advantage at the business, industry, or national level. While the rewards can be very high, the process of technological innovation (of which R&D is the first phase) is complex and risky. The majority of R&D projects fail to provide the expected financial results, and the successful projects (25 to 50 percent) must also pay for the projects that are unsuccessful or terminated early by management. In addition, the originator of R&D cannot appropriate all the benefits of its innovations and must share them with customers, the public, and even competitors. For these reasons, a company's R&D efforts must be carefully organized, controlled, evaluated, and managed.

OBJECTIVES AND TYPES OF R&D

The objective of academic and institutional R&D is to obtain new knowledge, which may or may not be applied to practical uses. In contrast, the objective of industrial R&D is to obtain new knowledge, applicable to the company's business needs, that eventually will result in new or improved products, processes, systems, or services that can increase the company's sales and profits.

The National Science Foundation (NSF) defines three types of R&D: basic research, applied research, and development. Basic research has as its objectives a fuller knowledge or understanding of the subject under study, rather than a practical application thereof. As applied to the industrial sector, basic research is defined as research that advances scientific knowledge but does not have specific commercial objectives, although such investigation may be in the fields of present or potential interest to the company.

Applied research is directed towards gaining knowledge or understanding necessary for determining the means by which a recognized and specific need may be met. In industry, applied research includes investigations directed to the discovery of new knowledge having specific commercial objectives with respect to products, processes, or services. Development is the systematic utilization of the knowledge or understanding gained from research toward the production of useful materials, devices, systems, or methods, including design and development of prototypes and processes.

At this point, it is important to differentiate development from engineering. Engineering is the application of state-of-the-art knowledge to the design and production of marketable goods. Research creates knowledge, and development designs and builds prototypes and proves their feasibility. Engineering converts these prototypes into products that can be offered to the marketplace or into processes that can be used to produce commercial products and services.

R&D AND TECHNOLOGY ACQUISITION

In many cases, technology required for industrial purposes is available in the marketplace—for a price. Before embarking on the lengthy and risky process of performing its own R&D, a company can perform a "make or buy" analysis and decide whether or not the new R&D project is justified. Factors that influence the decision include the ability to protect the innovation, its timing, risk, and cost.

Proprietary Character

If a technology can be safeguarded as proprietary—and protected by patents, trade secrets, nondisclosure agreements, etc.—the technology becomes exclusive property of the company and its value is much higher. In fact, a valid patent grants a company a temporary monopoly for 17 years to use the technology as it sees fit, usually to maximize sales and profits. In this case, a high-level of R&D effort is justified for a relatively long period (up to 10 years) with an acceptable risk of failure.

On the contrary, if the technology cannot be protected, as is the case with certain software programs, expensive in-house R&D is not justified since the software may be copied by a competitor or "stolen" by a disloyal employee. In this case, the secret of commercial success is staying ahead of competition by developing continuously improved software packages, supported by a strong marketing effort.

Timing

If the market growth rate is slow or moderate, in-house or contracted R&D may be the best means to obtain the technology. On the other hand, if the market is growing very fast and competitors are rushing in, the "window of opportunity" may close before the technology has been developed by the new entrant. In this case, it is better to acquire the technology and related know-how, in order to enter the market before it is too late.

Risk

Inherently, technology development is always riskier than technology acquisition because the technical success of R&D cannot be guaranteed. There is always the risk that the planned performance specifications will not be met, that the time to project completion will be stretched out, and that the R&D and manufacturing costs will be higher than forecasted. On the other hand, acquiring technology entails a much lower risk, since the product, process, or service, can be seen and tested before the contract is signed.

Regardless of whether the technology is acquired or developed, there is always the risk that it will soon become obsolete and be displaced by a superior technology. This risk cannot be entirely removed, but it can be considerably reduced by careful technology forecasting and planning. If market growth is slow, and no winner has emerged among the various competing technologies, it may be wiser to monitor these technologies through "technology gatekeepers" and be ready to jump in as the winner emerges.

Cost

For a successful product line with relatively long life, acquisition of technology is more costly, but less risky, than technology development. Normally, royalties are paid in the form of a relatively low initial payment as "earnest money," and as periodic payments tied to sales. These payments continue throughout the period of validity of the license agreement. Since these royalties may amount to 2 to 5 percent of sales, this creates an undue burden of continuing higher cost to the licensee, everything else being equal.

On the other hand, R&D requires a high front-end investment and therefore a longer period of negative cash flow. There are also intangible costs involved in acquiring technology—the license agreements may have restrictive geographic or application clauses, and other businesses may have access to the same technology and compete with lower prices or stronger marketing. Finally, the licensee is dependent upon the licensor for technological advances, or even for keeping up to date, and this may be dangerous.

MOVING AHEAD WITH R&D

R&D can be conducted in-house, under contract, or jointly with others. In-house R&D commands a strategic advantage: the company is the sole owner of the know-how created and can protect it from unauthorized use. R&D is also basically a learning process; in-house research thus trains the company's own research people who may go on to ever better things.

External R&D is usually contracted out to specialized nonprofit research institutions or to universities. These institutions often already have experienced personnel in the disciplines to be applied and are well-equipped. The disadvantages are that the company will not benefit from the learning experience and may become overly dependent on the contractor. The trans for the technology may turn out to be difficult and leaks to competitors may develop. Using university research is sometimes slightly less expensive than engaging institutes because graduate students rather than professionals do some of the work.

Joint R&D became popular in the United States after antitrust laws were relaxed and tax incentives were offered to R&D consortia. In a consortium, several companies with congruent interests join together to perform R&D, either in a separate organization or in a university. The advantages are lower costs, since each company does not have to invest in similar equipment; a critical mass of researchers; and interchange of information among the sponsors. The disadvantages are that all the sponsors have access to the same R&D results. However, because of antitrust considerations, the R&D performed must be "precompetitive," legalese meaning that it must be basic and/or preliminary. A company must take joint research beyond the "joint" stage to make money on it; it can use this type of result as the foundation, not as the innovation itself.

R&D PROJECT SELECTION, MANAGEMENT, AND TERMINATION

Industrial R&D is generally performed according to projects (i.e., separate work activities) with specific technical and business goals, assigned personnel, and time and money budgets. These projects can either originate "top down" (for instance, from a management decision to develop a new product) or "bottom up" (from an idea originated by an individual researcher). The size of a project may vary from a part-time effort of one researcher for a few months with a budget of thousands of dollars, to major five- or ten-year projects with large, multidisciplinary teams of researchers and budgets of millions of dollars. Therefore, project selection and evaluation is one of the more critical and difficult subjects of R&D management. Of equal importance, although less emphasized in practice, is the subject of project termination, particularly in the case of unsuccessful or marginal projects.

Selection of R&D Projects

Normally, a company or a laboratory will have requests for a higher number of projects than can be effectively implemented. Therefore, R&D managers are faced with the problem of allocating scarce resources of personnel, equipment, laboratory space, and funds to a broad spectrum of competing projects. Since the decision to start on an R&D project is both a technical and a business decision, R&D managers should select projects on the basis of the following objectives, in order of importance:

1. Maximize the long-term return on investment;
2. Make optimum use of the available human and physical resources;
3. Maintain a balanced R&D portfolio and control risk;
4. Foster a favorable climate for creativity and innovation.

Project selection is usually done once a year, by listing all ongoing projects and the proposals for new projects, evaluating and comparing all these projects according to quantitative and qualitative criteria, and prioritizing the projects in "totem pole" order. The funds requested by all the projects are compared with the laboratory budget for the following year and the project list is cut off at the budgeted amount. Projects above the line are funded, those below the line delayed to the following year or tabled indefinitely. Some experienced R&D managers do not allocate all the budgeted funds, but keep a small percentage on reserve to take care of new projects that may be proposed during the year, after the laboratory official budget has been approved.

Evaluation of R&D Projects

Since R&D projects are subject to the risk of failure, the expected value of a project can be evaluated according to a statistical formula. The value is the payoff anticipated—but discounted by probabilities. These are the probability of technical success, the probability of commercial success, and the probability of financial success. Assuming a payoff of $100 million and a fifty-fifty rate of technical success, a commercial success rate of 90 percent, and a financial probability of 80 percent, then the expected value will be $36 million—100 discounted by 50, 90, and 80 percent respectively.

Consequently, project evaluation must be performed along two separate dimensions: technical evaluation, to establish the probability of technical success; and business evaluation, to establish the payoff and the probabilities of commercial and financial success. Once the expected value of a project has been determined it can be compared with the projected cost of the technical effort. Given a company's usual rate of return on investment, the cost may not be worth the expected value given the risks.

Needless to say, such statistical approaches to evaluation are not silver bullets but as good as the guesses that go into the formula. Businesses use such evaluations, however, when many projects compete for money and some kind of disciplined approach is needed to make choices.

Management of R&D Projects

The management of R&D projects follows basically the principles and methods of project management. There is, however, one significant caveat in relation to normal engineering projects: R&D projects are risky, and it is difficult to develop an accurate budget, in terms of technical milestones, costs, and time to completion of the various tasks. Therefore, R&D budgets should be considered initially as tentative, and should be gradually refined as more information becomes available as a result of preliminary work and the learning process. Historically, many R&D projects have exceeded, sometimes with disastrous consequences, the forecasted and budgeted times to completion and funds to be expended. In the case of R&D, measuring technical progress and completion of milestones is generally more important than measuring expenditures over time.

Termination of R&D Projects

Termination of projects is a difficult subject because of the political repercussions on the laboratory. Theoretically, a project should be discontinued for one of the following three reasons:

1. There is a change in the environment—for instance, new government regulations, new competitive offerings, or price declines—that make the new product less attractive to the company;
2. Unforeseen technical obstacles are encountered and the laboratory does not have the resources to overcome them; or
3. The project falls hopelessly behind schedule and corrective actions are not forthcoming.

Due to organizational inertia, and the fear of antagonizing senior researchers or executives with pet projects, there is often the tendency to let a project continue, hoping for a miraculous breakthrough that seldom happens.

In theory, an optimal number of projects should be initiated and this number should be gradually reduced over time to make room for more deserving projects. Also, the monthly cost of a project is much lower in the early stages than in the later stages, when more personnel and equipment have been committed. Thus, from a financial risk management viewpoint, it is better to waste money on several promising young projects than on a few maturing "dogs" with low payoff and high expense. In practice, in many laboratories it is difficult to start a new project because all the resources have already been committed and just as difficult to terminate a project, for the reasons given above. Thus, an able and astute R&D manager should continuously evaluate his/her project portfolio in relation to changes in company strategy, should continuously and objectively monitor the progress of each R&D project, and should not hesitate to terminate projects that have lost their value to the company in terms of payoff and probability of success.

TAX ADVANTAGES FOR R&D

In the period 1981 through 2004, corporations had a R&D tax credit—had the ability to deduct research and development expenditures from income. The tax credit was renewed in 2004 and lasted through 2005, but the tax bill signed in May of 2006 left the provision out. This outcome no doubt pleased those who thought that government subsidies of corporate development were out of place—and energized those who saw the credit as nationally important to attempt to have the credit reinstated.

SMALL BUSINESS AND R&D

Research and development in public the public domain as well as in the media suggests big business, huge labs, vast testing fields, wind tunnels, and crash dummies flailing around as autos are crashed into walls. R&D is associated with the pharmaceutical industry, miracle cures, laser eye surgery, and super fast jet travel. To be sure, a vast amount of the money expended on formal research is expended by large corporations—often on relatively trivial improvements of products already doing quite a good job—and by government on weapons systems and space exploration. The glory and the power thus displayed before our eyes on television fail to remind us that the crucial research and development on which much else is based has been—and continues to be—the work of small entrepreneurs.

The explosive development of the oil industry was triggered by the invention of an effective kerosene lamp by Michael Dietz in 1859. Dietz ran a small lamp production business. Oil drilling began in earnest to support such lighting applications. An unwanted residue of kerosene refining was—gasoline, burned off as useless waste—until the first cars came along. The story of Thomas Edison is worth rereading occasionally to correct ones vision of modern R&D. Chester Carlson, the inventory of xerography, perfected his invention in part-time labors in a makeshift lab while working as a patent attorney. The computer revolution came about because two young men, Steve Wozniak and Steve Jobs, put together a personal computer in a garage and thus triggered the Information Age. Countless innovations large and small were made by tinkering individuals or small business people trying something new. The fact that many of these entrepreneurial, inventive, innovative, and persistent individuals are the fathers and mothers of great companies—indeed of whole industries—that now dominate formal R&D should not obscure their humble beginnings and catch-as-catch-can methods of discovering the new.

BIBLIOGRAPHY

Bock, Peter. Getting it Right: R&D Methods for Science and Engineering. Academic Press, 2001.

Dankbaar, Ben. Innovation Management in the Knowledge Economy. Imperial College Press, 2003.

Khurana, Anil. "Strategies for Global R&D: A study of 31 companies reveals different models and approaches to the conduct of low-cost R&D around the world." Research-Technology Management. March-April 2006.

Le Corre, Armelle, and Gerald Mischke. Innovation Game: A New Approach to Innovation Management and R&D. Springer, 2005.

Miller, William L. "Innovation Rules!" Research-Technology Management. March-April 2006.

                                Hillstrom, Northern Lights

                               updated by Magee, ECDI

INDUSTRIAL RESEARCH

INDUSTRIAL RESEARCH. The emergence and growth of industrial research and development during the twentieth century must rank as one of the most important economic developments in modern American history. There is no doubt that technological innovation is the primary driver of economic growth, and that it is the business firm that is at the core of the American system of technological innovation. Industrial research conducted by and substantially funded by business firms has thus played a key role in American prosperity. It was also key to the outcomes in both world wars, and arguably to the ending of the Cold War. What then, is the genius behind this system? How did it emerge, how does it work, and how did it change in the twentieth century?

Industrial research and development (R&D) is the activity in which scientific and engineering knowledge is used to create and bring to market new products, processes, and services. R&D encompasses several different activities that can occur in any order. There is basic research, which is aimed purely at the creation of new knowledge. Its purpose is to create new understandings of phenomena. Its core foundations are usually quite abstract. There is applied research, which is work expected to have a practical, but not a commercial, payoff. While basic research is aimed at new knowledge for its own sake, applied research has practicality and utility as its goal. There is also development, in which the product is honed for commercial application. Boundaries among these activities are quite fuzzy, and the manner in which they have been organized and linked has changed over time.

The roots of American industrial research can be found in the late nineteenth century when a discernible amount of science and technology began being applied to industry. This is the period when the science-based industries in dyestuffs, chemicals, electricity, and telecommunications began to emerge.

The first organized research laboratory in the United States was established by the inventor Thomas Edison in 1876. In 1886, an applied scientist by the name of Arthur D. Little started his firm which became a major technical services/consulting firm to other enterprises. Eastman Kodak (1893), B. F. Goodrich (1895), General Electric (1900), Dow (1900), DuPont (1902), Goodyear (1909), and American Telephone and Telegraph (AT&T; 1907) followed soon thereafter.

Growth of the Organized R&D Laboratory (1890–1945)

The industrial laboratory constituted a significant departure from an earlier period when innovation was largely the work of independent inventors like Eli Whitney (the cotton gin), Samuel Morse (telegraph), Charles Goodyear (vulcanization of rubber), and Cyrus McCormick (the reaper).

The founding of formal R&D programs and laboratories stemmed in part from competitive threats. For instance, AT&T at first followed the telegraph industry's practice of relying on the market for technological innovation. However, the expiration of the major Bell patents and the growth of large numbers of independent telephone companies helped stimulate AT&T to organize Bell Labs. Competition likewise drove George Eastman to establish laboratories at Kodak Park in Rochester, New York, to counteract efforts by German dyestuff and chemical firms to enter into the manufacture of fine chemicals, including photographic chemicals and film.

During the early years of the twentieth century, the number of research labs grew dramatically. By World War I there were perhaps as many as one hundred industrial research laboratories in the United States. The number tripled during the war, and industrial R&D even maintained its momentum during the Great Depression. The number of scientists and research engineers employed by these laboratories grew from 2,775 in 1921 to almost 30,000 by 1940. The interwar period also saw the industrial research labs produce significant science. In 1927 Clinton Davisson began his work at Bell Labs on electron defraction. His work led to a Nobel Prize in physics in 1937. At DuPont, Wallace Carothers developed and published the general theory of polymers, and went on in 1930 to create synthetic rubber; and then, a strong, tough, water-resistant fiber called nylon. These technological breakthroughs were in and of themselves of great importance, but it took time and money to leverage them into marketable products. For instance, over a decade elapsed to get from the beginning of research in super polymers to the production of nylon on commercial terms.

The Golden Era of "Big Science" (1945–1980)

Building on wartime success, including the Manhattan Project, the era of big science began, fueled by the optimism that well-funded scientists and engineers could produce technological breakthroughs that would benefit the economy and society. University scientists, working together with the engineers from corporate America, had indeed produced a string of breakthrough technologies including radar, antibiotics, the digital electronic computer, and atomic energy. The dominant intellectual belief of the immediate postwar period was that science-driven research programs would ensure the development of an endless frontier of new products and processes. The development of the transistor at Bell Labs gave strength to this view. Many firms augmented their commitments to industrial R&D. For instance, in 1956 IBM established a research division devoted to world class basic research.

As tensions increased during the Cold War, government funding increased considerably. In 1957, government funding of R&D performed by industry eclipsed the funding provided by the firms themselves. By 1967, it went back the other way, with private funding taking the lead. By 1975, industry funding of industry conducted R&D was twice the federal level, and the ratio was expanding. Government procurement was perhaps even more important to the technological development of certain industries, as it facilitated early investment in production facilities, thus easing the cost of commercialization. The newly emergent electronics industry in particular was able to benefit from the Defense Department's demand for advanced components and advanced products. By 1960, the electronics industry had come to rely on the federal government for 70 percent of its R&D dollars. Perhaps as an unfortunate consequence, the United States ceased to be the leader in consumer electronics as it became preoccupied with the requirements of the U.S. military, which was more performance-oriented in its requirements than the consumer markets.

By the early 1970s, however, management was beginning to lose faith in the science-driven view of industrial research and technological innovation, primarily because few blockbuster products had emerged from the research funded during the 1950s through the 1970s.

From the mid-1970s on, there has been a marked change in organization and strategy, as both industry and government have come to recognize that the classical form of R&D organization—with centralized research and a science driven culture—was simply not working, in part because new technology was not getting into new products and processes soon enough. Foreign competitors began undermining the traditional markets of many U.S. firms.

Many companies were confronted by the paradox of being leaders in R&D and laggards in the market. The fruit of much R&D was being appropriated by domestic and foreign competitors, and much technology was wasting away in many research laboratories. In telecommunications, Bell Lab's contribution to the economy at large far outstripped its contribution to AT&T. In the semi-conductor industry, Fairchild's large research organization contributed more to the economy through the spin-off

companies it spawned than to its parent. Xerox Corporation's Palo Alto Research Center made stunning contributions to the economy in the area of the personal computer, local area networks, and the graphical user interface that became the basis of Apple's Macintosh computer. Xerox shareholders were well served too, but most of the benefits ended up in the hands of Xerox's competitors.

Emergence of the "Distributed" Approach to Industrial R&D

Different modes of organization and different funding priorities were needed. The distinctive competence of firms was understood to depend upon knowledge diffused throughout the firm and embedded in new products promptly placed into the marketplace, rather than being confined to the R&D laboratory. A new way of conducting R&D and developing new products was needed.

By the 1980s and 1990s, a new model for organizing research became apparent. First, R&D activity came to be decentralized inside large corporations themselves, with the aim to bring it closer to the users. Intel, the world leader in microprocessors, was spending over $1 billion per year on R&D, but did not have a separate R&D laboratory. Rather, development was conducted in the manufacturing facilities. It didn't invest in fundamental research at all, except through its funding of Sematech and university research.

Second, many companies were looking to the universities for much of their basic or fundamental research, maintaining close associations with the science and engineering departments at the major research universities. Indeed, over the century the percentage of academic research funded by industry grew from 2.7 percent in 1960 to 6.8 percent in 1995. However, strong links between university research and industrial research is limited primarily to electronics (especially semiconductors), chemical products, medicine, and agriculture. For the most part, university researchers are insufficiently versed in the particulars of specific product markets and customer needs to configure products to the needs of the market. Moreover, in many sectors the costs of research equipment are so high that universities simply cannot participate.

Third, corporations have embraced alliances involving R&D, manufacturing, and marketing in order to get products to market quicker and leverage off complementary assets already in place elsewhere. (It is important to note, however, that outsourcing R&D is a complement,

Table 1

not a substitute, to in-house R&D.) Outsourcing and codevelopment arrangements had become common by the 1980s and 1990s (for example Pratt & Whitney's codevelopment programs for jet engines) as the costs of product development increased, and as the antitrust laws were modified to recognize the benefits of cooperation on R&D and related activities. The National Cooperative Research Act of 1984 and its amendment in 1993 provided greater clarity with respect to the likely positive treatment of cooperative efforts relating to technological innovation and its commercialization. Cooperation was also facilitated by the emergence of capable potential partners in Europe and Japan.

These developments meant that at the end of the twentieth century, R&D was being conducted in quite a different manner from how it was organized at the beginning of the century. Many corporations had closed their central research laboratories, or dramatically scaled back, including Westinghouse, RCA, AT&T, and Unocal to name just a few. Alliances and cooperative efforts of all kinds were of much greater importance.

Importantly, a transformation in industry structure brought about through venture capital funded "start-ups" was well under way. New business enterprises or "startups" were in part the cause for the decline of research laboratories; but in many ways the start-ups still depended on the organized R&D labs for their birthright.

The Role of Start-ups and Venture Capital

Beginning in the late 1970s, the organized venture capital industry, providing funding for new enterprise development, rose to significance. This was particularly true in industries such as biotech and information services. While venture capital in one form or another has been around for much of the twentieth century—the Rockefellers, Morgans, Mellons, Vanderbilts, Hillmans, and other significant families had been funding entrepreneurs for quite some time—institutional sources of money, including pension funds and university endowments, had become significant sources by the 1980s. This dramatically increased the funds that were available, as well as the professionalism by which "the money" provided guidance to a new breed of entrepreneurs, eager to develop and market new products.

As a result, venture funded start-ups have proliferated in many sectors. Thus while in the 1970s Apple Computer "bootstrapped" itself into the personal computer industry, in the 1980s Compaq and others received large infusions of venture capital to get started in the computer industry. In biotechnology, venture funding has also grown to great significance. However, it is extremely unusual for venture funds to support the efforts of companies making investments in early stage research. Rather, venture funding tends to be focused on exploiting research, not doing it. Successful start-ups frequently begin with an idea, and often personnel, that has been incubated to some level in a research program of an already established firm. Absent incumbent firms and their research programs, there would be far fewer start-ups. Figure 1 shows that significant venture funding was present in the early 1990s, and that it grew drastically from 1995 on, in part driven by the Internet boom. In 1995, however, it had risen to a level equal to 5.5 percent of the funds allocated by industry to R&D ($6 billion, compared to $108 billion). The comparison, however, should be used with care, because only a fraction of venture capital disbursements are likely to qualify as R&D. Nevertheless, the phenomena of venture funding is significant, as it is now a very important channel by which new products and processes come to the market.

Conundrum at the New Millennium

At least compared to half a century earlier, privately funded research had become more short run in its focus, and more commercial in its orientation at the millennium. International competition and the competition from spin-outs forced that outcome. The leakage of technology was such that the earlier stage the research was, the greater the chance one's competitors would also benefit from it. For example, half a century earlier, AT&T could rely on the Bell operating companies (BOCs) to each more or less pay their pro-rata share of the cost of Bell Labs; but the BOCs were divested in 1984. Their contracts to pay a fixed percent of revenues to supporting research and development were set aside in the breakup of AT&T. There no longer was an easy appropriability mechanism in place.

By 2000, it was easy in many cases to get a free ride on the efforts of others, scooping up from the public domain the product of R&D funded by others. Domestic and foreign rivals were so quick and capable that it was extremely difficult to justify the support for long-range research.

Industry and society was thus left with a deep concern—the concern that insufficient resources were being invested in the scientific "seed corn." Perhaps the solution would lie in more collective funding of research? Perhaps industrially relevant basic and applied research in universities could be expanded? The issues related more to the allocation of resources than to the amount. Clearly, as shown in Table 1, the federal government had continued throughout the postwar period to provide considerable resources to support industrial R&D. But whereas it was more than half of the total in 1960, it was only about 16 percent by 1995. A reallocation of resources from government labs to private and university labs would be one possible avenue to improve programs and augment prosperity.

BIBLIOGRAPHY

Chandler, Alfred D. Scale and Scope: The Dynamics of Industrial Capitalism. Cambridge, Mass.: Belknap Press, 1990.

Houndshell, David A. "The Evolution of Industrial Research in the United States." In Engines of Innovation: U.S. Industrial Research at the End of an Era. Edited by R. S. Rosenbloom and W. J. Spencer. Boston: Harvard Business School Press, 1996.

Mansfield, Edwin. The Economics of Technological Change. New York: Norton, 1968.

Moore, Gordon E. "Some Personal Reflections on Research in the Semiconductor Industry." In Engines of Innovation: U.S. Industrial Research at the End of an Era. Edited by R. S. Rosenbloom and W. J. Spencer. Boston: Harvard Business School Press, 1996.

Mowery, David C. "The Emergence of Growth of Industrial Research in American Manufacturing, 1899–1945." Ph.D. diss., Stanford University, 1981.

Teece, David J. "Profiting from Technological Innovation." Research Policy 15, no. 6 (1986): 285–305.

———. "The Dynamics of Industrial Capitalism: Perspectives on Alfred Chandler's Scale and Scope (1990)." Journal of Economic Literature 31 (March 1993).

David J.Teece

See alsoAT&T ; Bell Telephone Laboratories ; Capitalism ; Laboratories .

Research and Development

ECONOMIC ANALYSIS OF R&D

R&D AS INVESTMENT

BIBLIOGRAPHY

Research and development (R&D) is the term commonly used to describe the activities undertaken by firms and other entities such as individual entrepreneurs to create new or improved products and processes. The broadest meaning of the term covers activities from basic scientific research performed in universities and laboratories all the way to testing and refining products before commercial sale or use. The performance of, the incentives for, and the contributions of R&D are topics that are widely studied in management, economics, and other social science disciplines. Total spending on R&D activities is also one of the most widely used indicators of the innovative performance of firms, industries, and countries.

Informal R&D has existed at least since the first person experimented with methods of knapping flint to make Stone Age tools. In a formalized sense, it became part of the arsenal of the modern corporation beginning with the creation of industrial labs in the late nineteenth century, and in the early twenty-first century it comprises about 2 to 3 percent of the gross domestic product (GDP) in advanced economies. (For a history of the rise of organized R&D in the United States, see Mowery and Rosenberg 1989.) Spending on and outcomes of R&D investments have become important enough to be the subject of a satellite account in the U.S. System of National Income Accounts that was introduced in 2006 and 2007 (U.S. Department of Commerce 2006) and to be considered for inclusion in the international standard for systems of national income accounts (United Nations Statistics Division 2006).

The Frascati Manual of the Organisation for Economic Co-operation and Development (OECD), first published in 1963, created an international standard for surveys of spending on R&D. This manual defines R&D as “creative work undertaken on a systematic basis in order to increase the stock of knowledge, including knowledge of man, culture, and society, and the use of this stock of knowledge to devise new applications” (Organisation for Economic Co-operation and Development 2002, p. 30). R&D is generally thought to consist of three main activities: basic research, applied research, and development. Basic research is research undertaken primarily to acquire new knowledge without a view to its application. Applied research is research directed toward a specific objective, and development is work drawing on existing research results and is directed specifically toward the creation of new and improved products and processes. In general more than two-thirds of R&D spending by firms or countries is directed toward development rather than research. The 2003 OECD Science, Technology, and Industry Scoreboard reports that in developed countries with high R&D intensities, basic research is less than one-fifth of total R&D spending.

ECONOMIC ANALYSIS OF R&D

In the theoretical economics literature, the term R&D is commonly used to describe the conscious choice of firms and individuals to invest in the invention and commercialization of new products and processes. Although the activity described is seldom made precise in these models, in practice this kind of investment is assumed to correspond roughly to the spending on R&D that is reported in firm accounts and to various governmental surveys. Important insights into the motives underlying investments in R&D were first developed in seminal papers by Richard Nelson in 1959 and Kenneth Arrow in 1962. These two authors clearly argued the economic policy case for subsidies to R&D investment that arise from the nature of its output.

Briefly stated, the argument is that because most inventions (processes and products) can be imitated once they are made and at a cost lower than the original cost of making them, the incentives for undertaking R&D directed toward the creation of such inventions are inevitably weaker than society would like. The performance of R&D therefore generates positive externalities, or “spillovers,” that benefit others. Nelson distinguished between basic research with wide applications (which is most likely to be insufficiently provided) and development expenditures targeted to particular products or processes (which are more easily protected by patenting and other means). Arrow made the case for underinvestment in R&D more broadly by setting it in the context of the then newly invented Arrow-Debreu general equilibrium model. He argued that the allocation of resources for invention (that is, R&D spending) was likely to be nonoptimal because the production of information about new products and processes failed all three of the assumptions required for perfect competition to achieve a Pareto optimum: that the good (information) be infinitely divisible, that it be tradable on the market (that is, that its returns are fully appropriable by the owner), and that there be no associated uncertainty.

All three of these characteristics (indivisibility, inappropriability, and uncertainty) have proved to be important in the case of R&D. Indivisibility implies returns to scale because information about new products and processes can be spread over many units at increasingly lower cost per unit, leading to monopolistic competition in R&D-intensive industries. As Arrow and subsequent economic theorists have shown (see, for example, Reinganum 1989), this can lead to either over- or underinvestment in R&D from a social perspective. In contrast, lack of full appropriability suggests that there will be underinvestment in R&D. In a 1992 article, Zvi Griliches surveyed the evidence on the existence of R&D spillovers, and based on the evidence on measured private and social returns to R&D from a wide number of empirical studies, he concluded that overall R&D spillovers were both “prevalent and important.” This result suggests that underinvestment dominates overinvestment, at least in the majority of sectors.

Finally, uncertainty about the nature of the information to be produced by R&D makes it a risky undertaking and sometimes implies that there will be an asymmetric information problem between its producers and those who might finance its production, again leading to potential underinvestment. Empirical evidence for the existence of difficulties in financing R&D investment is surveyed by Bronwyn H. Hall (2002).

R&D AS INVESTMENT

The term R&D is often followed by investment, which hints at one of its most important attributes. Research and development continues to benefit both those who undertake it and society at large into the uncertain future. Another attribute of R&D that has been emphasized in the modern economic growth literature is its cumulative nature, which can lead to increasing returns, both in the aggregate and also for individuals and firms. The idea is that the stock of knowledge created by doing R&D makes one more productive in acquiring additional knowledge. At the economy-wide level, this idea is the basis of the modern endogenous growth literature that is discussed in Philippe Aghion and Peter Howitt’s Endogenous Growth Theory (1998).

These attributes create some interesting problems for measurement and analysis. In general applied researchers have dealt with the intertemporal nature of R&D investment by treating it in the same way as ordinary investment in tangible assets, adding up expenditures to create an R&D stock and using a suitable depreciation rate to capture the fact that older research may become less useful over time. However, the aforementioned spillover effects make this exercise somewhat more speculative than in the case of ordinary assets, because R&D that has ceased to be useful for the production of private profit may still be useful to others in the production of new knowledge. The process that renders some R&D output obsolete is the same one that was termed “creative destruction” long ago by Joseph Schumpeter (1942). For a survey of the R&D depreciation problem and its implications for measuring the returns to R&D, see Hall (2006).

SEE ALSO Arrow, Kenneth J.; Investment; Schumpeter, Joseph Alois; Technological Progress, Economic Growth; Technology

BIBLIOGRAPHY

Aghion, Philippe, and Peter Howitt. 1998. Endogenous Growth Theory. Cambridge, MA, and London: MIT Press.

Arrow, Kenneth. 1962. Economic Welfare and the Allocation of Resources for Invention. In The Rate and Direction of Inventive Activity, ed. Richard R. Nelson, 609–625. Princeton, NJ: Princeton University Press.

Bush, Vannevar. 1945. Science: The Endless Frontier. Washington, DC: U.S. Government Printing Office.

Cohen, Wesley M., and Richard C. Levin. 1989. Empirical Studies of Innovation and Market Structure. In The Handbook of Industrial Organization, ed. Richard Schmalensee and Robert D. Willig, 1059–1107. Amsterdam: North-Holland.

Griliches, Zvi. 1979. Issues in Assessing the Contribution of R&D to Productivity Growth. Bell Journal of Economics 10: 92–116.

Griliches, Zvi. 1992. The Search for R&D Spillovers. Scandinavian Journal of Economics 94: S29–S47.

Hall, Bronwyn H. 2002. The Financing of Research and Development. Oxford Review of Economic Policy 18 (1): 35–51.

Hall, Bronwyn H. 2006. R&D, Productivity, and Market Value. http://www.econ.berkeley.edu/~bhhall/bhpapers.html.

Mowery, David C., and Nathan Rosenberg. 1989. Technology and the Pursuit of Economic Growth. Cambridge, U.K.: Cambridge University Press.

Nelson, Richard R. 1959. The Simple Economics of Basic Scientific Research. Journal of Political Economy 77: 297–306.

Organisation for Economic Co-operation and Development. 2002. Frascati Manual: Proposed Standard Practice for Surveys on Research and Experimental Development. Paris: Author.

Organisation for Economic Co-operation and Development. 2003. Science, Technology, and Industry Scoreboard. Paris:Author.

Reinganum, Jennifer F. 1989. The Timing of Innovation: Research, Development, and Diffusion. In The Handbook of Industrial Organization, ed. Richard Schmalensee and Robert D. Willig, 850–908. Amsterdam: North-Holland.

Schumpeter, Joseph. [1942] 1976. Capitalism, Socialism, and Democracy. New York: Harper and Row.

United Nations Statistics Division. 2006. National Accounts: Research and Development. http://unstats.un.org/unsd/nationalaccount/nadefault.htm.

U.S. Department of Commerce. 2006. R&D Satellite Account. News release, September 28. http://bea.gov/bea/newsrel/rdspendnewsrelease.htm.

Bronwyn H. Hall

RESEARCH AND DEVELOPMENT

Research and development (R&D) is the process by which scientific and technological breakthroughs are used to create new or improved products or technologies. R&D generally involves three basic processes. The first is "basic" or "pure" research, which is the kind of exploration scientists and other researchers perform when they are motivated not by a specific goal or end product but by the desire to advance scientific knowledge wherever it leads. "Applied research" is research that takes the findings of basic research and modifies or elaborates them for a specific purpose. Finally, "development" is the process by which the insights of basic and applied research are transformed into an actual product or new process. For example, basic research into the properties of light and silica-based materials might be applied toward developing fiber optics, which in turn might be developed into fiber-optic cables for carrying digital data. Basic and applied research is conducted in government, academic, and industrial research laboratories, and today most product development is performed in company-run development laboratories.

The first U.S. industrial laboratories were modest affairs set up in the 1860s and 1870s by companies like Cambria Iron Co. and the Pennsylvania Railroad. In the 1870s Thomas Edison's (1847–1931) Edison Electric Light Company became the first company to devote a substantial portion of its revenues to industrial laboratory research. In the years before World War I (1914–1918) such companies as General Electric, AT&T, Standard Oil, and Eastman Kodak also established R&D labs. World War I opened the eyes of many Western governments to the need for industrial research, a process that was hastened by the requirement for new technologies to fight World War II (1939–1945). After 1945 the United States was the preeminent technological power, and the federal government's R&D spending in the 1950s and 1960s was used by government, academic, and industrial laboratories to fuel an explosion of new electronics, computer, aeronautical, defense, and space technologies.

When the U.S. government cut back on R&D funding at the end of the Cold War in the early 1990s, private industry filled the gap. By 1997, for example, total R&D investments reached an all-time high of more than $200 billion, two-thirds of which was conducted by private industry. The primary factors for this intensified industrial R&D were advances in biotechnology, electronics, and software; global competition; and years of strong profit growth. By 1996, 25 U.S. firms—led by General Motors, Ford Motor Company, and IBM—were spending more than $1 billion annually on R&D.