W 109-180. General note.

Historical statistics on research and development expenditures and employment by various groups in the major sectors of the economy are of comparatively recent origin. Public interest in representing the input of research and development activity in terms of some widely used measure, such as funds expended or personnel employed, has been largely incidental to concern with major national issues. During the depression years of the 1930's this interest stemmed from the role that research played in the recovery of the economy. Groups such as the National Research Project of the Work Projects Administration (formerly the Works Progress Administration) and the National Resources Planning Board engaged in studies of the interrelationships among trends in research and development, technological change, unemployment, education, and other major economic and social factors. Their interest in measuring research and development was generally subsidiary to a larger preoccupation with such broad national issues as economic recovery, re-employment, and national planning. The research and development estimates which they published were intended to serve primarily as illustrative background materials.

The period of World War II and its aftermath dramatized the critical place of research and development in the Nation's military security program. Groups concerned with measuring research and development during this period included the Committee on Science and Public Welfare (Bowman Committee), the President's Scientific Research Board, and the Research and Development Board of the Department of Defense. Like the earlier groups, these organizations supplemented fragmentary data already on hand with special inquiries and analyses in order to develop background estimates on research and development trends.

The National Science Foundation, a Federal agency established in 1950, undertook as one of its functions the development of such factual data and related analyses on research and development. As a first step, the Foundation initiated an annual survey of Federal funds for research and development, starting with data on funds for scientific research and development at nonprofit institutions for fiscal 1951 and 1952 and moving thereafter to annual surveys of the funds comprising the "Federal Research and Development Budget."

In 1954, the National Science Foundation undertook the first

effort to measure the volume of research and development activity, in terms of funds and personnel, through surveys of all major types of organizations in the several sectors of the economy which were known to be performing or financing this activity. Out of this effort grew a continuing Foundation program of surveys designed to facilitate preparation of annual estimates on funds and personnel employed in research and development by the major sectors of the economy. Before the National Science Foundation undertook its first surveys, there was no general agreement on such fundamental matters as the definition of research and development; the distinction between the conduct of research and development and such related activities as academic instruction or industrial production; the distinction between basic and applied research and development; and the major characteristics distinguishing various types of research organizations.

National estimates. National estimates of funds spent on the performance of research and development by the four major sectors of the economy have been made by the National Science Foundation for 1953-70. National estimates of funds received from various sources for the performance of research and development have also been made for 1953-70. These series appear in table B-1 of the National Science Foundation publication, National Patterns of R&D Resources, 1953-72, Funds and Manpower in the United States, (NSF 72-300). An analysis of intersectoral flows of transfers of funds for research and development for 1970 appears in table I, below. This table is based on information obtained in the National Science Foundation surveys of funds for research and development in 1970 as presented in table B-1.

The limitations described below indicate that table I should be considered a general approximation rather than an exact statement of the extent to which the different sectors are participating in the financing and performance of research and development.

Detailed information on the scope and limitations of the various surveys appears in the National Science Foundation publications listed below. Generally speaking, the National Science Foundation surveys seek full enumeration of the various segments. The exceptions are industrial firms and the smaller nonprofit institutions for which sampling procedures are employed.

The data in table I are derived basically from survey responses by performers of research and development as to how much they spent

on this activity and where their funds originated. The estimates represent final through-transfers from source organizations financing research and development to performing organizations which ultimately used the funds. Every effort was made to net out intermediate transfers.

Research and development in these series consist of basic and applied research in the sciences (including medical sciences) and in engineering and activities in development, all defined below. In terms of fields, the natural sciences-life, physical, and engineering—as well as the social and psychological sciences are covered in the Federal, universities, and other nonprofit sectors. Industry coverage is limited, at present, to the natural sciences. Research and development excludes routine product testing, quality control, mapping surveys, collection of general-purpose statistics, experimental production, and activities concerned primarily with the dissemination of scientific information and the training of scientific manpower.

Research, which is made up of basic and applied, is systematic, intensive study directed toward fuller scientific knowledge of the subject studied.

Basic research. For three of the sectors-Federal Government, universities and colleges, and other nonprofit institutions-the definition of basic research stresses that it is directed toward increases of knowledge in science with "... the primary aim of the investigator being a fuller knowledge or understanding of the subject under study, rather than a practical application thereof." To take account of an individual industrial company's commercial goals, the definition for the industry sector is modified to indicate that basic research projects represent "original investigations for the advancement of scientific knowledge... which do not have specific commercial objectives, although they may be in fields of present or potential interest to the reporting company."

Applied research. The core definition in the NSF questionnaire sent to the universities and colleges is: “Applied research is directed toward practical application of knowledge." Here again, the definition for the industry survey takes account of the characteristics of industrial organizations—it covers "... research projects which represent investigations directed to discovery of new scientific knowledge and which have specific commercial objectives with respect to either projects or processes." By this definition, applied research in industry differs from basic research chiefly in terms of objectives of the reporting company.

Development. The NSF survey concept of development may be summarized as ". . . the systematic use of scientific knowledge directed toward the production of useful materials, devices, systems or methods, including design and development of prototypes and processes." Funds used for research and development refer to current operating costs, consisting of both direct and indirect costs including depreciation, insofar as this information is available to respondents. Capital expenditures are excluded by definition in both the industry and the other nonprofit sectors. Under the accounting practices of some Federal agencies, particularly the Department of Defense, data on the Federal R&D funds, which are available in detail only in terms of obligations rather than expenditures, do not include an allowance for depreciation but do include some obligations for capital items. A small amount of capital outlays is also included in the universities and colleges sector.

The National Science Foundation surveys include data on research and development by Federally Funded Research and Development Centers (FFRDC's). These are laboratories or similar research undertakings supported wholly or predominantly by the Federal Government but operated under contract by an industrial, university, or independent organization. Data relating to the performance of research and development at these centers are included within the appropriate sector in the estimates for 1956 and later years.

Several groups of organizations comprise the industry sector as represented in table I. Private industrial firms account for over 95 percent of the total funds for performance of research and development

in this sector. (Data for firms appear in series W 144-160.) The remaining groups are FFRDC's operated by industrial concerns, independent commercial laboratories, and engineering service firms. The colleges and universities sector consists of institutions of higher education with substantial research programs and of the FFRDC's operated under contract by educational institutions. Included in institutions of higher education are their affiliated research organizations, agricultural research centers, graduate and professional schools, and affiliated hospitals.

Other nonprofit institutions include privately endowed philanthropic foundations, nonprofit research institutes, voluntary health agencies, academies of science, professional societies, museums, zoological gardens, and arboretums, as well as several FFRDC's operated by independent organizations.

The data on transfers of funds were based on estimates from many institutions having somewhat different understandings of costs and expenditures. The estimates for Federal agencies, moreover, were based on obligations rather than expenditures, since information on transfers to the other sectors was available only for obligations. (Additional details appear in series W 126-143.)

For detailed information and trend data on R&D funds and scientific personnel, see the following publications: National Patterns of R&D Resources, Funds and Manpower in the United States (annual); Federal Funds for Research, Development, and Other Scientific Activities (annual); Research and Development in Industry (annual); Scientific Activities at Universities and Colleges (biennial); Scientific Activities of Nonprofit Institutions (periodic); American Science Manpower (biennial); Scientific, Technical, and Health Personnel in the Federal Government (annual); and Employment of Scientists and Engineers in the United States, 1950-66.

Early major efforts to estimate the volume of research and development. The methodology, scope, and limitations of the various series are often summarized in the publications cited below, and any use of these estimates in descriptive or analytical work should be preceded by a careful review of their limitations. Differences in concepts and scope of these earlier series make comparisons with the later NSF data not generally possible.

George Perazich and Philip M. Field, Industrial Research and Changing Technology, Work Projects Administration, National Research Project, Philadelphia, 1940, pp. 5-17 and 52-79. This report presents data on research personnel in industrial laboratories for 1920, 1921, 1927, 1931, 1933, and 1938. The data are based on the six directory listings on industrial research laboratories in the United States published by the National Research Council between 1920 and 1938.

National Resources Committee (later, National Resources Planning Board), Research, A National Resource, vol. 1, Relation of the Federal Government to Research, Report of the National Resources Planning Board Science Committee, 1938. Section 3, pp. 61-112, of this report presents estimates of Federal expenditures for research in 1937 and 1938. Table D, p. 91, summarizes from other sources a number of earlier estimates of Federal expenditures going back as far as 1901. Section 6, pp. 167-193, contains a discussion of research in American universities and colleges. It also provides a general estimate of the dollar volume of expenditures for research and development for 1935-1936.

National Resources Planning Board, Research, A National Resource, vol. II, Industrial Research, a report of the National Research Council to the National Resources Planning Board, 1941. Section IV, pp. 173-187 of this report, presents estimates of research personnel in industrial laboratories for 1940; and section II, part 7, pp. 120-123, shows research personnel and expenditures in 31 firms for 1937.

U.S. Senate, Committee on Military Affairs, Subcommittee on War Mobilization (Harley M. Kilgore, Chairman), Report on the Government's Wartime Research and Development, 1940-44, 1945. Part I of this report presents detail, and part II summarizes data on funds for research and development for each of 45 Federal agencies

and bureaus, with detail on the fiscal sources of funds and the major categories of recipients for fiscal years 1940 through 1944.

Vannevar Bush, Science, The Endless Frontier, A Report to the President, July 1945, appendix 3, "Report of the Committee on Science and Public Welfare" (Isaiah Bowman, Chairman). The Bowman Committee's report to Dr. Bush presents the first known national estimates of trends in scientific research and development expenditures in table I, p. 80. It also contains series on scientific research expenditures (based largely on performance of research) for the following major groups: (a) Industry-annual expenditures estimates for 1920-1940; (b) nonprofit industrial research institutesannual expenditures estimates for 1930-1942; (c) Government (Federal and State)-annual estimates for 1923-1932, 1934-1938, and 1940-1944; (d) colleges and universities-biennial estimates, 1930, 1932, 1934, 1936, 1938, 1940, and 1942; (e) research institutes (not connected with any industry nor an integral part of any university)-annual estimates for 1930-1940; and (f) total scientific research expenditures-total of the foregoing five series for 1930, 1932, 1934, 1936, 1938, and 1940.

The President's Scientific Research Board (John R. Steelman, Chairman), Science and Public Policy, A Report to the President, vols. I, II, and IV, 1947. Based on data in Vannevar Bush, Science, The Endless Frontier (cited above, vol. I, A Program for the Nation, presents for the even years of 1930-1940 estimated expenditures by the Federal Government, industry, universities, and others. Estimates are also made of the average annual expenditures by major groups for 1941-1945 and of expenditures for 1947. Vol. II, The Federal Research Program, presents estimates of Federal "expenditures for research and development in the physical and biological sciences" in fiscal year 1947 based on project reports from the individual agencies. Vol. IV, Manpower for Research, presents annual estimates and forecasts of scientists and engineers in industrial research laboratories for 1929-1956.

Helen Wood, Robert Cain, and Joseph H. Schuster, Scientific Research and Development in American Industry, A Study of Manpower and Costs, Bulletin No. 1148, U.S. Bureau of Labor Statistics, prepared in cooperation with the Department of Defense, 1953. Data in this publication are based on the first survey specifically designed to obtain research and development performance costs and personnel for private firms. The report presents estimates of expenditures for research and development performed in 1951 by firms reporting. Personnel data cover research and development scientists and engineers employed by these firms in January 1951 and 1952.

Office of the Secretary of Defense, The Growth of Scientific Research and Development, 1953. This publication presents annual estimates on sources of research and development funds and on performance of research and development for 1941-1952 for the Federal Government, industry, and nonprofit institutions including colleges and universities. Estimates of the number of scientists and engineers employed in research and development by these broad sectors are also shown for the same years. No methodological notes accompany the estimates. They are known to be based on materials in Wood, Cain, and Schuster (cited above); U.S. Senate Committee on Military Affairs, Subcommittee on War Mobilization (cited above); and other published and unpublished sources.

Office of Education, Statistics of Higher Education; Receipts, Expenditures, and Property, 1953-54, 1957. This report presents biennial estimates on expenditures for performance of "Organized research" by institutions of higher education, 1930-1954. See series H 732 in this volume. This is the oldest known current series on research expenditures.

More recently, beginning with Bureau of the Budget, The Budget of the United States Government, 1955, the Federal budget documents have carried a special analysis of "Federal Research and Development Programs" summarizing expenditures and/or obligations for research and development and R&D plant on an agency basis. In The Budget, 1972, this was Special Analysis R and covered fiscal years 1970, 1971, and 1972.

W 109-125. Funds expended for performance of research and development and basic research, by sector, and major function, 1953-1970.

Source: U.S. National Science Foundation, National Patterns of R&D Resources, 1953-72, Funds and Manpower in the United States (NSF 72-300).

The four-sector division followed by the National Science Foundation attempts to take account of both the legal nature and major functions of organizations active in financing and performing basic research, applied research, and development. However, grouping diverse types of organizations into discrete sectors requires certain arbitrary judgments because of the mixed nature of many organizations, particularly those in the university and other nonprofit sectors. The Federal sector is made up of the agencies of the Federal Government.

The industry sector consists of both manufacturing and nonmanufacturing companies. Manufacturing is surveyed in major industry groupings; and nonmanufacturing, which includes organizations such as those in selected service industries, is treated as a unit. Federally Funded Research and Development Centers (FFRDC's) administered by industrial firms are also included.

The universities and colleges sector is composed of all institutions of higher education, both public and private. The term "universities and colleges" is used to refer to the academic institutions as a group without the associated FFRDC's administered by the schools for various Federal agencies. The universities and colleges comprise the following:

Colleges of liberal arts; schools of arts and sciences; professional schools, such as engineering and medical schools, including affiliated hospitals; associated research institutions, and similar organizations, which are integral parts of the universities and colleges; agricultural experiment stations and associated schools of agriculture.

Funds used at the universities and attributed to the universities sector as a source consist of: (a) State and local government funds separately budgeted for research and development, (b) the direct or indirect costs of R&D performance sponsored by outside organizations that were defrayed in part by universities and colleges in accordance with cost sharing or other arrangements, and (c) unrestricted or general funds which the institutions themselves have been free to allocate for research either through their instructional or departmental budget or through their own separately budgeted research. Funds from the Federal Government, industry, or other nonprofit institutions, which are supplied in the form of grants or contracts for research or development at a university, are credited to the appropriate source in the performance of research and development by universities and colleges. Thus, research contracts from industry are treated as university performance funded by industry as the source, whereas funds given to the institution by industry for general educational purposes and used by the school, at its discretion, for research, are treated as university performance financed with the university's own funds.

Institutions in the other nonprofit sector fall into two general groups: (1) Organizations that are primarily granting in nature, namely private philanthropic foundations and voluntary health agencies, and (2) public and private organizations that are primarily involved in performing research and development, comprising separately incorporated nonprofit research institutes, professional societies, academies of science, museums, zoological gardens, botanical gardens, arboretums, nonprofit hospitals, and FFRDC's administered by nonprofit organizations.

In these series, both the university and the other nonprofit sectors contain private and public institutions--the latter either closely associated with or considered a part of State or local government. A number of organizations in both sectors, as well as in industry, also receive State and local government funds.