Science Indicators 1974 National Science Board 1975 NATIONAL SCIENCE BOARD (as of 1975) Dr. Norman Hackerman (Chairman, National Science Board), President, Rice University Dr. Russell D. O'Neal (Vice Chairman, National Science Board), Chairman and Chief Executive Officer, KMS Industries, Inc., Ann Arbor, Michigan Dr. W. Glenn Campbell, Director, Hoover Insti- tution on War, Revolution, and Peace, Stan- ford University Dr. H. E. Carter, Coordinator of Inter- disciplinary Programs, University of Ari- zona Dr. Robert A. Charpie, President, Cabot Cor- poration, Boston, Massachusetts Dr. Jewel Plum.mer Cobb, Dean and Professor of Zoology, Connecticut College Dr. Lloyd M. Cooke, Director of Urban Affairs and University Relations, Union Carbide Corporation, New York, New York Dr. Robert H. Dicke, Albert Einstein Professor of Science, Department of Physics, Princeton University Dr. David M. Gates, Professor of Botany and Director, Biological Station, Department of Botany, University of Michigan Dr. T. Marshall Hahn, Jr., Executive Vice President, Georgia-Pacific Corporation, Portland, Oregon Dr. Anna J. Harrison, Professor of Chemistry, Mount Holyoke College Dr. Hubert Heffner, Chairman, Department of Applied Physics, Stanford University' Dr. Roger VV. Heyns, President, American Council on Education, Washington, D.C. Dr. Mr. Dr. Dr. Dr. W. N. Hubbard, Jr., President, The Upjohn Company, Kalamazoo, Michigan Saunders Mac Lane, Max Mason Distin- guished Service Professor of Mathematics, University of Chicago William H. Meckling, Dean, The Graduate School of Management, The University of Rochester Dr. Grover E. Murray, President, Texas Tech University and Texas Tech University School of Medicine William A. Nierenberg, Director, Scripps Institution of Oceanography, Ihiiversity of California at San Diego Frank Press, Chairman, Department of Earth and Planetars' Sciences, Massachusetts Institute of Technology Dr. Joseph M. Reynolds, Boyd Professor of Phys- ics and Vice President for Instruction and Research, Louisiana State L'niversity Dr. Donald B. Rice, Jr., President, The Rand Corporation, Santa Monica, California Dr. L. Donald Shields, President, California State University at Fullerton Dr. Charles P. Slighter, Professor of Physics and in the Center for Advanced Study, University of Illinois at Urbana-Champaign^ Dr. H. Guyford Stever, Director, National Sci- ence Foundation Dr. F. p. Thieme, Professor of Anthropology, Uni\ersity of Colorado Dr. James H. Zumberge, President, Southern Methodist University Miss Vernice Anderson, Executive Secretary, National Science Board Science Indicators Unit Dr. Robert W. Brainard, Unit Head^ Dr. Robert R. Wright, Unit Head* Mrs. Susan Coady Kemnitzer, Program Analyst Mr. Thomas R. Mika, Program Analyst ' Deceased April I. 1975. ^ Appointed December 8, 1975. ' Through Augusi 1975. ' Appoinied Unit Head in .September 1975. W^ (ji^y* Science Indicators 1974 '^>^' ^^g_Q Report of the National Science Board 1975 m^S^ ^^° National Science Board National Science Foundation /Aanne Biological Laboratory Library Woods Hole, Massachusetts Gift of Bostwick H. Ketchum - 1976 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 ■ Price $4.60 Stocit Number 038-000-00263-8 Letter of Transmittal December 10, 1975 My Dear Mr. President: I have the honor of transmitting to you, and through you to the Congress, the Seventh Annual Report of the National Science Board. The Report is submitted in accordance with Section 4(g) of the National Science Foundation Act of 1950, as amended. In this Report, Science Indicators — 2 974, the Board presents the second step in the process begun with Science Indicators — 1972 of developing indicators of the state of science in the United States. Our goal is a periodical series of indices of the strengths and weaknesses of science and technology in the United States and the changing character of that activity. We hope that by contributing to the understanding of science itself we will strengthen its forward thrust, illuminate its significance, and assist in the examination of its problems. The indicators in this Report deal primarily with resources — human and financial — for research and development. Progress has been made in developing measures of the outcomes or impacts of research and development and the contributions made thereby to the welfare of the Nation. We are continuing as a high priority our study of indicators of the characteristics of science and technology and will describe our progress in successive Science Indicator reports. Respectfully yours. Norman Hackerman Chairman, National Science Board The Honorable The President of the United States 111 Contents INTRODUCTION vii INTERNATIONAL INDICATORS OF SCIENCE AND TECHNOLOGY 1 Resources for Research and Development 3 Scientific Research 7 Technological Invention and Innovation 16 Productivity and Balance of Trade 21 RESOURCES FOR RESEARCH AND DEVELOPMENT 29 National Resources for Research and Development 32 Federally Funded R&D in Functional Areas 36 Research Equipment and Facilities 44 Dissemination of R&D Results 46 BASIC RESEARCH 49 Resources for Basic Research 52 Basic Research in Universities and Colleges 59 Basic Research in Federally Funded Research and Development Centers Administered by Universities 67 Basic Research in Intramural Federal Laboratories 68 Basic Research in Industry 70 Basic Research in Nonprofit Institutions 70 Research Outputs and Applications 71 INDUSTRIAL R&D AND INNOVATION 81 Resources for Industrial R&D 83 Outputs from R&D and Innovation 94 Returns from R&D and Innovation 110 SCIENCE AND ENGINEERING PERSONNEL 113 Characteristics and Utilization of Science and Engineering Personnel 116 Research and Development Personnel 126 Unemployment Among Scientists and Engineers 131 Supply of Scientists and Engineers 132 Women and Minorities in Science and Engineering 138 PUBLIC ATTITUDES TOWARD SCIENCE AND TECHNOLOGY 143 APPENDIX 153 Introduction The National Science Board is charged by the Congress with providing an annual report of the status of science in the United States.' In this, its seventh report, the Board continues the development of a series of indicators assessing the condition of the Nation's scientific endeavor. These indicators are intended to measure and to reflect U.S. science — to demonstrate its strengths and weaknesses and to follow its changing character. Indicators such as these, updated regularly, can provide early warnings of events and trends which might impair the capability of science — and its related technology — to meet the needs of the Nation. The indicators can also assist those who set priorities for the enterprise, allocate resources for its functions, and guide it toward change and new opportunities. In these ways, communication about the issues of science is facilitated and considerations of new areas of public policy can be explored. The internal characteristics of science provide the most readily available data for indicators, including the human and financial resources involved, the education of research scientists, changes in the institutional structures which support research and development, advances m the fundamental understanding of science and the transfer of technology. Of equal importance are measures of the external impact of science, often called "output indicators". These in- dicators are difficult to devise because the translation of science into technology and the genesis of science in technological advances are both deeply embedded among complex economic and social variables. In addition, many of the applications of science are not immediately realized, occurring long after and often appear- ing unrelated to their origins in research. However, the present report represents an advancement in the development of indicators of the outputs of the research and development enterprise. The establishment of a comprehensive system of science indicators involves the investigation of potential indices, expansion of the underlying ' Section 4(g) of the National Science Foundation Act as amended by Public Law 90-407. 2 Science Indkaion— 7972, National Science Board (NSB 73- 1). data base, improvement of methods for measur- ing the impacts of science and technology, development of analytic approaches for inter- preting the measures, and demonstration of their utility across several audiences. The effort to develop a system of effective indicators should be regarded as a long-term process. A central concept of the effort is, therefore, an evolving set of indicators derived from continuing exploration, testing, and design. The set will be evaluated, expanded, refined, and updated regularly as new data become available, as our understanding of their nature improves, and as the science enterprise itself changes. Quantitative indicators are not a substitute for the experience and judgment of the scientific community. Indices, at their best, can only serve as supplements. The interpretation of indicators themselves — what they mean for the present and the future of the enterprise — requires the participation of the scientific community. The Report Indicators in this report include measures of basic research activity and industrial R&D, indices of scientific and engineering personnel and institutional capabilities, indicators of productivity and the U.S. balance of trade in high-technology products, and other aspects of the Nation's science and engineering activities. Compared to the first Science Indicators report of the National Science Board, ^ the present report contains substantially more indicators, expanded to fill some of the major gaps and reorganized to present a more current and integrated coverage of science and related technology. A new chapter discusses industrial R&D in the United States, and includes the results of a survey on the innovative process. Additions to other portions of the report provide new information on the role of basic science in advancing technology, international aspects of technological innovation, and changing attitudes of the public toward science. These indicators of the scientific enterprise are presented in six chapters, generally with a time span beginning in the early 1960's and extending through 1974 where possible. Data which appeared in Science huiicalors — 19 72 for earlier years are repeated here to encourage longitudinal comparisons and to make it un- necessary to refer to the previous report. Most of the indicators are presented in graphical form and are numbered to correspond with the numerical data tables in the Appendix. Each of the chapters is introduced by an "Indicator Highlights" section which briefly summarizes the major indices of that chapter. It should be noted that these highlights often omit important caveats and discussion contained in the text itself. The original data sources, many of which are publications of the Division of Science Resources Studies, National Science Founda- tion, are indicated throughout the report. Staff of the Division also took part in the development of charts and text. The challenge faced in creating and using indicators of complex social systems such as science and technology is substantial, and the present efforts to assess U.S. science are still only in the early stages of maturity. Apprecia- tion is due to the Social Science Research Council which, with NSF support, convened a seminar in 1974 on science indicators and which has recently established a Subcommittee on Science Indicators. The reports to follow in this series will aim to sharpen concepts, refine their treatment, and seek new measures of the state of science. It is hoped that all those interested in science indicators will participate in the search. International Indicators of Science and Technology International Indicators of Science and Technology INDICATOR HIGHLIGHTS The proportion of the Gross National Product (GNP) spent for R&D has declined steadily over the last decade in the United States, while growing substantially in the U.S.S.R., West Germany, and Japan; in 1973, the fraction of GNP directed to R&D was 2.4 percent in the United States, compared with 3.1 percent for the U.S.S.R., 2.4 percent for West Germany, and 1.9 percent for Japan. i The number of scientists and engineers engaged in R&D per 10,000 population declined in the United States after 1969 but continued to grow in all other countries studied; by 1973, this number was 25 per 10,000 for the United States, 18 for West Germany, 19 for Japan (1971), and 37 for the U.S.S.R.i All major R&D-performing countries re- duced their proportion of government R&D expenditures for national defense between 1961 and the early 1970's, while either maintaining or expanding expenditures for the advancement of science and economic development; the United States had the largest fraction of expenditures for national defense and the smallest for the latter two areas throughout the period. (Data for the U.S.S.R. are not available). The United States was the largest producer of the scientific literature sampled throughout the 1965-73 period in all fields except chemistry and mathematics, where its share was second to that of the U.S.S.R.; in recent years, however, U.S. research publications in the fields of chemistry, engineering, and physics have declined slightly in both absolute and relative terms. 1 Data regarding the U.S.S.R. should be treated as gross estimates; limited information and differences in basic definitions make international comparisons involving the U.S.S.R. particularly tenuous. (See the following text for discussion of this point). Citation indices of U.S. scientific research equal or exceed those of other major research-performing countries based on a large sample of the 1973 literature; the United States ranked highest in the fields of chemistry and physics. U.S. scientists have received a larger overall number of Nobel Prizes in the sciences (physics, chemistry, and physiology- medicine) than any other country; awards to U.S. scientists, however, declined after the 1951-60 decade, primarily as the result of fewer prizes for research in physics. The United States had a favorable but declining "patent balance" between 1966 and 1973; the decline of 30 percent was due primarily to increases in the number of patents awarded by the United States to Japan and West Germany, and to decreases in patents granted to the United States by Canada and the United Kingdom. A majority of a sample of major technological innovations of the past twenty years were produced by the United States; the proportion of innovations of U.S. origin, however, declined from a high of 80 percent in the late 1950's to some 55-60 percent since the mid-1960's, while other countries— particularly Japan and West Germany — increased their shares. The United States had an increasingly positive balance of payments from the sale of technical "know-how" (patents, licenses and manufacturing rights) over the 1960-73 period, with four to five times more technical "know-how" sold to other nations than purchased from them; the rising net receipts to the United States were due largely to purchases by Japan after the mid- 1960's. The level of U.S. productivity (Gross Domestic Product per employed civilian) exceeded that of the other major R&D- performing nations between 1960-74, although gains in productivity were larger in the latter countries; by 1974, the productivi- ty of France and West Germany was some 75-80 percent of the U.S. level, while Japan, with the largest gains in productivity, reached a level which was approximately 55 percent as high as U.S. productivity. D The United States has a large, favorable balance of trade in commodities produced by R&D-intensive industries, in contrast to the increasingly negative balance in non-R&D- intensive products; the 1974 balance in R&D-intensive products was large enough to offset petroleum imports for the same year. The favorable U.S. trade balance in R&D- intensive products depends primarily upon exports to developing nations and to Western Europe; a deficit balance developed with Japan in the mid-1960's and continued through 1973, due largely to imports in the areas of electrical machinery, professional and scientific instruments, and nonelectrical machinery. This chapter presents indicators of science and technology in an international context. The focus is on the United States and how it compares with other major developed nations in several aspects of science and technology. The indicators are directed primarily to four general aspects. The first of these relates to the absolute and relative levels of national resources utilized for research and development (R&D); this includes both human and financial resources, as well as the areas of application to which the R&D is aimed. The second topic centers around scientific research; the indicators here deal principally with the quantity and quality of scientific research in individual countries and the international dimensions of science. The third facet concerns the output from applied R&D and technological efforts; indices in this group include trends in invention and innovation, and international transactions in technology. Finally, the fourth aspect deals with productivity, economic competitiveness, and international trade; indicators in this area provide measures of the level and change in the productivity of nations and of the role of R&D in the U.S. trade balance. International indicators of science and technology suffer from several general deficien- cies. There is usually a paucity of data; the reliability of the data which are available is often unknown or less than desired; and information is frequently based upon concepts and methods which may differ substantially among countries. These place restrictions on both the aspects of science and technology which can be measured and the accuracy of the measurements themselves. For these reasons, the indicators and international comparisons presented in this chapter should be interpreted with considerable caution. RESOURCES FOR R&D The international comparisons presented here are based upon indicators of the human and financial resources directed to R&D by the major R&D-performing countries. These indicators are limited to measures of the magnitude of the national resources for R&D, and the general areas to which they are directed (e.g., defense, space, and health). Expenditures for R&D R&D expenditures as percentages of the Gross National Product (GNP) are shown in figure 1-1 for the six countries with the largest R&D expenditures. 2 This indicator expresses the proportion of a country's economic output which is directed to R&D and is a measure of the R&D intensiveness of a nation.' But because of differences among countries in the composition. 2 Expenditures reported for the U.S. and the U.S.S.R. are for the performance of R&D alone, while those for other countries include associated capital expenditures. -' For the classification of various countries according to their R&D intensiveness, see "A Comparative Study of Science Advisory Approaches of Selected Developed Coun- tries" in Federal Poliqi, Plans, ami Organizaimn for Science and Technology, Part II. U.S. Congress, House Committee on Science and Astronautics, 93rd Congress, 2nd Session, 1974. Figure 1-1 R&D Expenditures as a Percent of Gross National Product (GNP), by Country, 1961-74 (Percent) 1 o ix ^•*^ 3.0 ^ ,,, U.S.S.R. / / vf ' 2.8 / X 2,6 2.4 / West Germany /* ^ / 2.2 France ^^ \ ^•' 2.0 / .•****'^^*«..^ •"* 1.8 /,. /*s 1.6 - / .* Japan ^ 1.4 ^/ --'' 1.2 • • • • • 1 1.0 - nl 1 1 1 1 1 1 1 1 1 1 1 1 1961 '63 '65 '67 '69 71 73 74 (est.) SOURCE Organisation lor Economic Co-operation and Development; indi- vidual country sources, U.S.S.R. estimates by Robert W. Campbell. Indiana University. last 10 years, falling nearly one-fourth from its peak level in 1964. The decline, as discussed elsewhere in this report,* is due primarily to reduced growth of expenditures by the Federal Government for R&D in the defense and space areas; increases in R&D funds from all other sources combined kept pace with growth in the GNP. In the case of France, the only other country of those studied which showed a long- term decline in this indicator, the reduction appears to result largely from a slower growth in government R&D expenditures for national defense and nuclear energy. Both Japan and West Germany recorded substantial growth in the proportion of the GNP directed to R&D. Underlying their growth were continuous large increases in R&D funding from both industry and government. Total R&D expenditures by Japan increased at an average annual rate of 21 percent between 1963 and 1973, and those of West Germany by 15 percent, as compared with 6 percent for the United States. More recently, annual increases between 1969-73 averaged 24 percent for Japan, 16 percent for West Germany, and 4 percent for the United States. While industry is the prime source of R&D funds in Japan and West Germany, funds provided by the government have grown relatively more than those from industry. Government funds for R&D in these two countries are concentrated on advancement of science and, to a lesser extent, on general economic growth and nuclear energy. ^ For the U.S.S.R. this indicator is based upon limited information and should be regarded only as an estimate. The general upward trend in the proportion of the GNP devoted to R&D is believed to be valid, although the specific numerical values may differ significantly from the true values. Possible differences in the variety of activities regarded as R&D, as well as differences in GNP accounting, make inter- national comparisons involving the U.S.S.R. particularly hazardous. cost, and effectiveness of R&D — as well as inconsistencies in GNP accounting — the measure is relatively gross. Interpretations of the indicator, therefore, should focus on general trends rather than specific numerical values. The fraction of the GNP of the United States devoted to R&D has declined steadily over the R&D Personnel The human resources involved in R&D provide another comparison of the magnitude of national R&D efforts. The number of scientists J See the chapter in this report entitled "Resources for R&D". 5 Information on the distribution of government R&D expenditures among these and other areas is presented in a later section of this chapter. Figure 1-2 Scientists and Engineers' Engaged in R&D per 10,000 Population, by Country, 1963-73 (Number per 10,000 Population) 40 35 20 .-»^ U.S.S.R. ^^ United States Japan West Germany ,••• France 1963 '65 '67 '69 '71 ' Includes all scientists antj engineers (full-time equivalent basis) Data for the United Kingdom are not available. SOURCE Organisation for Economic Co-operation and Development; U S.S.R estimates by Robert W. Campbell, Indiana University. and engineers in R&D per 10,000 population is shown in figure 1-2 for the United States, the U.S.S.R., Japan, West Germany, and France. (Data for the United Kingdom are not available.) This indicator should be treated only as an approximate measure of the level and intensity of R&D because it fails to account fully for certain factors, such as national variations in the designation of scientists and engineers and their productivity. The United States is the only major R&D- performing nation in which this indicator declined over the period studied." For each of the " The U.S. decline is due in large part to decreases in the employment of scientists and engineers in space and defense- related Ri&D. See the "Industrial R&D and Innovation" chapter of this report for further details. other countries, the number of scientists and engineers engaged in R&D increased at a faster rate than the population. The United States is also unique among these nations in that a decline occurred in the number of scientists and engineers involved in R&D; this number fell from 558,000 in 1969 to 523,000 in 1973. ^ By comparison, the estimated number of such personnel in the U.S.S.R. increased from ap- proximately 700,000 in 1969 to more than 900,000 in 1973. (See Appendix table 1-2.) Government-funded R&D Governments provide funds for R&D in a variety of areas such as national defense, space exploration, public health, and economic development. The distribution of funds among these areas indicates the relative emphases of the R&D programs of different countries. Government expenditures for R&D are classified by the Organisation for Economic Co- operation and Development (OECD) into the following categories: National Defmise, encompassing all R&D directly related to military purposes, in- cluding space and nuclear energy activities of a military character; Space, including all civilian space R&D such as manned space flight programs and scientific investigations in space; Nuclear Energy, consisting of all civilian R&D primarily concerned with nuclear sciences and technology; Economic Development, which covers R&D in a wide range of fields including: agriculture, forestry, and fisheries; mining and manufac- turing; transportation, communications, construction, and utilities; Health, encompassing R&D in all of the medical sciences, and in health service management; Community Services, which includes R&D for such purposes as pollution control, educa- tion, social services, disaster prevention, planning and statistics; and Advancement of Science, consisting of funds for fundamental research in government and private laboratories, and for research and science instruction in universities. " For more current data, see the chapter in this report entitled "Resources for R&D". Figure 1-3 Distribution of Government R&D Expenditures among Areas by Country, 1961-73 United States United Kingdom (Percent! IP^^""" 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 I — I — \ — \ — r — 1 — I — I — \ — I r I I I ' ' ' ^ National defense France (Percent) 10 20 30 40 50 60 Nuclear energy Japan 10 20 30 40 50 60 70 West Germany 10 20 30 40 50 l3) less than 0.5 percent. SOURCE. Organisation lor Economic Co operation and Development The percentage of total government funds going to each of these areas is shown in figure 1- 3.*' The United States differs principally from other nations in the relatively large percentage of R&D funds channeled to defense and space exploration (71 percent in 1971-72, the latest years for which such data are available for international comparisons), and the small percentages for the advancement of science and economic development." In general, government R&D funds in other countries (except the United Kingdom) were concentrated in the latter two areas; this applied particularly to Japan and West Germany. Changes in the distribution of government- funded R&D over the 1961-71 period were similar for each country. Defense-related R&D decreased as a proportion of the total R&D expenditures, whereas the fraction for the advancement of science and economic develop- ment generally increased, as did the percentage for health and community services. Overall trends suggest a relative shift from military R&D to areas of domestic concern and the advancement of science. (The magnitude of R&D expenditures for national defense, how- ever, increased in absolute terms in all countries other than Japan.) Differences between countries in the distribu- tion of their R&D efforts arise from a variety of factors, such as the extent of a nation's military commitments and variations in the roles of government and the private sector. The pattern of R&D expenditures shown in figure 1-3 is based upon funding by governments only and does not include the large expenditures by the private sector, due to the lack of comparable data. SCIENTIFIC RESEARCH This section presents indicators of the inter- national character of science and various measures of the magnitude and quality of scientific research in major nations. Indicators of magnitude are based upon the number of research publications from each nation in several fields of science. Quality indicators are developed from the international pattern of citations associated with these publications, as ' Data are not available for the U.S.S.R. ' For current information on the distribution of U.S. Government expenditures for R&D, see the chapter in this report entitled "Resources for R&D" well as from the distribution of Nobel Prizes among nations and scientific fields. The internationalism of science Science by its very nature is international. The phenomena studied, the methods of investiga- tion, and the validity of research findings are independent of national boundaries. Researchers from all countries can contribute to the body of scientific knowledge, with con- tributions assessed on their scientific merit, not the country of their origin. The internationalism of science is based upon and fostered by a wide variety of formal and informal arrangements. Foremost among these are the publication of research findings in widely circulated journals and books, international meetings, joint research efforts, and informal correspondence among scientists. In addition to these, governments frequently sponsor inter- national travel for scientists to consult and collaborate on research, and enter into formal bilateral agreements for scientific cooperation and exchange among nations. The international scientific community is also served by the International Council of Scientific Unions, which encompasses an array of associations for the advancement of science and the exchange of information. Finally, the United Nations has created specialized scientific agencies nearly global in scope, which foster international cooperation in science and which in turn provide models for similar regional organizations. International scientific literature. The inter- national dimension of science may be seen in one of its more fundamental forms in the perfor- mance of research and the publication of its results. Current research builds upon the extant body of scientific knowledge, which is the combined product of researchers from all countries. The dependence upon research per- formed in other nations is expressed, ap- proximately, by a large sample of the citations in published research reports to scientific literature of foreign origin. This indicator is shown in figure 1-4 for eight major fields of science and engineering, as well as for all the fields combined. i° The indicator is based upon data from the six major R&D- performing nations identified in previous sec- ^^ Indicators of the Quantity ami Quality of the Scifntific Literature, Computer Horizons, Inc., 1975 (A study commissioned specifically for this report). Figure 1-4 Percent of the Citations in Scientific Literature' Citing Countries Other than the Author's Own Country, by Selected Fields,' 1973 Percent of foreign citations 10 20 30 40 50 60 70 I I I I I I I Cfiemistry Ptiysics Biology and biomedical research Engineering Clinical medicine Mathematics Earth and space sciences Psychology Eight field total < Based on articles in 2,121 journals in the 1973 Science Citation Indel from ttie US . United Kingdom, West Germany, France, USSR,, and lapan. = See Appendii table 7a for the description of fields The social sciences are excluded because comparable data are not available, SOURCE Computer Horizons. Inc, Figure 1-5 Participation in International Scientific Congresses by the United States and Other Countries, 1960-74 Number of participants 80,000 J 60,000 All scientists y. 40,000 -J^ y Non-U.S. scientists 20,000 U.S. scientists ^ 1 1 1 1969-71 1972-74 tions. The figure shows that almost 60 percent of all citations in the scientific literature of these countries, for the eight fields as a whole, were to research of foreign origin. Participation in international congresses. International meetings provide opportunities for scientists to exchange information and ideas through personal contact with foreign researchers. Among these are the international scientific congresses of those organizations constituting the International Council of Scien- tific Unions. The numbers of scientists from the United States and from other nations who have attended these congresses in recent years are shown in figure 1-5. Although the attendance of U.S. scientists has increased throughout the period, attendance of foreign scientists has grown even more rapidly. In the 1972-74 period, non-U. S. scientists represented 75 percent of all participants. (Peaks in the attendance patterns are due to the larger number of congresses held in certain years). SOURCE National Academy ot Sciences Scientific literature Research reports published in scientific and technical journals are one of the more direct outputs of scientific effort. n Such reports add to the body of scientific knowledge and may stimulate further research. The findings of the research, in addition, may be used in a variety of practical applications, many of which are unan- ticipated at the time the research is done. Although the reports may vary considerably in their theoretical and practical importance, the critical review which usually precedes publica- " For discussions of publications as measures of the output of science, see: G, Nigel Gilbert and Steve Woolgas, "The Quantitative Study of Science: An Examination of the Literature", Samce Studies Vol, 4 (1974), pp, 279-294; Henry Menard, Science: Growth ami Change (Cambridge: Harvard University Press, 1971); and Derek ], deSolla Price, Little Science, Big Science (New York: Columbia University Press, 1963), 8 tion helps to ensure that the reports have some degree of scientific or technical significance. Indicators based on research reports, how- ever, have several limitations when used for international comparisons: the quantity of such reports may be influenced substantially by the journals selected for examination,' 2 by national customs regarding the publishing of research papers, by the availability of funds for preparing and printing papers, by journal refereeing and publishing policies, etc. These and other limitations provide good reason for caution in interpreting such indicators. The indicators presented in this section provide measures of: (1) the proportion of the world's research literature in selected scientific areas produced by the United States and other major research-performing countries; (2) the distribution of research literature among fields of science in each country; and (3) the influence of the literature produced in each field by each country. National origins of scientific literature. Estimates of the literature produced by researchers in each country were based upon counts of articles, letters, and notes published in some 500 journals covered by the Science Cilation Index IS.C.}.)'^ over the period 1965-73, supplemented by data from various abstracting services. !■• The journals included in the set were those which were most highly cited in the total 1965 literature, regardless of field. The national origin of the literature was determined by the country of the first author of each scientific paper. The results are presented in figure 1-6.^5 The United States produced a larger propor- tion of the 1973 scientific literature in this sample of 492 journals than any other country in these fields: physics, engineering, psychology, molecular biology, and systematic biology. In the fields of chemistry and mathematics, however. 12 The representativeness of a journal set only ap- proximates the representativeness of the articles themselves because of the varying sizes of journals and other reasons. The next Science Indiccilan report will examme this represen- tiveness in detail- 1-' Published by the Institute for Scientific Information, Philadelphia, Pennsylvania. i" For details of the sample and methodology employed, see JmUcalon of the Quantihi and Quality of the Scu-ntiftc Literature, Computer Horizons, Inc., 1975. (A study commissioned specifically for this report), 15 An analysis of 2,121 journals included in the Science Citation Inda for 1973 yields similar results in the ranking of nations within fields, but comparable data for the larger set of journals are not available for earlier years. the U.S.S.R. led all countries, with the United States following as the second largest producer.!" The overall position of the United States, relative to the other countries, has changed little since 1965, the initial year of this indicator. For the seven fields as a whole, U.S. scientists and engineers published more than did those of any other country, followed by Soviet scientists and engineers. The United Kingdom, in these terms, ranks a distant third, while France, West Germany, and Japan cluster at a somewhat lower level. The international position of the United States may be declining in the fields of chemistry, engineering, and physics. The U.S. share of the literature in each of these fields declined slightly in both 1972 and 1973, as shown in figure 1-6. Furthermore, the absolute number of publications in these areas was lower in 1973 than in some previous years.'' (These declines may be related to trends in the funding of research in the three fields, as presented in the "Basic Research" chapter of this report). Although attention was focused above on the six countries producing the largest number of scientific publications, several other nations contribute significantly to the world literature. 's The largest contributors among these in 1973 were: Australia Canada Czechoslovakia India Israel Italy Netherlands Poland Sweden Switzerland Each of these countries ranked among the first 10 nations in the number of 1973 research publications of at least one of the eight fields of science. National research profiles. Countries differ in the emphasis they place on various fields of scientific research. The relative number of 1^ The Science Cilation Index for 1973 and earlier years did not include a number of important U.S.S.R. chemistry journals; the U.S.S.R. share of the chemistry literatuie, therefore, may be underestimated. 1- Similar publication trends in these fields, found in another study, are presented in the "Basic Research"chapter of this report 18 These and all subsequent data on scientific literature were developed from an analysis of 2,121 of the journals in the 1973 Science Citation Index, as described in Indicators of the Quantity and Quality of the Scientific Literature. Computer Horizons, Inc., 1975 (A study commissioned specifically for this report). 9 Figure 1-6 Scientific Literature' in Selected Fields' as a Percent of Total Literature, by Country, 1965-73 CHEMISTRY PHYSICS (Percent of Total) 60 50 40 30 US.S.R. U.S. Other ,,.»* miiiiiiiiii**!*"""""""""'**'*'*"*"*"' W. Germany 'ttttttttt' (Percent of Total) 60 50 - 40 20 ■^ Japan UK J L other ■ III! iiiiiiiiiiiiiiii*" U.S.b.R. U.K. W. Germany ___^_^________,;i,*i;4*« _L Japan 1965 ■67 ■69 ■71 72 ^73 1965 ■67 69 PSYCHOLOGY ■71 ^72 '73 (Percent ot Total) ENGINEERING (Percent of Total) 60 50 20 U.S. other IIIIIIIIIIIIIIIIIIIIIIIIIIIIMIIIIII>*II>>II*'*'"** ..«' ^% 10 ^' U.S.S.R. ■ ^ ■■ I ' U.K.' ' Japan France U.S. 40 - 30 ; 10 W Germany ....••V«»«*\***"** T other ,1111111"" 11,1 M........ ""••" U.K. I •••••• ■ W. Germany .1 r 4 1....' 1965 71 '72 '73 1965 ■71 72 '73 J Includes article!, letters anil notes from the 492 scientific journals winch were most tieavily cited in 1965. = The social sciences are excluded because comparable data were not available. SOURCE: Computer Horizons. Inc. 10 Figure 1-6 continued MOLECULAR BIOLOGY (Percent of Total) 60 40 US Other ^■"■■•in,,,,,„, ,,,,,,, Ill iiiii'"" U.K. ___ france W.Germany U.S.S.R. jgpg^ :/miiiiii)l ,11 r-— "T — '-T — I — (Percent of Total) 60 SYSTEMATIC BIOLOGY Otfier uiner o ; .O'" \ US 1965 ■67 71 72 73 1965 France ^^^^^» W. Germany— -^'"^" 71 72 73 MATHEMATICS (Percent of Total) bU 50 - 40 - 30 """■•■1.... "'^^^ U.S. _ , ^- -X.^ U.S.S.R. 20 10 _ W. Germany ^A»»%H ***** ttj France J^P^" 1 , ,UK , 1965 71 72 73 SOURCE Computer Horizons, Inc. 11 publications in each field provides an ap- proximate profile of a country's research effort. These 1973 profiles, based on the 278,894 S.C.J. publications in seven fields, i° are shown in figure 1-7.20 for purposes of comparison, the profile of publications produced by all countries combined is shown also. The 1973 profile of the United States was most similar to that of West Germany and the United Kingdom in the relative proportion of the total literature in each field, although chemistry was emphasized somewhat more by the latter two countries. The profile of France's scientific research also resembles the United States, except for a smaller proportion of engineering research on the part of France and a larger fraction of literature in chemistry. The country with the profile which differs most from that of the United States in the literature studied appears to be the U.S.S.R. The life sciences (biology, biomedical research, and clinical medicine) represent nearly 55 percent of the U.S. literature compared with just over 20 percent of the Soviet scientific and technical literature; conversely, engineering and the physical sciences (chemistry and physics) ac- count for some 20 percent of the U.S. literature whereas the U.S.S.R. published almost 60 percent of its literature in these fields. Literature citations. The significance of a nation's scientific literature is more important than mere counts of publications. One indicator of quality is the recognition that the research reported was dependent on published accounts of earlier investigations. Such a "citation index" is based on the belief that the most significant literature will be more frequently cited. In support of this assumption are a number of studies which demonstrate high correlations between citation counts and other measures of '" These data employ a somewhat different taxonomy of fields of science than that used for the 492-journal set; see Appendix table l-7a for the detailed taxonomy of the fields described in Indicators of the Quantili/ ami Quality of the Scientific Lileralure, Computer Horizons, Inc., 1975. ^o Because of the way this broad sample was selected, some fields may be understated, such as Russian mathematics However, the scope of the Science Citation Index is determined by a 20-member, international editorial board consisting of two Soviet scientists; one is an expert in the information and documentation sciences area, the other is a mathematician. In recent years, the Science Citation Index has been expanded to include "0 percent of the 1,000 journals most highly cited by articles in some 2,100 journals and 100 percent of the 575 most highly cited. scientific importance, such as judgments of researchers in the field. -^ The quality of scientific research is far more difficult to measure than its quantity. The use of citation indicators is one such approach, but one which requires considerable caution. Some articles may fail to be noticed because scientists do not have access to them, although this characteristic of the availability of a nation's scientific literature is itself an important aspect of the internationalism of science. Articles may be heavily cited only for the criticisms they provoke, or because they deal with minor improvements in methodology. Authors in some countries may cite only a few outstanding references for reasons such as journal space limitations, while similar scientists in other countries may give more complete citations. The particular choice of a sample of journals to be examined can have an effect on international comparisons if countries do not have appropriate representation in the sample. Because some nations concentrate more on applied research than on basic research, they may be un- derrepresented in the scientific literature. The data source for this indicator was the Science Citation Index, as augmented for improved coverage of certain fields and countries, com- prising 2,121 journals — virtually all of those included in the S.C.J, for 1973. The index was created by comparing the actual fraction of the world's total citations in a given field with the expected proportion based on that nation's share of the total publications in that field. The resulting citation indices are shown in the table below for six fields. An index of 1.0 means that there were as many citations to a country's literature in the field as would be expected from its share of the world's publications; a larger index indicates a proportionally higher level of citation to the literature produced by a country than could be accounted for simply by the volume of its publications. -> See "Citation Analysis: A New Tool for Science Administrators", Science. Vol. 188 (1975), pp. 429-432; Jonathan R. Cole and Stephen Cole, Social Stratification in Science. (Chicago: University of Chicago Press, 1973); Eugene Garfield, "Citation Analysis as a Tool in Journal Evaluation", Science. Vol. 178 (1972), pp. 471-478; J. Margolis, "Citation Indexing and Evaluation of Scientific Papers", Science. Vol. 155 (1967), pp. 1213-1219; and C. Roger Myers, "Journal Citations and Scientific Eminence in Contemporary Psy- chology", American Psychologiit. Vol. 25 (1970), pp. 1041-1048. 12 Citation indices of selected scientific literatures^ by selected fields and countries, 1973 Citation Field Country indices 1.3 13 .6 .5 .5 .2 1.3 1.2 .8 .8 .6 .3 1.5 1.5 1.4 .7 .7 .4 1.4 1.0 .9 .8 .7 .6 1.1 1.1 1.0 1.0 .9 .8 1.3 1.0 .7 .7 .6 .3 The United States ranks first or ties for first place on this measure in each of the eight fields. The U.S. lead is greatest in physics, followed by the earth and space sciences. Each country tends to have higher citation indices for its own scientific literature than it has for the literature of other countries (see Appendix table l-7b). This is particularly true for the U.S.S.R. and France. The United States, Clinical medicine United States United Kingdom Japan West Germany France U.SS.R. Biology and biomedical research United States United Kmgdom Japan West Germany France U.S.S.R. Chemistry United States West Germany United Kingdom Japan France U.S.S.R. Physics United States West Germany United Kingdom France Japan U.S.S.R. Engineering France United States United Kingdom U.S.S.R. West Germany Japan Earth and space sciences United States United Kingdom Japan West Germany France U.S.S.R. on the other hand, cites its own literature less than other countries cite theirs, except in the fields of chemistry and physics where its domestic citation indices are higher than those of the other five countries. Nobel Prizes in science International prizes for scientific achieve- ment, although awarded to individuals rather than countries, provide a gross indication of the relative position of nations in scientific research. Foremost among such awards are the Nobel Prizes. These prizes were established by a bequest of Alfred Bernhard Nobel, and give international recognition to achievements in the fields of physics, chemistry, and physiology /medicine. 2 J The Nobel Prizes from the first year awarded, 1901, are shown in figure 1-8 in terms of the number awarded to scientists in each of five countries which together account for a majority of the awards, and in relationship to the population of these countries.-^ Data are presented by year of award which, on the average, is some 15 years after the time of the research itself. Scientists in the United States have received the largest number of awards over the 1901-74 period as a whole, surpassing all other countries since the 1931-40 decade. Prizes going to the U.S. scientists, however, declined after the 1951- 60 decade, primarily as a result of a smaller number of prizes in the field of physics. In relationship to population, however, U.S. scien- tists received a smaller fraction of prizes than the United Kingdom over the last three decades. ^^ 22 The relatively high citation ratios associated with the United States and the United Kingdom may reflect, in part, the growing use of English as the language of scientific publication. Nevertheless, when citations made by U.S. and U.K. authors were excluded from these indices, the United States still had the highest citation ratios for chemistry, physics, mathematics, and the earth and space sciences. -' Nobel, the Man and His Priifs. (Stockholm: Nobel Founda- tion, 1962). Nobel also established prizes in the fields of literature and peace. Later, in 1969, the Nobel Foundation instituted the prize in economics and since then, 4 prizes have been awarded to U.S. economists, and single prizes to economists in Austria, the Netherlands, Norway, Sweden, and the United Kingdom. In some other areas of science which are not within the scope of the Nobel Prizes, there are similar international distinctions awarded for eminent accomplishments; for example, the Fields Medal for Mathematics was established in 1936 and since that time, U.S. mathematicians have received 35 percent of the quadrennial awards, largely after 1958. -' The apparent decline in 1971-74 is partially explained by the shorter time interval covered in this period. 25 Other countries, such as the Netherlands and Switzerland, have received a greater number of Nobel Prizes in respect to population size than either the United Kingdom or the United States. 13 The number of awards in individual fields of science are presented in figure 1-9. Over the 1901-74 period as a whole, the United States has thelargest total number of awards in physics and in physiology/medicine, and is surpassed only by Germany in the number of prizes received in chemistry. In the most recent period, 1971-74, the United States received 56 percent of the awards in physics, 57 percent of those in chemistry, and 44 percent of those in physiology/medicine. These represent a smaller fraction of the prizes in each category than was received by the United States in the 1951-60 period. Figure 1-7 Percent Distribution of Scientific Literature' by Selected Field,' for Each Country, 1973 • = World averages UNITED STATES U.S.S.R. (Percent) 10 30 40 JAPAN WEST GERMANY ' Includes articles, letters and notes trom a sample of 2.121 journals. ' The social sciences are excluded because comparable data were not available. SOURCE Computer Horizons. Inc. Clinical medicine Biology & biomedical research Physics Engineering Chemistry Earth & space sciences Clinical medicine Biology & biomedical research Physics Engineering Chemistry Earth & space sciences Clinical medicine Biology & biomedical research Physics Engineering Chemistry Earth 8. space sciences (Percent) 10 FRANCE UNITED KINGDOM 14 Figure 1-8 a Nobel Prizes Awarded in Science, for Selected Countries, 1901-1974 (Number of Prizes) 30 I Figure 1-8 b Nobel Prizes in Science per 10 Million Population for Selected Countries, 1960-1974 (Number) 3.0 • • • 2.5 — • • • Germany ' • 2.0 • • _ ^..•* ''.^ United Kingdom ,•*._ 1.5 V '^••/>1 X / •, W United StatesX 1.0 -*•. > J» \ 1 ^W^ France ^r '*. .•*, \ • • ^W. ^^ •* « 1 • ^^^^^ • ' .5 ' -* X U.S.S.R. X \» *^w^>\ 1901- 1911- 1910 1920 1921- 1930 1931- 1940 1941- 1950 1951 1960 1961- 1971- 1970 1974 1 After 1945, excludes ttie German Democratic Republic. SOURCE The Nobel Foundation. Figure 1-9 Nobel Prizes Awarded by Field for Selected Countries, 1901-74 (Number of Prizes) PhysiCS 15 1 United States , United Kingdom ^ ■••• ^ • ••••■•••••••** • • J \ IvL 1901- 1910 '11- '20 '21- '30 ■31- •40 •41- ■50 '51 '61- •71- '60 '70 '74 (Number of Prizes) 15 Chemistry United Kingdom (Number of Prizes) Physiology/ medicine 15 United States^r^ I 10 ^ United Kingdom* 1 5 • • • • • -> ^r \Germany' •; .^^^i.'** T 1 -.L.-- 1 1 ••• 1901- 1910 '11- '20 •21- •30 '31- '40 •41- '50 '51- '61- '71- '60 '70 '74 ' After 1945, excludes tiie German Democratic Republic, SOURCE: Ttie Nobel Foundation, 15 TECHNOLOGICAL INVENTION AND INNOVATION This section presents indicators of inter- national trends in technological invention and innovation, as well as transactions in technology involving the United States. Indicators of inventions are based upon patent awards in the United States and abroad, and include the identification of areas of technology in which recent patenting activity by foreign countries in the United States was especially high. Innova- tion indicators are based on major new products of a technological nature, and include trends in the proportion of such innovations produced by each major nation, the time between invention and market introduction, and the "radicalness" of the innovations. Transactions in technology, measured in terms of international sales of technical "know-how", are used as an ap- proximate indicator of the relative state of U.S. technology. The "patent balance" Inventions of new and improved products and processes may represent actual or potential advances in technology. Those inventions which are of sufficient originality to be patented provide a basis for indicators of the inventive output of countries. The use of patent statistics for this purpose, however, has several limitations. Some inventions — even major ones — are not patented. And those which are patented vary greatly in their technical and economic significance, with only a small propor- tion of the total number of inventions ultimately reaching the market. In addition, the criteria for awarding patents differ from country to coun- try; not only does the rigor of tests for originality vary, but so does the extent of protection afforded by patents. The latter factors deter- mine the relative ease and value of obtaining patents in different countries. The number of patents granted in individual countries is not an adequate measure of inven- tiveness for purposes of international com- parisons. A more meaningful measure relates the number of patents granted to nationals with those granted to foreigners in each country. Such an index-" reflects the relative success of -" When applied to the United States, the index is the number of patents granted to U.S. nationals by foreign countries minus the number of patents granted to foreign nationals by the United States. countries producing inventions of sufficient potential significance to warrant international patent protection. Since it is generally more costly to obtain such protection, the index tends to focus on those inventions which are thought to be most important. Figure 1-10 presents the total number of patents granted to U.S. nationals by ten coun- tries (Canada, West Germany, Japan, U.S.S.R., United Kingdom, and five other European Economic Community countries, including Belgium, Denmark, Ireland, Luxembourg, and the Netherlands); the number granted to nationals of these countries by the United Figure 1-10 Patents Granted to U.S. Nationals by Foreign Countries and to Foreign Nationals by the United States, 1966-73 (Thousands) 50 U.S. balance 30 — 10 — _ US, patents to foreign nationals' 1966 •67 ■69 I 70 I 72 73 ' including Canada, West Geimany, Japan, Uniled Kingdom USSR Denmark, Ireland Luxembourg, and the Nettierlands SOURCE World Intplleclual Property Organization 16 States; and the resulting U.S. balance. These 10 countries were responsible for nearly 70 percent of all foreign patent transactions with the United States during 1966-73. (Data are not available for Italy, and are not reliable for France for use in this report). The "patent balance" of the United States fell by about 30 percent between 1966 and 1973, as shown in figure 1-10. The decline was due both to an increasing number of U.S. patents awarded to foreign countries and a decline (in 1973) in the number of foreign patents awarded to U.S. citizens. Overall, foreign patenting increased in the United States during the period by over 65 percent, and by 1973 represented more than 30 percent of all U.S. patents granted. This suggests that the number of patentable ideas of inter- national merit has been growing at a greater rate in other countries than in the United States. The United States has a favorable but declining patent balance with each country except West Germany and the U.S.S.R.^" (figure 1-11). The favorable balance with Japan has declined steadily since 1968, as its patenting of inventions in the United States increased some threefold. The U.S. balance with Canada dropped sharply after 1972 as a result of a 30 percent reduction in the number of patents granted by Canada to U.S. inventors. Foreign origin patents by product area. The rapid growth of foreign patenting in the United States has occurred in a broad spectrum of product areas and technologies. The number of such foreign patents granted in these areas can be used to identify the products and technologies in which the foreign impact is greatest. For this purpose, all U.S. patents granted during 1963-73 were assigned to 15 major product areas according to the probable areas of application of the invention.-** The percentage of foreign origin patents within each of these areas in 1973 is presented in the table below. In 1963, the proportion of foreign origin patents in 12 of the 15 areas was less than 20 percent; only one area — petroleum refining and extraction — had less than 20 percent foreign patents in 1973. In studies of more specific fields and technologies, the U.S. Patent Office has iden- tified a number of areas in which the foreign share of U.S. patents is particularly high and increasing rapidly.-" Listed below are some of these areas and the corresponding foreign share of patents during 1972: Areas Piezoelectric compositions Magnetic field responsive resistors Automatic transmissions Superconductors Vinyl halide polymers Ground effect machines Semiconductor internal structures Magnetic sound recording and reproducing structures Magneto-hydrodynamic generators Ignition timing controls Percent of U.S. patents to foreign countries 78 72 69 60 56 54 52 52 49 49 Japan, West Germany, and the United Kingdom received the greatest proportion of foreign patents awarded by the United States in these areas. Percent of total U.S. patents granted to foreign countries by major product area, 1973 Product area Percent Product area Percent Product area Percent Drugs and medicines . . Aircraft and parts Textile mill products . . . Chemicals, except drugs Primary metals 44 Food and kindred products 39 Machinery, except 37 electrical 35 Electrical equipment, 34 except communications . Professional and scientific instruments . . . Communication equipment and electronic components . . 33 30 29 29 28 Motor vehicles and other transportation equipment.. Rubber and miscellaneous plastics products Stone, clay and glass products Fabricated metal products . . . Petroleum refining and extraction 28 28 27 25 17 2' The U.S.S.R. accounted for only one percent of all the patent transactions considered. ^» Indualon of thi Patent Output of U.S. Industry, Office of Technology Assessment and Forecast, U.S. Patent Office, 1974 (A study commissioned specifically for this report). -" This information was taken from a series of reports of the Office of Technology Assessment and Forecast, U.S. Patent Office, April 1973-January 1975. 17 Figure 1-11 U.S. Patent Balance with Selected Countries, 1966-73 (Thousands) 20 18 - »***""•.., Canada 16 ^»' 12 - 10 - >>'• United Kingdom Other E.E.C, countries ' ■■"■iiiiii,,, 4 ^^ ^ ^^ Japan 2 V WestGermany ..^ -2111 *!*** I I 1966 '67 '68 '69 70 71 72 I Other European Economic Community (E E CI countries include Betgium, Denmarl(, Ireland, Luxembourg, and ttie Nettierlands Data are not available for Italy, and are not reliable for France for use in this study. SOURCE: World Intellectual Property Organization. International trends in technological innovation Technological innovation is a complex process culminating in the introduction of new and improved products and processes. Several steps are involved in bringing a new product into the market, including successful research and development which provide the technical and engineering foundation for innovation. Technological innovation is, in turn, one of the more important factors in determining the productivity, economic growth, and inter- national position of developed nations. •'° The indicators presented here concerning international trends in technological innovation are based upon a study conducted specifically for this report. The study investigated 500 major technological innovations (i.e., new products or processes embodying a significant technological change) which were introduced into the com- mercial market'i between 1953-73. The 500 innovations studied were those receiving the highest ratings among 1,300 major innovations produced by Canada, France, Japan, the United Kingdom, the United States^^ and West Ger- many. An international panel of experts rated the innovations based on their technological, economic, and social importance. J-' The present indicators should be interpreted with their several limitations in mind. The number of innovations on which the indicators are based is relatively small, particularly for countries other than the United States, with the result that the national trends presented are somewhat tenuous. Furthermore, only the most important innovations are represented by the indicators, even though the more numerous innovations of a less significant nature may have a greater overall impact. Moreover, the measures do not go beyond the initial introduc- tion of the innovations into the market and, thus, do not include information on factors such as the economic benefits accrued by the in- novating nations nor the international diffusion of the innovations. Finally, the indicators do not account for the negative impacts — such as job displacement, environmental pollution, or in- ^0 For further discussion of these relationships see: The Condilions for Success m Techtwlogical Innoviilimi, Organisation for Economic Co-operation and Development, 1971, and Robert Gilpin, Technology. Economic Growth, ami Internahonal Com- petiliveness. U.S. Congress, Joint Economic Committee, ^ath Congress, 2nd Session, 1975. ^' Some innovations were brought into the commercial market after havmg been first introduced in the government market. 32 The U.S. innovations are more fully analyzed in the "Industrial R&D and Innovation" chapter of this report. -" For further information on the methodology and results of the study see Indicators of International Trends in Technological Innovation. Gellman Research Associates, Inc., 1975. (A study commissioned specifically for this report). Other topics investigated in the study but not discussed here include: the characteristics of the innovating companies, the role of basic and applied research in the development of each innovation, and the utilization of patents and licensing in acquiring the technology associated with each innovation. 18 creased energy consumption — which may be associated with the innovations. The innovations included in the study repre- sent a wide range of product areas and industrial sectors. Examples of the innovations are listed below: Nuclear reactors Oral contraceptives Urethane foams Electron beam welding High voltage electric cables Automatic optical readers High speed electric trains Integrated circuits Lasers Weather satellites The innovations were classified according to the type of market which the innovating company intended for the innovation^"": producer goods, consumer goods, or the govern- ment (viewed as both a producer and a consumer market). The innovations in total were aimed principally at the producer-goods market (65 percent of all innovations), followed by the government (19 percent), and the consumer- goods market (16 percent). The following table shows the distribution of innovations among the three types of markets for each of the five countries:-'^ Percent distribution of innovations by type of market and country, 1953-73 Type of market Country Producer Consumer goods Government goods United States . . United Kingdom Japan West Germany . France 62 89 77 69 45 19 2 16 7 10 19 9 7 24 45 Major innovations by selected countries. The proportion of the 492 innovations produced by each of the five countries is shown in figure 1-12. The United States leads each of the other nations by a wide margin in the percentage of major innovations produced. The U.S. lead, however, declined steadily from the late 1950's to the mid- 1960's, falling from 82 to 55 percent of the innovations. The slight upturn in later years -' The innovation may have been introduced subsequently into other markets; e.g., innovations initially directed to the government may have been introduced later into another market -'-' Innovations originating in Canada were omitted from this report because they are small in number and therefore cannot be analyzed in detail. represents a relative rather than an absolute gain, and results primarily from a decline in the proportion of innovations produced in the United Kingdom, rather than an increase in the number of U.S. innovations. The largest actual gains were recorded by Japan, although its share of the innovations reached only some 10 percent by the early 1970's. The innovations as a whole covered a wide range of product areas, but U.S. innovations were concentrated primarily in the most R&D- intensive industries, particularly: electrical equipment and communications, chemicals and Figure 1-12 Major Technological Innovations, by Selected Countries, 1953-73 (Percentage of Total) 90 80 ^^ ^^^^^—^United States 60 \^_^^ 50 40 30 — 20 10 United Kingdom •*••, / •-• •-.. West Germany "^'" I!!rrr5^'^K^!^ — 1^ 19 53- 1956- 1959- 1962- 1965- 1968- 197 55 58 61 64 67 70 7 1- 3 SOURCE: Gellman Research Associates. Inc. 19 allied products, machinery, and professional and scientific instruments. In the United Kingdom, aircraft was the principal area in which the innovations were found, whereas those of West Germany were primarily in machinery. In- novations originating in Japan were most often in primary metals or in the broad area of electrical equipment and communication. French innovations were least concentrated, tending to occur in a variety of areas. Invention and innovation. The inventions (i.e., the first conception of the innovations) originate, for the most part, in the same country as the innovation; 91 percent of all the in- novations included in this study were based on domestic inventions. The proportion of each country's innovations which resulted from its own inventions ranged from a high of 100 percent in France to a low of 79 percent for West Germany, with the United States at 93 percent. The time between invention and innovation ranged from less than one year to 81 years among the present set of major new products and processes. The mean numbers of years in the invention-innovation interval are shown in figure 1-13 for the various countries. (It should be noted that the date of invention is often difficult to determine precisely). Figure 1-13 Interval Between Invention and Innovation, for Selected Countries, 1953-73 (Number of Yeats) 12 3 4 5 6 7 8 9 United States Japan West Gemany France United Kingdom 1953-62 ' M 1963-73 ' Refers to the date of the innovation - Sampie size does not affovv calculation of the time interval SOURCE; Gellman Research Associates. Inc. In the most recent period, 1963-73, Japan had the shortest period between invention and market introduction (3.6 years), followed by West Germany (5.6 years), the United States (6.4 years), France (7.3 years), and the United Kingdom (7.5 years). "Radicalness" of the innovations. Innovations may embody technologies which range from imitations of existing technologies to radical breakthroughs. To investigate this aspect, each innovation was classified by the innovating company into one of the following five categories: "no new knowledge required", "imitation of existing technology", "improve- ment of existing technology", "major technological advance", and "radical breakthrough". Only 22 of the 369 innovations for which such data were acquired were assigned to the first two categories; these innovations are excluded from the following analysis. The distribution of the remaining innovations among the other three categories is presented in figure 1-14. The largest proportion of innovations in the five countries combined were classified as major technological advances (37 percent), followed by improvements in existing technology (35 per- cent), and radical breakthroughs (29 percent). The innovations originating in the United States were a relatively balanced mix of the three types, whereas innovations of the United Kingdom were most often characterized as radical breakthroughs. France, West Germany, and Japan were similar in that their innovations were most often considered to be major technological advances. These indicators are particularly inexact for all countries other than the United States because of the small number of innovations involved. Furthermore, only the U.S. innovations were numerous enough to permit the determination of trends, which indicate that the percentage of radical innovations declined nearly 50 percent between the 1953-59 and 1967-73 periods, while those representing major technological advances doubled. The decline in radical innovations was due to a smaller number of such innovations from the electrical equipment and communica- tion, and the machinery industries. Technical "know-how" The extent to which nations purchase the technical "know-how" (e.g., patents, licenses. 20 Figure 1-14 "Radicalness" of Innovations, by Selected Countries, 1953-73 (Percentage ot each Country's Innovations) 10 I 20 I 30 40 I 50 I 60 I Improvement of existing technology Major technological advance Radical breakthrough United States United Kingdom France West Germany Japan SOURCE- Gellman Research Associates, Inc. and manufacturing rights) of a country is one indicator of the technological position of that country vis-a-vis other nations. Several other factors, however, may influence the volume of such purchases, such as the economic develop- ment policies of the nations involved and the trading arrangements among them. Information on payments and receipts for technical "know-how" is available for transac- tions between multinational companies and their foreign affiliates as well as between independent organizations. The latter informa- tion was selected for use primarily on the assumption that purchases by independent enterprises are more likely to be based on the technical merit of all available "know-how". The omission of transactions between corporations and their foreign affiliates, however, results in a substantial understatement of the extent of technology transferred. In addition, a significant amount of "know-how" is transferred through the exchange of technical and management personnel, and through informal agreements which are not reflected in the financial data presented here. The dollar value of U.S. receipts, payments, and the resulting balance (i.e., receipts minus payments) for exchange of technical "know- how" is shown in figure 1-15. Over the 1960-74 period, U.S. receipts from the sale of "know- how" grew exponentially while its payments grew more linearly, resulting in an increasingly large positive balance of payments in this area. Increases in the U.S. balance are due principally to purchases of U.S. "know-how" by Western Europe and Japan (accompanied by relatively small purchases of Japanese "know-how" by the U.S.). From 1970 onward, for example, nearly 45 percent of U.S. net receipts were associated with Japan, and 30 percent with Western Europe (including the United Kingdom). The developing countries are increasingly important purchasers of U.S. "know-how", accounting for 15 percent of the U.S. balance in 1974. U.S. purchases of foreign "know-how" are primarily from Western Europe. Approximately 80 percent of U.S. payments in 1974 went to these countries, with nearly 35 percent going to the United Kingdom alone. Although considerably more technical "know- how" appears to flow from the United States than to it, the volume of foreign technology acquired by the United States is substantial and expanding in various areas. Machine tools is one such area in which the advanced "know-how" of foreign countries has been acquired for use in the United States. In plastics, the European developments in polyethylene have impacted significantly on American industry. Imported technology and "know-how" have also had substantial influence in the optical equipment area. 36 PRODUCTIVITY AND BALANCE OF TRADE This section presents indicators of inter- national trends in productivity, as well as measures of the contribution of R&D to the U.S. balance of trade. Trends in the level of national ■*° International Economic Report of the President. Council on International Economic Policy, 1975. 21 Figure l-15a U.S. Receipts and Payments for Patents, Manufacturing Rights, Licenses, Etc., 1960-74 (Millions of Dollars) 800 700 , y 600 - us, Receipts^^^ 500 - J^ 400 - ^/y 300 ^^ ^;^i/ 200 U.S. Paytnents^^^ 100 - ^^^^ 1 1 1 1 ! 1 1 i 1 1 1 i 1960 '62 '64 '66 '68 70 72 Figure l-15b U.S. Net Receipts for Patents, Manufacturing Rights, Licenses, Etc., by Selected Countries, 1960-74 (Millions of Dollars) 74 300 250 Japan XX X • • 200 / ^r / ^^ 150 M ^^^^^^^r Western Europe g ^r^ .^^^V ^ ^ 100 K,^^^^^^ ^^^^ ^*'***^ / > *• — — ' ^^^^^^^^^^^ 50 ^ "" ^ ^^"^^^ Developing Nations 1 1 1 1 1 1 1 1 1 1 1 1 1 1960 '62 SOURCE US Department of Commerce. '68 '72 '74 productivity (i.e.. Gross Domestic Product per employed civilian) and grov^?th in manufacturing productivity (i.e., output per man-hour) are presented for each major developed country. An approximate indicator of the role of R&D in the U.S. trade balance is developed through an analysis of U.S. exports and imports of manufac- tured products, in terms of the R&D intensity of the products involved. The indicator is used also to determine the balance of trade in R&D- intensive products between the United States and other specific nations. Productivity The level of productivity and its rate of growth can greatly influence the economic strength of nations and affect living standards, costs and prices, and international trading and monetary arrangements — as shown by the experience of many countries in recent years.-'" Productivity expresses the relationship between the quantity of goods and services produced (output) and the quantity of labor, capital, land, energy, and other resources (input) used to produce them. Over time, productivity tends to grow as new knowledge and new technology are embodied in capital investments, as the educational levels of labor forces rise, and as management skills become more effective. While the effect of R&D on productivity growth is not known precisely, the general conclusion based on a large number of studies is that the impact of R&D is "positive, significant, and high".-'* The measurement of productivity is difficult, particularly when measures are sought for the purpose of international comparisons. Problems arise from a diversity of sources, such as differences in concept and methodology and the availability of data. For these reasons, small reported differences in productivity — between nations and over short periods — may not be significant; interpretation of the indicators, therefore, should be confined to general trends. A relatively general and approximate measure of productivity is the "real Gross Domestic ^" Information on the role of productivity in the inter- national area may be found in Productivit)/: An Internalional Perspfclirc. U.S. Department of Labor, Bureau of Labor Statistics, 1''74. ■'* Reiearch and Development and Economic Growlh/ Produclivity, Papers and Proceedings of a colloquium, National Science Foundation (NSF 72-303). For a discussion of this relationship, see the chapter entitled, "Industrial R&D and Innovation" in this report. 22 Figure 1-16 Real Gross Domestic Product per Employed Civilian, for Selected Countries Compared with the United States, 1960-74 (Indexes, United States = 100) 100 90 80 70 60 40 — 20 United States United Kingdom ^^ ^ Japan J_ I I I I 1960 '65 SOURCE: U.S. Department of Labor. 72 Product per employed civilian". Measured in these terms, the level of U.S. productivity exceeded that of France, Japan, West Germany, and the United Kingdom throughout the 1960- 74 period (figure 1-16). Gains in productivity, however, were larger in the four other coun- tries, with the result that the U.S. lead diminish- ed significantly. By 1974, the productivity levels of France and West Germany were only 20-25 percent lower than the United States. Japan gained the most in productivity, but was still some 40-45 percent below the U.S. level in 1974. Trends in productivity are more commonly measured in terms of output per man-hour. The use of this index does not imply that labor alone is responsible for productivity growth; output per man-hour may also be influenced by factors such as technological advances, scale of produc- tion, and management effectiveness. This index is developed for each country separately, and is used to measure the change in productivity over time in that country; it does not permit comparisons of the actual productivity levels of different countries. This indicator is presented in figure 1-17 for manufacturing industries in the five countries. The U.S. productivity gain between 1960-74 is the smallest of these five countries (60 percent) Figure 1-17 Productivity' in Manufacturing Industries, by Selected Countries, 1960-74 (Index 1960 = 100) 400 1 1 350 f f Japan | 1 1 / 300 / i i i 250 1 1 i m France^ 200 # ^^^ W. Germany 150 • ••^^^^-^•••••-^^ 100 1960 '61 '62 '63 '64 '65 '66 '67 '68 '69 '70 '71 '72 '73 '74 (Pre.) ' Output per man-hour. SOURCE: U.S. Department of Labor. 23 and nearly five times less than increases in Japan, which recorded the largest gains. However, starting from a relatively high level of produc- tivity in 1960, the United States might not be expected to sustain the same high proportional gains as countries starting from a lower produc- tivity base. The effectiveness of a nation's productivity level is perhaps best indicated by the measure of "unit labor cost" (i.e., hourly labor costs divided by output per man-hour). If gains in productivity exceed increases in the cost of labor, then unit labor costs drop, products can be produced at less cost, and sold at lower prices, placing a nation in a favorable competitive position in the inter- national market.-'" Trends in this index for manufacturing industries are shown for the five countries in figure 1-18. It can be seen that productivity gains in the U.S. were sufficient to offset increases in labor cost from 1960 through 1965 and again from 1970 through 1973. Productivity rises in 1974 were negligible, however, while hourly labor costs had the largest yearly gain of the entire period. As a result, unit labor costs in manufacturing industries rose more rapidly than in any other year since World War II. Gains in hourly compensation in 1974 ex- ceeded advances in productivity in other coun- tries also, and by even wider margins than in the United States. Thus, unit labor costs increased to an even greater extent in foreign manufac- turing. The 1973-74 increase in Japan was nearly 30 percent and in the United Kingdom nearly 20 percent, both of which were the largest year-to- year gains in unit labor costs experienced by these countries during the 1960-74 perit^d. Figure 1-18 Unit labor cost' in Manufacturing Industries, by Selected Countries, 1960-74 llndex 1960 = 100) 230 220 210 200 190 180 170 160 150 140 130 120 110 100 90 ^1 Japan ^ t W. Germany V«* •A* >^- --/.•• I I I I I I \ I \ I 1 L I960 '61 '62 '63 '64 '65 '66 '67 '68 '69 '70 '71 '72 '73 '74 (Pre.) ' In national currency unadiusted for inflation. SOURCE: US. Oepartment of Labor Balance of trade in R&D-intensive products The U.S. position in world trade depends upon a variety of factors, including the price of its products, the effectiveness of its international marketing, trading arrangements with other countries, and its performance in technological innovation. Such innovation, as discussed elsewhere in this report, depends significantly upon research and development. -'" For a discussion of recent trends in these factors, see Patricia Capdevielie and Arthur Neef, "Productivity and Unit Labor Costs in the United States and Abroad", Monthly hihor Ra^iew, July 1975; for a detailed analysis of the role of these factors in international trade, see Compeliliveness of U.S. Industries. United States Tariff Commission, 1972. The precise role of R&D and technological innovation in U.S. trade have not been deter- mined, although recent studies suggest that it is substantial. ""o Some indication of this is provided by analyzing the U.S. trade balance in terms of the products involved, with the latter classified according to the relative level of R&D invest- ment of the industries which produce the products. For this purpose, products from industries-*' with (a) 25 or more scientists and ^° Raymond Vernon (ed). The Teihitology f:ictor in Inffr- national Trade, (New York: Columbia University Press, 1970). Federally Funded Research and Development Centers administered by universities SOURCE: National Science Foundation. Scientific and technical information (S&TI) activities consist of: (1) documentation, reference, and information services; (2) publica- tion and distribution; (3) symposia and audio visual media; (4) R&D in information sciences; and (5) information systems, techniques, and devices. Federal support for these activities increased six fold over the 1960-74 period i" (figure 2-17); in terms of constant dollars, however, 1968 was the year of highest funding followed by a leveling off through 1974. The ratio of total S&TI obligations to Federal R&D grew from .010 in 1960 to .025 in 1970, and remained approximately at that level through 1974 (figure 2-17). Several agencies support programs in this area. Those which account for most of the Federal support are indicated in figure 2-18, which presents the obligated funds from each. The Department of Defense, through such programs as the Defense Documentation Center, supplied one-third of all Federal funds for S&TI in 1974. The Department of Com- merce provided 20 percent of the total, much of which is accounted for by the National Technical Information Service. A similar amount comes from a variety of programs in the Department of Health, Education, and Welfare, a major one being the National Library of Medicine. Figure 2-17 Federal Obligations for Scientific and Technical Information Activities, Compared with Total Federal R&D Obligations, 1960-74 (Millions of Dollars) (Percent) 500 400 - Current dollars „i>«' Ratio of S&TI to — all Federal obligations - (Ratio) .030 .020 .010 1980 '62 '64 '66 '68 '70 '72 '74 (est.) lalGNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE; National Science Foundation, '" Available data reflect only a portion of the Federal support for all S&TI activities, in that they include only the direct obligations for S&TI and not the support provided through R&D grants and contracts. 47 Figure 2-18 Federal Obligations for Scientific and Technical Information Activities, by Selected Agency, 1960-74 Current dollars Constant 1967 dollars' (Millions of Dollars) 160 (Millions of Dollars) 160 ^ Vllllllll / ,*, ,»''' Librarv of Congress / ..-•• -X'** •...• 1960 '62 72 74 1960 '62 '64 (est.) i^'GNP implicit price deflators user! to convert current dollars to constant 1967 dollars SOURCE National Science Foundation, 72 '74 (est,) 48 Basic Research 49 Basic Research INDICATOR HIGHLIGHTS The Nation's total expenditures for basic research rose continually during the 1960- 74 period in current dollars; in constant 1967 dollars, funds for basic research in 1974 were equal to the 1965 level, and almost 13 percent lower than the peak year of 1968. Universities accounted for approximately 55 percent of the Nation's total expenditures for basic research in 1974 (versus 37 percent in 1960), followed by the Federal Govern- ment and private industry at some 15 percent each, and other sectors with the remainder. The Federal Government provided the largest share of support for basic research during the 1960-74 period, increasing from nearly 60 percent of all such funds in 1960 to almost 70 percent in 1974; industry's share declined from 28 percent in 1960 to 15 percent in 1974, and the universities' share increased from 6 to 11 percent over this period. Funds provided by the Federal Government for basic research increased each year (except for 1971) in current dollars, but declined 13 percent between 1968 and 1974 in constant dollars; the largest reductions in constant dollars were recorded in the physical sciences which declined ap- proximately 25 percent between 1969 and 1974. University expenditures for basic research (from all sources of support) rose con- tinuously in current dollars between 1960- 74, but declined some 5 percent in constant dollars between 1968 and 1974; this decline is due to reduced growth of Federal support in combination with inflation. Basic research expenditures by academic institutions in 1974 were concentrated in the life sciences (51 percent of all expen- ditures), engineering (12 percent), physical sciences (13 percent), social sciences (8 percent), and the environmental sciences (7 percent). Federal support for basic research in univer- sities, which accounted for 70 percent of all such funds in 1974, increased in current dollars between 1964-74 in the broad fields of science and engineering; the level of research effort as reflected by constant dollar expenditures, however, was lower in each field in 1974 than in previous years, with the largest reductions occurring in engineering and the physical sciences. Federal support for universities in 1974 was provided primarily through six agencies — NSF, HEW, DOD, USDA, AEC, and NASA — with no more than two agencies supplying at least 70 percent of all Federal basic research support in each major field of science; the NSF provided either the largest or second largest amount of funding among these agencies in each field. Expenditures for basic research per scientist and engineer in doctorate-granting in- stitutions were almost 30 percent lower in constant dollars in 1974 than in 1968; the largest decline was in physics, where reduc- tions were nearly 40 percent from 1966 to 1974. Federal laboratories accounted for 16 per- cent of the total national expenditures for basic research in 1974; current dollar expenditures by these laboratories increased throughout most of the 1960-74 period, but the level of research effort in terms of constant dollars was some 20 percent lower in 1974 than in 1970, the year of highest real expenditures. Private industry was responsible for 16 percent of the total national expenditures for basic research in 1974; although current dollar expenditures have risen, particularly since 1972, inflation reduced real expen- 50 ditures in 1974 to approximately the same level as 1961. D The number of research publications from major fields of science increased generally throughout the 1960's, but leveled off in several fields in the early 1970's; publication output in chemistry, engineering, and physics, for example, has remained at a nearly constant level in recent years. D Universities are by far the largest producers of published research reports with some 75 percent of the total in 1973, followed by the Federal Government and private industry with approximately 10 percent each, and other nonprofit institutions with 5 percent. D Basic research contributes increasingly to technological innovation, as reflected by the growing number of citations to research in patents associated with major advances in technology; the frequency of such citations increased 17 percent between the 1950's and 1960's, while citations to other patents declined by almost 25 percent. □ Research performed in universities is most frequently cited as the origin of patented technological advances, accounting for almost 55 percent of the cited research in recent years and replacing industry as the prime sector in which such research is performed. Basic research is the quest for fundamental understanding of man and nature, in terms of scientific observations, concepts, and theories. Such research is generally motivated by curiosi- ty and the desire to advance scientific knowledge, with the opportunities for its advancement determined primarily by the existing state of scientific understanding itself, rather than by practical need or potential application. As an activity, this research ranges from efforts of teams of scientists working with large facilities such as particle accelerators to the efforts of individual scientists using little or no research equipment. And basic research, being international in its nature, joins the activities of scientists from many countries. ^ Although curiosity is frequently the prime motive of the individual scientist for performing research, potential applications often underlie the private and public support of basic research. There is as yet, however, no method for correlating the cost of such research with its total returns — intellectual, economic, and social. But the many and varied uses of basic research suggest that the benefits may be substantial, particularly in comparison with the relatively small investment involved. The findings of basic research represent much of the objective knowledge of the physical and social world which forms a major part of the educational ' For further discussion of international aspects of science, see the chapter entitled, "International Indicators of Science and Technology" in this report. curriculum of the general population, while both the results and the conduct of such research constitute the core of advanced education in the sciences and engineering. Basic research provides the fundamental knowledge on which modern technology increasingly depends. This research, in addition, supplies indispensable knowledge for planning and directing the rest of the R&D effort. Finally, the maintenance of a wide spectrum of basic research can provide the new knowledge needed for responding to challenges in the future — challenges which may not be foreseen at present. Indicators of the state of basic research presented in this chapter consist largely of the financial resources committed to research and preliminary measures of outputs and their application in industrial technology. The "input" indicators provide information on national expenditures for basic research, the extent of research performed in universities and other sectors, and trends in expenditures for basic research in the various fields of science. "Out- put" indicators include publications of scientific research produced by different sectors in major fields of science, and measures of the extent to which such research underlies advances in technology. The present set of indicators are deficient in a number of major aspects. They do not encom- pass substantive aspects of basic research, such as advances in knowledge achieved in the various scientific disciplines. The indicators, further- 51 more, do not identify the wide applications made of the results of this research. Nor do they represent the economic and social returns from the varied uses made of its cumulative findings. The present indicators, in addition, do not include measures of the effectiveness, or productivity, of the research activity. Besides these deficiencies, there are other limitations in regard to the data used for the present indicators. There is, for example, uncertainty regarding the precision with which "basic" research can be distinguished from "applied" research. A particular research effort may be identified as basic or applied, depending on whether the classification is made by the sponsor of the research or by the organization performing it. Furthermore, differences among sectors in the assignment of costs to basic research make it difficult to compare expen- ditures and the magnitude of research efforts among the sectors. Industrial firms, for example, include in their reported expenditures for basic research an annual depreciation cost of the facilities used in the research; universities and Federal laboratories do not. The construction costs of large. Government-financed research facilities such as the National Accelerator Laboratory are not included as basic research expenditures, whereas NASA, in figuring the costs of research using expendable space probes, includes the costs of spacecraft and launch vehicles (in compliance with NSF reporting requirements). Figure 3-1 Basic Research Expenditures, 1960-74 (Dollars in Millions) $4500 4000 3500 — current dollars 3000 2500 2000 1500 1000- 500- 1960 I 62 I I I I 64 66 I I 70 74 (est.) ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE National Science Foundation RESOURCES FOR BASIC RESEARCH The Nation's total expenditures for basic research increased continuously during the 1060-74 period, rising from $1.2 billion to $4.0 billion in current dollars (figure 3-1). In recent years, however, this growth has not been large enough to offset the eroding effect of inflation. As a result, the actual level of basic research activity — as reflected approximately by expen- ditures in constant dollars — peaked in 1968 and declined in subsequent years.- By 1974, expen- ditures for basic research were at their 1965 level in constant dollars, and 13 percent less than in 1968. - The use of constant 10o7 dollar expenditures to approximate the level of research activity is discussed in the preceding chapter entitled, "Resources for Research and Development." The proportion of all R&D expenditures reported for basic research has remained essen- tially constant at some 13 percent since 1965, after rising during the early 1960's.-' Expenditures by performer There are four major sectors of the research community which perform basic research: private industry. Federal laboratories, univer- sities and colleges (and the Federally Funded Research and Development Centers they ad- minister), and other nonprofit institutions which conduct R&D. Because these sectors have differing missions and purposes, two different definitions of basic research are used for data -' Nitlwtujl Pullfrui ol R&D Resourcci, I^S-J-ZS, National Science Foundation (NSF 75-307). 52 collection. For all but the industry sector, the definition of basic research stresses that such activity be 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. "^ For the industrial sector, to take account of an individual company's commercial goals, basic research is defined as ". . .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. "-i ' Ihul The varying levels of basic research expen- ditures from 1960 to 1974 are shown in figure 3- 2 for the R&D-performing sectors. It should be noted that the growth in current dollar expen- ditures between 1908-74 was not sufficient to compensate for inflation in any of these major sectors. Constant dollar expenditures for basic research leveled off in the late 1960's for most sectors, and fluctuated around that level in subsequent years. The largest proportional declines between the year of peak funding and 1974 were in industry (31 percent), whereas the smallest percentage decline (9 percent) occurred in universities and colleges. Figure 3-2 Basic Research Expenditures, by Performer, 1960-74 Current dollars Constant 1967 dollars' (Dollars in Millions) 4500 4000 (Dollars in Millions) 3500 h- 3000 2500 — 2000 1500 1000 ' GNP impNcit price deflators used to convert current dollars to constant 1967 dollars. - Federally Funded Research and Development Centers administered by universities SOURCE: National Science Foundation 53 The distribution of the total funds expended for basic research changed significantly among the sectors during the 1960-74 period. The fraction of the total accounted for by universities and colleges increased from 37 percent in 1960 to 54 percent in 1974, while industry's fraction fell from 32 to 16 percent. There was little change in the distribution of such expenditures among the other sectors. Basic research support by source of funds The sources of expenditures for basic research are the Federal Government, industry, univer- sities, and nonprofit institutions. Funds supplied for such research by these sources are shown in figure 3-3. Basic research support from all sources increased in current dollars throughout most of Figure 3-3 Basic Research Expenditures, by Source, 1960-74 Current dollars Constant 1967 dollars ' (Dollars in Milllonsl (Dollars in Millions) 4200 " '' 3900 y - 3600 ^ - 3300 Total ^^ 3300 - 3000 / 3000 Total ^^ "^^^^^ V 2700 M Federal ^^ ^r 2700 / X 2400 / /^"^^^"^ 2400 t Federal 2100 / y 2100 / y^ ^^^ "^ 1800 / /^ 1800 / / X 1500 / / 1500 / / 1200 / / 1200 / 900 / 900 / 600 300 _ Industry ^^"^ ^ -*- — Universities ^^ 600 300 ~ Industry - and colleges =^,^^^ . .^4^^^» Nonprofit institutions n 1 1 1 1 1 1 1 1 1 1 1 «• and colleges "^^^^.. --■>••• ?,^" . • • • Nonprofit institutions 1 1 1 1 1 1 1 1 1 1 1 1 1 19 60 '62 '64 '66 68 70 72 7 (es 4 19 t.) 60 '62 64 '66 '68 '70 '72 1 (es 4 t.) " GNP implicit price deflators used to convert current dollars to constant 1967 dollars - Includes state and local government sources. SOURCE' National Science Foundation. 54 the 1060-74 period, although the annual in- crements were smaller after the late 1960's — the same years in which inflation grew fastest. As a result of these trends, funding by all sources except nonprofit institutions declined in con- stant dollars with the largest absolute reductions occurring in Federal Government support. Funds from this source in 1974 were down 16 percent in comparison with the peak funding year of 1968. Funds supplied by universities* continued tooutpace inflation through 1972, but declined more than 13 percent between then and 1974. Industry's funding for basic research peaked in 1966 in constant dollars, then fluc- tuated around a somewhat lower level through 1974. Universities, on the other hand, raised their share of support from 6 percent in 1960 to 11 percent by 1974. Federal support, as a percentage of the total national expenditures, increased from 59 percent in 1960 to a high of 72 percent in 1967 before declining to 68 percent of the total in 1974. Federal support of basic research The Federal Government assumed prime responsibility for support of basic research after World War II. This policy recognized the decisive role played by scientific knowledge in the war effort, and sought to strengthen the Nation's basic research capability for peacetime pursuits. Over the past 30 years, the policy has come to be predicated on the broad and varied role of basic research in advancing the country's defense, economy, health, and technology, as well as upon its general cultural value, in education and in the intellectual life of the Nation. During this period, many Federal agencies came to support basic research as an instrument in fulfilling their missions, and a new agency — the National Science Foundation — was created for the ex- press purpose of supporting scientific research and strengthening such capability. 5 Includes funds from State and local governments, as well as the universities and colleges themselves. " Federal obligations for basic research may differ from federally provided expenditures in the same year for a number of reasons. A sector which performs research, for example, may report expenditures for research projects which it regards as "basic research," whereas the Federal agency providing the support may report the same projects as consisting of "applied research." In addition, obligations made in a given year may actually extend over several later years in terms of the availability of the funds for expen- diture. Moreover, the withholding of obligated funds may produce discrepancies between obligations and reported expenditures. Six of these agencies accounted for 95 percent of all Federal obligations" for basic research in Fiscal Year 1974." Percent of total Federal obligations for basic research, by agency, 1974 Federal agency Percent basic National Aeronautics and Space Administration (NASA)' 29 Department of Health, Education, and Welfare (HEW) 23 National Science Foundation (NSF) . 16 Atomic Energy Commission (AEC) .. 11 Department of Defense (DOD) 10 Department of Agriculture (USDA) . 6 Basic research and total R&D. Basic research funded by each of these Federal agencies, and performed intramurally or by other sectors, is a part of the overall R&D effort of that agency. The magnitude of the basic research component, in relationship of the total R&D program, suggests the relative importance assigned to basic research by the agency. This ratio is shown in figure 3-4 for each of the six agencies. For all agencies as a whole, the ratio has increased slowly, reaching 15 percentof all R&D obligations in 1974. Obligations for basic research increased 20 percent between 1971-74, compared with a 14 percent increase for all R&D obligations. The NSF has the largest ratio by far, as would be expected in view of its designated role in the support of basic research. Recent declines in this agency's concentration on basic research — down from just over 90 percent of its total R&D obligations in the mid-1960's to approximately 80 percent in 1974 — are due to initiation of such new and largely applied research programs as "Research Applied to National Needs." Two other agencies — NASA and HEW — show sizable changes in recent years in the fraction of their total R&D expenditures which is directed to basic research. The fraction for NASA has fetieral fundi [or Research, Devt'lopmeut, ami Other Scientific AcUvtlies, Fiscal Yean 1973. 1974 ami 1975. Vol. XXIII, National Science Foundation (NSF 74-320-A). * NASA considers all of its activities to be R&D, or in support of R&D. The agency's obligations for basic research (as well as for applied research and development) include the related costs of spacecraft, launch vehicles, tracking and data acquisition, and the pro rata costs of ground operations and agency administration. 55 Figure 3-4 Federal Obligations for Basic Research as a Percent of Each Agency's R&D Obligations, by Agency, 1960-74 Percent (basic research) 100 70 50 40 •>,^' NSF ^ , USDA HEW DOD Other agencies J \ L J I \ L I960 62 '64 '86 '58 70 72 74 (est.) SOURCE: National Science Foundation ranged from a low of 11 percent in the mid- 1960's to some 25 percent in the 1972-74 period, with much of the latter growth coinciding with reduced obligations for the manned-space program. Basic research obligations by HEW show a long-term decline, as a percentage of the agency's obligations for all R&D; increase in life sciences research "targeted" toward specific disease areas accounts in part for the declining fraction of basic research obligated in recent years by this agency. Basic research obligations. Obligations for basic research alone are shown in figure 3-5 for each of the six agencies, as well as for all other agencies combined. Current dollar obligations were higher in 1974 than 1973 in each of the six agencies other than DOD and NASA. In contrast, constant dollar obligations declined in all agencies other than HEW. The principal scientific disciplines supported by each of these agencies" and the agency missions which generated the need for basic research in 1974 were: NASA. The physical and environmental sciences receive some 75 percent of all NASA's basic research obligations, primarily in connection with lunar and space explora- tion. HEW. Some 80 percent of HE W's obligations for basic research are directed to the life sciences, principally for biomedical research, and almost 6 percent to the social sciences for research in areas such as education and drug abuse. NSF. Over 30 percent of this agency's basic research obligations are for the physical sciences, with 23 percent for the en- vironmental sciences, lb percent for the life sciences, and 11 percent for engineering. The broad purpose of the research is to advance the state of basic scientific knowledge. AEC. The physical sciences receive almost 80 percent of AEC's basic research obligations and the life sciences nearly 13 percent — principally in high energy physics and in nuclear sciences. The purpose of this research is to generate the foundation for the development and utilization of atomic energy. DOD. Engineering accounts for 29 percent of DOD's obligations for basic research, physical and environmental sciences 22 percent each, and the life sciences about 12 percent. The prime aim of the research is to provide the fundamental knowledge needed for developing future military systems and improved operations. USDA. The life sciences receive some 70 percent and the physical sciences nearly 15 " federal Funds /or Research, Devehprnent, and Other Siientific Adwities. Fiscal Years 1973, J 974 and 1975, Vol. XXIII, National Science Foundation (NSF 74-320-A). 56 Figure 3-5 Federal Obligations for Basic Research, by Agency, 1960-74 Current dollars (Millions ol Dollars) 800 750 - l-\ 700 - 1 650 " NASA" X 600 " 550 - y^ 500 - J • 450 -- 1 HEW / 400 — 1 • 350 1 ./■ 300 — 1 >-i^^fe>*i: 250 1 >s / / >^ 200 — ^t, \r / NSF 150 i •' ^^ Other ,^. ,.•*** # / 100 F- • ^^09^^^^^ 0^ ^ ^^if^ USDA 50 ^«»*"' »«***** 1 1 1 1 1 1 1 1 1 1 1 1 1 1960 •64 ■66 ■68 ■74 (est.) Constant 1967 dollars' unions of Dollars ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars -Ttie large amounts reported by NASA for tiasic researcti are due to the substantial cost of support equipment sucti as spacecraft and launcti veliicles peculiar to space exploration, and tlie statistical proration ol costs for tracking and data acquisition. SOURCE National Science Foundation. percent of all USDA's basic research obligations, as a part of the agency's R&D aimed at improving animal and plant produc- tivity and enhancing the use of natural resources related to agriculture. The proportion of total Federal obligations for basic research provided by each of these agencies shifted considerably in the period 1960-74. The Department of Defense provided 28 percent of the total basic research obligations in 1960, compared to 10 percent in 1974. This decline may be due, in part, to the "Mansfield amend- ment" which restricted the DOD to the funding of research related directly to its mission. The proportion supplied by NASA declined from a high of 33 percent in 1964 to 29 percent in 1974, reflecting both changes in the mission of this agency and the faster growth of basic research obligations in some other agencies. 57 Accompanying the declines in DOD and NASA were recent increases in the fractions provided by HEW and NSF, with the former accounting for 23 percent of total obligations in 1974 (versus 17 percent in 1960), and the latter 16 percent (versus 11 percent in 1960). Much of the growth in HEW's share during the period occurred in 1973 and 1974 in connection with increased funding for cancer research; similarly, a large part of the growth in NSF's share took place in the years after 1970, as a result of increasing obligations for basic research in virtually all major scientific disciplines. Basic research obligations in scientific areas. An overview of the distribution of Federal support for basic research by scientific area is presented in figure 3-6.^° The five broad areas shown in the figure accounted for 95 percent of If See Appendix table 3-6 for disaggregated data for certain disciplines and Appendix table 3-6a for a listing of the scientific disciplines encompassed in these broad fields. Figure 3-5 Federal Obligations for Basic Research, by Selected Fields of Science, 1960-74 Current dollars Constant 1967 dollars' (Dollars in Millions) 900 800 (Dollars in Millions) Physical sciences 700 600 500 400 300 200 100 800 / / V' / r Engineering ^^^^ ^— >^.^^ Environmental sciences jT'^ / ^- ^'^' s^j- \.^ Social sciences I i I I L 100 - • »••«..»»** Engineering ,^^ .^ Social sciences T' I I I960 '62 '64 '66 '68 70 72 74 (est) ' GNP implicil price deflators used to convert current dollars to constant 1967 dollars. SOURCE National Science Foundation. I960 '62 '64 '66 '68 70 72 74 (est.) 58 Federal obligations for basic research in Fiscal Year 1974. Three of the areas— life, physical, and social sciences — had reached their highest level of current dollar obligations in 1974, whereas obligations for the environmental sciences and engineering declined after 1972. In constant dollars, basic research obligations for all areas other than the life sciences were lower in 1974 than in some previous year. The largest decline occurred in the physical sciences, where con- stant dollar obligations decreased by 24 percent between 1969 and 1974. A major and rapid shift in the distribution of basic research obligations among these areas of science occurred in the life and physical sciences. The proportion of obligations for the life sciences increased from 27 percent of the total obligations in 1969 to 34 percent in 1974. Over the same period, the fraction of total basic research obligations for the physical sciences dropped from 39 percent in 1969 to 32 percent in 1974. This shift from the physical to the life sciences is due to reductions or relatively slow growth in basic research obligations from DOD, NASA, and the AEC — the major sources of funding for the physical sciences — coupled with substantial increases in HEW's obligations for the life sciences (figure 3-5). Within these broad areas, large changes have occurred in individual fields in recent years (Appendix table 3-6). In the area of physical sciences, for example. Federal obligations for basic research in physics were at their highest level in 1967 in constant dollars before declining 28 percent by 1974 when obligations were approximately at a pre-1963 level. In the life sciences, basic research obligations for the biological sciences grew steadily, whereas Clinical medical sciences declined 58 percent in constant dollars between the peak funding year of 1967 and 1974. BASIC RESEARCH IN UNIVERSITIES AND COLLEGES Universities and colleges perform the bulk of the Nation's basic research. They accounted for 54 percent of the total national expenditures for such research in 1974 (figure 3-2). The presently dominant position of these institutions in fundamental research is the culmination of a long-term trend. In 1953, universities and colleges accounted for only 26 percent of the total expenditures for basic research, compared with 35 percent for industry and 24 percent for intramural research by the Federal Government. As funding of basic research rose over the years — primarily as the result of increasing Federal support — the fraction of the total going to universities and colleges grew rapidly, much more rapidly than funding in the industry and Federal intramural sectors. In consequence, the percentage of the total funds for basic research accounted for by these two sectors had declined to 16 percent each in 1974. There was little change in the share of basic research expen- ditures accounted for by the nonprofit in- stitutions and the university FFRDC's, with each accounting for some 7 percent of basic research expenditures throughout the last decade (Appendix table 3-2). The significant role of universities and colleges in basic research is reflected also in the fact that scientists and engineers employed by these institutions are responsible for a large proportion of all U.S. scientific research reports — approximately three-fourths of the total in 1973 (Appendix table 3-21). The research performed by these institutions, moreover, is increasingly the basis for advances in technology (figure 3-25). Basic research in universities and colleges ranges from the efforts of individual scientists and engineers to those of large research teams which often are organized around the use of unique equipment and facilities. Most of the research takes place in universities which have graduate-level programs offering doctorate degrees; these institutions reported 98 percent of all academic basic research expenditures in 1974 11 Jhjs concentration reflects, in part, the close relationship between research and graduate education in science and engineering. Research is an integral part of graduate educa- tion in these areas and, indeed, students are involved in performing much of the research. Graduate students in chemistry, for example, were coauthors of 56 percent of the research reports published in 1971 by institutions awar- ding doctorate degrees in that field. i- Expenditures by universities and colleges for basic research (from all funding sources com- bined) increased continuously from 1960 to 1974 " Expemiilures for Scientific and Engineering Aclivilies al Univer- sities and Colleges, FY 1973, National Science Foundation (NSF 75-316-A). ■' Directory of Graduate Research. American Chemical Society, 1971. 59 Figure 3-7 Basic Research Expenditures in Universities and Colleges, by Source, 1960-74 Constant 1967 dollars' Current dollars (Millions of Dollars) 2,200 2,000 (Millions of Dollars) 1,800 1,600 1,400 1,200 1,000 200 Industry -^-^^-'■^-'-^■■4-f T-rT' I960 74 (est.) ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE: National Science Foundation. 200 Other sources Industry ->-^H" 1960 ■62 ■64 '66 '68 '70 '72 '74 (est.) in current dollars, although the annual rate oF growth diminished after 1968 (figure 3-7).'^ This decline in the growth rate, coupled with rising inflation, produced a level of constant dollar expenditures which changed little during the 1968-72 period. Small constant dollar increases in 1971 and 1972 were succeeded by larger decreases in the two following years, with the result that basic research expenditures in 1974 were 9 percent lower than in 1972, the year of highest constant dollar funding. 13 These expenditure data are for R&D which has been sponsored by other agencies and organizations, as well as R&D supported by an institution's own funds which it allocates to separately organized institutes, divisions, or specific R&D projects. They do not include the expenditures for research /teaching assignments of the faculty (departmental research). Expenditures associated with FFRDC's administered by universities are treated later in this chapter. The leveling off and decline in constant dollar expenditures for basic research is due mainly to reduced growth of funding by the Federal Government (figure 3-7), in combination with inflation. The scientific fields most affected by these declines were the physical sciences (par- ticularly physics) and clinical medicine (see figure 3-9 and Appendix table 3-9). Sources of funds for basic research The sources of financial support for basic research in universities and colleges are shown in figure 3-7. The largest of these— the Federal Government — provided substantial annual in- creases in current dollars between 1960-68, but reduced significantly the average annual in- crements in later years. Translated to constant 60 dollars. Federal funding for basic research reached a maximum in 1968 and declined a total of 13 percent by 1974. In spite of the slowed growth in current dollars, the Federal Govern- ment provided 70 percent of all funds expended by the academic sector for basic research in 1974 — down, however, from the high of 77 percent which prevailed between 1964-67. Funds provided by "All other sources"' ■■ for basic research in figure 3-7 increased in both current and constant dollars until 1972 — thus replacing stime of the reduced Federal support — before declining 1 1 percent in constant dollars by 1974. These sources of support accounted for 27 percent of the total support for basic research in these institutions in 1974. was in these fields. About one-fourth of the total expenditures was divided almost equally between engineering and the physical sciences (principally physics and chemistry), while the social sciences received 8 percent and en- vironmental sciences 7 percent of the total (Appendix table 3-8). In current dollars, basic research expenditures increased between 1973 and 1974 in all areas except engineering. In constant dollars, however, a reduction in basic research spending was recorded in all fields other than the environmental sciences and clinical medicine, with the largest declines occurring in engineer- ing and the biological sciences. Basic research in fields of science Estimates of total academic expenditures for basic research in selected fields of science are presented in figure 3-8. i^ These estimates are based upon a survey conducted by the National Science Foundation in which universities and colleges report their total research and develop- ment expenditures for each of several fields of science, as well as the percentages of the total R&D expenditures (over all fields combined) which are given to basic research, applied research, and development. This information is correlated with other factors — such as the source of the research support and the type of academic institution which performed the research — in deriving the estimates of expen- ditures for basic research in the individual scientific fields. Because these data are es- timates, and may differ from actual expen- ditures, they should be regarded only as ap- proximations.'" The six broad areas of scientific research indicated in figure 3-8 received almost 90 percent of all expenditures for basic research in universities and colleges in 1974.15 Expenditures for fundamental research in these institutions are concentrated in the life science fields of clinical medicine and the biological sciences; 51 percent of all basic research expenditures in 1974 " This includes universities and colleges, State and local governments, and other nonprofit institutions. 15 See Appendix table 3-8a for a listing of the scientific disciplines encompassed by these broad fields and Appendix table 3-8 for more detailed data for certain disciplines. >" The feasibility of obtaining data directly on basic research expenditures in individual fields of science is being investigated and may be attempted in future NSF surveys. Federal Government support of basic research Current dollar expenditures from Federal Government sources for basic research in universities and colleges increased throughout most of the 1964-74 period for each of the six broad fields of science and engineering, except for a 14 percent decline in engineering expen- ditures from 1973 to 1974 (figure 3-9). i^ Increases in the level of support after 1968, however, were less than increases in inflation in all fields other than the environmental and biological sciences. As a result, the magnitude of the federally funded research effort — as measured by constant dollar expenditures — was lower in 1974 than in some previous year in each of the six fields. The fields with the largest reductions were engineering, the physical sciences, and clinical medicine, which recorded declines of 26, 30, and 10 percent, respectively, between 1968 and 1974 (see Appendix table 3-9). The Federal Government, as noted earlier, provided 70 percent of all funds expended by universities and colleges for basic research in 1974. The dependence on this source of support, while varying from field to field, declined over the last decade in all fields other than the biological sciences, as shown below: !■ These data are estimates based on the same NSFsurvey as the total expenditures for basic research in academic institutions presented in figure 3-8. 61 Federal support as a percent of all basic research expenditures Selected fields All fields Physical sciences Chemistry Physics Environmental sciences Life sciences Clinical medicine ... Biological sciences . . , Engineering Social sciences Other fields 1964 76 93 89 97 91 69 84 53 82 61 80 1974 70 82 78 87 74 69 77 59 69 58 72 Support by Federal agencies. The six Federal agencies mentioned earlier accounted for 98 percent of total Federal obligations to univer- sities and colleges for basic research in 1974, with the NSF and HEW alone providing 74 percent of all such obligations. The individual Federal agencies differ greatly in the proportion of their total obligations for basic research which they direct to universities and colleges. i* Of the '» Ffi/mi/ funds for Reieanh. DevelopmenI, ami Other Scientific Activities, FiscalYears 1973. J974ami 1975, Vo/. XXIJJ National Science Foundation (NSF 74-320-A), and earlier volumes in this report series. Figure 3-8 Estimated Basic Research Expenditures in Universities and Colleges, by Field of Science, 1964-74 Current dollars Constant 1967 dollars ' (Dollars in Millions) (Dollars in Millions) • • • • • • • 500 • • Clinical ,•* ^^ medicine ,• ^r / 400 ^•'^r^ Biological ,'^r sciences 300 / ../ Physical ^^^^^ ^^ i t / ^.^--^i^'' • 200 y r ^ Engineering ( y Social , . • sciences ^,»' ^ ^"^ 100 - ...•• ^^ ^.••* ^^ Environmental ^ sciences n "^ 1 1 1 II 500 - ::— ' ^ Environmental 1964 •66 74 (Prelim.) 1964 72 74 (Prelim.) I GNP implicit price detlatofs used to convert current dollars to constant 1967 dollars SOURCE- National Science Foundation. 62 six agencies noted earlier, NSF and HEW allocated the largest fraction of their total basic research obligations to educational institutions in 1974 (84 and 70 percent, respectively), followed by DOD (44 percent), USDA (24 percent), AEC (21 percent), and NASA (6 percent). Comparably large variations exist among the agencies in respect to the allocation of basic research obligations for broad fields of science and engineering at universities and colleges, as shown in the following table. Figure 3-9 Estimated Federal Basic Research Expenditures in Universities and Colleges, by Field of Science, 1964-74 Current dollars Constant 1967 dollars' (Dollars in Millions) 450 (Dollars in Millions) 400 - 350 300 - 250 200 150 100 • • • Clinical ...•••• medicint • • • • • • • • • • • • • — • • • • m^^^ • ^^^^ « • Biological M • • • sciences X »* Physical M •" sciences i^^^^^^*^^^^^^^^-^^ y^- .. ^ ^ ^r ^^ Engineering ^^^ ^^^ ~^^V ^^ • >_ Social ^-^??^ .^^^^' ^*ii*— *• science .... 1 1 Environmental sciences 1 1 1 350 .•.. Clinical • ^ medicine • * . 300 • • * • • • ••..•• • • • • • • 250 • • Physical sciences Biological sciences 200 ^^^^^^^^^^ 150 ^' Engineering ^^ y > • 100 — Social sciences • • • . .^«t^ . . • • • .W" ! Environmental sciences 1 1 1 50 n 1 1 1964 ■66 74 (Prelim.) 1964 ■66 ■68 ■72 '74 (Prelim.) ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE. National Science Foundation. 63 Percentage of total basic research obligations directed to universities and colleges, by field, 1974''' Fields of science^o NSF HEW DOD USDA Life sciences 83 73 30 24 Physical sciences 82 84 36 8 Environmental sciences 63 — 54 22 Engineering 97 92 40 12 Social sciences 90 37 — 56 AEC NASA 29 3 19 11 — 6 25 12 These fields of science and engineering are supported by various combinations of Federal agencies, as indicated in figure 3-10 which presents the proportion of Federal obligations provided by each of the six agencies to univer- sities and colleges for basic research in each major field. The figure indicates that either one or two agencies alone provided at least 70 percent of all Federal obligations for basic research in each field. The NSF and HEW together, for example, provided nearly 90 percent of all federally obligated dollars for basic research in the life sciences in 1974, almost 83 percent of the obligations for psychology and the social sciences, and approximately 75 percent for chemistry. Similarly, two agencies (DOD and NSF) accounted for more than 85 percent of the six agencies' obligations for the environmental sciences and some 80 percent of those for engineering, while the AEC and NSF in com- bination provided nearly 80 percent of all obligations for physics research in universities and colleges. The fact that the NSF in 1974 provided either the largest or next largest amount of basic research obligations in the several fields — and nearly 35 percent of all obligations from the six agencies — underscores the extent of dependen- cy on that agency by universities and colleges for support of basic research. Institutional concentration of basic research Basic research is concentrated in institutions which award advanced degrees in science and engineering. The 280 universities which grant doctorate degrees in the sciences and engineer- ing accounted for 98 percent of academic basic research expenditures in 1974, with 82 percent of the total expenditures concentrated in 100 such institutions.-' Little change occurred in this pattern of institutional concentration during the 1964-74 period as shown in the table below, although there were considerable shifts in the positions of specific institutions. Percentage of expenditures for basic research by groups of institutions ranked in order of expenditures, 1964 and 1974 Year First First First First First First 10 20 40 60 80 100 1964 1974 25 41 60 72 NA NA 24 39 59 72 81 86 The institutional concentration of R&D expenditures varies among the five broad scientific areas (figure 3-11)." The life sciences exhibited the least concentration in 1974, and the environmental sciences the greatest. The social sciences, physical sciences, and engineer- ing had similar patterns of distribution or concentration, although varying considerably among individual institutions. The ten academic institutions with the largest R&D expenditures in the life sciences, for example, reported 23 percent of the total for all universities in 1974, compared with a concentration of 47 percent of all environmental science R&D expenditures in the first ten institutions for that field. No university ranked among the first ten in all five fields, and only one university held this position in four of the fields — reflecting a diversity of field concentration patterns even within the major research universities. " Ihid.. and special tabulations. 2° See Appendix table 3-6a for descriptions of these fields. -' Expi^ntiiturt'i tor Scientifii lirid Etignteeriug Activiiies ai llniver- itliei ami Collegei, FY 1973. National Science Foundation (NSF 75-316-A), and special tabulations. -^ Data on basic research expenditures alone are not available for separate fields of science and individual institutions. An approximation is available, however, in the form of total R&D expenditures by these institutions in scientific fields, the largest component of which is basic research. 64 Figure 3-10 Federal Obligations for Basic Research in Universities and Colleges, by Selected Supporting Agencies and by Selected Fields, 1973-74 (Dollars in Millions) ALL FIELDS 50 100 150 200 250 300 350 400 450 ENVIRONMENTAL SCIENCES 10 20 30 40 50 60 1 1 1 1 > USDA IIIM/J II l'"-l DOD 1 1 1 1 HEW 1 1 1 1 AEC 1 1 1 \ NSF 1 1 r' 1 NASA 1 1 1 1 PHYSICS 10 20 30 40 50 60 USDA DOD HEW AEC NSF NASA 5 973 974 USDA g 19" DOD HEW AEC NSF NASA 3 ^ XI USDA DOD HEW AEC NSF NASA 1 1 1 1 1 j| 1973 1 ' ^1 157.1 1 1 1 1 1 1 1 1 LIFE SCIENCES 50 100 150 200 250 300 350 USDA DOD 1 1 1973 -Jll 1974 3 HEW 1 1 1 1 AEC a NSF 1 1 1 1 NASA fl ) 5 ENGINEERING 15 25 35 45 USDA ff ' 1 DOD 1 1973 1. 1 1Q7/| HEW AEC NSF 1 1 1 NASA n 11 SOCIAL SCIENCES 5 10 15 20 25 1 1 1 1 1 1 )■ -J 1973 r ^]Q7A r' h _i 1 1 1 II CHEMISTRY 10 15 20 25 30 35 USDA DOD HEW AEC NSF NASA USDA DOD HEW AEC NSF NASA -al973 3-3 1974 ■±^_Ij ' I I MATHEMATICS 5 10 15 20 25 1 1 1973 T^ 1 iq74 5 t 1 I 1 1 GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE National Science Foundation, I Constant 1967 dollars ^ I Current dollars 65 Figure 3-11 Concentration of R&D Expenditures at the 100 Universities and Colleges with the Greatest Expenditures in Selected Fields, 1974 first First First First First 10 20 40 60 80 Institutions rant^ed by expenditures in each field 1 Based on total R&D expenditures in individual fields SOURCE: National Science Foundation. Basic research expenditures per scientist and engineer Basic research expenditures in doctorate institutions^-' reached their highest level in 1972 in constant dollars and then dropped nearly 15 percent over the next two years (see Appendix table 3-12), while the number of scientists and engineers in these institutions rose continuously through 1974. This increase of scientists and engineers was due partially to an expanding number of institutions awarding doctorate degrees in a science or engineering field — 224 such universities in 1969 compared with 280 in 1974 — as well as increases in the number of such personnel at existing doctorate-level in- stitutions. These trends— an increase in the number of scientists and engineers and a drop in real expenditures for basic research— have produced a reduction of almost 30 percent in constant dollar expenditures per scientist and engineer in Figure 3-12 Basic Research Expenditures per Scientist and Engineer , in Doctorate-granting Institutions, by Source, 1966-74 (in constant 1967 dollars -) (Dollars per FTE Scientist and Engineer) 12000 11000 1 0000 9000 8000 7000 6000 5000 4000 3000 2000 1000 Federal I Non-Federal 1966 1968 1970 1972 1973 1974 i-' Those granting doctorates in at least one science or engineering field. > Full-time-equivalent tjasis - GNP implicit price deflators used to convert current dollars to constant 1967 dollars- SOURCE: National Science Foundation. 66 doctorate institutions since 1968 (figure 3-12). A slight shift from expenditures for basic to applied research occurred after 1972 and is one reason for this decline, the inclusion of scientists and engineers from the new doctorate in- stitutions is another, and the reduction in constant dollar expenditures, particularly those supported by the Federal Government, is a third factor. Federal funds for basic research per scientist and engineer declined almost 30 percent between 1968 and 1974. Funds from other sources decreased by a similar percentage after 1972, but the reduction in absolute terms was much less than the Federal declines. The reductions in real expenditures for basir research per scientist and engineer have oc curred in several fields,-^ as shown in figure 3- 13. The largest decline was recorded in physics, where such expenditures dropped almost 40 percent between 1966 and 1974. Decreases in this field were due primarily to declines in funding, rather than to increases in the number of physicists. BASIC RESEARCH IN FEDERALLY FUNDED RESEARCH AND DEVELOPMENT CENTERS ADMINISTERED BY UNIVERSITIES Federally Funded Research and Development Centers (FFRDC's) are organizations financed exclusively or primarily by the Federal Govern- ment to perform R&D in relatively specific areas, or in some instances to provide facilities at universities for research and associated training purposes. The Centers usually have a direct and long-term relationship with their funding agency, making it possible for them to maintain instrumentation, facilities, and operational support beyond the capabilities of single educational or research institutions. Non- Federal organizations — academic, industrial, or nonprofit — administer the FFRDC's. In 1974, FFRDC's administered by universities accounted for 7 percent of the Nation's total ^•i The actual cost of conducting research differs substan- tially from field to field, reflecting in part the extent to which research depends upon special equipment, facilities, and technical support staff. Figure 3-13 Estimated Basic Research Expenditures in Doctorate-granting Institutions per Scientist or Engineer' by Selected Fields, 1966-74 (In constant 1967 dollars)-' 15,000 10,000 5,000 Physics Engineering Chemistry ,,•• '••.^ Social sciences Mathematical sciences 1966 ■68 70 Fiscal years 74 (Prelim.) 1 Full-time equivalent basis - GNP implicit price detlators used to convert current dollars to constant 1967 dollars. SOURCE: National Science Foundation. basic research expenditures, ^^ and 86 percent of the Federal obligations for all FFRDC's. 2" These Centers and their sponsoring agencies are: Atomic Energy Commission Ames Laboratory Argonne National Laboratory Brookhaven National Laboratory " Nnhoii.jf Pallems of R&D Resources, I95J-75, National Science Foundation, (NSF 75-307). 2" Federal Funds for Research, Derelopment, and Olher Scientific Activities. Fiscal Years 1973, 1974 anii 1975, Vol. XXIIl, National Science Foundation (NSF 74-320-A). 67 Cambridge Electron Accelerator Fermi National Accelerator Laboratory E.O. Lawrence Berkeley Laboratory E.O. Lawrence Livermore Laboratory Los Alamos Scientific Laboratory Oak Ridge Associated Laboratory Plasma Physics Laboratory Stanford Linear Accelerator Department of Defense Applied Physics Laboratory Applied Research Laboratory Center for Naval Analyses Lincoln Laboratory National Aeronautics and Space Administration Jet Propulsion Laboratory Space Radiation Effects Laboratory National Science Foundation National Astronomy and Ionosphere Center Cerro Tololo Inter-American Observatory Kitt Peak National Observatory National Center for Atmospheric Research National Radio Astronomy Observatory In current dollars, expenditures by university- managed FFRDC's for basic research were at their highest level in 1973 and declined slightly in 1974 (figure 3-14). In constant dollars, however, basic research expenditures in 1974 were almost 25 percent less than those of the 1968 peak year and approximately equal to expenditures in 1964. Data are not available on expenditures for specific scientific fields, but it is apparent from the above listing of the Centers, and the Federal agencies involved, that the basic research is predominantly in the physical sciences and engineering. The proportion of all R&D expenditures in FFRDC's reported as basic research has remained at nearly 35 percent in the last few years. 2^ Although some of the FFRDC's are permitted to receive support from sources other than the Federal Government, such funds amounted to less than 1 percent of their total funding in 1974. Figure 3-14 Basic Research Expenditures at Federally Funded Research and Development Centers Administered by Universities, 1964-74 (Dollars in Millions) ?nn JUU current dollars ^^^^ 250 y/xT^V 200 '^r 1967 dollars '^V^^ ^V 150 - 100 - 50 - 1 1 1 1 1 1 1 1 1 1964 '66 '68 70 72 74 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE National Science Foundation :i-.sj;.\i 2" See Appendix table 3-14. BASIC RESEARCH IN INTRAMURAL FEDERAL LABORATORIES Several agencies of the Federal Government operate their own R&D laboratories as part of their effort to meet the research needs associated with their agency mission and program objectives. Intramural laboratories were responsible for 16 percent of the total basic research expenditures and 23 percent of all federally supported basic research in 1974. About 94 percent of all such research in 1974 was undertaken by the six agencies indicated in figure 3-15. Examples of such laboratories are the Goddard Space Flight Center of NASA, the National Animal Disease Laboratory of USDA, 68 Figure 3-15 Federal Obligations for Intramural Basic Research, by Selected Agencies, 1960-74 Current dollars Constant 1967 dollars ' (Dollars in Millions) 250 I 200 - 150 - (Dollars in IVIillions) 250 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. - The large amounts reported by NASA tor basic research are due to the substantial cost of support equipment sucti as spacecraft and launch vehicles peculiar to space exploration, and the statistical proration of costs for tracking and data acquisition. SOURCE National Science Foundation and the R&D Institute at the National Cancer Institute of HEW." Current dollar funding for basic research in these laboratories increased steadily from 1960 -' For further information on the utiHzation of intramural Federal laboratories see: U.S. Congress, House Committee on Appropriations, Agriculture-Environmental and Con- sumer Protection Appropriations for 1975; Part 7, bweilignlive Report on Ihe Ulilizalion of FeiiernI iahoralories - 93rd Cong., 2nd Sess., 1974. to 1970, and then after a slight decline in 1971 had risen again by 1974.2'' jp constant dollars, however, 1974 funding was approximately equal to that of 1965, and down 22 percent from the peak year of 1970. The constant dollar decline in intramural basic research funding is evident in all major agencies -" See Appendix table 3-15. 69 except the Department of the Interior (figure 3- 15). In the case of NASA, HEW, and Commerce, the year of highest funding of intramural basic research support was 1970, after which funding decreased in each of the agencies. By 1974, NASA's funding had declined more than 25 percent over its 1970 level, HEW by 40 percent, and Commerce by almost 70 percent. Basic research in DOD and USDA intramural laboratories received the highest level of con- stant dollar support in 1972. The DOD program declined by just over 20 percent while the USDA program remained fairly constant through 1974. In contrast, the Department of the Interior obligations for intramural basic research reached their highest level in 1974. BASIC RESEARCH IN INDUSTRY'o Basic research consists of original in- vestigations for the advancement of scientific knowledge which has no specified commercial objective, although the research may be within the general area of a company's interest. Such research, which is conducted largely by manufacturing industries, may provide a technical basis for product improvement, expan- sion or new business, and a defense against technological obsolescence. Expenditures for basic research in industry represented 16 percent of the total national funds spent for basic research in 1974, but only 3 percent of all R&D expenditures in industry.-^' Although the current dollar total from all sources has risen, particularly since 1972, the effect of inflation has been to reduce the 1974 basic research expenditures in industry to approximately the same level as 1961 (figure 3- 16). Federal support has dropped 12 percent since 1971 in constant dollars compared to a 3 percent increase of non-Federal basic research expenditures. The proportion of basic research in industry which has had Federal support has been about 22 percent for the last three years, compared to 32 percent in 1967. Over three-fourths (78 percent) of the 1973 basic research expenditures in industry were accounted for by only four industries (figure 3- 17): chemicals and allied products (37 percent), electronic equipment and communications (28 50 A more comprehensive discussion of R&D in industry is found in a later chapter entitled, "Industrial R&D and Innovation." 31 Naiioml Pallenii of R&D Resources. 1953-75. National Science Foundation (NSF 75-307). percent), aircraft and missiles (9 percent), and machinery (4 percent). For the most part, basic research in industry is concentrated in the physical sciences and engineering (some 80 percent in 1973). Expen- ditures in the physical sciences, however, have declined significantly since the late 1960's, in both current and constant dollars (figure 3-18), while engineering expenditures reached their highest level in 1973 in current dollars. Constant dollar expenditures in the life sciences, on the other hand, grew substantially in the late 1960's before peaking in 1971 and then declining. BASIC RESEARCH IN NONPROFIT INSTITUTIONS Independent nonprofit institutions are organizations other than educational in- stitutions chartered to serve the public interest, and include research institutes, hospitals, private foundations, science exhibitors, professional societies, trade associations, and FFRDC's administered by such nonprofit in- stitutions. Although the largest single category is the research institutes, the others generally perform other services in addition to research, such as patient care or charitable activities. These institutions were responsible for 7 percent of the Nation's expenditures for basic research in 1974, a fraction which changed little during the 1960-74 period. Current dollar expenditures for basic research in nonprofit institutions reached their maximum in 1974 (figure 3-19). In terms of constant dollar expenditures, funds for basic research in 1974 were comparable in magnitude to the funding level of earlier years (1971 and 1962-63), and approximately 20 percent lower than the year of highest funding which was 1966. Federal sources provide the greatest part of support for basic research in these institutions and have a large impact on the total level of funding in any given year. In 1974, Federal support accounted for 53 percent of all basic research expenditures in nonprofit institutions, compared to 58 percent in 1966 and 50 percent in 1960. In contrast to the fluctuating Federal funding, support from other sources rose comparatively steadily, although slowly. Over the 1960-74 period, basic research as a proportion of total research and development expenditures by these institutions declined from 38 percent to 22 percent. 70 Figure 3-16 Industrial Basic Research Expenditures, by Source , 1960-74 Current dollars Constant 1967 dollars ' (Millions of Dollars) (Millions of Dollars) 800 700 Total 700 - Total 600 y^"^^^""^^^ 600 y >^^^\^ 500 400 300 ^^^ Industry Federal 500 400 300 / Industry ^^^^ Federal 200 ^^•^•^'^^ 200 " ^ ^*\_ 100 -/^ 100 J ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 19 60 '62 '64 '66 '68 '70 '72 *7 4 19 60 '62 '64 '66 '68 '70 '72 '74 (est.) (est.) ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE: National Science Foundation. RESEARCH OUTPUTS AND APPLICATIONS Subsequent sections of this chapter present the results of experimental studies aimed at measuring a part of the output of research and a portion of its applications. The studies repre- sent, at best, small steps in these directions. Output of scientific research literature Information on the quantity and sectoral origin of published research reports from several fields of science was obtained from a study conducted by the National Federation of Abstracting and Indexing Services. ^^ The study, 3- \niutiiori of Ihe Output of Scientific Research, National Federation of Abstracting and Indexing Services, 1974 (A study commissioned specifically for this report, and funded largely by the Office of Science Information Service, National Science Foundation). in brief, involved the selection of a set of scientific and engineering journals which was representative of the total literature in each field. This was accomplished largely through the guidance of the Federation's member services and by advice from experts active in the fields. On a sampling basis, individual reports in the journals were examined to determine the first author's institutional affiliation: academic, government, industry, or other nonprofit organization. The sample of reports was restricted to those whose first authors were affiliated with U.S. institutions. The data obtained from the study were used to develop preliminary measures of the relative growth of several fields of science and engineer- ing in terms of their publication output, the roles of the different sectors in the overall research effort of each field, and the relationship between the research output and financial inputs. 71 Figure 3-17 Expenditures for Basic Research in Industry, by IVIajor Performing Industries, 1960-73 Current dollars Constant 1967 dollars ' (Dollars in Millions) 300 250 200 - 150 100 50 Chemicals and allied products All others / N # -7^ V # /C J^ J^V^ \ J / \ ^ 'Ca \, r .•^ ^-.. ^ .^'— ^ V^ ^^^^^ / r ^^ Electrical eqpt. and i / communications / J Aircraft ^^^ and missiles ▼ Machinery 1 1 1 1 1 1 1 1 1 1 1 1 1960 ■62 '64 72 73 1 GNP implicit price deflators used to convert current dollars to constant 1967 dollars, SOURCE: National Science Foundation. (Dollars in Millions) 300 250 _ 200 All others » \ Chemicals and allied products 150 \ ' m Electrical eqpt. and # communications V, 100 - / 50 V'--'"-- Machinery Aircraft and missiles 1 1 1 1 1 1 1 1 1 1 i 1 1960 ■62 ■64 ■68 ■72 '73 Growth in research output. The extent and pattern of the relative growth in research publications are shown in figure 3-20 for each of 13 fields of science and engineering. The fields listed in this figure are presented in descending order with respect to the magnitude of their relative growth in publications during the 1960- 73 period. The fields included in the top part of the figure grew by more than 200 percent during the period, those in the second plot by more than 100 but less than 200 percent, those in the third by more than 75 but less than 100 percent, and those in the bottom plot by less than 75 percent. The fields differ considerably in their pattern of growth. For example, research publications in physics, chemistry, and engineering's (third plot from top) have remained at nearly a constant level since the late 1960's, whereas astronomy and biology (second plot from top) grew continuously throughout the period. The field of oceanography exhibits one of the more complex and unusual growth patterns; research publications in that field rose rapidly until 1969, but declined in most subsequent years. ^^ 3-' These fields, as noted later in this chapter, play a large role in technological innovation. ^* For information on the U.S. output of scientific research in an international context, see the chapter entitled, "International Indicators of Science and Technology" in this report. 72 Figure 3-18 Expenditures for Basic Research in Industry, by Selected Fields, 1967-73 Current dollars Constant 1967 dollars ' (Dollars In Millions) 220 (Dollars in Millions) 200 - 40 20 Clinical medical sciences Mathematics ,•*••••••• Environmental sciences [ L_ 1967 •69 » GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE- National Science Foundation. 40 20 Clinical medical sciences Mathematics Environmental sciences I L 73 1967 ■73 Research output by sectors. The research publications produced by each sector — university, government, industry, and other nonprofit organizations — are shown in figure 3- 21 for five selected fields. (Data for each of the 13 fields are presented in Appendix table 3-21). Universities were by far the largest producers of published research reports, followed by government, industry, and nonprofit organizations. The predominant role of academic institutions increased throughout the 1960-73 period covered by the study. By 1973, universities were responsible for an average of almost 75 percent of the publicatioi.s in the 13 scientific fields, compared with some 60 percent of the total in 1960. The share of publications accounted for by the academic sector rose in all fields during the period, with the largest increases occurring in sociology, physics, chemistry, geology, and mathematics (including computer sciences). 73 Figure 3-19 Basic Research Expenditures in Nonprofit Institutions,' by Source, 1960-74 Current Dollars Constant 1967 dollars' (Dollars in Millions) (Dollars in Millions) 300 250 Totaly^^^ / ^^ Total ZOO y/^ ^ 200 M ^V ^ - 150 1 >^^ Federal ^ 150 ^^^^^^ Federal 100 50 /s. 100 50 ^Jf Own funds ' ^ ^^> J- — Own funds '•'' Industry Industry ■«.^'* •* "" 1 1 1 1 1 1 1 1 1 1 1 1 1 ■"7 1 1 1 1 1 1 1 1 1 1 1 1 1= 60 '62 '64 '66 '68 •70 '72 '74 1960 '62 '64 '66 '68 '70 (est.) 11 '74 (est.) ' Includes State-administered tiospitals. - GNP implicit price deflators used to convert current dollars to constant 1967 dollars. ^ Includes State and local government funds. SOURCE; National Science Foundation The Federal Government was the second largest producer of published research reports in 1973, with an average of 11 percent of the total reports from the 13 fields. The proportion of the total research publications produced by this sector declined, however, in all fields between 1960 and 1973, with the largest decreases occurring in oceanography, chemistry, and physics. In 1973, this sector accounted for a significant share of total research publications in the fields of astronomy (29 percent), ocean- ography (21 percent), geology (18 percent), and astronomy (16 percent). Private industry's share of research publications in 1973 averaged 10 percent among all fields as a whole, with the largest fractions in the fields of engineering (44 percent), chemistry (18 percent), and physics (16 percent). The proportion of total publications for which this sector was responsible declined in all fields — other than the atmospheric sciences and oceanography — between 1960 and 1973. The largest declines were recorded in engineering, physics, and chemistry. Nonprofit organizations produced the remaining 5 percent of total publications in 1973, down from some 10 percent in 1963. It is clear from these indicators that academic institutions are predominant in the production 74 (Percent) Figure 3-20 Relative Growth in Scientific Researcli Publications, by Selected Fields of Science, 1960-73 (Percent growth after 19601 GROUP I 250 > s 200 — Sociology y^ J^^ g £ .' Psychology' ' 150 100 * r"y' ^^^^^ AtmosDheric bU n - ^^^ Sciences 1 1960 '62 (Percent) ■64 '66 GROUP II '72 '74 Mathematics ^ 16U 100 ^^^' .>"*'^ Biology 50 n ^^^ Astronomy ^T**i 1 1 1 1 1 1 1 1 1 1 1 of research publications, and that their role v'ls-a- vis other sectors is increasing. The extent of their publication output appears high in relationship to the fraction of total financial resources for research which is expended by these in- stitutions. (See figure 3-2 for the research expenditures by this and other sectors.) Research publications and research expen- ditures. Publications in the five fields shown in figure 3-21 which were produced by universities were compared with the reported R&D expen- ditures for these fields. Expenditures in constant dollars were used for this purpose, with a "lag time" of two years between the expenditures and publications. (The limited available data on expenditures restricted the correlation to a short period of time, and did not permit exploration of alternate "lag times"). The results are presented in figure 3-22. A relatively close fit between lagged expenditures and publication output was found for the fields of biology, engineering, and mathematics. On the other hand, relatively large deviations between input and output were obtained in chemistry and physics, particularly in later years. I960 '66 '72 GROUP (Percent) 150 1 100 50 GROUP IV Geology ^ Oceanography ttPolitical Science 1960 1 Data not available for 1973. SOURCE^ National Federation ot Abstracting and Indexing Services. Basic research and technology The relationships between basic research and eventual applications in modern technology are complex and difficult to trace. Certain aspects of these relationships were the subject of a special study upon which the data presented here are based. The study centered around 179 major advances in technology which occurred in the United States during the 1950-73 period. The patent documentation associated with each of the advances was examined to determine characteristics of the research which were cited as the origin of the invention. The sample of 179 major advances covers ten broad areas of technology. These areas and examples of specific advances included in the study are shown below, on page 78 75 Figure 3-21 Publication Output for Selected Fields of Science, Percent of Yearly Totals by Sectors, 1960-73 Physics Biology (Percent of Total) 100 40 University Industry (Percent of Total) 100 1 University 20 J Government - Other ^ Industry 'r--^-:-i^H-i< 74 1960 '66 '68 70 Engineering (Percent of Total) 100 University ^ ^^ (Percent of Total) 100 Industry (Percent of Total) 100 60 University 20 Industry __ Other ^^ Government 1960 '62 '64 '66 ■72 74 SOURCE, National Federation of Abstracting and Indexing Services, 1974, 76 Figure 3-22 Research Publications and R&D Constant 1967 Dollar Expenditures' in Universities and Colleges, 1964-72 Mathematics Engineering Biology 200 U3 U7 .^n Expenditures Publications 1966 Expenditures 1964 72 73 1966 70 71 1964 Chemistry Physics 100 Publications 1966 Expenditures 1964 - Publications V*^ *m„^ i / X / 1 Expenditures :v f 1 ^ 1966 1964 •68 ■66 72 70 73 71 1 GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE National Science Foundation and National Federation of Abstracting and Indexing Servicese. 77 Technological areas Chemicals Electronic components Nonelectrical machinery Communication devices Scientific, photographic & optical equipment Computers & electronic data processors Metals and alloys Transportation systems & devices Pharmaceuticals Ceramics and other nonmetals .. Total Examples Number of advances 20 25 22 20 19 17 16 16 12 6 17" Organo-phosphoric acids Oral antidiabetic agent Thermoelectric devices Permutation decoder Tunnel diode Permanent magnetic materials Wavefront reconstruction Low energy electron sterilization Processing of nuclear reactor fuel elements Multiple speed transmission Dependence on basic research. One important indicator of the relationship between basic research and technology is the extent to which new technologies or major advances in existing ones depend upon results from basic research. A measure of the incidence of such relationships is shown in figure 3-23. These findings show that other patents were cited more frequently than published research, but that differences between the two in citation frequency have narrowed considerably. The frequency of citation (number of citations per basic patent) increased by 17 percent for the basic and by 8 percent for the combined basic and applied research categories from the first to the second decade. On the other hand, the frequency of citation to other patents decreased almost 25 percent. These results suggest that more recent technological advances may depend increasingly on new scientific discoveries reported in the research literature. Seven different fields of science and engineer- ing were represented in citations to the research literature (figure 3-24). Almost an equal percentage of basic patents cited research in Each technological advance is represented by a single "basic patent" in which the fundamental concept or idea embodied in the invention is presented for the first time in a patent applica- tion. The documentation provided with the application, as well as information added in the patent examination process, was reviewed in order to identify the research which was cited as the basis for the advance. Of the 179 examples, slightly more than 50 percent of the associated basic patents cited published research literature and/or other patents. ^5 Xhe data presented here are based on those patents in the sample which contained such citations.^" ^5 The absence of citations in the remaining basic patents may have several causes, including the possible lack of candid disclosure by the patent applicant. Failure to make required disclosure has, in fact, resulted in a doubling of the number of patent invalidations over the past tvi'nety years, -'" For further information on the methodology of the study, see Imiicalon of Ihe Role of Science in Patenleii Technology. Franklin Pierce College Law Center and the PTC Research Foundation, 1974 (A study commissioned specifically for this report). Figure 3-23 Citations per Basic Patent, by Type of Citation, 1950-61 and 1962-73 (Citations per Basic Patent) 2.2 2.0- 1.6- 1.4- 1.2- 1.0- 0.8 0.6- 0.4- 0.2 0.0- I 1950-61 1962-73 Basic research Basic and/or applied researcii Other patents SOURCE: FranHin Pierce College Law Center and the PTC Research Foundation. 78 Figure 3-24 Percent of Basic Patents Citing Research Literature, by Field of Science and Engineering, 1950-61 and 1962-73 (Percent) 10 20 30 I I I Electrical Engineering Chemistry Physics 40 I Metallurgy Mechanical Engineering Medicine SOURCE: Franklin Pierce College Law Center and the PTC Research Foundation, 1950-61 1962-73 chemistry and electrical engineering (29 and 31 percent) over the 1950-73 period. Following these were physics (22 percent), biology (14 percent), metallurgy (8 percent), mechanical engineering (8 percent), and medicine (4 per- cent). Sectors producing cited research. For each research citation, the institutional sector in which the cited research was performed was identified (figure 3-25). In the 1950-61 period, most of the research cited in the sample was performed in corporate laboratories (57 per- cent). In the 1962-73 period, however, corporate research was cited least frequently, accounting for only 15 percent of the research citations. Universities, on the other hand, rose from second place (28 percent) in 1950-61 to first place in 1962-73, with 54 percent of the cited research being performed in this sector. Research in academic institutions also accounted for most of the basic research citations in both periods, and applied research in the second period. These results should be considered, however, in respect to the total literature output of each of the four sectors. While most academic research is published without restraint, it is generally believed that research reports of corporate and government-affiliated scientists may be published less frequently because of their proprietary or national security character. Time between research and application Many of the results from basic research are not immediately incorporated into applied technologies. Often a long period of time is required to synthesize research results, or to await an economic or social need for a particular application in technology. In the present study, the time between the research and its utilization in technology was defined as the interval between the publication date of the cited research and the date of patent application. The average time was found to decrease from seven to six years from the first to the second half of the 1950-73 period. The most recent period covered in the study (1970-73) has an average time interval of only three years, suggesting an increasingly rapid utilization of research results in modern technology. 79 Figure 3-25 Percent of Citations in Basic Patents, by Type of Citation and Source, 1950-61 and 1962-73 c 10 20 30 40 (Percent) 50 60 70 80 90 10 citations Government --■—-. 1 1 1 1 1 i ^" 1950-61 ^H 1962-73 1 1 1 ^^B Universities ' ^^M ^H Corporations ^ ^ ^H I^^H ■^H I^^H Basic research citations Government Basic and/or applied research citations Patent citations Corporations Government Universities Corporations 1 Includes nonprofit institutions which account for less than 3 percent ot ail the citations- SOURCE- Franltlin Pierce College Law Center and the PTl^ Research Foundation 80 Industrial R&D and Innovation 81 Industrial R&D and Innovation INDICATOR HIGHLIGHTS Total expenditures for industrial R&D more than doubled between 1960-74, with one- third of the growth occurring after 1971; the large increases in recent years came almost entirely from industry's own funds, raising the total expenditures for industrial R&D to more than $22 billion in current dollars in 1974. Adjusted for inflation, total expenditures in 1967 constant dollars for industrial R&D were $15.2 billion in 1974, which was 11 percent lower than in 1968-69, the years of highest funding, and approximately equivalent to the funding level of 1965; in 1974, development activities accounted for 79 percent of total industrial R&D expen- ditures, compared with 18 percent for applied research and 3 percent for basic research. The total number of scientists and engineers engaged in industrial R&D increased in 1973 and 1974 to 360,600, following a decline from a peak employment level in 1969 of 387,000; such personnel supported by industry's own funds increased throughout the 1960-74 period, while the number supported by the Federal Government declined to pre-1960 levels. Expenditures for applied research and development in industry are focused on six product areas: communications equipment and electronic components, aircraft and parts, guided missiles and spacecraft, machinery, motor vehicles and other transportation equipment, and chemicals; these areas comprised nearly 70 percent of all such expenditures in 1973. Industrial R&D is concentrated in a few manufacturing industries and in a relatively small number of large companies within those industries; five industries accounted for some 80 percent of all industrial R&D expenditures in 1973 and a similar propor- tion of all R&D personnel, while the 100 companies with the largest R&D programs spent nearly 80 percent of all industrial R&D funds. Improvement of existing products was the reported goal of one-half of all industrial R&D in 1974, compared with approximately 35 percent for developing new products, and 15 percent for new processes. The R&D intensity! gf manufacturing industries declined steadily after 1964 as a result of reduced Federal support for industrial R&D (primarily in the aircraft and missiles industry); in terms of industry support alone, however, the level of R&D intensiveness has changed little since the early 1960's. The most R&D-intensive industries were the largest producers of patented inven- tions, accounting for over 67 percent of all patents granted during the 1963-73 period; the majority of patents were for inventions in six major product fields: machinery, fabricated metals, electrical equipment, chemicals, professional and scientific in- struments, and communications equipment. The most R&D-intensive industries produc- ed the majority of a sample of major technological innovations during the 1953- 73 period; these industries accounted for 66 percent of the innovations, followed by intermediate level industries with 24 per- cent, and the least R&D-intensive industries with 10 percent. Large companies (those with 10,000 or more employees) produced a greater number of the sample of innovations between 1953-73 than companies with less than 100 em- ployees, but a smaller number than firms ' The proportion of net sales devoted to R&D and the number of R&D scientists and engineers relative to total company employment. 82 employing less than 1,000; small firms (those with less than 100 employees and those with 100-999 employees) produced more innovations per unit sales than larger firms throughout the period. D The largest percentage of the sample of technological innovations produced during the 1953-73 period represented im- provements in existing technology (41 percent), followed by those representing major technological advances (32 percent) and radical breakthroughs (27 percent); the fraction of radical innovations declined 50 percent between 1953-59 and 1967-73, while those rated as major technological advances increased proportionately. D The most frequently cited sources of the underlying technology for the major in- novations were research (applied and basic), followed by the transfer of technology from existing product lines of the innovating firm, licensing, and the purchase of technical "know-how" from other firms. D Basic research was more often involved in product innovations characterized as radical breakthroughs (68 percent) than in those rated as major technological advances (48 percent) or improvements in existing technology (45 percent); applied research occurred with nearly equal frequency in all categories of the innovations studied. Research and development is increasingly the basis and impetus for technological innovation in industry. The results of innovation are new and improved products, processes, and services. These are the elements of technological progress, through which many of the advances in the Nation's productivity, economic status (domestic and foreign), and standard of living take place. While R&D is increasingly important in innovation, it is not sufficient by itself. Innova- tion is a complex process which occurs within a broad economic and social context, and which requires successful efforts in areas such as product design, engineering, manufacturing, and marketing. Although the innovation process is complex, expensive, and risky, the failure of a firm or an industry to be innovative may mean failure of the firm or industry itself, with consequent implications for the general economy. As an activity, industrial R&D ranges from basic research, consisting of original in- vestigations for the acquisition of scientific knowledge — to development, which attempts to translate acquired knowledge into new and improved products and processes. The character and extent of industrial R&D activity vary considerably, both in terms of the industry and size of the company involved. In general, R&D is viewed as an investment which competes for funds and other resources with alternative investments. For many firms, R&D is regarded as a necessary investment whose returns are believed to be competitive with those from other areas of potential resource allocation. Indicators of the state of industrial R&D and innovation presented in this chapter consist of selected financial and human resources invested in R&D and measures of the outputs from such investment. The "input" indicators deal primari- ly with expenditures and scientific and engineer- ing personnel involved in R&D, including trends in the R&D intensity of particular industries. "Output" indicators include measures of patents and technological innovations produced by R&D-performing industries, as well as factors which influence these activities. These measures, combined with R&D intensity and other characteristics, provide indicators of the relative inventiveness and innovativeness of different industries. The chapter concludes with a summary of the major findings from studies of the relationship of R&D and innovation to productivity and economic growth. The present set of indicators provides a more comprehensive description of the state of industrial R&D than was provided by the first report in this series. The indicators, however, are still deficient in several respects, as discussed in later sections of this chapter. RESOURCES FOR INDUSTRIAL R&D Financial and human resources directed to industrial R&D represent principal "inputs" to R&D as well as approximate indicators of the 83 magnitude of the effort. Expenditures for R&D are presented initially for the total industrial R&D effort, followed by information on the source of funds and expenditures by specific industries. Trends in the number of scientists and engineers engaged in R&D are presented in terms of the Nation's overall effort in industrial R&D. These are followed by indicators of the division of R&D resources among the categories of basic research, applied research, and develop- ment, as well as the product fields on which the effort focuses. Data are presented also on certain institutional characteristics of industrial R&D — the distribution of R&D expenditures among companies of different size and among specific industries. The section concludes with trends in the R&D intensity of U.S. industries. Financial and human resources for R&D represent only a small part of the total invest- ment which industry makes for technological innovation, the principal aim of its R&D. Although little empirical data are available regarding total expenditures for innovation, estimates have been made of the typical distribu- tion of costs among the several steps in the innovation process.- These estimates, which apply to successful innovations only, are shown in the table below. Typical distribution of costs in successful product innovations Activity Percent Research (advanced development-basic invention) 5-10 Engineering and designing the product . 10-20 Tooling (manufacturing engineering) . . . 40-60 Manufacturing start-up expenses 5-15 Marketing start-up expenses 10-25 Although R&D (which encompasses all of the first step and most of the second) is estimated to account, on the average, for no more than 15-30 percent of the total costs of innovation, it is especially significant in that R&D often initiates and provides the basis for the subsequent steps in the innovation process. Expenditures for industrial R&D The total national expenditures for industrial R&D3 are comprised of funds from both the Federal Government and private industry. The combined funding is shown in figure 4-1, in current and constant dollars. Total expenditures in current dollars more than doubled between 1960 and 1974, with one-third of the growth occurring after 1971. The average annual increase of just over $1.0 billion during that latter time was greater in absolute terms than in any other three-year interval between 1960 and 1974, and came almost entirely from an increase in funding by industry. Total expenditures in current dollars for industrial R&D were over $22 billion in 1974. The growth in current dollar funding, however, was less than increases in inflation in recent Figure 4-1 Industrial R&D expenditures, 1960-74 (Billions of Dollars) 20 Current dollars ^^ / 16 ^^^^r Constant 1967 dollars' ^ 12 i:::^^^ 8 - 4 - 1 1 1 1 1 1 1 1 1 1 1 i 19 60 '62 '64 'ee '68 70 72 7 4 1 GNP implicit price deflators used to convert current dollars to constant 1967 dollars. (est) SOURCE: National Science Foundation, - Tfi/iMo/o^ira/ Innovation: Us Environment and Management. Department of Commerce, 1967. For a discussion of other estimates of the distribution of costs associated with innovation, see Edwin Mansfield, et. al., Research and Innovation in the Modem Corjjoration, (New York: W. W. Norton, 1971). •" Industrial RiStD expenditures presented in this report include all costs incurred in support of Ri&D (i.e., salaries, laboratory equipment, overhead, etc.), but do not include associated capital expenditures. See Research and Development in Industry. 1973. National Science Foundation (NSF 75-315) p. 81, for further information on the scope of these costs. 84 years. As a result, total expenditures in constant dollars for industrial R&D were 11 percent lower in 1974 than in 1968-69, the years of highest funding, and approximately equivalent to the funding level of 1965. Some perspective on the size of the invest- ment in industrial R&D can be obtained by comparing it to other major investments by industry, such as those for new plant and equipment and for advertising. Such comparison is not intended to imply that identical factors determine levels of investment among the three areas. Indeed, the mix of investments in these areas varies from industry to industry. Trends in expenditures for the three purposes are shown in figure 4-2 for manufacturing industries which, as discussed later in this chapter, account for almost all industrial R&D expenditures. Total funds for industrial R&D were close in size to those for new plant and equipment during the early 1960's, but the latter grew more rapidly in subsequent years and by 1973 had exceeded total R&D expenditures in industry by approximately one-third. Expen- ditures for R&D from industry's own funds and for advertising-* were closely comparable throughout the 1960-73 period. Sources of funds for industrial R&D As a consequence of increasing funds from industry and a leveling off of Government funds, industry replaced the Federal Govern- ment after 1967 as the major source of funds for industrial R&D (figure 4-3). By 1974, industry supplied 62 percent of all such funds, compared with only 42 percent in 1960. Federal funds for industrial R&D — principally from the Depart- ment of Defense (DOD) and the National Aeronautics and Space Administration (NASA)— reached a plateau in the late 1960's, dropped some 10 percent in the early 1970's as NASA and DOD support declined, and recovered in later years as DOD funding rose. These changes were reflected most prominently in the aircraft and missiles industry, and to a lesser extent in the electrical equipment and communication industry. The extent of Federal support for industrial R&D differs substantially from one industry to Figure 4-2 Expenditures for industrial R&D, advertising, and new plant and equipment by manufacturing industries, 1960-73 Current dollars (Billions of Dollars) JU 25 New plant and equipment M 20 15 rZ" 10 .^ ^^' 5 Advertising ^^^^^^ ^^^^^ Industry-funded R&D 1 1 1 1 1 1 1961 63 65 '67 '69 71 1 3 1 Includes funds from all sources SOURCE, National Science Foundation and Department of Commerce. ■^^::.,:i^r^^:^.^^s^;^>r^ '• Includes expenditures by manufacturing corporations for newspaper, radio, television, magazine, and other miscellaneous local and national forms of advertising. 85 Figure 4-3 Expenditures for industrial R&D, by source of funds, 1960-74 Current dollars (Dollars m Billions) 14 1960 '62 '64 '66 '68 '70 1 GNP implicit price deflators usei) to convert current ilollars to constant 1967 dollars SOURCE National Science Foundation. I I I I I 1 '74 (est) Constant 1967 dollars' (Dollars in Billions) — 1 14 Federal Government Industry ^^^ I I I I I I I I I L^ I960 ■62 •66 •72 '74 (est) another, as shown in the table below. s Federal funds as a percentage of total industrial R&D expenditures, by industry, 1973 Industry Percent Aircraft and missiles 78 Electrical equipment & communication 50 Professional & scientific instruments 20 Motor vehicles and other transportation equipment 17 Machinery 16 Rubber products 12 Chemicals and allied products 10 Fabricated metal products 5 Primary metals 4 Petroleum refining and extraction 3 Stone, clay, and glass products 2 Textiles and apparel 2 Food and kindred products 1 Paper and allied products 1 5 Federal support for nonmanufacturing industries amounted to 56 percent of their total R&D expenditures in 1973. Industrial R&D personnel Another indicator of the magnitude of in- dustrial R&D is the number of scientists and engineers engaged in such activities. Trends in this measure are shown in figure 4-4, for the total of such personnel as well as for those who are supported by industry itself and those by R&D funds from the Federal Government. The total number of these personnel rose to a high of some 387,000 in 1969, declined 10 percent over the next three years, and rose slightly in both 1973 and 1974. The decline in 1970-72 was concentrated among those scientists and engineers, principal- ly the latter, supported by Federal R&D funds. The reductions, corresponding to the pattern of declines in Federal funding of industrial R&D described above, were primarily in the aircraft and missiles industry and secondarily in the electrical and communication, machinery, and chemicals industries. Some 70 percent of the reduction in numbers of federally supported R&D scientists and engineers was in these four industries. 86 Figure 4-4 Scientists and engineers engagedin industrial R&D, by source of funds, 1960-74 (Thousands) 400 .,^X_ 350 «yr ^^^^ 300 ^ 250 200 " Industry ^i 150 ^^r Federal Government ^V 100 50 - ^ 1 1 1 1 1 1 1 1 i 1 1 i 1 19 60 '62 64 66 '68 70 72 7 4 (January) 1 Full-time equivalent basis- SOURCE National Science Foundation. Trends in the total number of R&D scientists and engineers in industry paralleled constant dollar expenditures for industrial R&D throughout most of the 1960-74 period (see figures 4-1 and 4-3). Such a correlation might be expected since the cost to industry for these personnel — a cost which represents a large fraction of the total cost of industrial R&D — has increased at approximately the same rate as inflation. Thus, the similarity of the two trends provides support for the use of the GNP implicit price deflator as a gross adjustment of current dollar expenditures to reflect more accurately the real level of financial input and the magnitude of effort. ° R&D expenditures by specific industries The extent to which a specific industry invests in R&D is dependent upon a diversity of factors, including competition within the industry and from other industries, government regulations requiring improved performance of products, the need of substitutes for and the conservation of natural resources, and the availability of funds and personnel for R&D. Nearly all manufac- turing industries engage in some type of R&D activity. A few nonmanufacturing industries^ also perform R&D, but their effort represents less than 5 percent of all industrial R&D spending.* Expenditures for R&D in current dollars rose steadily in most industries from 1960 to 1973. In recent years, R&D spending has grown at a rate comparable to the early and mid-1960's, averag- ing over 7 percent per year between 1971 and 1973. The industries principally responsible for this growth are electrical equipment and com- munications, motor vehicles and other transpor- tation equipment, machinery," and chemicals and allied products. These four industries accounted for approximately 80 percent of the total increase in current dollar expenditures for R&D between 1971 and 1973. Trends in expenditures for R&D in these and other major R&D-performing industries are shown in figure 4-5. In five of the seven specific industries, 1973 was the peak funding year for R&D in both current and constant dollars. The only major industry experiencing a large decline in R&D spending in either current or constant dollars during the 1960-73 period was aircraft and missiles, which dropped sharply after the late 1960's. Other industries, not shown in the figure, in which R&D expenditures in current and constant dollars wereat their highest level in 1973 are: drugs and medicine, rubber products, fabricated metal products, communication equipment and electronic components, and optical, surgical and photographic instruments. " A more complete discussion of the use of deflators for R&D expenditures appears in the chapter entitled "Resources for R&D" in this report. ' These include, but are not limited to, agriculture, public utilities, finance, insurance, business services, medical and dental laboratories, and engineering and architectural serv- ices. * Research and Development in Induilry, 1973, National Science Foundation (NSF 75-315). ' Includes office, computing, and accounting machines; metal-working machinery; engines and turbines; farm machinery; construction, mining, and materials handling machinery. 87 Figure 4-5 R&D expenditures, by selected industries 1960-73 Current Dollars Constant 1967 Dollars' (Dollars in Millions) 6.000 Aircraft & missili 5,000 4,000 3,000 2,000 1,000 -•*' Motor vehicles and other transportation equipment Chemicals Sallied products All other m3ni]l,i(:tu(in3 inrtustfes _J ^ \ \ 1_ Mjchinety Professional & scientific instruments Noiinidnufacturing industiies '"— -Petioleun^ :HlinmL' ,^ extraction I I i \ . L_ (Dollars in Millions) 6,000 5,000 4,000 3,000 2,000 1,000 Motor vehicles and other transportation equipment Chemicals 8 allied products \ Machinery professional & scientific All other manufacturing industries instruments Nonmanufacturing industries \ I i I L Petroleum letmmg & extraction J \ \ I I I960 '64 '88 72 73 1960 66 72 73 1 GNP implicit price deflators used to convert current dollars to constant 1967 dollars- SOURCE National Science Foundation, Expenditures for R&D in nonmanufacturing industries changed little after 1970.1° Industries differ substantially in the size of recent changes in their R&D expenditures. Industries with the largest relative growth in R&D spending between 1971 and 1973 are shown in figure 4-6. •" R&D expenditures for all industries are presented in Appendix table 4-5. The overall pattern of R&D funding shown in figures 4-5 and 4-6 as well as elsewhere in this chapter, indicates a general shift in the Nation's industrial R&D effort. One aspect of the shift is that industry itself, rather than the Federal Government, has become the prime source of funds for industrial R&D. A second and related aspect is that the R&D is directed increasingly to "civilian" areas, i.e., to areas other than defense and space such as the development of new sources of energy, conservation of resources. 88 Figure 4-6 Industrial R&D expenditures, percent change, 1971-73 Current dollars Constant 1967 dollars' (Percent) -10 I Motor uetiicles and otiier transportation equipment Rubber products Professional and scientific instruments Machinery Electrical equipment and communication Fabricated metal products Lumber, wood products, and furniture Ctiemicals and allied products Stone, clay, and glass products Food and kindred products Textiles and apparel Paper and allied products Aircraft and missiles 10 10 20 30 40 3 GNP implicit piice deflalois used to convert current dollars to constant 1967 dollars SOURCE: National Science foundation. and improvement of the quality of the en- vironment. ^ 12 Energy is one of the civilian areas in which R&D expenditures have grown and are expected to increase still further in the years ahead. The exploration and development of new and alternative sources of energy for their own needs and the needs of the Nation as a whole have become important for many industries. As a result, expenditures by industry for energy- related R&D have risen almost 50 percent since " A similar stiift of federally funded R&D toward civilian areas is discussed in more detail in "Resources for R&D" in this report. 1- Historically, the Federal Government's role in industrial R&D dates from World War II, during which the principal emphasis was on defense-related R&D. Prior to that time. Federal support for industrial R&D was miniscule. (See Helen Wood, Scienlific Research and Development in American Industry. Bureau of Labor Statistics, 1953; and Vannevar Bush, Science — The Endless Frontier, a report to the President, 1945.) 1972, reaching an estimated $1.1 billion in 1974. iJ The petroleum industry is the leading per- former of energy R&D, with expenditures of $325 million in 1973 and an estimated increase of 25 percent in 1974. The electrical equipment and communication industry is the second largest performer; these two industries combined accounted for over 65 percent of all energy- related R&D activities in industry in 1973. Advances in technology in the use of fossil fuels (particularly oil and coal) and nuclear energy are the principal objectives of the industrial R&D effort in this area. ' ' "20-Percent Increase in Energy Activity Paces Industrial R&D Spending in 1973", Science Resources Studies Highlights. National Science Foundation (NSF 74-319), December 4, 1974. 89 R&D expenditures by character of work Development activities receive by far the largest portion of total expenditures for in- dustrial R&D, followed by applied and basic research . The proportion going for development efforts has ranged between 75 and 80 percent of total expenditures during the 1960-74 period, compared with nearly 20 percent for applied research and some 3 to 4 percent for basic research (figure 4-7). The emphasis on development efforts reflects the general nature of industrial R&D, which is usually focused on specific product lines and relatively short-range goals in terms of the time Figure 4-7 Industrial R&D expenditures for basic research, applied research, and development, 1960-74 (Billions of Dollars) 18 Basic research J \ L I960 1 GNP implicit price deflators used to convert cuirent dollars to constant 1967 dollars. SOURCE National Science Foundation. J 74 between R&D and expected returns from the investment. These tendencies are strengthened by the usually large proportion of total corporate R&D resources — funds and personnel — which are controlled by divisional managers of firms whose major focus is often on existing product lines and processes. Expenditures in current dollars for applied research and development generally increased each year between 1960 and 1974, whereas funding of basic research has remained at a relatively fixed level of some $600 million since 1965. Constant dollar expenditures, on the other hand, declined after 1969 for development efforts, due primarily to reductions in Federal funds, whereas those for applied research have changed little since 1964. Funding of basic research in constant dollars has fallen since the mid-1960's, reaching a level in 1974 which is approximately equal to that of 1961. The distribution of funds among these categories of R&D differs according to the sources of funds, with industry providing most of the funds for basic and applied research and a lesser, but still the largest, share of development funds. In 1^74, for example, industry funded 78 percent of its own basic research and 75 percent of its applied research, compared to 59 percent of its development. '^ Applied R&D in product fields Over the last two decades there has been a rapid expansion and diversification of firms into new product lines, markets, and technical fields. Thus, R&D data reported by a firm may include expenditures in several product fields, in addi- tion to the single, major field which determines the broad industrial category to which the firm is assigned. Therefore, R&D expenditures in terms of product areas, rather than industries, are more indicative of the actual composition and focus of the national effort in industrial R&D. Expenditures for applied research and development's are concentrated in 6 of the 15 broad product fields used for classification purposes.!"^ These 6 fields, and the percentage of n Nalkvuil Paltmii of R&D Resouras. 1953-75, National Science Foundation (NSF 75-307). '5 Expenditures for basic research are excluded since such research, by definition, is not directed toward specific products. lo See Appendix table 4-12 for a listing of these product fields. 90 the total applied R&D expenditures each received in 1973, are shown in the table below. i^ Distribution of applied R&D expenditures, by selected product field, 1973 Product field Percent Communication equipment and electronic components 17 Aircraft and parts 12 Guided missiles and spacecraft 12 Machinery 11 Motor vehicles and other transportation equipment 10 Chemicals"* 7 Substantial changes in applied R&D expen- ditures have occurred in more specific product fields in recent years. Fields with an overall increase or decrease in constant dollar expen- ditures of 10 percent or more during the 1971-73 period are cited below. Concentration of industrial R&D The U.S. industrial R&D effort is concen- trated within relatively few industries, and within a small number of large companies within these industries. Throughout the 1960's and early 1970's, over 80 percent of all industrial R&D expenditures and over 11 percent of industrial scientific personnel engaged in R&D were concentrated in only five industries — aircraft and missiles, electrical equipment and communications, chemicals and allied products. machinery, and motor vehicles and other transportation equipment (figure 4-8). The largest change over the period occurred in the aircraft and missiles industry, where R&D expenditures declined significantly in relative terms after the mid-1960's. Similar trends are evident in the concentration of R&D scientists and engineers in these same five industries. The aircraft and missiles in- dustry is seen to account for a declining proportion of the total industrial R&D per- sonnel resources beginning in 1963. This industry, however, in combination with the electrical equipment and communications in- dustry employed over 46 percent of all scientists and engineers engaged in industrial R&D in 1973 (figure 4-9). To a significant degree, the concentration of industrial R&D in a few industries reflects the influence of Federal R&D contract work, primarily in the defense and space areas. In 1973, for example, almost 92 percent of all federally funded R&D in industry went to these five industries. Federal funds to these industries ranged from about 10 percent of the total R&D expenditures in the chemicals and allied products industry, to some 50 percent in electrical equipment and communications, and to over 80 percent in the aircraft and missiles industry. Together, Federal funding for R&D in these five industries represented over 35 percent of the total expenditures for industrial R&D in 1973. A similar pattern is observed in regard to the concentration of scientific personnel. The five industries cited above employed 90 percent of all Increases of more than 25 percent Percent change in constant dollar applied R&D expenditures, by product field, 1971-73 Increases of 10-25 percent Decreases of 10 percent or more Ferrous metals & products Transportation equipment, except motor vehicles . . . Textile mill products Motor vehicles & equipment Professional & scientific instruments Engines & turbines Rubber & plastics products 87 Electrical industrial apparatus S^ Farm machinery & equipment 36 Office, computing, and accounting machines 35 Ordnance, except guided missiles 32 Stone, clay, & glass 30 products 26 Communication equipment & electronic components 1" For additional data on these and other product fields, see Research ami Development in Imiuslry, 1973, National Science Foundation (NSF 75-315). 1' Except drugs and medicine. 24 19 19 11 10 10 Guided missiles Metalworking machinery and equipment Nonferrous metals & products Agricultural chemicals . . Plastics materials & synthetic resins -20 -20 -19 -19 -10 91 Figure 4-8 Distribution of industrial R&D expenditures among selected industries, 1960-73 (Percent of Total) 40 ftircraft and missiles Chemicals and allied products Motor vehicles and other transportation equipment 4 i< " — "^. -#.*r T.**^ M<»-' Machinery J LJ I L _LJ L^ I960 '62 '64 'Se SOURCE National Science Foundation. 70 72 73 Figure 4-9 Distribution of scientists and engineers engaged in industrial R&D among selected industries, 1961-74 (Percent of Total) 40 Aircraft and missiles Chemicals & allied products *••** Machinery ^^^m Motor vehicles and other transportation equipment I I I I I I L^ J \ L 1961 '63 '65 '67 '69 I Full-time equivalent basis. (January) SOURCE: National Science Foundation. 73 74 federally supported R&D scientists and engineers in industry as of January, 1974, representing almost 30 percent of all R&D scientists and engineers employed by industry. Most of the R&D activity in industry is further concentrated within a small number of large companies. i" Of a total of over 11,000 •" The words "company" and "firm" are used in- terchangeably in this report, even though they may have slightly different meanings in other contexts. Each term denotes a business organization consisting of one or more establishments under common ownership or control. R&D-performing companies in 1973, the 300 companies with over 10,000 employees accounted for more than 80 percent of all industrial R&D expenditures. Thirty-one of these companies reported R&D programs costing more than $100 million, for a total of almost $12 billion, or more than 60 percent of all R&D expenditures by industry. Thus, even small percentage changes in the level of R&D activity in a few large companies can have a substantial effect on the overall U.S. industrial R&D effort. When viewed against the totality of companies which comprise industry, or even the manufac- 92 turing sector akine, the number oi individual companies with a formal R&D program is comparatively small. For example, in 1967 (the latest year for which Census data on the total number of manufacturing companies is available) only 11,200 companies, or less than 5 percent of ail manufacturing companies, reported having any R&D program. Further- more, the proportion of companies conducting R&D differed substantially by company-size groups. In 1^67, only 4 percent of all manufac- turing companies with under 1,000 employees conducted any R&D, while 55 percent of companies employing between 1,000 and 5,000 persons, and 88 percent of companies with 5,000 or more employees reported such efforts. Among the performers of R&D is a subset of the small company group which consists of "high technology" firms whose main objective is the performance of R&D and the development of new products. These new research-based enter- prises represent only a small percentage of the total industrial R&D effort, but they have often evolved into large firms that dominate market segments and, in some cases, entire industries. These science-based firms are predominantly located in industries such as electronics, com- munications, computers, aircraft, and nuclear and medical instruments. ^o R&D intensity The proportion of an industry's human and financial resources which are utilized for R&D may be regarded as a measure of the "R&D intensity" of that industry. The indices used frequently for quantifying the level of R&D intensity are (1) total and company funds expended for R&D as a percentage of net sales and (2) the number of R&D scientists and engineers per 1,000 employees. Based on these indices, each of the 15 largest R&D-performing industries in the manufacturing sector was Measures of R&D intensity, by industry, 1961-72 Mean over the 1961-72 period Industry Group I Chemicals & allied products Machinery Electrical equipment & communications Aircraft & missiles Professional & scientific instruments Mean for group I Group // Petroleum refining & extraction Rubber products Stone, clay & glass products Fabricated metal products Motor vehicles & other transportation equipment Mean for group II Group 111 Food & kindred products Textiles & apparel Lumber, wood products & furniture Paper & allied products Primary metals Mean for group III -" For further information on R&D in small companies, see Thomas Hogan and John Chirichiello, "The Role of Research and Development in Small Firms", in The Vital Majority: Small Business m the American Economy, Small Business Administra- tion, 1974. R&D scientists Company funds & engineers Tot il funds for for R&D as a per 1,000 R&D as a percent percent of employees of net sales-i net sales^i 37.8 4.0 3.5 25.9 3.9 3.1 47.2 8.5 3.6 88.6 20.9 3.5 33.9 5.9 4.1 47.1 8.2 3.5 15.8 17.8 10.7 12.8 19,4 16.4 7.2 3.1 4.7 8.5 5.6 6.0 0.9 2.0 1.6 1.3 3.3 1.9 0.4 0.5 0.5 0.9 0.8 0.6 0.9 1.7 1.5 1.2 2.5 1.6 0.4 0.5 0.4 0.8 0.8 0.6 -' Total net sales by Group I industries over the entire 1961-72 period were only 25 percent larger than sales by industries in Group II and approximately 50 percent larger than those of Group III industries. 93 placed into one of three groups according to its relative level of R&D intensity over the 1961-72 period. These groups, each consisting of five industries, are shown in the table below along with their R&D intensity indices, with Group I industries being the most R&D-intensive and Group III the least. As the table shows, the level of R&D inten- siveness among industries within each group is within a relatively close range regardless of the specific index chosen. Furthermore, the R&D intensity of each group is separated from the next by approximately a factor of three. The R&D intensity of manufacturing in- dustries overall has declined steadily since 1964. As shown in figure 4-10, the declines occurred almost exclusively in the most R&D-intensive industries (Group I), and were caused primarily by reductions in the level of Federal support for industrial R&D after the mid-1960's. Aside from the changes produced by declining Federal support, the R&D intensity of each group of industries has changed little since 1961. The assignment into three major groups of industries exhibiting approximately the same level of R&D activity provides a convenient and direct method for relating R&D to the outputs and returns from such effort, as shown in the next section of this chapter. OUTPUTS FROM R&D AND INNOVATION Earlier sections of this chapter dealt largely with the resources employed in industrial R&D and structural aspects of the system. This section attempts to provide indicators of some of the outputs and returns from R&D— aspects which are considerably more difficult to measure than inputs. The principal indicators in this section deal with trends in technological invention and innovation. Measures of invention are in terms of patents and include indicators of the areas and magnitude of inventive output, sources of invention, product areas involved, and relationships with R&D intensity. Indicators of innovation are based upon new industrial products embodying major advances in technology and include characteristics of in- novating organizations, innovativeness of different industries, relationship to investment in R&D, time between invention and innovation, and the role of research. The measures of output presented here represent only a step toward the array of indicators needed for the industrial sector. Present measures are small in number, broad in scope, and restricted generally to relatively direct outcomes of R&D. Lacking are indicators such as reduced costs, gains in an industry's productivity or increases in sales which result from R&D. The measures, in addition, do not encompass the qualitative improvements in industrial products and processes which often constitute the major form of return from R&D investment. The present indicators, furthermore, do not specify the separate and distinct contribution of R&D to invention and innovation. As noted earlier, invention and innovation result from a complex of interacting factors— economic, social, and technical. Analytical efforts to date have not been successful in determining the precise contribution of the individual factors, including that of R&D. Thus, indicators presented here should be regarded as ap- proximate measures of the relationship between this factor and invention or innovation. The impact of technological innovation on productivity and economic growth is, in turn, understood only in general terms. Present knowledge of the causal connection between innovation and economic returns is not suf- ficient for developing quantitative indicators of the relationship. In lieu of such indicators, the major conclusions derived from studies in this area are summarized. Finally, the indicators in this chapter do not include measures of the negative impacts and side-effects of technological innovation. These costs may be extensive in human and social terms, ranging from the loss of jobs to the pollution of our environment. The determina- tion of these costs and their assessment relative to benefits, is necessary for the wise manage- ment of innovation. Valid indicators of these costs, however, are exceedingly difficult to develop; it is for this reason, rather than a lack of recognition of their need, that such indicators are not provided in this report. Objectives of industrial R&D In general, industry views R&D as a means to remain competitive and profitable. Most in- dustrial R&D is an attempt to focus science and technology on improving existing products and 94 Figure 4-10 R&D intensity of U.S. manufacturing industries, 1961-72 (a) Total R&D as a percent of net sales (b) Company funds for R&D as a percent of net sales (Percent) (Percent) 11 c 10 - •J ^^^^r ^^^ Group 1 4 Group 1 ^ 8 ^^^^ ^ ^^— ^-^^^X^^^^*^^**-^^ >*^^ 3 - 6 - - 2 _ 4 - ■^_ ^^ Group II 2 Group II 1 — n _ Group III n ••••*• Group III 1 1 1 1 1 1 I 1 1 ....................... .... ...... U ' »:~^-n-rTri-r' 1961 '63 65 SOURCE U.S. Patent Office. 71 73 1973, foreign residents were granted over 30 percent of all U.S. patents, as compared with 16 percent in 1960. (The significance of foreign patent activity in the U.S. is discussed in the chapter entitled "International Indicators of Science and Technology" in this report). The patent output of U.S. industry (i.e., patents assigned to U.S. corporations) accounted for the largest proportion of total patents granted throughout the 1960-73 period. ^s The " A recent report, A Review of Patent Ownership, Office of Technology Assessment and Forecast, U.S. Patent Office, lanuary 1975, identified specific companies involved in active technological areas. 97 actual number of patents assigned to U.S. corporations increased 73 percent between 1960 and 1973. "Assignment," however, cannot be equated completely with the actual source of the invention. Some patents granted to individuals may be assigned subsequently to corporations, and some patents assigned to the Federal Government have their origins in federally funded R&D performed in industry. Nevertheless, it is clear that industry is the major producer of patented inventions in the U.S. Patent output by product field. In addition to the sources of patents, information was obtained on the product fields in which the patents were most likely to be applied. Through a concordance developed between the patent classification system of the U.S. Patent Office and the Standard Industrial Classification (SIO^" system, it was possible to categorize U.S. patents granted between 1963 and 1973 into 30 SIC- based product fields,-'" with respect to the fields in which the patents were most likely to be applied. These product fields encompass most of the manufacturing sector of industry, and include 96 percent of all U.S. patents granted during the period. All patents granted to U.S. citizens, cor- porations, and the Federal Government were assigned to these product fields on the basis of the area of their probable use.-'i The six product fields with the highest patent activity are shown in figure 4-12. The greatest number of patents during the 1963-73 period were applicable to the machinery product field, and within this field to the construction, mining and materials handling machinery subfield. Following machinery in the number of patents were fabricated metals, chemicals (particularly basic industrial chemicals), electrical equipment, communication equipment, and professional and scientific instruments. These fields include many of the areas with a high output of major innovations. (See the later section of this chapter entitled 'Technological innovation".) -" Slnndani Industrial Classification Manual, Executive Office of the President, Office of Management and Budget, 1972. 5» Indicators of the Patent Output of U.S. Industry. Office of Technology Assessment and Forecast, U.S. Patent Office, 1974. (A study commissioned specifically for this report). 3> Because of the possible utilization of the technology represented by a given patent in more than one product field, many patents w^ere counted more than once. For this reason, prcxluct field totals do not correspond with the patent totals presented in the previous section. Figure 4-12 U.S. patents granted for inventions,' by selected major product field, 1963-73 (Thousands) ^3 24 ,.-\ 23 _ ,• •. (Vlachinery 22 — • • • • 21 • \ : \ • • - • . • • • •- • \ 20 / V \ / K 19 1 18 • • 17 ..••• 16 - 15 - 14 ^^^^^ Fabricated metals 13 / X 12 — M Electrical ^ ^k ^^ # equipment ^ ^^\ M ^^ 11 / ^^^^^ V^ A^ 10 >* / /^ /^A 9 g If If V ''•'/ Chemicals ' 7 ' ;mI Pfofessional 6 -r / scientific ? instruments , '\ 5 Communications equipment 4 " 3 - 2 - 1 - ! 1 1 1 1 1 I 1 1 1963 64 '65 66 '67 '68 '69 '70 '71 '72 '73 (est.) 1 Patents originating in the United States. SOURCE Office of Tectinology Assessment and Forecast, US, Patent Office 98 Patent output by product field and R&D intensity. Patents can also serve as an indicator of the inventive output of specific industries. An approximate correspondence exists between product fields of patent activity and the in- dustries which produce the patented invention. The correspondence is less than perfect, since many companies in a specific industry may be active in a number of diverse product areas. An invention produced by the electrical equipment industry, for example, may have its principal application in the aircraft product field. -'^ The relationship between patent output and R&D intensity is shown in figure 4-13.'-' Those industries which devote the largest proportion of their resources to R&D (Group I industries) are by far the largest producers of patented inventions, accounting for 67 percent of all U.S. patents granted between 1963 and 1973. Group II industries — lower in R&D intensity than Group I — produced about 29 percent of the patents over the period, while the least R&D- intensive industries (Group III) produced only 4 percent.-'^ During the same period (1963-73), Group I industries were responsible for 80 percent of the total expenditures for industrial R&D; Group II industries, 16 percent; and Group III industries, 4 percent. Technological innovation Technological innovation occurs when new or improved products, processes, or services em- bodying advances in technology are introduced into the market. Although R&D has a major role in the process, innovation takes place in a broad context in which economic, social, and political factors may be crucial. '^ It has been estimated, for example, that of every ten products emerg- -'- The lack of correspondence, however, was reduced by grouping industries according to their R&D intensity; these groups produced patented inventions which tended to be utilized by industries within the same group. '-' For a concise review of the relationship between R&D and patents, see Dennis Mueller, "Patents, Research and Development and the Measurement of Inventive Activity," ]oumal of buiuilnal Economics. Vol. 15 (November 1966), •'■' The patent totals upon which these percentages are based include some multiple counts, but only those which occur across the three groups. The extent of this multiple counting is approximately 8 percent of the total patents granted. 35 P. Kelly, et al.. Technological Innovation: A Cnlical Review of Current Knowledge, (Atlanta: Georgia Institute of Technology, 1975). Figure 4-13 U.S. patents granted for inventions in major product fields by groups of R&D-intensive industries, 1963-73 (Percent of Patents) 100 90 60 40 30 20 10 — Group I Group II Group III T---r' J \ [ \ L 1963 '65 '67 '69 71 73 SOURCE Office of Tecfinology Assessment and Forecast, U.S. Patent Office. ing from R&D, five fail in product and market tests, and only two become commercial successes.'"^ Technological innovation is integral to the operation of many industries and crucial to their survival and growth. Innovation in these industries may be the principal means for acquiring new markets and maintaining existing ones, as well as for improving internal produc- tion processes and reducing costs.'" Other industries, while producing few major in- novations themselves, purchase goods which embody innovations from the first group of 3" E. A. Pessemier, New Product Decision: An Analytical Approach. (New York: McGraw-Hill, 1966). '" For a review of factors which determine a firm's effectiveness in innovation, see James M. Utterback, "Innovation in Industry and the Diffusion of Technology," Siifiiif, 'Vol. 183 (February 15, 1974), pp. 620-626. 99 industries; these goods may enable the purchas- ing industries to increase their productivity, improve the quality of their products, or develop new and improved products and services for their own markets. The computer is one of the most obvious examples of an innovation from the first group of industries which is used extensively by other industries. In addition to its importance at the firm and industry levels, technological innovation is acknowledged as a prime source of the Nation's economic progress, contributing to productivity and economic growth.-'* The capability for such innovation, moreover, is regarded as a major comparative advantage which the United States has in international relations — political, military, and economic.-" The present indicators focus on major technological innovations. The vast majority of innovation efforts, however, seek or attain modest improvements in products and processes, rather than major advances. The results of these efforts are not captured by present measures even though the cumulative impact of the more numerous minor advances may often exceed that of major innovations. -to Furthermore, no indicators are provided of the extent to which the innovations replace or represent advances over existing products and processes. In addition to these limitations, the indicators do not specify the economic and social benefits — and costs — associated with the in- novations. Indicators of trends in innovation presented in this section are based, for the most part, on a study conducted specifically for this report. The study provides information on 500 major product innovations which were introduced into the market during the 1953-73 period by leading industrialized nations. -"i The innovations were selected by an international panel of experts as representing the most significant new industrial products and processes, in terms of their -"" These aspects are discussed later in this chapter in the section entitled, "Returns from R&D and Innovation" '" See the chapter "International Indicators of Science and Technology" in this report for indicators of the role of technology in international trade '" Jacob Schmookler, Patents. Imentwn. and Economic Change, (Cambridge: Harvard University Press, 1972). " This information was used in devising indicators of the relative innovativeness of the United States and other major developed nations; these indicators are presented in an earlier chapter of this report, entitled "International Indicators of Science and Technology." technological importance and economic and social impact.''- Information on a subset of these innovations — a total of the 319 produced by U.S. industries — was used to develop the measures of innovation presented below. The innovations on which the indicators are based span a wide range of technologies and all major manufacturing and nonmanufacturing industries. The diversity is suggested by the following innovations which were among those included in the study. Integrated circuits Lasers Microwave transmission Cortisone synthesis Permanent magnetic alloys Weather satellites Double-knit synthetics Computer time-sharing Light-emitting diodes Textured granular protein Innovation and company size. A topic of enduring concern is the relationship between size of firm and technological innovation. ■•-' This topic was examined through the use of the major innovations described above. The results are shown in figure 4-14, in terms of the percentage of innovations produced by companies in each of five size categories. These data, which are based on a total of 277 innovations, ^-i show that large manufacturing companies (those with 10,000 or more employees) produced the greatest propor- tion of major innovations, followed by firms in the two smallest size categories. Companies of intermediate size (1,000-4,999 and 5,000-9,999 employees) accounted for the fewest in- novations. The data also show that the number of innovations from large companies has in- creased over time, in both absolute and relative terms. Small firms, however, are sometimes regarded as those with less than 1,000 employees. By this J- For details of the methodology employed in the study, see huiicalon ol Internatumal Trends in Technological Innovation, Gellman Research Associates, Inc., 1975. (A study com- missioned specifically for this report). ^■' For a discussion of factors related to firm size which may influence innovation, see Sumner Myers and Donald G. Marquis, Successful Industrial Innovations, A Studu of factors Underlying Innovation w Selected Firms, National Science Founda- tion (NSF 69-17). •■^ Here, and elsewhere, the number of innovations used for analysis may be less than the total 319 innovations mentioned earlier because of the unavailability of specific data on all the innovations, or the consideration of only those originating from manufacturing industries. 100 Figure 4-14 Distribution of major U.S. innovations, by size of company, 1953-73 [Percent of Innovations) 45- 30 — 25- 20- 5 - I 1953 59 1960-66 1 1967-73 I 1-99 SOURCE: Gellman Research Associates, inc. 100-999 10004999 5000-9999 Company size (number of employees) 10000-1- definition, the small firm — rather than the large one — was the site of the greatest number of major innovations during the 1953-59 and 1960- 66 periods, but not in the 1967-73 period. The matter of firm size and innovation can be viewed also with respect to the sales volume associated with companies of different size. When this aspect is considered, the smallest companies are found to produce proportionally more innovations per unit sales than larger companies (figure 4-15). Small firms, further- more, have maintained a relatively higher level of innovative output per unit sales than larger companies in each of the time periods for which sales data are available. ■'-^ (The decline in the '^ Data on sales and receipts of manufacturing industries, in terms of company size, are available only for the years 1958, 19o3 and 1967 and are taken from Enterpme Slatislics. Department of Commerce, Bureau of the Census, 1968 and 1972. number of innovations per unit sales, observed in each company size category, results from a combination of increasing company sales and a relatively constant number of innovations.) These indicators shed some additional light on the question of the relationship between firm size and technological innovation. The in- dicators, however, are dependent on the partic- ular set of innovations selected for study. Furthermore, all industries are treated as if they were alike, even though differences among them with respect to innovativeness may exceed the differences between small and large firms. Finally, the indicators offer no insights regard- ing the attributes, causal factors, and dynamics which determine the relative innovativeness of various size companies. For these and other reasons, interpretations of indicators presented here should be limited and tentative. 101 Figure 4-15 Major U.S. innovations per $10 billion in sales, by size of company, 1953-73 (Number per $10 Billion) 3.5 — 199 100-999 1000-4999 5000-9999 Company size (number of employees) SOURCE Geltman Research Associates, Inc., and Department of Commerce 10000 + Innovation and R&D intensity. The most R&D-intensive industries (Group I industries) produced the largest fraction of the major U.S. innovations in the manufacturing sector — 182 of the 277 innovations included in this analysis — during the 1953-73 period, followed by in- novations from industries in Groups II and III (figure 4-16). Innovations by Group I industries comprised 66 percent of the total, compared with 24 percent in Group II and 10 percent in Group III industries. Group I industries accounted for 80 percent of the total industrial R&D expenditures over roughly the same period (1956-73), compared to 16 percent from Group II, and 4 percent from Group III. Over the 1953-73 period as a whole, the number of innovations from the most R&D- intensive industries increased to a greater extent than those from the other two industrialgroups. After the 1965-68 period, however, the number of innovations in Group I industries declined in relative terms. Within these industry groups, the largest number of innovations — 171 — are in four of the most R&D-intensive industries: electrical equip- ment and communication; chemicals and allied products; machinery; and professional and scientific instruments (figure 4-17).-'° It should be noted that innovations in the defense and space areas are not included unless they were introduced into the commercial market; this may account, at least in part, for the relatively small number of innovations from the fifth Group I industry — aircraft and missiles. Innovations in the manufacturing sector were examined to identify the major areas of in- novative activity and the shifts among these areas during the 1953-73 period. For this purpose, innovations were classified in terms of their product fields through use of the Standard Industrial Classification (SIC). The product fields with the largest number of innovations are listed below for each of three time periods. The fields are described briefly in terms of their corresponding three-digit SIC designations, and ranked in approximate order of the number of associated innovations. ^'> See Appendix table 4-17 for the number of major innovations in each of the 15 manufacturing industries. 102 Figure 4-16 Percent distribution of major U.S. innovations, by groups of R&D intensive industries, 1953-73 (Percent) 90 - 80 Group 1 ^^^^^^^^ 70 ^^1 1 '"^^ ^^^ 60 - ^r 50 - 40 - 30 20 ^V Group II 10 _ ••,_ Group III 1 1 1 1953-56 1957-60 1961-64 1965-68 1969-73 SOURCE: Gellman Research Associates, Inc. Figure 4-17 Major U.S. innovations in selected industries, 1953-73 Industry Electrical equipment and communication Chemicals and allied products Professional and scientific instruments (Number of Innovations) 5 10 15 20 25 30 35 40 45 50 55 I I I I I I I I I I Motor vetiicles and ottier transportation equipment ' Innovations in ttie defense and space areas are not included unless ttiey were introduced into the commercial market SOURCE Gellman Research Associates, Inc. 1967-73 1953-59 Medicinal chemicals & pharmaceutical products Industrial organic chemicals Electronic components and accessories Electronic calculating and computing machinery Metalworking machinery and equipment Machinery for specific industries Photographic equipment and supplies 1960-66 Electronic components and accessories Communications equipment Electronic calculating and computing machinery Synthetic materials Plastic films, sheets, and cellulose products Medical instruments and supplies Abrasives, asbestos & nonmetallic mineral products Electronic components and accessories Photographic equipment and supplies Motor vehicles and other transportation equipment Machinery for specific industries Abrasives, asbestos & nonmetallic mineral products Communications equipment Synthetic materials The prominent role of electronics is evident in each of the three periods, particularly during the early 1960's. The relatively large number of innovations in this area is due, in part, to significant advances in scientific knowledge in fields closely related to electronics.''^ In- •" Richard R. Nelson, et al.. Technology. Economic Growth, and Public Policy, (Washington, D.C.: The Brookings Institution, 1967). 103 novations in product fields associated with chemicals appear to have declined somewhat since the 1950's. Another set of technological innovations was used in developing an additional indicator for this report. ■»» This set consisted of "IR-100" award winners, one hundred of which are selected annually by the Editorial Advisory Board of Industrial Research magazine. The awards, begun in 1963, identify significant technological advances and recognize innovators and organ- izations responsible for such developments. The innovations are selected on the basis of their importance, uniqueness, and usefulness from a technical standpoint. They are chosen, in general, from advances in technology which have particular interest for the industrial research community; for this reason, the innovations tend to concentrate in areas such as scientific instruments, electronic apparatus, and new industrial materials. The innovations, therefore, represent a somewhat limited seg- ment of the total U.S. innovation activity, and do not reflect market success nor economic impact. Each of the more than 1,200 "IR-100" award- winning innovations chosen over the 1963-74 period was classified according to the SIC designation of the industry of origin and grouped in terms of its industry's R&D intensi- ty. c5ver 75 percent of the innovations were found to originate in industries included in the three groups of R&D-intensive industries. (The remainder originated primarily in nonmanufac- turing industries, academic institutions, or U.S. Government agencies.) As shown in figure 4-18, the most R&D-intensive industries (those of Group I) were responsible for the largest share of innovations over the twelve year period, accounting for about 62 percent of all the "IR- 100" awards. Industries in Group II claimed approximately 10 percent and Group III in- dustries some 4 percent of the total innovations. The preponderance of innovations in Group I results primarily from the large number of innovations originating in the electrical equip- ment and communications industry and the professional and scientific instruments industry. Together, these two industries accounted for over 45 percent of all the "IR-100" awards given over the twelve year period. Figure 4-18 "IR-100" award-winning innovations', by groups of R&D-intensive industries, 1963-74 (Percent of Total Awards) 100 "» Indicators of the Output of New Technological Products from Industry. Battelle Columbus Laboratories, 1975. (A study commissioned specifically for this report). 1963 1 Industrial Research magaiine's annual awards for the 100 "most significant new technological products of the year." SOURCE: Battelle Columbus laboratories. Time between invention and innovation. The innovation process — extending from the "first conception" of the innovation to "first realization"— may cover a long period of time. This interval may be necessary, among other things, to conduct research, determine the technical feasibility of the potential innovation, design and test engineering prototypes, develop the required manufacturing capability, and perform market analyses. The period is difficult to define precisely, since invention and innova- tion usually occur as stages in the process, rather than as discrete events. Roughly, invention occurs when initial determination of the technical feasibility of a new idea is made, while innovation corresponds to the actual commercial development and marketing of the new product or process. The invention-innovation intervals are approximate, and are usually not strictly comparable from one study to the next. 104 In a study of a selected number of major innovations,^" the interval between invention and innovation appeared to decrease over time, as suggested by the following gross historical trends. 5'^ Average time between invention and innovation Time period Years Early 20th century (1885-1919) 37 Post-World War I (1920-1944) 24 Post-World War II (1945-1964) 14 It is generally advantageous for an industrial firm to minimize the time between invention (i.e., the first conception of an innovation) and its introduction into the market. Competing firms may introduce similar products earlier, giving them a favored position in the market; the cost of capital may increase; or a loss in sales and profits may be experienced due to a lag in the introduc- tion of innovations into the market. These and other considerations usually encourage rapid innovation in order to reduce risk and increase profitability. Trends in the time between invention and innovation were calculated from a set of 277 innovations^! associated with the three groups of manufacturing industries which varied in respect to R&D intensity. The invention- innovation intervals, which ranged from less than one year to 82 years, were determined for each industry group and for all manufacturing industries combined. The results are shown in figure 4-19 in terms of the mean number of years between invention and innovation for three time periods between 1953 and 1973. These data suggest that the invention- innovation interval was shorter in recent years (7.2 years during the 1967-73 period) than in the ■'" Frank Lynn, "An Investigation of the Rate of Develop- ment and Diffusion of Technology in Our Modern Industrial Society", Report of the National Commiiswn on Technology. Automation, and Eiononuc Progress. 1966. 5" For other studies of the invention-innovation interval, see Edwin Mansfield, The Economics of Technological Change. (New York: W. W. Norton, 1968), and Interactions of Scienceamt Technology in the Innopative Process: Some Case Studies. Battelle Columbus Laboratories for the National Science Foundation, March 1973. 5' The 277 U.S. innovations are among those identified in Indicators of International Trends in Technological Innovation. Gellman Research Associates, 1975. Figure 4-19 Mean time in years between invention and innovation, by groups of R&D-intensive Industries, 1953-73 (Mean Time in Years) 16- ■I All manufacturing Industries 14 _ t Group I Group II ■■Group I 1953.59 1960.66 ■ Insufficient number of innovations for delermininE mean of Group III industries SOURCE Gellman Research Associates, Inc 1967-73 1950's (7.8 years), but generally somewhat longer than in the early 1960's (6.9 years). '^ ^^ Furthermore, the time between invention and innovation appears to correlate with R&D intensity. Industries with the largest fraction of financial and human resources for R&D tend to translate inventions into innovations more quickly than industries which are less R&D- intensive. In each of the three periods, the mean invention-innovation interval for industries of Group I was shorter than the interval for Group II which, in turn, was shorter than that for Group III industries. ^^ It has been suggested that the actions of Federal regulatory agencies may be responsible, in part, for the lengthening of the invention-innovation interval in recent years. -^-' These data differ from those presented for the U.S. in the chapter, "International Indicators of Science and Technology"; the invention-innovation intervals in that chapter are based upon all industries rather than the smaller set of selected manufacturing industries encompassed in this section. 105 "Radicalness" of the innovations. Innovations may range from imitations of existing tech- nologies to developments of radically new technologies and products. At one end of the spectrum, little or no new knowledge may be involved in an innovation, while at the other end, new and fundamental advances in knowledge may constitute the basis for the innovation. The distribution of innovations along this spectrum was estimated by obtaining ratings of the "radicalness" of the innovations. These ratings were made by the innovating organizations themselves. Although inherently subjective, such ratings may provide some valid insights regarding trends in industrial innovation. Each innovation was assigned to one of five categories which together form the "radicalness" continuum: "no new knowledge required", "imitation of existing technology", "improvement of existing technology", "major technological advance", and "radical breakthrough". S'' Of the 225 innovations for which ratings were obtained, only 17 were rated in the first two categories; these innovations were omitted in subsequent analyses. Included among the innovations rated as radical were integrated circuits, permanent magnetic alloys, and L-Dopa, which is used in the treatment of Parkinson's disease. Innovations regarded as representing major technological advances in- cluded hand-held solid state calculators, Ketalor (an anesthetic), and an ultrasonic process for the joining of synthetic fibers. Improvement of existing technology was represented by such innovations as a high-speed phototypesetting machine, resin catalysts, and Pyroceram, a hard, light-weight, and heat-resistant material. Innovations involving the improvement of existing technology were most prevalent, followed in order by those which constitute a major technological advance and the set which represents radical breakthroughs (figure 4-20). Over the 1953-73 period as a whole, 41 percent of the 208 innovations included in the analysis were rated as improvements in existing technology, compared with 32 percent in the category of major technological advance and 27 percent in the radical class. The most significant change in this distribution during the period -•■ The "radicalness" of innovations, it may be noted, does not determine their economic or social significance. In- novations which represent improvements or even imitations of existing technologies may have greater economic returns or social consequences than more radical innovations. Figure 4-20 "Radicalness" of major U.S. innovations, 1953-73 (Percent of Innovations) 50 — ^1 improvement of existing technology 1 Major technological advance *!^ Radical breakthrough 40- 20- 1953-59 1960-66 SOURCE: Gellman Research Associates. Inc. 1967-73 centered on the latter two categories. The number of innovations rated as radical breakthroughs declined nearly 50 percent between 1953-59 and 1967-73, while the number representing major technological ad- vances doubled during the same period. As a result of these changes, radical innovations accounted for 18 percent of the innovations in the 1967-73 period, down from 35 percent in 1953-59. The overall decline in radical innovations (and the corresponding increase in innovations representing a major technological advance) is due primarily to reductions in the number of such innovations from the most R&D-intensive industries (figure 4-21). Radical innovations in these industries decreased from 23 percent of the innovations in 1953-62 to 14 percent in 1963-73, whereas the proportion involving major technological advances rose from 20 percent to 30 percent over the same periods. Research and innovation. The technology embodied in an industrial innovation may be 106 Figure 4-21 "Radicalness" of major U.S. innovations, by groups of R&D-intensive industries, -irf lOCJTJ Mjr glUU^O Wl ItW^fc^ III 1953-62 and 1963-73 (Percent of Innovations) 30- 25- 20- 15- 10- 5- — 1953-62 I Improvement of existing technology I Major technological advance ■ Radical breakthrough J^ i Group I Group II Group III 1963-73 (Percent of Innovations) ^1 Improvement of existing technology 25- ^1 Major technological advance 20- ■■ Radical breakthrough 15- 10- 1^ 5- ^■^'k'.^ ll 1^. Group I Group II Group III SOURCE: Gellman Research Associates, Inc. obtained through a variety of means. These include basic research, applied research, licens- ing, merger or acquisition of other concerns, and the transfer of technology from another product line. Various combinations of these means may be involved in the case of a single innovation. For example, the underlying technology for the light-emitting diode v^as acquired through a combination of internally generated basic and applied research, coupled v^^ith the transfer of technology from one of the firm's existing product lines. The modes by which the technology was acquired for the innovations in this study are shown in figure 4-22. These data, supplied by the innovating firms, represent the number of innovations in which the various acquisition modes were involved. Each mode is counted separately, even if the underlying technology involved a combination of sources; and each mode, and all instances of its occurrence, are treated as equally important in the innovation process. It should be noted also that the information collected regarding the underlying technology applies only to the period between conception and realization of the innovation and does not include prior research activities. These several limitations require that the indicators be regarded as gross measures only. The dependence of innovation on researches — applied and basic — is evident from figure 4-22; applied research was involved in almost 75 percent of the innovations, and basic research in almost 40 percent. Aside from research, the only other acquisition mode of significance was the transfer of technology from existing product lines. Actually, research is involved even more extensively in innovation than the figure Figure 4-22 Sources of technology underlying major U.S. Innovations, 1953-73 (Frequency'} 20 60 100 140 180 220 I I I I I I I I I I I Applied researcli Basic research Technology transfer-' Licensing Purchase of "know-how" ' Multiple responses were accepted. - From wtlfiin the innovating company. SOURCE. Gellman Research Associates, Inc. -^5 See the chapter in this report entitled "Basic Research" for additional information on the relationship between research and innovation. 107 suggests. First, much of the transferred technology itself is based on prior research. But even more important is the contribution from the total body of knowledge gained from centuries of scientific research — knowledge upon which innovation in general draws. The research directly underlying the in- novations was reported by the firms to have been performed primarily by the innovating companies themselves. This was particularly true for applied research, but significantly less so for basic research. In some 96 percent of the cases, applied research was performed within the innovating firm, compared with 73 percent for basic research. Although no attempt was made to determine where the external portion of the basic research was performed, it may be presumed to have been performed largely in the university sector. As indicated in the "Basic Research" chapter of this report, industrial innovation (as represented by major patented technological advances) depends heavily upon basic research performed in universities — a dependency which has increased over the years. Research figures prominently in the sampled innovations of all industries, the least R&D- intensive as well as the most intensive (figure 4- 23). Applied research was involved in some 70 percent of the innovations in each of the three groups of industries (Groups I, II, and III) which vary from high to low in their R&D inten- siveness. Basic research, on the other hand, was more frequently associated with the innovations of Group I industries than with those of Groups II and III — 44 percent versus 32 and 28 percent, respectively.-^" A similar pattern of dependency was found between research and the "radicalness" of innovations (figure 4-24). Applied research was reported with nearly equal frequency for innovations representing radical breakthroughs, major technological advances, and im- provements in existing technologies. Basic research, however, was more often involved in innovations characterized as radical breakthroughs than it was in the other two categories. Such research was reported as a source of innovation in 68 percent of the new products and processes regarded as radical innovations, compared with less than 50 percent in the case of other innovations. Figure 4-23 Research underlying major U.S. innovations, by groups of R&D intensive industries, 1953-73 (Percent of innovations in each group in which applied and basic research were involved)' 80 90 100 I 1 I Applied research Group Group II Group I ■ Multiple responses were accepted. SOURCE: Gellman Research Associates. Inc. Figure 4-24 Research and "radic ( Improvement ot existing technology Major technological advance Radical breakthrough ' Multiple responses w SOURCE Gellman Rese jnderlying major U.S. innovations ainess" of innovations, 1953-73 (Percent of innovations in each category in which applied and basic research were involved)' ) 20 40 60 80 1 1 1 1 1 y Applied research ^ Basic research 1 )0 ere accepted. arch Associates, Inc. -^'' The percentages reported in this section are based on the innovations within each group of R&D-intensive in- dustries. The U.S. innovations included in the study conducted by Gellman Research Associates, Inc., were examined to identify the more specific fields of science which had contributed in a major way to the realization of these innovations. These fields are listed below, along with the principal industries and product areas to which the fields contributed most directly. Associated with each field is a sample of the innovations in which the specific field played a significant role. 108 Scientific field Industries and products Illustrative innovations Polymer chemistry Atomic electron and molecular physics Metallurgy Inorganic chemistry Optics Organic chemistry Solid-state physics Acoustics Biology and bioengineering Pharmacology Chemicals and allied products Plastic materials and synthetic resins Industrial organic chemicals Rubber and miscellaneous plastics products Electrical and electronic machinery and equipment Communication equipment and services Electronic components and accessories Electrical industrial apparatus Machinery Office, computing, and accounting machines Special industrial machinery Professional and scientific instruments Primary metals Fabricated metal products Machinery Transportation equipment Chemicals and allied products Industrial inorganic chemicals Stone, clay, glass, and concrete products Electrical and electronic machinery and equipment Electrical industrial apparatus Professional and scientific instruments Optical instruments and lenses Surgical, medical, and dental instruments Photographic equipment and supplies Electrical and electronic machinery and equipment Electronic components and accessories Chemicals and allied products Industrial organic chemicals Agricultural chemicals Drugs and medicines Food and kindred products Electrical and electronic machinery and equipment Electronic components and accessories Communication equipment and services Electrical industrial apparatus Machinery Office, computing, and accounting machines Professional and scientific instruments Machinery Special industrial machinery Office, computing, and accounting machines Electrical and electronic machinery and equipment Electronic components Communication equipment and services Professional and scientific instruments Professional and scientific instruments Surgical, medical, and dental instruments Electrical and electronic machinery and equipment Drugs and medicines acrylic adhesives double-knit synthetics polyoryl ether phenolic adhesives epoxy cement microwave transmission lasers weather satellites magnetic computer cores video tape permanent magnetic alloys transparent stainless steel superconducting magnets Niobium Beryllium Borazon oil slick emulsifiers synthetic cryolite Pyroceram synthetic diamonds optical scanners Polaroid camera holography via laser fiber optics liquid chromatography textured granular protein benzene — phenol process phenyldimethylurea (herbicide) light-emitting diodes minicomputers printed circuits integrated circuits silicon-controlled rectifiers long range sonar Xerox Telecopier II sonic pile drivers ultrasonic sealers acoustic couplers (telephone-computer) kidney transplants heart pacemakers muscle-activated prosthetics surgery by laser Ketalar (anesthetic) L-Dopa (Parkinson's disease) Terramycin cortisone synthesis 109 Virtually all major fields of science contribute to technological innovation, but certain fields are particularly significant, as indicated in the table above. The physical sciences (especially chemistry and physics) are of general impor- tance across the entire spectrum of industrial innovation. The significance of the biological sciences and medicine has increased considerably in the last decade, both in their direct and indirect contributions to innovation. RETURNS FROM R&D AND INNOVATION The contribution of R&D and innovation to the economy and society is presently understood in broad and general terms only. Existing knowledge of the subject is fragmented and tenuous, to an extent which prohibits the development of indicators of the kind presented elsewhere in this report. Several studies, how- ever, have been conducted in the area, par- ticularly in the last decade. Some of the major findings of these studies are summarized below in the form of tentative conclusions based upon the collective results of these investigations. The findings from the various studies, in- cluding estimated rate of returns from invest- ment in R&D and innovation, are not strictly comparable. The studies employ different con- cepts, assumptions, and methodologies; each has limitations regarding the specification of inputs, the level of aggregation and the availability of data, and the method and degree of attribution of calculated outputs. They, in addition, have one major limitation in common — the inability of conventional measures (such as the Gross National Product and output per man-hour) to capture the full impact of technological innova- tion on the economy and on society. For these and other reasons of a methodological nature, findings regarding the contributions and returns from R&D and innovation appear to be un- derestimated in general (l).*^ The contribution of R&D to economic growth nnd productivity is "positive, significant, and high"(2). This contribution occurs through technological in- novation consisting of enhanced production processes and new and improved products and services. These may expand economic output, increase productivity, or reduce unit costs. Such 5" These numbers refer to the references provided at the end of this chapter. innovation is regarded as an important — possibly the most important — factor in the economic growth of the United States in this century (3-5). Investment in R&D and innovation yields a rate of return as high — and often higher — than the return from other investments. This applies to investments for specific innovations by both the public and private sectors and to R&D investments by individual industries. Rates of return from specific innovations are estimated, conservative- ly, to average between 10 and 50 percent per year (6-11), while returns to innovating in- dustries in the form of productivity growth range from 30 to 50 percent (12-21). The benefits to industries which purchase new and improved products from innovating firms may equal or exceed the direct returns to the innovating firms themselves. These benefits occur particularly in the form of reduced costs or prices per unit of output in the industries which purchase and use the in- novations. The rate of return to these industries is estimated to range from 20 to 80 percent per year (22-24). Industry may underinvest in R&D and innovation with respect to the probable returns to the firm and the benefits to society (25-27). Firms may invest less than the average returns to them would warrant because of the uncertainty and risk associated with specific innovation efforts, as well as the lengthy time before returns can be expected, and the scale of investment which is often involved in innovation. Although the potential benefits to society may often exceed the cost of innovation, a firm may not be able to translate enough of these benefits into profits to justify the necessary investment. "This is particularly true of basic research, where the output frequently occurs. . . not as a marketable product but rather as an advance in basic knowledge that can subsequently be used in applied research and development by a wide and often unforeseeable range of firms" (27). Standard indices of economic performance reflect only part of the contribution which R&D and innovation make to the economy and society (28). Technological innovation sometimes results in new products (e.g., antibiotics and the airplane) which satisfy material needs and wants not satisfied previous- ly. The value of such innovations may far exceed the price paid for the products, although only the latter is counted in standard economic measures. In addition, the effects of qualitative im- provements in products and services (e.g.. 110 machinery requiring less maintenance or longer- lasting automobile tires) may not be represented adequately in common economic indices. In fact, innovations of this kind may contribute less to economic growth as commonly measured than was contributed by the unimproved products. Finally, in present economic accounting, goods and services provided to the public sector through nonmarket channels are valued at cost, rather than at market prices. Thus, benefits from R&D and innovation in areas such as public education and national defense may be un- derestimated by a considerable margin in conventional economic indices. While the benefits from innovation are only partially accounted for by economic indicators, little if any of the associated societal costs are reflected. These costs in human and social terms, as discussed earlier in this chapter, may be substantial, especially when the full range of adverse effects such as loss of jobs and potential health hazards are considered. References 1. Jorgenson, D., Griliches, Z., and Denison, E., The Measurement of Productivity (Washington, D.C.: The Brookings Institution, 1972). 2. Research and Development and Economic Growth/ Productivity, Papers and Proceedings of a Colloquium, National Science Founda- tion (NSF 72-303). 3. Solow, R. M., "Technological Change and the Aggregate Production Function", Review of Economics and Statics, Vol. 39 (August 1957). 4. Denison, E. F., The Sources of Economic Growth in the United States and the Alternatives Before Us (New York: Committee for Economic Development, 1962). 5. Mansfield, E., The Economics of Technological Change (New York: W. W. Norton, 1968). 6. Griliches, Z., "Research Costs and Social Returns: Hybrid Corn and Related In- novations", Journal of Political Economy, Vol. 66 (October 1958). 7. Peterson, W. L., "Returns to Poultry Research in the United States", Journal of Farm Economics, Vol. 49 (August 1967). 8. Ardito-Barletta, N., Costs and Social Benefits of Agricultural Research in Mexico. Unpublished Ph.D. dissertation (Chicago: University of Chicago, 1971). 9. Eastman, S. E., The Influence of Variables Affecting the Worth of Expenditures on Research or Exploratory Development: An Empirical Case Study of the C-141A Aircraft Program. Institute for Defense Analysis, unpublished memoran- dum (1967). 10. Weisbrod, B. A., "Costs and Benefits of Medical Research: A Case Study of Poliomyelitis", Journal of Political Economy, Vol. 79 (May/June, 1971). 11. Freeman, R., "Effects of R&D: Social and Private Rates of Return, Investment Oppor- tunities", in Supporting Studies for Alternate Federal Policies Affecting the Use of Technology, J. H. Holomon (ed.), (Cambridge: Massachusetts Institute of Technology, Center for Policy Alternatives, 1971). 12. Mansfield, E., Industrial Research and Technological Innovation (New York: W. W. Norton, 1968). 13. Minasian, J. R., "The Economics of Research and Development", in The Rate and Direction of Inventive Activity, R.R. Nelson (ed.). National Bureau of Economic Research (Princeton: Princeton University Press, 1962). 14. Minasian, J. R., "Research and Development, Production Functions, and Rates of Return", American Economic Review, Vol. 59 (May 1969). 15. Griliches, Z., "Research Expenditures, Education, and the Aggregate Agricultural Production Function", American Economic Review, Vol. 54 (December 1964). 16. Evenson, R., The Contribution of Agricultural Research and Extension to Agricultural Production. -Unpublished Ph.D. dissertation (Chicago: University of Chicago, 1968). 17. Terleckyj, N. E., Sources of Productivity Advance. Unpublished Ph.D. dissertation (New York: Columbia University, 1960). 18. Terleckyj, N. E., "Comment", in The Theory and Empirical Analysis of Production, M. Brown (ed.). National Bureau of Economic Research (New York: Columbia University Press, 1967). 19. Kendrick, J. W., Productivity Trends in the United States, National Bureau of Economic Research (Princeton: Princeton University Press, 1961). 20. Brown, M. and Conrad, A. H., "The In- fluence of Research and Education on CES Production Relations", in The Theory and Empirical Analysis of Production, M. Brown (ed.). National Bureau of Economic Research (New York: Columbia University Press, 1967). Ill 21. Mansfield, op. cit., Industrial Research and Technological Innovation. 22. Griliches, Z., "Research Expenditures and Growth Accounting", in Science and Technology in Economic Growth, B. R. William (ed.), (New York: J. Wiley, 1973). 23. Mansfield, E., et al.. Social and Private Rates of Return from Industrial Innovations. An un- published paper presented before the Eastern Economic Association, October 26, 1974. 24. Terleckyj, N. E., Effects of R&D on the Productivi- ty Growth of Industries: An Exploratory Study (Washington, D.C.: National Planning Association, 1974). 25. Arrow, K., 'The Comment", in The Rate and Direction of Inventive Activity, R.R. Nelson (ed.). National Bureau of Economic Research (Princeton: Princeton University Press, 1962). Nelson, R. R., "The Simple Economics of Basic Scientific Research: A Theoretical Analysis", Journal of Political Economy, Vol. 67 (June 1959). Economic Report of the President and The Annual Report of the Council of Economic Advisors, 19 7 2, p. 126. Nelson, R. R., Science, The Economy, and Public Policy (Santa Monica: The Rand Corporation, 1964). 26 27 28 112 Science and Engineering Personnel 113 Science and Engineering Personnel INDICATOR HIGHLIGHTS D The total number of scientists and engineers employed in these occupations in 1974 was approximately 1.7 million, which is nearly the same as in 1970, with engineers representing nearly two-thirds of the total. D The number of scientists and engineers with doctorates reached approximately 245,000 in 1973, representing almost 15 percent of all scientists and engineers; life and physical scientists each accounted for one-fourth of the doctoral total. D The majority of doctoral scientists in 1973 were employed in educational institutions (64 percent), and primarily engaged in teaching, while doctoral engineers tended to be concentrated in business and industry (49 percent) and were primarily involved in R&D. D Employment of scientists and engineers in universities and colleges increased between 1965 and 1974 by more than 60 percent, with most of the growth occurring prior to 1972; the largest increases in employment occurred for life and social scientists, bringing the total number of scientists and engineers employed in higher education to just over 288,000 in 1974. D In recent years the proportion of young doctoral faculty in doctorate-level science and engineering departments has declined from approximately 42 percent in 1968 to some 28 percent in 1974; concurrently, median ages have increased from 41 to 44 years, and the proportion of faculty with tenure has risen from 47 percent to 65 percent. D The largest number of the Nation's scien- tists and engineers were employed in industry, with engineers accounting for nearly 80 percent of the total in 1974; approximately 25 percent of the engineers were involved in R&D and its management, compared with some 35 percent of the industrial scientists. D The Federal Government supported less than one-fourth of all industrial scientists and engineers in 1974, down from nearly 30 percent in 1972; most of the support was provided by DOD and NASA which together accounted for nearly 70 percent of all such Federal support. D The number of scientists and engineers employed by the Federal Government declined in 1973 for the first time since the 1950's; employment in this sector comprised 10 percent of all employed scientists and engineers in 1973, with some 30 percent of the Federal total involved in R&D. D The total number of scientists and engineers engaged in R&D (on a full-time equivalent basis) was 530,000 in 1974, down by more than 30,000 from the high in 1969; 68 percent of the total were employed in industry, 13 percent in the academic sector, and 12 percent in the Federal Government. D Approximately 40 percent of all doctoral scientists and engineers were involved in R&D in 1973; in universities, physical and life scientists comprised the majority of doctorates who were involved primarily in basic and applied research; in the industrial sector, most doctorates were engineers working on development-related activities. D The number of R&D scientists and engineers in industry increased in 1973 and 1974, reaching almost 360,000 (on a full- time equivalent basis) but nearly 7 percent less than the number employed in the peak year of 1969; the decline occurred primarily in the aircraft and missiles industry, and was confined mainly to those scientists and engineers supported by Federal funds. n Academic R&D was conducted by 67,000 scientists and engineers (on a full-time equivalent basis) in 1973, and was heavily focused on research (basic and applied). Of all the doctoral faculty involved in R&D, the proportion of young investigators decreased 114 for all science and engineering fields by 14 percent between 1968 and 1974. D Annual awards of bachelor's and first- professional degrees in the sciences and engineering doubled between 1963 and 1972; as a fraction of first degrees awarded in all fields, however, those in science and engineering remained essentially constant at nearly 30 percent during the period, due in large part to a rapid growth in the number of social science degrees awarded. Awards of master's level degrees in science and engineering followed a similar trend, but declined in recent years to 21 percent of all master's degrees awarded. D Annual awards of doctoral degrees in science and engineering began to level off in 1971, decreasing for the first time in a decade to a level in 1974 of approximately 18,000; the largest declines occurred in the number of physical science doctorates awarded; science and engineering doctorates as a fraction of all doctorates declined from 64 percent in 1965 to 56 percent in 1974. D The proportion of science and engineering graduate students receiving Federal support declined from 42 percent in 1967 to 25 percent in 1974; this decrease was compen- sated primarily by increases in self-support (up 13 percent) and institutional support (up 6 percent). D Women comprised 5 percent of the persons employed in science and engineering oc- cupations in 1974, and were primarily involved in psychology, social sciences, and mathematics; in the academic sector, women represented 15 percent of all scientists and engineers employed full-time in 1974. D The predominant proportion of all scientists and engineers in 1972 were Caucasian (96 percent), while 2 percent were Asian, and 1 percent each were Black or were of other nonwhite background; the smallest propor- tional representation of minorities is in engineering (3 percent) and the largest is in mathematics (8 percent). The country's scientists and engineers are an important national asset. They provide instruc- tion and training in the various fields of science and engineering, conduct basic research to advance the understanding of nature, and perform applied research and development in a diversity of areas such as health, defense, energy, and industrial technology. In addition, persons trained in the sciences and engineering are employed throughout the economy — from industrial management to agricultural production — to provide the knowledge and skills which are essential in a technologically advanced society. The role of scientists and engineers in helping to meet the changing needs of the Nation, coupled with the extended time and high cost involved in their training, requires that continuous attention be given to trends and patterns in the production and utilization of such personnel. This chapter presents information on the magnitude and characteristics of the Nation's population of scientists and engineers. It con- siders trends in the supply and utilization of these personnel and examines developments which may affect their future supply. Scientists and engineers, in this chapter, are defined as persons actually engaged in scientific or engineering work at a level which requires knowledge of the physical, life, social, mathematical, or engineering sciences equivalent at least to that acquired through completion of a four-year college program with a major in one of these fields, regardless of whether a college degree is actually obtained in the field. In regard to data presented on employment, enrollments, and degrees awarded, the health professions are not included under "science and engineering", unless otherwise indicated. Throughout the chapter, information is limited to certain quantitative aspects of scien- tists and engineers. These measures, it is recognized, provide only a partial indication of the characteristics of such personnel. Lacking are measures of the quality of their work, extent of "underutilization", and the increasingly important concerns of productivity and output. Furthermore, little is known about motivational factors that affect the supply and utilization of scientists and engineers, such as considerations which lead students to enter science and 115 engineering, or influence those already in these fields to move from one type of employment to another. The present lack of such indicators, it is hoped, will be remedied in the future as improved methodologies are developed for measuring these aspects. The measures of quantitative characteristics presented here are themselves less than com- plete. ' In the case of the utilization of scientists and engineers, for example, data are not available after 1970 with respect to industrial employment. Data are also lacking on new baccalaureates and masters entering the market since 1970. Information on the specific activities of scientists and engineers, especially those in the academic sector, are limited by the current inability to obtain full-time equivalent (FTE) data on major activities such as R&D and teaching, by field of science. The surrogate measure of numbers of scientists and engineers "primarily involved" in an activity provides a useful but relatively crude measure of this factor. In the case of supply, the latest data on the production of baccalaureate and masters degrees from the National Center for Educational Statistics covers the 1972-73 period. CHARACTERISTICS AND UTILIZATION OF SCIENCE AND ENGINEERING PERSONNEL Employment of scientists and engineers Employment of scientists and engineers stabilized in the first years of the 1970's, after increasing substantially for several decades. - During the 1950's, the number of scientists and engineers doubled, rising from about 600,000 to nearly 1.2 million. In the 1960's, employment grew by almost as much in absolute terms, from about 1.2 to over 1.7 million; the relative gain, however, was only about half that of the 1950- 60 decade. Furthermore, between 1960 and 1970 the number of scientists grew significantly faster than the number of engineers (75 and 38 percent respectively), partially as a result of substantial gains in social science fields. 1 Some of these deficiencies are expected to be corrected during thie next year througfi the new National Science Foundation — Bureau of the Census surveys of industrial employment and the complete implementation of the National Science Foundation Manpower Characteristics System. 2 EmploumenI of Scientists and Engineers, 1950-70, Bureau of Labor Statistics, Department of Labor, lo/3. Beginning in 1969, growth in total employ- ment of scientists and engineers slowed and then remained relatively level until about mid-1972. The factors underlying these changes include cutbacks in defense, space, and associated R&D spending in these areas, the general economic climate, and the beginning of a slowdown in academic hiring. Though employment in some sectors continued to increase — namely, higher education and government (particularly State and local) — little if any growth occurred in industry, the major sector of employment for scientists and engineers. Unfortunately, no specific measurements of industrial employment of scientific and technical personnel have been taken since 1970, though a survey is being reinstated by the National Science Foundation. However, by using information on past trends and relationships and several related sources of information, it has been possible to prepare estimates of the probable level of industrial employment of scientists and engineers for 1972. Using these estimates, together with information on nonindustrial sectors of employ- ment, it is thus possible to estimate the total number of scientists and engineers employed in 1972. The available data do not permit estimates to be made for more recent years. In 1972, estimated overall employment of scientists and engineers stood at about 1.7 million, approximately the same as in 1970. In the first years of the 1970's, employment in the sciences continued to grow slightly while engineering employment declined somewhat between 1970 and 1972. These overall patterns of change include minor shifts in the sectoral distribution, with university and college and Government employment gaining while the proportion declined for the industrial sector. Although recent information is not available on an overall and detailed basis for scientists and engineers, selected information about such personnel is provided by the National Science Foundation's National Sample of individuals in science and engineering jobs. A sample was drawn from the 1970 Census of Population and used for the 1972 and 1974 surveys; the results provide information on a large portion of the Nation's science and engineering personnel. An estimate of the distribution of these personnel among fields of science and engineer- ing was obtained from the 1974 survey, and is shown below. 116 Distribution of the 1970 science and engineering labor force by field, 1974 Field Percent Engineers 64 Physical scientists 14 Life scientists 7 Computer scientists 5 Social scientists 4 Psychologists 3 Mathematical scientists 3 The following are some additional characteristics of the 1970 science and engineer- ing labor force, surveyed in 1974:-' — Approximately 35 percent of the employed scientists and engineers are engaged in work supported with Federal funds. — Industry and business are the employers of most scientists and engineers, 65 percent of the total in 1974. — Scientists and engineers holding doc- toral degrees account for some 15 percent of the scientific and engineering population, master's and professional degree holders almost 25 percent, and baccalaureates nearly 60 percent. — Management or administration was the most common work activity in which scientists and engineers were engaged in 1974; this activity was reported by about 30 percent of the sample, with about one-third (10 percent of the total employed group) involved in the management or administration of R&D. — Research and development was the primary work activity of almost 30 percent of the employed scientific and engineering population, with almost 10 percent involved in research (applied and basic). Doctoral scientists and engineers Those scientists and engineers holding doc- toral degrees represent, as a group, the most highly trained men and women in their professions. The investment of resources in their education and training is significant in both monetary terms and in the amount of time involved. The characteristics and activities of this group warrant careful monitoring, since doctoral level scientists and engineers provide leadership for the entire scientific community. It is estimated that in 1973 there were 245,000 doctoral scientists and engineers in the United States. ^ This number is over twice that reported in 1963, and represents about 14 percent of all scientists and engineers. Approximately 9 percent of the 245,000 doctoral scientists and engineers were women and 6 percent were foreign citizens. Scientists and engineers of oriental background made up 5 percent of the doctorate science and engineering population; Blacks, 1 percent; and other groups, 94 percent. ^ The physical and life sciences were the two largest fields represented in the 1973 population of doctoral scientists and engineers, as shown below." Distribution of doctoral scientists and engineers, by field, 1973 Field Percent Life scientists 26 Physical scientists 26 Engineers 15 Social scientists 13 Psychologists 12 Mathematical scientists 6 Computer scientists 1 Among doctoral scientists, the proportion accounted for by physical scientists declined over the 1966-73 period while the life scientists' share increased. Other fields remained relatively constant over the period in terms of their relative proportions (figure 5-1). Sectors of doctoral employment. The pattern of employment of doctoral scientists and engineers in 1973 is shown in figure 5-2. Doctoral scientists are predominantly employed by educational institutions (64 percent); within ^ "National Sample of Scientists and Engineers: Changes in Employment 1970-72 and 1972-74", Science Resources Studies Highlights. National Science Foundation (NSF 75-309), May 19, 1975. ■' Doctoral Scientists and Engineers in the United Stales, 1973 Profile, National Academy of Sciences, 1974. -■' For further information on this topic, see the subsequent section in this chapter entitled "Women and Minorities m Science and Engineering." ^ Characteristics of Doctoral Scientists and Engineers in the United States, 1973. National Science Foundation (NSF 75-312). 117 Figure 5-1 Estimated percent distribution of doctoral scientists, by field, 1966-73 (Percent) 50 45 ^ 1966 ■i 1968 40 - ■1 1970 1973 35 - 30 - ' 25 - 20 - ■ 15 - 1 ■ 10 - 1 1 1 5 - 1 1 1 1 u . m ■ i ■ Physical Mathematical Life Psychologists scientists scientists scientists Social scientists SOURCE; National Science Foundation- Figure 5-2 Percent distribution of employed doctoral scientists and engineers, by employment sector, 1973 Education • Business and industry Government All other employers Scientists and engineers (Percent! 20 40 r Engineers (Percent) 20 40 60 I I I I I ' Universities and colleges account for 99 percent of the doctoral scientists and engineers in education, SOURCE: National Science Foundation. 118 this group 61 percent are employed by four-year colleges and universities and 2 percent by two- year colleges. During the period 1966-70, there was a shift in the proportion of doctoral scientists employed by business and educational institutions, the former declining and the latter increasing. However, in view of enrollment trends and financial problems of institutions of higher education, this shift is not expected to continue. Doctoral engineers as a separate group exhibit a different pattern of employment from scien- tists, with nearly half of them employed in the industrial sector. Primary work activities of doctoral scientists and engineers. The activities in which doctoral scientists and engineers were primarily involved are indicated in figure 5-3. The data do not show the time allocated among the several activities of doctoral scientists and engineers, but rather the activity reported as occupying the largest portion of their time. Teaching and R&D represent the primary work activities of doctoral scientists, the majority of whom are employed in universities and colleges. A declining proportion of the doctoral scientists, however, were in- volved in R&D as a primary work activity during the period 1966-73. This decline of about 10 percent was accompanied by a relatively larger increase in the fraction reported as primarily teaching^ (figure 5-4). Of the 32 percent of the doctoral scientists primarily engaged in R&D, over one-half were working in basic research, over one-third were involved in applied research, and only a small percentage in development and design. The number of such scientists primarily engaged in management activities, however, was nearly the same as the total number primarily involved in basic research (30,851 versus 31,213). The preceding discussion concerning the utilization of doctoral scientists and engineers provides, for the most part, a description of the characteristics of these doctorate holders in 1973. Over time, however, there has been movement from initial doctoral disciplines into other fields of science, while others have shifted ' This topic is discussed in more detail later in this chapter. Figure 5-3 Percent distribution of employed doctoral scientists and engineers, by primary work activity, 1973 Teaching Research and development Management or administration Sales Consulting Other activities Scientists and engineers (Percent) 10 20 30 I I Scientists (Percent) 10 20 Engineers (Percent) 40 SOURCE: National Science Foundation. Key: Types of R&D Types of management and administration I I Basic research I I R&D management and administration I I Applied research i i NonR&D management and administration I I Development i i Both types of management and administration 119 Figure 5-4 Estimated percent distribution of doctoral scientists, by primary worl< activity, 1966-73 (Percent) 40 35 30 k , 1968 ■W 1970 i^- 1973 25 - 20 Management or administration I Teaching Other SOURCE: National Science Foundation, to nonscience occupations. Between 10-30 percent of the doctorates in each field are employed in fields different from their doctorate field. The fields of bioscience, mathematics, and psychology experience the highest retention rates, with approximately 90 percent of the doctorates in these fields still employed in the same field of their doctorate, while physics and chemistry have the lowest retention rates (approximately 70 percent). Data on shifts to nonscience occupations show 11 percent of the doctoral social scientists changing fields, com- pared to 6 percent of the doctoral chemists and doctoral psychologists.* Academic employment of scientists and engineers Universities and colleges employed about 288,100 scientists and engineers in 1974 (in- cluding full-time and part-time personnel), an increase of 61 percent over the 178,900 ' Doctoral Scimlisli and Engineen in the Uniled Stales, 1973 Profile. National Academy of Sciences, 1''74. employed in 1965. Most of the growth occurred between 1965 and 1971, with increases in all scientific disciplines. The average annual rate of growth in academic employment of scientists and engineers between 1971-74 was only 1.7 percent compared with 7.3 percent during 1965- 71. In absolute terms, the largest growth occurred in the employment of life scientists and social scientists, which together accounted for more than three-fifths of the overall increase between 1905-74 (figure 5-5). In two fields, engineering and social sciences, there were small declines in employment from 1973 to 1974. As demand slackened for academic employ- ment during the early 1970's, the attainment of the doctoral degree in the sciences and engineer- ing became increasingly important as a requisite for employment in this sector. Since 1971, academic scientists and engineers with Ph.D.'s or health profession doctorates increased about 10 percent, compared with small declines in the employment of persons with master's or bachelor's degrees (figure 5-6). Between 1965 and 1974, employment of doctorates in univer- sities and colleges increased by more than 60 120 Figure 5-5 Scientists and engineers employed in universities and colleges, by field of employment, 1965-74 (Thousands) 120 1 60 40 20 Social Scientists Physical Scientists Engineers ^ ^ ^ *" ,■•••••••• , . • • • * "^Mathematicians and , •••* computer scientists ^^» • Psychologists _L _L 1965 '67 '69 '71 (January) SOURCE: National Science Foundation. '73 '74 percent, with the result that by 1974, 65 percent of all academic scientists and engineers had doctorate degrees, compared with 60 percent in 1965. Primary work activities among academic scientists and engineers have shifted toward more teaching and less R&D (figure 5-7). In 1974, 17 percent of all science and engineering professionals working in institutions of higher education were primarily engaged in R&D, compared with 22 percent in 1965. A part of this shift is due to the rapid growth of two-year academic institutions where teaching is the primary activity of almost all the faculty. Other academic institutions, including the large research universities, also experienced the shift toward more teaching. During 1969-74, four- year institutions reported an average annual percentage rise of 4.7 percent in the number of scientists and engineers working primarily as teachers, compared with only a 0.4 percent average annual growth of those working primarily in R&D." This shift in utilization occurred at the same time as the reduction in the rate of growth in Federal support for academic R&D. From 1968 to 1974, annual increases in Federal R&D support to universities and colleges have not kept pace with increases in inflation; such support in constant dollars has declined about 8 percent in this six-year period. The financial status of R&D in this sector might have been worse except for substantial increases in separately budgeted R&D support by the institutions themselves, and by State and local governments. Funds from the latter increased some 6 percent annually in constant dollars between 1968 and 1974 for a total growth of nearly $70 million. Support from the in- stitutions' own funds rose at an average annual rate of 2.5 percent over the period for a total of almost $35 million. Federal support over the same period, on the other hand, declined by more, than $60 million in constant dollars. 'o Selected characteristics of higher education faculty. Significant changes have occurred in recent years in the characteristics of the faculty of academic institutions, in terms of their median age, tenure status, and number of years since receipt of doctorate. Between 1968 and 1974, the overall propor- tion of young 11 doctoral faculty in doctorate- level science and engineering departments decreased substantially, dropping from 42 percent to 28 percent of the total doctoral faculty. 12 For the fields shown in the table below, the total number of full-time faculty increased " Manpower Resources for Scienlific Activities at Universities and Colleges, lanuary 1974, Detailed Statistical Tables, National Science Founciation (NSF 75-300-A), and earlier volumes. 1° "Separately Budgeted Academic R&D Expenditures Decline in Real Terms in FY 1974", Science Resources Stuiiies Highlights. National Science Foundation (NSF 75-306), April 21, 1975. ' ' Those who had held doctorates for seven years or less at the time of each study. 1^ Young and Senior Science ami Engineering Faculty, 1974: Support, Research Participation, and Tenure, National Science Foundation (NSF 75-302). 121 40 120 I Figure 5-6 Employment of academic scientists and engineers, by level of attainment, 1965-74 (Thousands) 160 200 I I ^_^^^^_^^^^_^^^_^_^^^^^____^_^^_ 1965 Total Ph.D. or Sc.D, 320 I 1974 M.D., D.D.S. Master's Bachelor's or equivalent Level Average annual percent change 1965-71 1971-74 Total Ph.D.. Sc.D. M.D., D.D.S. Master's Bachelor's or equiv. 7.3% 8.8 5.6 7.1 4.8 1.7% 4.0 .9 -.9 -.3 SOURCE: National Science Foundation. by 8 percent from 1968 to 1974. However, the young doctorate faculty proportions declined in all seven fields, while even the absolute numbers of young doctorate faculty decreased in 5 of the fields, biology and psychology being the only exceptions. Proportion of young'^ doctoral faculty in doctoral level science and engineering departments in universities and colleges, by selected fields, 1968 and 1974 Young doctoral faculty as a percent of all doctoral faculty Selected fields 1968 1974 Biology 32 27 Cfiemistry 35 21 Economics 43 34 Electrical engineering . . 52 27 Mathematics 52 36 Physics 40 19 Psychology 43 37 '■> Those who had held doctorates for seven years or less at the time of each study. Another recent trend is the substantial increase in the proportion of faculty with tenure. An American Council on Education study" found in 1973 that 65 percent of the faculty in all fields were tenured, compared with 47 percent in 1969. Figure 5-8 presents more recent NSF data which show the 1974 proportions of tenured faculty in doctoral level science and engineering departments for 15 fields. Overall, 70 percent of these faculty have tenure, with the proportions ranging from a high of 81 percent in chemical engineering to 59 percent in physiology. Between 1969 and 1973, the median age of faculty in science and engineering fields employed in doctorate-granting institutions rose from 41 to 44 years. '^ Changes in the " Alan E. Bayer, Teaching Faculty in Academe: 1972-73, {Washington, DC; American Council on Education, 1973). >5 Bayer, Ihid.. and Alan E. Bayer, College and University Faculty: A Slatiilical Descrifilwn. (Washington, D.C.; American Council on Education, 1970). 122 Figure 5-7 Academic scientists and engineers, by primary work activity, 1965-74 (Thousands) 280 ^^^ ■^^ 260 ~ Total ^^^ H 240 ^r 1 220 >^ ^ , 200 V^ ^^'^^ , 180 ^^^ Teaching 1 160 ^^^ n 140 -y/^ 1 120 - ■ 100 - 1 80 - I 60 R&D ■U-'-^/':\ 40 20 ^^^ Other activities 1 1 1 1 19 65 '67 '69 71 73 7 4 SOURCE: National Science Foundation. median age of faculty for several science and engineering fields are presented in the table below. Median ages of science and engineering faculty in doctorate-granting universities and colleges, 1969 and 19731'' Field 1969 1973 All science and engineering fields ... 41 44 Biological sciences 43 46 Chemistry 39 43 Earth sciences 40 43 Engineering 41 47 Mathematics 37 39 Physics 39 43 Psychology 39 43 Social sciences 39 43 '" National Science Foundation and American Council on Education, special tabulations. Figure 5-8 Tenured faculty as a percent of all faculty in doctorate-level science and engineering departments, by field, 1974 (Percent) 10 20 30 40 50 60 70 80 90 SOURCE: National Science Foundation. It should be noted that age distributions among academic science and engineering doc- torate faculties do not differ greatly from those for doctoral scientists and engineers in other employment sectors (figure 5-9). While the older age pattern displayed by science and engineering faculty is related to the decreasing number of new faculty appointments, the relatively small proportion of employed doctorates under the age of 30 may be accounted for in part by the time required to attain a doctorate. In recent years, the median time-lapse between the 123 Figure 5-9 Age distribution of doctoral scientists and engineers, by type of employer, 1973 (Percent) 5 10 15 20 I I I 25 under 30 30-34 years 35-39 years 40-44 years 45-49 years 50-54 years 55-59 years 60-64 years 65 or over Four-year colleges and universities Business and industry Federal Government r SOURCE National Science Foundation. baccalaureate and the doctorate has been approximately 7 to 8 years. i^ Inherent in the age and tenure data is an implication that the pattern in many fields over the next 5 to 10 years may be that of a relatively senior faculty, the great majority of whom will be tenured. However, it is possible that for some fields these recent trends could be reversed if in the future, replacement openings for science and engineering faculty caused by death and retire- ment are filled extensively with new junior faculty on academic staffs. Utilization of postdoctoral personnel. Between 1967 and 1974, the number of science and engineering postdoctorals, as indicated by a survey of representative science and engineer- ing departments in doctorate-granting in- stitutions, increased by 21 percent, reaching almost 17,000 in 1974 (figure 5-10). is The reasons for these increases may have changed midway during this period. In the late 1960's, universities provided increasing numbers of Figure 5-10 Postdoctorals and research assistants in science and engineering departments at doctorate-granting institutions, 1967-74 Index: 1967 = 100 140 1967 '68 '69 70 SOURCE: National Science Foundation. ■71 ■73 1' Doclorale Recipients from U.S. Unwersilies: Summary Report, National Academy of Sciences, annual series. i» The indices for 1967-71 are estimates based on applications submitted to NSF for its departmental traineeship program. Indices after 1971 were collected by the "Survey of Graduate Science Student Support and Postdoc- torals" for matched departments. 124 postdoctoral appointments as part of the general academic science expansion. Between the late 1960's and the early 1970's, however, academic science funding from the Federal Government leveled off (in constant dollar terms), while the employment market for new doctorates, es- pecially in academic institutions, declined markedly. Although fewer academic R&D funds were available, there are at least two possible reasons for increases in the postdoctoral population between 1967 and 1974. Both new doctorates and the universities at which they worked may have used the postdoctoral appointments as a "holding pattern" until new Ph.D.'s could find desirable positions. This reason was given by over 35 percent of science and engineering postdoctorals in a recent study. ^'^ A second possible reason is the interest of academic researchers in maximizing their research effec- tiveness in periods of financial stress. One way to accomplish this may have been to substitute postdoctorals for research assistants during the earlier years of the 1967-74 period, a pattern which is suggested by the data in figure 5-10. A high in the number of academic postdoctoral appointments was reached in 1972 followed by a decline resulting primarily from decreases in the number of postdoctoral appointments of 're- cent" doctorates (i.e., those who had earned their doctorates within four years of the study). In 1972 these recent degree recipients comprised 72 percent of all postdoctorals, compared with 58 percent in 1974. Industrial employment of scientists and engineers The industrial sector is by far the largest employer of scientists and engineers. -f TTiere was, however, some fluctuation in the level of employment during the 1970-74 period, reflec- ting first the layoffs of scientific and technical personnel in industry in 1971-72, and then the general upturn in the economy during late 1973 and early 1974. Scientists and engineers employed in industry in 1970 constituted almost two-thirds of all such personnel employed in I' Charactenslics of Doctoral Scienhsis and Engineers in the United States. 1973. National Science Foundation (NSF 75-312-A). 2° "National Sample of Scientists and Engineers: Changes in Employment, 1970-72 and 1972-74", Science Resources Studies Highlights. National Science Foundation (NSF 75-309), May 19, 1975. that year, but increased in the two subsequent years so that the level in early 1974 was near that of 1970. Engineers accounted for nearly 80 percent of the scientists and engineers employed in the industrial sector in 1974. Physical scientists (including those in the environmental sciences) accounted for 11 percent of the total and computer scientists, 6 percent. In 1974, R&Dand its management constituted the largest primary activity of industrial scien- tists and engineers, involving almost 30 percent of the total group. However, as shown in the table below, there were some differences between the activity patterns of scientists and engineers. A larger fraction of industrial scien- tists were primarily engaged in R&D and management of R&D (36 percent) than was the case for engineers (26 percent). The next most common activity of industrial engineers was management of non-R&D activities, while for scientists it was the area of computer applications. Percent distribution of the 1970 science and engineering labor force employed in industry, by primary work activity, 1974* Primary work activity Total Scientists Engineers R&Dand R&D management 29 36 26 Management of non-R&D activities ..19 15 20 Production and inspection 16 13 17 Design 14 NA 18 Computer applications 6 19 2 Other activities 16 17 17 ^ NSF, special tabulations The Federal Government provided support for 23 percent of all industrial scientists and engineers in 1974 versus 28 percent in 1972. This decrease was evident among most fields. In both years, much of the Federal support was for industrial R&D activities. The estimated relative level of support varied widely among the different science fields. In 1974, 26 percent of the engineers and 22 percent of the mathematical scientists received Federal support; the same was true for only 10 percent of the physical scientists and approximately 5 percent of the life and environmental scientists. Over half — 52 percent — of all Federal support of scientists and engineers in 1974 came from 125 the Department of Defense, with another 17 percent provided by the National Aeronautics and Space Administration. With the exception of life and environmental scientists, engineers and scientists in industry received their major Federal support from the Department of Defense. Employment of scientists and engineers in the Federal Government Nearly 10 percent of all scientists and engineers are employed by U.S. Government agencies. The number of Federal scientists and engineers in 1973 declined by 3 percent over 1972 to 162,000, the first sizable annual reduction since data were initially collected in 1954." The major agencies employing scientists and engineers are shown below in terms of the percentage of the total employed by each during 1973. Distribution of Federal scientists and engineers, by agency, 1973 Agency Percent DOD .. . 45 USDA .... 15 Interior . . 8 NASA .... 7 Commerce 4 HEW 4 Ail other agencies 16 Of all Federal scientists and engineers, some 30 percent were employed in R&D positions in 1973. Those engaged in research consisted of nearly 19,000 scientists and some 4,000 engineers, whereas development activities employed nearly 19,000 engineers and over 6,000 scientists. Employment of scientists and engineers in nonprofit institutions These institutions-- employ only about 1 to 2 percent of the national total of scientists and engineers. By 1973, employment of scientists and engineers in this sector reached ap- proximately 26,300, an increase of some 20 percent since 1965. ^-^ In contrast to trends reported in the academic sector, virtually all of the increase in independent nonprofit in- stitutions was attributable to personnel who worked primarily in research and development; this group of personnel comprised nearly 90 percent of all scientists and engineers employed in such institutions. RESEARCH AND DEVELOPMENT PERSONNEL Total scientists and engineers in R&D An estimated 530,000 scientists and engineers were engaged in R&D activities on a full-time equivalent basis in all sectors of the economy in 1974. This number accounts for approximately one-third of all employed scientists and engineers, ^•i Over the past two decades, the employment of these R&D scientists and engineers grew at an average annual rate of 4.1 percent, 1.6 times faster than the rate of growth of total civilian employment. In 1969-70, however, the long- term growth trend was reversed as the number of R&D scientists and engineers declined and national R&D expenditures in constant dollars decreased. Between 1973 and 1974, R&D scientist and engineer employment increased by nearly 5,000, reversing the downward trend slightly. The 1974 employment level, however, was over 30,000 short of the peak employment level reached in 1969. Doctoral scientists and engineers in R&D Approximately 90,000 of the science and engineering doctorates in the 1973 U.S. labor force cited R&D or R&D management as their primary work activity. --°^ While some one-third of all scientists and engineers were engaged in R&D, the proportion of doctorates primarily ^1 "Federal Scientific and Technical Personnel Decline in 1973", Science Reiources Sludies Highlights. National Science Foundation (NSF 74-316), October 18, 1974. -~ Which include research institutes, hospitals, and Federally Funded Research and Development Centers administered by nonprofit institutions. 23 R&D Acliviliei of Independent Nonfirolil Institutions, 1973, National Science Foundation (NSF 75-308). 2^ See Appendix table 2-2 and National Patterns of R6D Resources, 1953-75, National Science Foundation (NSF 75- 307). 25 Characteristics of Doctoral Scientists and Engineers m the United States. 1973, National Science Foundation (NSF 75-312). 126 involved in R&D-related work exceeded 40 percent. The 1P73 distribution of these scientists and engineers by field of science and work activity is shown in figure 5-11. Engineers represent the major portion of those with development as a primary work activity, while physical and life science doctorates constitute the major portion of those involved in research. A relatively large proportion of R&D doctorates spend the major part of their time in R&D administration. The 1973 distribution of R&D doctorates by type of employer is shown in figure 5-12. In contrast to the pattern for all R&Dscientists and engineers, the doctorates are about equally concentrated in industry and educational in- stitutions, for all fields combined. Information on the distribution by type of employer for major fields of science is presented in figure 5-13. Physical science and engineering R&D doc- torates are most heavily concentrated in in- dustry, while life scientists, mathematical scien- tists, and social scientists are located predominantly in educational institutions. The concentration of R&D doctorates by employment sector varies considerably. Almost three-quarters of the doctorates employed in industry are engaged primarily in R&D or R&D management, while the R&D involvement of doctorates employed in government is slightly higher. In academic institutions, where teaching is the chief activity, only one-fourth of the doctorates work primarily in R&D. R&D in the academic sector Some 67,000 or 13 percent of the Nation's full- time equivalent R&D scientists and engineers were employed in universities and colleges in 1974;^" approximately 26 percent (18,000) of these are graduate students working as scien- tists and engineers. In contrast to other sectors of employment, university and college personnel involved in R&D are usually primarily engaged in teaching. Thus, the actual number of faculty members engage(i in R&D may be considerably greater than the reported FTE number of 67,000. A 1973 survey of U.S. science and engineering doctorates showed that about 80,000 of the science and engineering doctorates employed by universities and colleges considered -" Natumal Patlems of R&D Resources, 1953-75. National Science Foundation (NSF 75-307). Figure 5-11 Distribution of doctoral scientists and engineers within R&D activities, by field, 1973 Total Research (Percent) 10 20 1 1 30 1 40 50 60 1 1 1 N = 97,820 70 80 1 1 90 1 100 1 ^^^HHi^^^^^™ Physical scientists Engineers Mathematicians N = 60,929 tife scientists Social scientists Development Administration of R&D N = 8,580 N = 28.311 SOURCE; National Science Foundation. 127 Figure 5-12 Doctoral R&D scientists and engineers, by type of employer, 1973 other employer Scientists and engineers (Percent) 20 Scientists (Percent) Engineers (Percent) 20 40 SOURCE: National Science Foundation. Figure 5-13 Distribution of doctoral R&D scientists and engineers, by field and type of employer, 1973 Total doctoral R&D scientists and engineers Physical scientists Mathematical scientists Environmental scientists ' Engineers Life scientists 10 I Percent 20 3D 40 50 III N = 97.820 Business and industry 60 70 I I Educational institutions N = 27,445 80 90 100 I I Other employers^ Federal, state and other government Social scientists IN = 5,661 ' IncluiJes earth scientists, oceanographers, and atmospiienc scientists. SOURCE: National Science Foundation. 128 themselves involved in R&D as a primary or a secondary activity.-^ The type of R&D carried out by these academic scientists and engineers is heavily focused in the research area. An indicator of the extent of this concentration is R&D expen- ditures; in 1974, 96 percent of academic R&D funding was expended for research activities (basic and applied), with only 4 percent reported for development activities. ^^ The extent of the involvement in research by scientists and engineers who have recently received doctoral degrees is indicated in figure 5- 14. This figure applies to those faculty spending 20 percent or more of their time in R&D; young investigators are defined as those who had held their doctorate seven years or less at the time of each of the studies. The proportion of young investigators in relationship to the total number of faculty investigators has decreased significantly. These decreases, however, match the overall changes in faculty age distribution, regardless of activity. The physical sciences and electrical engineering have been most affected, while the decreases in young investigators have been least pronounced in biology and psy- chology. In 1974, more than one-half of the faculty investigators28 in the fifteen fields listed in the table below were performing R&D directly connected with project grants and contracts awarded by Federal agencies. -"This represents a considerable decrease from 1968, when two- thirds of faculty investigators were involved in Federal projects. There were large differences among the several scientific fields, however. For 2" National Science Foundation, special tabulations 2* "Investigators" were defined as those spending at least 20 percent of their time in research. 2" Youtig and Senior Science and Engineertng faculty, 1 974: Support, Participation, and Tenure, National Science Foundation (NSF 75- 302). Figure 5-14 Young doctorate faculty' investigators' as a percent of all faculty investigators, 1968 and 1974 (Percent) 60- 1968 11974 40- 30- 20- Biology Chemistry Economics I Those who had held doctorates seven years or less at the time of each study. - Spending 20 percent or more of their time in research. SOURCE: National Science Foundation. n Mathematics Physics Psychology 129 example, more than three-fourths of the faculty investigators in biochemistry, but only one- fourth of those in sociology, were doing research connected with federally supported projects in 1974. Proportion of faculty investigators performing R&D connected with Federal grants and contracts, by field, 1974 Field All fields Biochemistry Physiology Microbiology Physics Electrical engineering Chemical engineering Biology Geology Chemistry Zoology Psychology Mathematics Botany Economics Sociology Percent whose research was federally supported 56 78 75 74 72 71 65 62 59 58 52 43 42 42 30 26 R&D in industry The number of R&D scientists and engineers (on a full-time equivalent basis) in industry was at its highest level in 1969, declined in later years through 1972, and then increased in 1973 and 1974, bringing the number to approximately its 1971 level (360,000, or 68 percent of all R&D scientists and engineers).-'" The recent increases occurred primarily in the chemical, machinery, and electrical equipment industries; the largest decline since 1969 occurred in the aircraft and missiles industry.-'' These four industries are among the leading industrial employers of R&D scientists and engineers, accounting for almost 70 percent of the industrial total in 1974. The Federal Government is a major source of support for industrial R&D activities; 32 percent of industrial R&D scientists and engineers were supported by Federal funds in 1974 (figure 5-15). ^° These and other aspects of industrial R&D are covered more fully in another chapter in this report entitled, "Industrial R&D and Innovation". ^' See Appendix table 4-9b. Figure 5-15 R&D scientists and engineers' employed in industry, by source of R&D funds, January 1967 and 1974 Federally supported aiiousands) Company supported 170 150 130 110 90 70 50 30 10 10 3D 50 70 90 110 130 150 170 190 210 230 250 I I I 1 I ' Fuli-time equivalent basis. SOURCE: National Science Foundation. 130 However, this represents a significant decrease from 1967 when the Federal share amounted to 44 percent. The relative decrease in federally supported R&D scientists and engineers is most evident in the machinery, aircraft, and motor vehicle industries. As shown in the figure, almost 80 percent of the federally supported R&D scientists and engineers are employed in the electrical equipment and aircraft and missiles industries, both of which are heavily involved in space and defense R&D. UNEMPLOYMENT AMONG SCIENTISTS AND ENGINEERS Employment of scientists and engineers during most of the 1960's rose substantially in all sectors. Unemployment was low, ranging around 1 percent, and for most of the period remained about three-fourths of the level for all professional, technical, and kindred workers, and no more than one-fourth the rate for all workers in the country (figure 5-16). However, starting in the early 1970's, changes in the labor market for both scientists and engineers were brought about by a series of factors — cut-backs in defense and other R&D programs, the general economic downturn, and the beginning of the decline in academic recruiting. Thus, unemploy- ment rates for scientists and engineers reached a level of around 3 percent at the beginning of 1971. At that point, the rate was nearly as high as that for all professional workers but only one- half that for all workers. Early in 1972 the employment situation began to improve. The unemployment rate for engineers alone dropped from 3.2 percent in the first quarter of 1971 to under 1 percent at the end of 1973 — a rate similar to that of the mid-1960's. In mid-1974 the unemployment rate for a sample of scientists and engineers was 1.1 percent.-^- Of those employed, 97 percent held full-time positions while 3 percent were working '- "National Sample of Scientists and Engineers: Changes in Employment 1970-72 and 1972-74", Science Resources Studies Highlights. National Science Foundation (NSF 75-309), May 19, 1975. Figure 5-16 Average unemployment rates, 1963-74' (Percent) 4 7 6 5 4 3 2 1 - All workers ^^ ^^^^ _^r^ _ Professional and a« ^ ^ .^ Technical workers _ Engineers ^^^^^^^ Scientists • ^ 1 1 1 1 ■ ,.' — ■■'•••••i^Sc ■^ ° Doctoral scientists °'":t°^3' ^"^ineers • 1 1 1 1 1 1 19 63 '64 '65 '66 '67 1 Not seasonally adjusted. SOURCE: Bureau of Labor Statistics and National Science Foundation. ■68 '69 '70 '71 11 11 1 131 part-time. In late 1974, the unemployment level for engineers alone was still only 1.9 percent. Unemployment rates express only a part of the overall situation. The national unemploy- ment rate, for example, is expressed in terms of occupation last held. In some cases an individual scientist or engineer may have previously taken a nonscience or nonengineering job before becoming unemployed and would therefore not be reported as a scientist or engineer. Unemploy- ment levels, furthermore, do not indicate the extent of employment (part-time employment may be involuntary) nor the degree of un- derutilization in positions requiring lesser skills than individuals possess. In addition, in most instances it has not been possible to measure the difficulty or the length of time required for obtaining employment for scientists and engineers who are first entering the job market or for those who are changing jobs. SUPPLY OF SCIENTISTS AND ENGINEERS Early student engineering interest science and Information concerning occupational preferences of college freshmen provides an early indicator of student interest in science and engineering. 3-' In recent years, interest has decreased in the occupations of research scien- tist, engineer, and educator, while increasing in those of medical doctor, nurse, and non-M.D. health professional (figure 5-17). A second indicator of early interest in science and engineering is the choice of college majors by National Merit Scholars as they enter college (figure 5-18). The proportion of these students planning to enter science and engineering increased from 62 percent to 70 percent between 1966 and 1974. Between 1972 and 1974, however, there was a decline of two percentage points in the proportion of National Merit Scholars choosing science as a major, while over this same period, there was an increase of nearly three percentage points in those planning to major in engineering. The earliest information about undergraduate enrollments by major field is obtainable in a Figure 5-17 Occupational preference of college freshmen, 1968-74 (Percent) ^k Educator 20 \ 15 ^ Business ^ '», \ / '"-■■■■ 1 II II Id II III II I II III if^^ 10 Engineer Health ^^*1*' 6 •^™ — ™^^ Nurse , Farmer or forester Research scientist 1 1 1 1 1 1968 '69 70 71 72 73 74 Source; American Council on Education and University of California, Los Angeles. Student's junior year. One study shows that total junior-year undergraduate enrollment increased by 3.2 percent in the fall of 1972 over the fall of 1971. J-i The number of students majoring in various science and engineering fields increased about 4.5 percent. Life science majors increased by more than 12 percent. Social science majors increased about 6 percent, and in the fall of 1972, they accounted for 47 percent of the science and engineering majors. Fewer students chose majors in engineering, mathematical sciences, and physics, while small increases occurred in chemistry and other physical science majors in the fall of 1972. " The American freshman: National Norms, American Council on Education and University of California, Los Angeles, annual series. '■I J. E. Dutton and B. A. Blandford, Enrollment of Junior-Year Students (1971 and 19721. (Washington, D.C.: American Council on Education, 1973). 132 Figure 5-18 College majors of National Merit Scholars, 1966-74 50 40 20 Science All other fields and undecided Engineering •«..^* J \ \ \ \ L 1966 '67 '68 '69 70 71 72 73 74 Selected fields of science (Percent) 18 16 ^ .•••••. ..•' -• Mathematics ••••• 14 "' ••• .••.. / • Social sciences >'' '.' 12 •- •• • • • 10 / \ 8 _ X '**«»^ ''""''"^ X Pre-medicine 6 '"'^'X^^^^ 4 Ctiemistry ^^113 2 - 1 1 1 1 1 1 1 1966 '67 '68 '69 '70 '71 Source: National Merit Scliolarsliip Corporation. Bachelor's degrees awarded Annual awards of bachelor's degrees in the sciences and engineering are shown in figure 5- 19 for the years 1960 through 1972, the last year for which National Center for Educational Statistics data are available. Over the 1960-72 period, the annual recipients of science and engineering degrees doubled, including a tripling of the number of recipients of social science degrees. Social science degrees — as a proportion of all bachelor's degrees in science and engineering — rose from about 26 percent in 1960 to almost 50 percent in 1972. Bachelor's degrees in science and engineering, as a fraction of bachelor's and first-professional degrees-^5 jn all fields, remained essentially constant at approximately 30 percent between 1960 and 1972. The large increases in annual recipients of social science degrees were respon- sible for maintaining the fraction at a constant level; engineering degrees, as a proportion of degrees in all fields, declined continuously from 10 percent to 5 percent during the period and the physical sciences fell from 4 percent to 2 percent. Graduate enrollments in science and engineering Enrollments in the various fields at the graduate level are affected by many complex factors, including population trends, attitudes and aspirations (such as the increasing career interests of women), military draft regulations, employment outlook, and financial capability of the students. The availability or lack of Federal support for fellowships, traineeships, and train- ing grants has an obvious, though not precisely measurable influence on graduate enrollments in science and engineering. Enrollments for advanced degrees in science and engineering fields, as shown by annual data from the National Center for Educational Statistics, have grown considerably over the long term, doubling from 1960 to 1972 (figure 5- 20). Within the science and engineering fields, engineering had the largest enrollment from 1960 through 1968, but declined in later years. During the 1960-72 period, however, the most rapid growth in enrollment for advanced degrees occurred in fields other than science and ■" M.D., D.D.S., D.V.M., etc. 133 Figure 5-19 Bachelor's degrees in science and engineering, 1960-72 Number (Thousands) 280 240 200 160 120 80 40 All sciences and engineering , Social sciences Life sciences Engineering Physical sciences Mathematical sciences i 1 I I I I I I I I \— As a percent of all bachelor's and first-professional degrees (Percent) 35 25 All sciences and engineering Social sciences > ^^ Engineering Life sciences I I I Mathematical sciences ^^ Physical sciences I I I \ I ^^ 1960 '62 '64 '66 '68 '70 '72 I960 '62 SOURCE: National Center for Educational Statistics and National Science Foundation. '64 '70 '72 engineering. As a result, enrollment for ad- vanced degrees in science and engineering fields as a proportion of all advanced degree enroll- ment declined from 38 percent in 1960 to 28 percent in 1972 (figure 5-20). Engineering and the physical sciences accounted for most of this decline. Related data, though not strictly comparable to those of the National Center for Educational Statistics, illustrate the direction of more recent trends in graduate enrollment. Data collected by NSF from institutions granting science and engine ering doctorates indicate that the number of full-time graduate students in these fields decreased steadily from 1969 to 1974. Data from this fall 1974 survey indicate that full-time graduate science enrollment increased about 5 percent over fall 1973, the first increase since 134 Figure 5-20 Enrollment for advanced degrees in science and engineering, 1960-72 Number 100 60 40 Social sciences Engineering^ « I I I I I ' Life sciences Physical sciences Mathematical sciences J I I I I 1960 '62 '64 '66 '68 70 '72 SOURCE: National Center for Educational Statistics and National Science Foundation. As a percent of enrollment for advanced degrees In all fields (Percent) 40 1 36 32 28 24 16 - Total science and engineering Engineering '^^^ Social sciences ^^--. Life sciences Mathematical sciences J I L ^_--, J \ L I960 '62 '68 '70 1969. The life sciences accounted for almost all of the overall increase.^" Master's degrees awarded Annual awards of master's degrees in science and engineering for 1960 through 1972 are '" "Graduate Science Enrollment in 1974 Shows First Increase Since 1969", Science Resources Studies Highlights, National Science Foundation (NSF 75-328), October 22, 1975. shown in figure 5-21. The number of these degrees awarded annually increased by over 150 percent between 1960 and 1972, with the largest percentage increases occurring in the mathematical sciences (307 percent) and the social sciences (263 percent), and the smallest in the physical sciences (86 percent). As a fraction of master's degrees in all fields, sciences and engineering degrees declined from a high of 30 135 Figure 5-21 Master's degrees in science and engineering, 1960-72 Number (Thousands) 55 I 45 All sciences and engineermg Engineering ^ ^ ^ * ^^ Social sciences «• • ■ , • * Life sciences Physical sciences Mathematical sciences J \ \ \ \ I \ I I As a percent of all master's degrees (Percent) 30 r All sciences and engineering Engineering ^-%. Social sciences Mathematical sciences J \ \ \ \ L 1960 '62 '64 '66 '68 70 '72 1960 '62 '64 '66 '68 '70 SOURCE: National Center for Educational Statistics and National Science Foundation. percent in 1965 to 21 percent in 1972; the largest proportional declines occurred in engineering and the physical and life sciences. Doctoral degrees awarded Annual awards of doctorates are shown in figure 5-22. Science and engineering degrees accounted for the majority of all doctorates awarded between 1965 and 1974, but their share fell from a high of 64 percent in 1964 to 56 percent in 1974. The number of men receiving doctoral degrees decreased in 1974, and although there was an increase in women doctorates it was not great enough to offset the drop for men. Changes in major areas of science over the 1965-74 period are shown in figure 5-22. The physical sciences exhibited the slowest growth throughout the period and the largest decline in recent years; the number of physical science doctorates awarded dropped almost 20 percent from 1971 to 1974. Much of this decline is due to 136 Figure 5-22 Doctoral degrees awarded, 1965-74 (Thousands) 35, 30- 25 - .^^ Another J L 1965 '66 '67 '68 '69 70 71 72 73 74 Science and engineering doctorates, by field, 1965-74 (Thousands) Social sciences Life sciences Engineering Mathematical sciences J L J L J I 1965 '66 '67 '68 '69 '70 '71 '72 '73 '74 SOURCE: National Academy of Sciences. the sharp drop in physics and astronomy doctorate recipients, down 23 percent from 1971 through 1974, and to a nearly 20 percent decrease in chemistry doctorates over the same period. Graduate student support During the 1967-74 period, there were significant shifts in the patterns of support of graduate science students. In 1974, Federal support for full-time graduate science students in doctorate-granting institutions was provided at a level only slightly more than one-half that of 1967. While Federal support was being reduced, institutional and self-support increased, as shown in the table below. Percent distribution of full-time graduate science students in doctorate departments, by source of major support, 1967 and 1974" Major source of support 1967 1974 Federal support 42 25 Institutional support 34 40 Other outside support 10 9 Self-support 14 26 Among the various Federal programs for financial aid to graduate students, major reduc- tions occurred in the number of awards for fellowships and traineeships.^s By 1974, the number of graduate science students on federal- ly supported fellowships and traineeships was reduced to approximately one-third of the 1967 level. There was also a decrease in the employ- ment of graduate students on research projects, with the result that research assistants receiving Federal support declined by almost 20 percent during the same period. Since Federal R&D obligations to academic institutions rose 11 percent in constant dollars from the base year of 1967, it appears that occupational categories other than research assistants were given greater priority by these institutions. There have been marked changes in patterns of Federal support of fellowships, traineeships. '~ Graduate StmienI Support and Manpower Resources in Science Education. 1969, National Science Foundation (NSF 70-40) and Graduate Science Education: Student Support and Postdoctorah, Fall 1974, National Science Foundation (NSF 75-322). " Graduate Science Education: Student Support and Postdoctorah, National Science Foundation, annual series, and special tabulations. 137 and training grants in recent years. s" Rather than providing direct student aid, there has been a tendency to rely more heavily on graduate student participation in federally funded research projects that support areas of national concern. Thus, Federal obligations for fellowships, traineeships, and training grants declined from $421 million in 1971 to $287 million in 1973. These funds rose again in 1974 to $327 million, largely because approximately $85 million of funds impounded in 1973 were released to HEW in 1974. Among Federal agency programs affected by the shifts in funding were the Office of Education's student programs under the National Defense Education Act, NSF's traineeship program, and NASA's traineeship program. As a result, obligations by the Office of Education declined from $52 million in 1971 to $41 million in 1972, and after the termination of National Defense Education Act awards, to $10 million in 1973. NSF's support of fellowships and traineeships dropped from $42 million in 1971 to $16 million in 1973, and NASA's traineeship program was virtually eliminated. Reductions in Federal support of fellowships, traineeships, and training grants were spread across all fields of science. The largest absolute decrease occurred in the life sciences, which dropped from $225 million in 1971 to $179 million in 1973. Immigrant scientists and engineers Another source of supply of scientists and engineers are those persons achieving im- migrant status in the United States. Ap- proximately 6,600 scientists and engineers immigrated to the United States in 1973. These numbers (see the table below) represent a reduction from the high 1966-72 yearly inflows resulting from revisions in October 1965 in the national immigration laws. Scientists and engineers immigrating to the United Stales, annual average, 1949-73'" Natural Social Period Total Engineers scientists scientists 1949-65 ., 4,053 2,851 1,048 154 1966-72 ., 11,531 7,993 2,973 565 1973 6,632 4,443 1,790 399 In February 1971, the existing system of "precertification" of prospective immigrants came to an end under U.S. Department of Labor regulations. This change did not bring about an immediate reduction in immigration because large numbers of foreign scientists and engineers, in anticipation of this legislation, had become precertified for immigration and eligible to enter the United States. There were enough of these scientists and engineers "in the pipeline" to maintain a high inflow of immigration through 1972, but the number of immigrant scientists and engineers has fallen sharply since that year. Over the period 1966-68, the largest numbers of immigrant scientists and engineers came to the United States from developed nations. After that time, the situation changed, with by far the largest numbers coming from the developing nations. WOMEN AND MINORITIES IN SCIENCE AND ENGINEERING Women employed in science and engineering Increasing interest has been expressed in recent years in the opportunities for participa- tion of women in science and engineering. Despite the widespread interest, however, relatively little information is available on the subject, particularly those that allow the ex- amination of trends over time. This section presents some data concerning the employment of women in science and engineering oc- cupations, women receiving doctorates in these areas, and women enrolled for advanced degrees in the sciences and engineering. In 1974, women comprised 5 percent of the persons employed in science and engineering occupations, compared with 39 percent of the total civilian work force, and 41 percent of the professional and technical workers.^' Large differences exist in the level of employment of women among the various fields of science and engineering, as shown in the table below. " Feiieral Support to Univcrsiiiei, Colleges, ami Selected Nonprofit Institutions. National Science Foundation, annual series. '" "Immigration of Scientists and Engineers Orops Sharply in FY 1973; Physician Inflow Still Near FY 1<'72 Peak", Science Resources Studies Highlights, National Science Foundation (NSF 74-302), March 29, 1974, and earlier reports of the series. " The category of professional and technical workers includes occupations such as accountant, lawyer, nurse, physician, and teacher. In 1970 (the most recent year for which comparable data are available), the proportions of all lawyers who were women (5 percent) and the proportion of all physicians who were women (9 percent) were relatively similar to that for scientists and engineers (5 percent). 138 Percent distribution of women scientists and engineers, by field, 1974''^ Field Percent of total Psychologists 28 Social scientists 21 Mathematical scientists 15 Life scientists 13 Computer scientists 12 Physical scientists 8 Environmental scientists 3 Engineers 1 Women scientists and engineers were most likely to be involved in psychology and the social sciences, and least likely to work in engineering and in the environmental and physical sciences. A somewhat different pattern of employment of women scientists and engineers exists in the academic sector. In 1974, 15 percent of the scientists and engineers employed full-time''^ at colleges and universities were women; 16 percent of the scientists and 1 percent of the engineers. The proportion of women in each field of science varies widely, as shown in the table below. Women comprise 21 percent of both the life scientists and the psychologists, but less than 10 percent of both the physical and environmental scientists. In the case of doctorate-granting institutions alone, the level of employment of women is somewhat lower than in colleges and universities as a whole. Women in graduate education An increasing number of women are pursuing advanced studies in science and engineering (figure 5-23). Between 1965 and 1974, the number of women receiving doctoral degrees in science and engineering increased by almost 250 percent, from 744 to 2,590. This absolute growth also represents an increase in the share of science and engineering doctorates earned by women, the proportion growing from 7 percent in 1965 to 14 percent in 1974. By 1974, women were awarded 24 percent of the doctorates in the social sciences, and 18 percent in the life sciences, but 10 percent or less in the mathematical sciences, physical sciences and engineering. ■»<' The proportion of women students enrolled for Full-time women scientists and engineers employed by universities and colleges, by field, 1974''^ Field All scientists and engineers Engineers Physical scientists Chemists Physicists Other physical scientists Environmental scientists'^ . Mathematical scientists . . . Life scientists Agriculture Biological Medical Psychologists Social scientists All institutions Doctorate institutions Number Percent women Number Percent women of women in each field of women in each field 35,083 15 20,896 14 311 1 260 2 1,912 7 801 5 1,378 10 526 7 392 4 197 3 142 7 78 6 319 5 195 5 2,825 13 856 9 19,264 21 14,605 19 1,796 13 1,757 15 5,550 18 3,379 16 11,918 25 9,469 22 3,067 21 1,132 17 7,385 15 3,047 13 <2 National Science Foundation, special tabulations. ■>■' Data for part-time women scientists and engineers are not available. " Manpower Resources for Scientific Aclwilies al Unwersilies and Colleges. January 1974, Detailed Statistical Tables, National Science Foundation (NSF 75-300-A). '5 Includes earth scientists, oceanographers, and at- mospheric scientists. '" For further, more recent information on this topic see Joseph L. McCarthy and Dael Wolfle, "Doctorates Granted to Women and Minority Group Members", Science, Vol. 189, (1975), pp 856-859. 139 Figure 5-23 Women as a percent of total science and engineering doctorate recipients, by field, 1965-74 (Percent) 30 20 Social sciences — '^^ Engineering JHLJ^-^LMJ^LHJM-H-^lJ Physical sciences J L 1965 '66 -67 '68 '69 '70 '71 '72 '73 '74 ' Includes environmental sciences. SOURCE: National Academy of Sciences. advanced degrees in science and engineering, as reported by the Office of Education, also increased markedly, by 73 percent overall between 1966 and 1972. In 1972 (the latest year for which data are available) women represented varying proportions of the total enrollments of each of the fields below. Proportion of women enrolled for advanced degrees, by field, 1966 and 1972 Percent of total Field 1966 1972 All science and engineering fields ... 13 19 Social sciences 24 31 Life sciences 20 24 Mathematical sciences 18 22 Physical sciences 8 12 One factor which may affect the participation of women in science and engineering is the substantial difference in salary levels for men and women in science occupations. Among doctoral scientists and engineers, the 1973 median salary for women ($17,600) was 17 percent lower than that for men ($21,200). Women's salaries are consistently below men's at each age level, but the gap widens con- siderably after age 40. ■'^ Racial minorities in science and engineering Information concerning the racial identifica- tion of members of the scientific community has been made available only in recent years. Data are presented here concerning the racial com- position of the national pool of scientists and engineers, the characteristics of minority doc- toral scientists and engineers by field, and the representation of minority students in each field of graduate science study. Caucasians represent the predominant por- tion of all scientists and engineers (96 percent); those of Asian background account for over 2 percent. Blacks comprise about 1 percent, and other nonwhites (e.g., American Indians) the remainder (figure 5-24). The field of mathematics has the largest proportion of racial minorities (8 percent), followed by the physical sciences (6 percent) and the life sciences (6 percent). Blacks have the highest level of participation in mathematics, ■>" Doctoral Scienttils and Engineers in the U.S.. 197} Profile, National Academy of Sciences, 1974. 140 representing 5 percent of all mathematicians. Orientals in the physical sciences (4 percent), and other races in the life sciences (1 percent). The largest absolute number of minorities, by total and for each group, are found in engineer- ing, although minorities have the smallest proportional representation in this field. The representation of minorities among doctoral scientists and engineers in 1973 is shown in the table below.''* *^ Ch(iracterislii's of Docioral Scientists and Engineers in the United Stales, 1973, National Science Foundation (NSF 75-312). ■''' Less than 0.05 percent. Proportion of minority doctoral scientists and engineers, by field, 1973 Percent in each field American Field Black Indian Asian All scientists and engineers 0.8 (^°) 4.5 Physical scientists 8 (<^) 4.7 Mathematical scientists ... .8 NA 4.8 Environmental scientists . . .2 NA 2.8 Engineers 2 i^") 8.4 Life scientists 9 (»») 4.3 Psychologists 9 {"") 1.1 Social scientists 1.1 1 3.6 Figure 5-24 Minority representation among scientists and engineers, by field, 1972 (Percent) 12 3 4 5 1 I I I I I All minorities All science and engineering fields Engineers Mathematical scientists Computer scientists Life scientists Physical scientists Social scientists and psychologists Blacks SOURCE: National Science Foundation. 141 Among the black doctoral scientists and engineers, the largest proportion is involved primarily in teaching activities (40 percent), followed by administration (19 percent), and research and development (16 percent). This pattern of activity applies in each of the fields. Black doctoral scientists and engineers are employed for the most part by universities and four-year colleges (61 percent), with the next largest proportions employed by industry (13 percent) and the Federal Government (7 per- cent). This pattern is consistent across all fields. In comparison, about one-half of the white doctoral scientists and engineers are employed by universities and four-year colleges, with the next largest proportion (21 percent) employed by industry, and 8 percent employed by the Federal Government. Doctoral scientists and engineers who are American Indians also are primarily involved in teaching (69 percent) and employed by universities and four-year colleges. Asian doctoral scientists and engineers exhibit quite different characteristics. They are primari- ly involved in research and development (41 percent), teaching (29 percent), and administra- tion (7 percent). Compared with the other minorities, a greater proportion of Asians are employed by industry: 51 percent in universities and four-year colleges, 28 percent in industry, and 5 percent in the Federal Government. These data suggest that there are characteristic patterns of involvement in science for selected minorities. Black scientists and engineers, for example, tend to be involved in social science and health science fields, and predominantly in teaching activities. In contrast, Asian Americans tend toward the physical sciences and engineering, and involvement in R&D activities. An indication of the current participation of minority students in science and engineering graduate study is presented in the following table. 5" It should be pointed out that these data do not represent national totals, but they were reported by a significant proportion of doctorate-granting institutions. Proportion of minorities in science and engineering graduate studies, by field, 1973 Percent in each field Field Black All science and engineering 2.5 Physical sciences 1.4 Mathematical sciences .... 2.5 Engineering 1.2 Life sciences 1.5 Health professions 5.5 Social sciences and psychology 4.1 American Indian Asian 0.3 .2 .2 .1 .2 .6 2.1 2.6 2.1 3.3 1.9 2.0 1.1 In analyzing the proportion of black students enrolled in each field, it is apparent that the health professions and social sciences attract the largest percentage of black graduate students, while engineering, physical sciences, and life sciences attract the lowest proportion. In contrast, the Asian graduate students enroll in higher proportions to study engineering and the physical sciences, and are less involved in the social sciences. 50 Elaine H. El-Khawas and Joan L. Kinzer, EnrotlmenI of Minorihi Graduate Students at Ph.D. -Granting Imtitutions, (Washington, D.C.: American Council on Education, 1974). 142 Public Attitudes Toward Science and Technology 143 Public Attitudes Toward Science and Technology INDICATOR HIGHLIGHTS The belief that science and technology have changed life for the better was expressed by 75 percent of the public in 1974, compared with 70 percent in 1972; 5 percent saw the change as for the worse, down from 8 percent in 1972. In the public's ranking of nine professions and occupations, scientists were second only to physicians in both 1972 and 1974, with engineers in third place. Science and technology were believed to have done "more good" than "more harm" by 57 percent of the people in 1974, compared with 54 percent in 1972; 31 percent in both years saw the impact as about evenly divided between good and harm. Among people who believe science and technology do more good than harm, the largest group (59 percent in 1974 and 54 percent in 1972) cited improvements in medicine and medical research as the leading benefit; among those having the view that science and technology do more harm than good, 'lack of concern for the environment" was the most frequently mentioned example (25 percent in 1974 and 27 percent in 1972). Science and technology were thought to have caused some of our problems by approximately half of the respondents in both 1972 and 1974; a smaller group (approximately 37 percent) believed that few or none of our problems were so caused, while a still smaller group (less than 8 percent) thought that science and technology were responsible for most of the problems. The pace of change produced by science and technology was viewed as "about right" by some 50 percent of the public in both 1972 and 1974, as too fast by about 20 percent of the people, and as too slow by a slightly smaller percentage. The public expects science and technology to solve, eventually, many of our major problems, although the fraction expecting most problems to be so solved declined from 30 percent in 1972 to 23 percent in 1974. Areas in which the public felt they would most like to have taxes spent for science and technology were health care, crime reduc- tion, education, prevention of drug addic- tion, and pollution control; areas in which they would least like to have taxes spent for science and technology were "space explora- tion" and "developing and improving weapons for national defense." Demographic analysis of selected questions in the survey suggests that the most positive attitudes toward science and technology were held by men, persons between 30-59 years of age, those with some college education, and by people whose family income was $10,000 or more. 144 Public attitudes affect science and technology in many ways. Public opinion sets the general environment and climate for scientific research and technological development. It is influential in determining the broad directions of research and innovation, and through the political process, the allocation of resources for these activities. In addition, public attitudes tov^'ard scientists and engineers and their efforts affect the career choices of the young by influencing their decision to enter these fields. The survey of public attitudes toward science and technology summarized in Science Indicators — 1972 was repeated for this report.' The 1974 replication of the earlier survey serves both as a check on the findings of the previous survey and as the beginning of a time series of data for tracking trends in attitudes and opinions. A personal interview survey was conducted in July and August 1974 among 2,074 persons 18 years of age and older. The sampling techniques used in the survey permit the results to be pro- jected to the entire U.S. population. The survey was designed to explore three aspects of public attitudes and opinions: the public's regard for science and technology; the public's sense of the impact of those activities; and the public's expectations and desires regard- ing the role of science and technology in dealing with national problems. Results are reported first for the total sample of respondents, and then for demographic groups. TOTAL GROUP RESPONSES Public regard for science and technology Three aspects of attitudes were explored under this heading: how the public feels science and technology have affected the quality of life; the general emotional reaction associated with science and technology; and where scientists and engineers rank in prestige among nine professions and occupations. Percen ( 972 1974 70 75 8 5 11 11 2 3 9 6 In 1974, 75 percent of the public felt that science and technology have changed life for the better, compared with 70 percent in 1972. This gain is concurrent with a decline in the "worse" and the "no opinion" responses. Do You Feel That Science and Technology Have Changed Life for the Better or for the Worse? Response Better Worse Both No effect No opinion The reaction of "satisfaction or hope" to science and technology was expressed by 56 percent of the people in 1974, versus 49 percent in 1972. In both years, a reaction of "excitement or wonder" was shared by 22 to 23 percent of the public. Fewer respondents expressed "No opin- ion" in 1974 than in 1972. Which One of These Items Best Describes Your General Reaction to Science and Technology? Response Satisfaction or hope . . Excitement or wonder Fear or alarm Indifference or lack of interest No opinion For a further indication of the regard for science and technology, people were asked to rate each of nine professions and occupations in terms of the "prestige or general standing that each job has." The rating categories used were "excellent," "good," "average," "below average," and "poor." These categories were assigned weights, and the resulting rankings are shown below, not only for the 1972 and 1974 surveys butalso for comparable studies in 1947and 1963. Percet ( 1972 1974 49 56 23 22 6 5 6 7 16 11 1 Both surveys were conducted by the Opinion Research Corporation, Princeton, N.J. For more complete information concerning the survey results and methodology, including a description of the reliability of the results and the differences required for statistical significance, see: Alliliuies of the U.S. Public Toward Science and Technology, Study U, Opinion Research Corporation, 1974 (A study commissioned specifically for this report). 145 Rankings of Occupations 1947" 1963" 1972 1971 Physician 1 1 ^ 1 Scientist 2 2 2 2 Engineer 7 6 3.5 3 Minister 4 5 3.5 4 Architect 5.5 4 6.5 5 Lawyer 5.5 3 5 6 Banker 3 7 6.5 7 Accountant 9 8 8 8 Businessman 8 9 9 9 ^R W Hodge, f/al. "Occupational Prestige in the United States. XtZi-bi," Amtrtan journal of Sotlolosy, Vol 70 (1964), pp 286-302 In both 1972 and 1974, scientists held their relative ranking among occupations, second only to physicians, with engineers third. Against 1963 ratings, all occupations remained lower in both 1972 and 1974. Impact of science and technology This part of the survey explored several facets of the impact of science and technology as perceived by the public, including whether the overall impact is more positive than negative; identification of the science and technology activities which the public regards as good or harmful; the extent to which it feels science and technology cause problems; and whether the pace of change induced by science and technology is desirable. Following these questions, the public was asked to assess the adequacy of control that is exercised over science and technology. Slightly more than half of those interviewed believed that science and technology do more good than harm. About one-third saw the extent of good and harm as being nearly the same, and only a negligible percentage said "more harm." Changes from 1972 to 1974 were slight. Overall, Would You Say That Science and Technology Do More Good Than Harm, More Harm Than Good, or About The Same Each? Response More good About the same More harm No opinion Those responding "more good than harm" or "about the same" were asked, without prompt- 1 °eneni 1972 1974 54 57 31 31 4 2 11 10 ing, to mention some "good thing" they thought science and technology had done, and the responses were then categorized. The results summarized below show that "medical ad- vances" was by far the most frequently men- tioned benefit, followed by "new and improved products" and "space research". Benefits from Science and Technology (Cited by group responding "More good than harm") Percent ciUng " Response 1971 1974 Medical advances 54 New and improved products 10 Space research 12 Environment and natural resources 6 Living and working conditions 5 Food and agriculture 4 Energy 1 Other 4 Don 't know 4 59 11 9 Benefits from Science and Technology (Cited by group responding "About the same") Percent citing " 1974 48 15 9 6 5 2 2 6 19 Response 1972 Medical advances 50 New and improved products 8 Space research 9 Living and working conditions 5 Environment and natural resources 6 Food and agriculture 3 Energy ('') Other 3 Don't know 1^ 1 Multiple responses were accepted. Less than 0,5 percent, The group which believed that science and technology do about equal amounts of good and harm was asked, without prompting, to mention "one of the harmful things." These results, summarized below, show that "lack of concern for the environment" was most frequently mentioned as harmful, followed by "develop- ment of military weapons," "space research," and "dangerous drugs and medicines." Almost one-third of this group failed to offer an example of a harmful result from science and technology, whereas less than 20 percent of the same group failed to provide an example of a "good" result. (See the table just above). 146 Harmful Effects of Science and Technology (Cited by group responding "About the same") Percent ciling Response 1972 1974 Lack, of concern for the environment 27 25 Development of military weapons 9 11 Space research 16 9 Dangerous drugs and medicines 3 9 Depletion of natural resources 2 2 Other 16 19 Don't know 27 32 Multiple responses were accepted As shown in the following three tables, the public's views remained stable over the 1972-74 period on questions regarding the relationship of science and technology with society. Science and technology are thought to cause some of today's problems by about half the public, and as the source of few or none of the problems by some 40 percent. Do You Feel that Science and Technology Have Caused Most of our Problems, Some of our Problems, Few of our Problems, or None of our Problems? Percen ( '972 1974 28 28 48 46 7 8 17 18 technology should "remain as it is," and nearly 30 percent felt that greater control was needed. Do You Feel That the Degree of Control that Society Has Over Science and Technology Should be bicreased. Decreased, or Remain As It is Now? Response Should be increased . . Remain as it is Should be decreased . . , No opinion Expectations and directions for science and technology About three-fourths of the public remained confident that science and technology will eventually solve at least some of the major problems, examples of which were named in the question. But the expectation that most problems would yield to such solution declined, falling from 30 percent in 1972 to 23 percent in 1974. The trend toward a lower level of confidence is evident in the larger percentage of those who expect science and technology to solve only "some" and "few" such problems. Percen / 972 1974 7 6 48 50 27 29 9 9 9 6 Response Most Some Few None No opinion A slight majority of the public continued to feel that science and technology produce change at a pace "about right". Remaining opinion is almost evenly divided between "too fast" and "too slowly". Do You Feel That Science and Technology Change Things Too Fast, Too Slowly, or Just About Right? Response Too fast About right Too slowly No opinion Almost half of those polled felt that the extent of control society should have over science and Percen 1 972 1974 22 20 51 53 16 18 11 9 Percent 1971 I 1974 30 23 47 53 16 20 7 4 Do You Feel That Science and Technology Will Eventually Solve Most Problems Such as Pollution, Disease, Drug Abuse, and Crime, Some of These Problems, or Few, if Any of These Problems? Response Most problems Some problems Few problems No opinion Areas in which the public would "most like" to see their tax money for science and technology sjjent are "health care," "reducing crime," "reducing and controlling pollution," "prevent- ing and treating drug addiction," and "im- proving education."^ Two major shifts in public preferences occurred in these areas between 1972 and 1974: "reducing and controlling pollution" declined considerably in the frequen- cy of selection, whereas "improving education" increased. Among the less highly ranked areas. - Selection was made from a list of 12 areas snown in the next tabulation. 147 "improving the safety of automobiles" fell from the choice of 38 percent of the public in 1972 to 29 percent in 1974. Areas in which the public in 1974 indicated they would least like their taxes spent for science and technology were "space exploration," and "developing or improving weapons for national defense." period. An increasingly large percentage of the public believed that science and technology had changed life for the better; a substantial and growing fraction expressed a feeling of satisfac- tion and hope with respect to science and technology; and scientists and engineers re- ceived high rankings among other occupations and professions, although the rankings of all groups were relatively high. In Which of the Areas Listed Would You Most Like (and Least Like) to Have Your Taxes Spent for Science and Technology? Response Improving health care Reducing and controlling pollution Reducing crime Finding new methods for preventing and treating drug addiction Improving education Improving the safety of automobiles Developing faster and safer public transportation for travel within and between cities Finding better birth control methods Discovering new basic knowledge about man and nature Weather control and prediction Space exploration Developing or improving weapons for national defense No opinion Multiple responses were accepted Percent choosing area Mosi like Least like 972 1974 1972 1974 65 69 1 1 60 50 3 3 59 58 2 2 51 48 4 4 41 48 4 3 38 29 5 8 23 26 14 13 20 18 18 23 19 21 15 14 11 14 19 16 11 11 42 37 11 11 30 30 6 3 13 7 These opinions should be interpreted with caution . The relevance of science and technology for alleviating or solving the problems involved was not considered explicitly. Thus, the responses may reflect areas of general concern to the public without regard for the possible specific role of science and technology in dealing with them. Furthermore, the actual words used in describing the various areas may have a biasing effect; e.g., the word "weapons" in "developing or improving weapons for national defense" may have a negative connotation which accounts in part for the low preference for science and technology in this area. Summary of the total group responses The results of the survey provide reasonably clear answers to the three general questions addressed to the public. The regard for science and technology appears to be relatively high and to have grown slightly during the 1972-74 The results regarding the impact of science and technology are somewhat less positive, and differ little in the two surveys. A small majority expressed the belief that science and technology overall did more good than harm — although they were held responsible for at least some of our problems — while almost one-third thought the impact was about equally divided between beneficial and harmful effects. The extent of social control over science and technology, however, should remain as it is according to almost half those surveyed, whereas the need for greater control was expressed in nearly 30 percent of the responses. The predominant expectation is for con- siderable achievement by science and technology in solving major problems, even though the level of expectation declined somewhat between 1972-74. In both years, slightly more than 75 percent of those surveyed expected science and technology to solve some or most of our current problems. 148 DEMOGRAPHIC RESPONSES The responses of demographic groups, although similar, were not identical. Examina- tion of these differences is limited in this report to two of the questions covered in the survey. The pattern of responses to these two is similar to the attitudes and opinions expressed by the demographic groups to the other survey questions. The response of "no opinion" is relatively high in all groups,-' but is especially so among the oldest, lower income, and least educated sub- groups. Such responses mask differences in expressed opinion toward science and technology and for this reason, comparisons of subgroups in the following two tables are based on percentages of those expressing an opinion. Differences of sex and age Responses of men were somewhat more positive than women to science and technology in both questions. Men appear to judge past contributions of science and technology more favorably, and to express more confidence in future accomplishments. Both groups, however, were less confident in 1974 that science and technology would "eventually solve most problems." Overall, Would You Say That Science and More Harm Than Good, or In general, people between 30-59 years of age expressed the most favorable attitudes toward science and technology, followed by the young (18-29 years), and the older group (60 and above). All age groups recorded less confidence in 1974 in expecting problems to be solved by science and technology. (Major differences between responses of the youngest group and those of the total are noted below for all questions in the survey.) Differences in education and family income Attitudes and opinions toward science and technology appear to correlate closely with education: the greater the amount of formal education, the more favorable the response. For example, 54 percent of those with less than a high school education felt in 1974 that science and technology do more good than harm, compared with 67 percent of those who had completed high school, and 71 percent of those with some college education. Attitudes and family income appear to cor- relate to some extent on both the overall impact of science and technology and future con- tributions toward solving problems. Some 70 Technology Do More Good Than Harm, About the Same of Each? Percentage of group expressing "More good" J972 1974 All 61 63 Men 64 67 Women 59 59 18-29 yrs 55 59 30-39 69 71 40-49 66 64 50-59 60 67 60 + 57 55 Less than high school 51 54 High school 63 67 Some college 74 71 Family income; Under $5,000 44 56 $5,000-$6,999 47 53 $7,000-$9,999 54 59 $10,000-$14,999 61 71 $15,000 or over 71 69 -' A high frequency of "no opinion" responses occurs typically in surveys concerned with science and technology, as discussed in Amitai Etzioni and Clyde Nunn, "The Public Appreciation of Science in Contemporary America," Daedalus. Vol. 103 (1974), pp. 191-206. "About same" "More harm" "No opinion" 1972 1974 1971 1974 1972 1974 35 35 4 2 11 10 32 31 4 2 8 8 38 39 3 2 13 12 39 38 5 3 8 4 29 28 2 1 7 10 29 33 5 3 7 8 35 31 4 2 9 8 39 41 4 4 19 20 43 43 6 3 18 19 35 31 2 2 5 4 22 27 4 2 5 4 39 41 7 3 18 21 40 41 4 6 Ife 10 34 39 6 2 10 7 34 28 2 1 4 5 27 30 2 1 3 6 149 percent ot those with a family income of $10,000 or more in 1974 felt that science and technology do more good than harm, compared with an average of 58 percent for the groups having a lower income. With regard to solving problems in the future, groups with higher incomes tended to expect solutions from science and technology to a greater extent than the lower income groups. All groups generally expressed more satisfaction with science and technology in 1974 than in 1972, but felt less confident in their ability to solve major problems in the future. in prestige than did the total sample. On the other hand, a somewhat larger percentage of the young in both surveys felt that science and technology have caused some of our problems-- 56 percent versus 50 percent of the total sample in 1974. There are other differences, however, between the young and the total sample, but these do not bear so directly on matters or attitudes as on differences in concern and priority. The young in both surveys expressed For The Most Part, Do You Feel That Science and Technology Will Eventually Solve Most Problems Such as Pollution, Disease, Drug Abuse, and Crime, Some of These Problems, or Few if Any of These Problems? Percentage of group expressing All Men Women 18-29 yrs 30-39 40-49 50-59 60 + Less than high sc'iool High school Some college Family income: Under $5,000 $5,000-$t.,999 $7,000-$Q,999 $10,000-$14,999 .. $15,000 or over . . . "Most" "Some" Tew" "No opinion" 1972 1074 1972 1974 1972 1974 1972 1974 32 24 51 55 17 21 7 4 36 26 47 54 17 20 5 2 29 23 54 59 18 20 9 5 28 24 55 56 17 20 5 2 33 25 54 59 13 16 3 3 31 25 53 56 16 19 4 3 37 27 43 53 20 20 7 2 33 22 46 53 21 25 13 9 33 21 47 54 20 25 12 8 29 25 55 56 16 19 4 2 35 28 51 58 14 14 3 1 35 22 47 51 18 27 14 8 23 24 57 52 20 24 6 5 33 24 49 58 18 18 6 2 33 25 52 56 15 19 3 3 33 26 51 59 16 15 1 1 Attitudes of the young The belief that young people of the Nation have negative attitudes toward science and technology gained considerable credence begin- ning in the late 1960's. To examine the current validity of this belief, responses of the young (18-29 years of age) to all questions of the survey were compared with responses of the total sample. For the most part, attitudes of the young were closely similar to those of the total sample. Major differences from the sample as a whole were found in only two areas, one of which suggests a more positive attitude toward science and technology on the part of the young, whereas the other indicates a more negative assessment. In the first case, the young group (in both 1972 and 1974) rated "scientists" significantly higher consistently more concern for the environment. The reduction and control of pollution was specified by 60 percent of the young group in 1974 as an area where they would most like to see their tax dollars spent, compared with 50 percent for the sample as a whole. "Lack of concern for the environment" was listed in 1974 as one of the "harmful" effects of science and technology by 31 percent of the young versus 25 percent of the total sample. The young differed from the total sample in 1974 in their choice of areas for efforts in science and technology. "Improvement of education" was selected by 58 percent of the young compared with 48 percent of the total sample; "discovering new basic knowledge" was chosen by 29 percent versus 21 percent; and "finding better birth control methods" was selected by 25 percent versus 18 percent. In listing "least liked" 150 areas for expenditures, 45 percent of the young group cited "developing and improving weapons for national defense," as against 30 percent of the total group. Similarly, "weather control and prediction" was cited as "least liked" by 24 percent of the group versus 16 percent of the total sample. OTHER SURVEYS Surveys on public attitudes toward science and technology were recently reviewed by Etzioni and Nunn."* Results from the survey conducted for this report appear to be consistent with earlier studies, to the extent that direct comparisons can be made. The results of the present survey (1972 and 1974) with respect to the public's general regard for science may be placed in a broader context by reference to comparable surveys : a Harris poll in 1972 and one by the National Opinion Research Center (NORC) in 1973 and a replication in 1974. These surveys explored levels of public confidence in "the people who are running" 11 institutions. 5 In the Harris Poll, science as an institution ranked second among the 11 in terms of the percentage of the public indicating "a great deal of confidence." In 1973, the NORC survey" also showed science ranking second, with education, in public confidence. The percentage expressing a great deal of confidence in science rose from 37 to 45 in 1974, but because of an even larger gain for education, from 37 to 49, the rank of science dropped to third among the 11 institutions in 1974. A more recent survey by LaPorte and Metlay^ found a "reasonably high degree of cor- respondence" in responses to several items which were included in the survey reported in Science Indicators — J 972. Similar attitudes, for example, were found in both surveys regarding the confidence and prestige associated with scientists and engineers, the desired extent of social control of science and technology, and ratings of benefits in different areas, such as health and space exploration. TTie LaPorte and Metlay survey, in addition, found that attitudes toward science differ from those toward technology; "there was considerable agreement that scientific activities are intrinsically beneficial and should not be controlled", whereas "the public reaction to the impact of technology upon society is one of wariness and some skepticism". ■> Ibid. 5 Tine institutions for which data were available over the three years included medicine, science, education, the military. Supreme Court, Federal executive branch. Con- gress, major U.S. companies, the press, television, and labor. " Codehook of the General Social Survey. National Opinion Research Center, 1973 and 1974. ~ Todd LaPorte and Daniel Metlay, "Technology Ob- served; Attitudes of a Wary Public," Science. Vol. 188 (1975), pp. 121-127. 151 Appendix 153 CD (J, O) «C ^^ cvj Z ^ Z Z O ir> CO t£) csj ^ csi csj CM •—••—< CNl ■—« CM o r-^ r-; z T CM CM •— i CM CM CO CT> O «C <: ^ <: < 9> ^ 00 CD ^ 00 CO O Oi CM Z liS CO ^- CO 00 ^ a> CM 00 06 a> z ^ CM CM — ' — • Ol •— « r- op ^ «a: eT> o to to r-^ Z liS ^ CM •— t ■— ' ^^ ^^ c> CM lo «a; *a: (T> r^ CM ^ •— < CM CO iri ^ CM ■— « ro cri CM ^H -^ CO r^ CO «:r CD ^ CO o ■— < ^ r-^ CM ^r O CO CM CO CM oi ■— « tb r-^ CM — ' O o^ ^ r-- oi ^ CO ^ ^ 06 ' Z to CM ^^ — * op op •-« Z r^ ^ LO CO ^ C3 tp -I oS CO CM O CO LO CM CD P-- to CM CO CM C3 CD r- CD *3- «i: ^^ ^T CM IT) O CD ^ o c -o to DC E.E ^ aJ E °-<^ cs. E >-> o c -o ,_ . TO ClO S ■ E E ^ ^ aj ==^ -O a> C:? TD ^ -ti c: to -ti ^ «3 -< o Oj DO E E c: «5 tu tz E 0" ■= .« ^ 5. ^°^|| s O) « ^ -otSo: Q. O. Q= w = a; — 00 154 Table 1-2. Scientists and engineers' engaged in R&D, by country, 1963-73 Scientists and engineers' engaged in R&D per 10,000 population Country 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 United States NA 24.7 25.4 NA NA 27.4 27.5 26.8 25.6 25.0 24.9 USSR 18.8 20.3 21.6 25.2 25.8 27.3 29.1 30.7 32.6 34.3 37.2 Japan 12.0 NA NA NA 15.8 NA 16.9 NA 18.9 NA NA West Germany NA 5.7 NA NA 10.3 NA 12.5 NA 14.9 16.2 17.8 France 6.7 NA NA NA 9.9 NA 10.9 NA 11.1 NA NA Scientists and i engineers engaged in R&D ( in thousands) United States NA 474.5 494.1 NA NA 550.4 558.2 549.5 529.7 521.5 523.1 USSR 422.8 463.2 499.4 558.4 605.6 651.5 698.9 746.2 797.8 848.8 931.0 Japan 114.8 NA NA NA 157.6 NA 172.0 NA 198.1 NA NA West Germany NA 33.4 NA NA 61.6 NA 76.3 NA 90.0 100.0 110.0 France 32.2 NA NA NA 49.2 NA 54.7 NA 56.7 NA NA Population (in thousands) United States 189,242 191,889 194,303 196,560 198,712 200,706 202,677 204,875 207,045 208,842 210,396 U.S.S.R 225,060 228,150 230,940 233,530 235,990 238,320 240,550 242,760 245,090 247,460 250,000 Japan 95,900 96,900 97,950 98,850 99,870 101,000 102,200 103,390 104,650 105,990 108,710 West Germany 57,610 58,290 59,040 59,680 59.870 60,170 60,840 60,650 61,290 61,690 61,970 France 47,820 48,310 48,760 49,160 49,550 49,910 50,320 50,770 51,250 51,700 51,915 ' Includes all scientists and engineers (full-time equivalent basis) Data for the United Kingdom are not available. SOURCE Organisation for Economic Co-operation and Development, International Survey of Resources Devoted to R&D by OECD Member Countries, for 1963. 1964. 1967. 1969. and 1971. United Nations. Demographic Yearbook. 1972 and UN estimates for 1973. U.S.S.R. estimates by Robert W. Campbell, Department of Economics. Indiana University. 155 Table 1-3. Distribution of government R&D expenditures among areas by country, 1961-73 National objectives United States National defense Space Nuclear energy Economic development ... Health Community services Advancement of science . . United Kingdom National defense Space Nuclear energy Economic development Health Community services Advancement of science . . France National defense Space Nuclear energy Economic development Health Community services Advancement of science . . West Germany National defense Space Nuclear energy Economic development ... Health Community services Advancement of science . . Japan National defense Space Nuclear energy Economic development ... Health Community services Advancement of science . . , National currency in millions Percent distribution 1961-62 1966-67 1971-72 1961-62 1966-67 1971-72 7.338.5 1,225.9 755.0 339.1 500.6 99.9 118.2 8,264.8 5,307.0 875.0 792.3 968.8 321.1 308.6 8,584.7 2,957.6 838.0 1,322.1 1,379.8 729.2 465.4 70.7 11.8 7.3 3.3 4.8 1.0 1.1 49.0 31.5 5.2 4.7 5.7 1.9 1.8 52.6 18.1 5.1 8.1 8.5 4.5 2.9 1961-62 1966-67 1972-73 1961-62 1966-67 1972-73 248.6 2.7 56.5 37.9 5.7 0.7 26.0 260.4 21.4 65.2 71.0 13.0 1.3 57.8 335.0 11.9 67.3 177.6 32.8 4.5 119.9 64.8 0.7 14.7 9.9 1.5 0.2 6.8 52.3 4.3 13.1 14.3 2.6 0.3 11.6 44.0 1.6 8.8 23.3 4.3 0.6 15.8 1961 1967 1972 1961 1967 1972 1,310.0 16.5 735.0 231.6 13.0 12.7 592.3 3,082.0 522.8 1,723.2 1,381.0 116.1 81.0 1,758.1 3,050.0 730.0 1,600.0 2,200.0 200.0 170.0 2,800.0 44.2 0.6 24.8 7.8 0.4 0.4 20.0 34.9 5.9 19.5 15.6 1.3 0.9 19.9 27.8 6.7 14.6 20.1 1.8 1.6 25.5 1961 1966 1971 1961 1966 1971 381.0 NA 267.0 NA NA NA 639.0 803.0 177.0 693.0 NA NA NA 1,488.0 1,180.0 522.0 1,230.0 1,057.0 195.0 133.0 3,190.0 22.3 NA 15.6 NA NA NA 37.4 19.0 4.2 16.4 NA NA NA 35.3 15.0 6.6 15.6 13.4 2.5 1.7 40.6 1961-62 1965-66 1969-70 1961-62 1965-66 1969-70 3,162.0 NA 5,881.0 25,446.0 724.0 1,071.0 47,321.0 4,495.0 141.0 4,944.0 44,898.0 3,679.0 2,818.0 103,163.0 6,523.0 2,083.0 22,539.0 69,987.0 5,492.0 7,254.0 185,376.0 3.7 NA 7.0 30.1 0.9 1.3 55.9 2.7 0.1 3.0 27.2 2.2 1.7 62.5 2.2 0.7 7.5 23.2 1.8 2.4 61.4 SOURCE- Organisation for Economic Co-operation and Development, Changing Priorities tor Government R&D. July, 1973. 156 Table 1-4. Percent of the scientific literature' citing countries other than the author's own country, by selected fields,^ 1973 Fields Chemistry Physics Biology and biomedical research . Engineering Clinical medicine Mathematics Earth & space sciences Psychology Eight field total Total citations from the 6 major countries Citations to countries other than the author's country Citations to the author's country Percent foreign citations 338,993 233,176 170,985 246,471 43,523 320,158 13.355 35,384 1 1 ,934 1,074,986 105,817 99,105 193,744 37,131 284,700 12,404 37,864 30,364 801,129 69 270,090 63 440,215 56 80,654 54 604,858 53 25,759 52 73,248 48 42,298 28 1,876,115 57 ' Based on 2.121 of the journals in the Science Citation Index for 1973 Included is the literature of the first six countries ranKed by the number of their scientific publications. United States. United Kingdom. West Germany, France. US S.R. and Japan ' See Appendix table l-7a for the description of fields and subfields The social sciences are excluded because comparable data are not available. SOURCE: Computer Horizons. Inc., Indicators ol ttie Quantity and Quality ol the Scientific Literature, 1975 (A study commissioned specifically for this report) Table 1-5. Participation in international scientific congresses, by the United States and other countries, 1960-74 Year Total U.S. Non-U.S. participants participants participants 1 960-62 1963-65 1966-68 1969-71 1972-74 33,082 9,033 24,049 37,964 10,012 27,952 59,748 12,297 47,451 55,711 12,956 42.755 73,819 18,630 55,189 SOURCE: National Academy of Sciences, special tabulations, 157 Table 1-6. Scientific literature' in selected fields' as a percent of total literature, by country, 1965-73 Selected field and year Total Percent of total literature United United West Other (number) States Kingdom Germany' France U.S.S.R. Japan countries 34,657 25.9 7.7 8.2 3.9 30.9 4.1 19.3 39,730 24.5 7.9 8.4 5.9 28.8 5.3 19.2 43.362 24.2 8.2 7.9 5.7 28.5 6.4 19.2 45,052 239 8.4 6.8 6.2 29.0 5.9 19.8 45,665 22.4 7.0 5.4 6.0 30.1 6.0 23.2 45,778 21.2 6.4 5.4 5.8 32.4 6.3 22.6 10,006 49.9 11.2 4.7 1.4 12.6 2.4 17.8 11,968 48.8 11.3 5.6 1.8 12.5 2.8 17.2 13,222 48.3 11.0 6.2 1.8 12.5 2.9 17.4 13,765 49.7 9.0 6.1 2.2 11.8 3.9 17.3 11,992 44.6 9.7 6.4 2.4 11.4 3.9 21.7 12,690 43.7 10.9 7.0 2.5 9.4 4.5 21.9 "2,971 23.9 6.6 6.3 5.6 22.4 4.2 31.8 4,298 23.9 4.6 6.4 4.5 26.4 4.7 29.8 3,024 26.9 6.4 6.0 6.9 20.0 5.2 27.1 3,739 27.8 3.9 6.5 6.0 22.2 7.0 26.7 3,599 29.3 3.9 6.8 5.6 28.6 5.0 20.8 4,844 23.6 4.4 7.0 5.5 30.3 4.2 25.1 24,321 46.6 9.5 4.8 9.4 3.0 4.2 22.4 25,858 48.6 11.0 5.4 7.4 2.1 4.8 20.7 29,359 47.6 9.3 5.5 9.0 1.8 4.9 21.8 30,148 48.7 8.9 5.1 8.9 1.8 5.0 21.6 31,032 45.9 9.7 4.4 9.6 1.8 5.4 23.3 33,619 46.7 9.4 4.3 8.0 1.9 5.9 23.9 23,224 41.3 8.2 7.4 4.8 15.7 4.4 18.2 27,121 42.1 8.6 7.5 5.3 13.8 5.2 17.5 29,353 41.0 8.3 7.2 5.4 14.6 5.1 18.4 29,824 42.4 8.1 5.8 5.1 13.8 6.0 18.7 31,031 38.5 7.7 5.2 6.1 15.1 5.6 21.7 31,548 38.4 7.2 5.7 5.7 14.4 6.5 22.0 3,537 79.3 8.1 0.5 0.2 0.5 11.4 3,967 79.2 6.4 0.5 0.2 — 0.6 13.2 4,308 76.6 7.8 1.6 0.1 0.1 0.4 13.5 4,075 76.5 7.9 0.8 0.2 — 0.5 14.1 4,091 74.4 8.5 0.7 0.2 0.2 0.9 14.9 4,443 74.4 8.1 0.6 0.6 — 0.8 15.5 "6,201 35.8 6.1 4.1 4.4 8.3 3.5 37.8 6,101 29.4 6.0 4.9 4.8 9.0 5.0 41.0 7,050 33.3 7.2 5.3 5.2 6.4 4.3 38.2 5,192 31.2 6.5 4.4 5.5 2.5 5.0 44.9 3,342 30.8 6.5 5.1 5.3 10.4 4.6 37.4 Chemistry 1965 1967 1969 1971 1972 1973 Engineering 1965 1967 1969 1971 1972 1973 Mathematics 1965 1967 1969 1971 1972 1973 Molecular biology 1965 1967 1969 1971 1972 1973 Physics 1965 1967 1969 1971 1972 1973 Psychology 1965 1967 1969 1971 1972 1973 Systematic biology 1967 1969 1971 1972 1973 158 ' Includes articles, letters and notes from the sample of 492 scientific journals most heavily cited in 1965. ^ The social sciences are excluded because comparable data are not available ^ Prior to 1972, data for East Germany were included in the fields of chemistry, engineering, molecular biology, physics, and psychology. * For mathematics and systematic biology, these numbers are the size of the literature samp/e from which the percent distributions were derived, and should not be used as counts of articles NOTE: Percents may not add to 100 because of rounding. SOURCE Computer Honzons, Inc.. Indicators of the Quantity and Quality of the Scientific Literature. 1975 (A study commissioned specifically for this report). Table 1-7. Percent distribution of scientific literature' by selected field,' for eacfi country, 1973 Fields United United West States Kingdom Germany France U.S.S.R. Japan Other World countries total Percent In each field All fields 100.0 100.0 100.0 Clinical medicine 29.9 31.1 31.9 Biology and biomedical research 24.9 23.8 20.1 Chemistry 9.8 13.9 16.9 Physics 10.8 11.1 11.9 Earth and space sciences 5.1 3.8 2.3 Engineering 10.8 12.0 13.2 Psychology 5.1 1 .8 0.4 Mathematics 3.8 2.5 3.5 Total count of literature 109,320 25,462 16,461 Percent distribution 39.2 9.1 5.9 100.0 28.7 100.0 9.9 100.0 18.5 100.0 28.5 100.0 27.3 25.0 12.4 20.4 25.1 23.2 20.0 33.1 29.6 18.7 16.5 12.8 25.5 16.4 11.7 12.8 3.8 4.7 1.7 3.9 4.2 5.3 13.4 10.8 7.8 10.2 0.5 0.1 0.3 1.5 2.6 4.0 0.9 2.4 2.9 3.1 Total number of publications 15,184 24,435 14,309 5.4 8.8 5.1 73,723 278,894 26.4 100.0 ' Includes 278,894 articles, letters and notes from a sample of 2,121 scientific journals Because of the way in which this sample of journals was chosen, these profiles may understate certain fields; eg., Russian mathematics articles may be understated here. ^ See Appendix table 1-7a for the description of fields and subfields The social sciences are excluded because comparable data are not available NOTE Percents may not add to 100 because of rounding SOURCE Computer Horizons. Inc . Indicators of the Quantity and Quality of the Scientific Literature. 1975 (A study commissioned specifically for this report). 159 Table 1-7a. Fields and subfields of scientific literature, 1973 Clinical medicine General and internal medicine Allergy Anesthesiology Cancer Cardiovascular system Dentistry Dermatology & venereal diseases Endocrinology Fertility Gastroenterology Geriatrics Hematology Immunology Obstetrics & gynecology Neurology & neurosurgery Ophthalmology Orthopedics Arthritis & rheumatism Otorhinolaryngology Pathology Pediatrics Pharmacology Pharmacy Psychiatry Radiology & nuclear medicine Respiratory system Surgery Tropical medicine Urology Nephrology Veterinary medicine Addictive diseases Hygiene & public health Miscellaneous clinical medicine Biology and biomedical research Biomedical research Physiology Anatomy & morphology Embryology Genetics & heredity Nutrition & dietetics Biochemistry & molecular biology Biophysics Cell biology cytology & histology Microbiology Virology Parasitology Biomedical engineering Microscopy Miscellaneous biomedical research General biomedical research Biology General biology General zoology Entomology Miscellaneous zoology Marine biology & hydrobiology Botany Ecology Agriculture & food science Dairy & animal science Miscellaneous biology Chemistry Analytical chemistry Organic chemistry Inorganic & nuclear chemistry Applied chemistry General chemistry Polymers Physical chemistry Physics Chemical physics Solid state physics Fluids & plasmas Applied physics Acoustics Optics General physics Nuclear & particle physics Miscellaneous physics Earth and space science Astronomy & astrophysics Meteorology and atmospheric science Geology Earth & planetary science Geography Oceanography & limnology Engineering and technology Chemical engineering Mechanical engineering Civil engineering Electrical engineering & electronics Miscellaneous engineering & technology Industrial engineering General engineering Metals & metallurgy Materials science Nuclear technology Aerospace technology Computers Library & information science Operations research & management science Psychology Clinical psychology Personality & social psychology Developmental & child psychology Experimental psychology General psychology Miscellaneous psychology Behavioral science Mathematics Algebra Analysis & functional analysis Geometry Logic Number theory Probability Statistics Topology Computing theory & practice Applied mathematics Combinatorics & finite mathematics Physical mathematics General mathematics Miscellaneous mathematics 160 Table 1-7b. Citation indices of scientific literature' in selected fields,^ by selected countries, 1973 Citing country Field of science Cited country Clinical medicine Biology and biomedical research Chemistry Physics Earth and space science Engi- neering Psy- chology Mathe- matics United States United States World Non-U.S 1.52 1.31 1.15 1.45 1.29 1.15 1.97 1.54 1.38 1,60 1,38 1.26 1.42 1.31 1.18 1.49 1.07 0,90 1.05 1,03 0,97 1.25 1.17 1.07 United Kingdom United Kingdom ... World Non-U. K 2.26 1.26 1.13 1.91 1.17 1.09 2.11 1.40 1.30 1.37 0.93 0.87 1.59 0.97 0.92 1,75 1,04 0,86 1,87 0,93 0.86 1.80 1.08 0.99 West Germany West Germany World Non-W. Germany . . 2.31 0.55 0.41 1.82 0.79 0.73 3.62 1.46 1.29 1.54 1.03 1.00 1,72 0,71 0,68 4,03 0,87 0,58 (3\ (3) 1.85 0.91 0.83 France 2.62 0.50 0.36 2.10 0.64 0.58 2.14 0.66 0.56 1.14 0.80 0.78 1,56 0,57 0,53 10,37 1,08 0,67 /3\ /3\ 2.24 France" World 0.77 Non-France 0.73 USSR. U.S.S.R World Non-U.S.S.R 9.44 0.18 0.05 4.78 0.29 0.14 2.54 0.42 0.16 2.51 0.61 0.32 3,14 0.35 0.21 6,18 1,02 0,12 /3\ /3\ (3\ 13.25 0.53 0,31 Japan Japan World 4.66 0.57 3.62 0.82 1.48 0.69 2.14 0.75 3.75 0.73 3,58 0,76 (3\ /3\ 3,61 0.65 ' Based on 278.894 articles, letters and notes in a sample of 2.121 journals ' See Appendix table 1-7a for a description of ttie fields The social sciences are excluded because comparable data are not available. ' Because these countries had less than 2 percent of the world's literature total in psychology, reliable citation ratios cannot be calculated. ' Although French scientific journals may have severe space restrictions which discourage complete citations, the articles themselves tend to be more specific, covering less substantive material. For this reason, there may be an inflation of French publications in the scientific literature, making them more subject to citation than their significance warrants. SOURCE Computer Horizons, Inc , Indicators ol the Quantity and Quality ot the Scientilic Literature, 1975 (A study commissioned specifically for this report) 161 Table 1-8. Nobel Prizes awarded In science, for selected countries, 1901-74 Date United United States Kingdom Germany' France U.S.S.R. Num iber of Nobel Prizes per 10 million popL ilation 0.12 1.24 2.80 1.47 0.15 0.20 0.71 2.00 1.01 — 0.26 1.56 2.21 0.75 — 0.70 1.49 2.05 0.49 — 0.98 1.42 0.91 — — 1.74 1.75 0.59 — 0.20 1.35 2.23 0.91 1.05 0.13 0.63 1.08 0.17 — — Number of Nobel Prizes awarded 1 5 12 6 2 2 3 7 4 — 3 7 8 3 — 9 7 8 2 — 14 7 4 — — 29 9 3 — 4 26 12 5 5 3 13 6 1 — — 97 56 48 20 9 Population (In millions) 84.25 40.31 42.81 40.79 133.06 99.44 42.58 35.00 39.43 147.89 114.77 44.79 36.25 39.95 167.15 127.84 47.05 39.05 41.23 187.00 142.43 49.42 44.22 41,52 187.50 166.47 51.56 50.54 43.71 197.20 192.77 53.80 54.91 47.59 230.05 207.64 55.52 59.29 50.71 247.85 1901-1910 1911-1920 1921-1930 1931-1940 1941-1950 1951-1960 1961-1970 1971-1974 1901-1910 1911-1920 1921-1930 1931-1940 1941-1950 1951-1960 1961-1970 1971-1974 Total ... 1901-1910 1911-1920 1921-1930 1931-1940 1941-1950 1951-1960 1961-1970 1971-1974 ' Before 1946, includes East Germany SOURCE: The Nobel Foundation. Les Pnx Nobel, annual series. 162 Table 1-9. Nobel Prizes awarded by field for selected countries, 1901-1974 Date Total United States France Germany' U.S.S.R. United Kingdom Other 1901-1910 1911-1920 1921-1930 1931-1940 1941-1950 1951-1960 1961-1970 1971-1974 Total 1901-1910 1911-1920 1921-1930 1931-1940 1941-1950 1951-1960 1961-1970 1971-1974 Total 1901-1910 1911-1920 1921-1930 1931-1940 1941-1950 1951-1960 1961-1970 1971-1974 Total Number of prizes in physics 14 1 4 3 2 4 10 — — 4 — 3 3 12 2 2 3 — 2 3 10 3 — 1 — 3 3 8 2 — 1 — 3 2 20 12 — 1 3 2 2 18 9 2 2 3 — 2 9 5 — — — 4 — 101 34 8 15 6 19 19 Number of p irizes in chemistry 10 1 5 — 2 2 8 1 3 3 — — 1 10 — — 4 — 3 3 12 2 2 4 — 1 3 11 4 — 3 — 1 3 13 5 — 1 1 5 1 15 4 — 2 — 6 3 7 4 — 1 — 1 1 86 20 6 23 1 19 17 Number of prizes in physiology/medicine 12 — 1 4 2 1 4 6 1 1 — — — 4 11 1 1 1 — 2 6 13 4 — 3 — 3 3 17 8 — — — 3 6 19 12 — 1 — 2 4 26 13 3 1 — 6 3 9 4 — — — 1 4 113 43 6 10 2 18 34 ' Before 1946. includes East Germany SOURCE. The Nobel Foundation, Les Prix Nobel, annual series. 163 Table 1-10. Patents granted to U.S. nationals by foreign countries' and to foreign nationals' by tfie United States, 1966-73 Patents granted 1966 1967 1968 1969 1970 1971 1972 1973 U. S. balance 36,066 34,441 36,045 35,887 33,697 31 ,445 30,520 25,306 Patents granted to U.S. nationals by foreign countries 45,633 44,350 45,168 47,825 45,918 47,311 47,359 41,186 Patents granted to foreign nationals by tfie United States 9,567 9,909 9,123 11,938 12,221 15,866 16,839 15,880 ^ Including Canada, West Germany, Japan. United Kingdom, US S R , Belgium, Denmark. Ireland. Luxembourg, and the Nettierlands. SOURCE: World Intellectual Properly Organization, Industrial Properly. Geneva 1966-73 (December issues). Table 1-11. U.S. patent balance with selected countries, 1966-73 Selected country 1966 1967 Canada: Balance 15,676 16,592 Granted to U.S 16,614 17,583 Granted by U.S 938 991 West Germany: Balance -248 -360 Granted to U.S 3,733 3,406 Granted by US 3,981 3,766 Japan: Balance 3,561 2,008 Granted to U.S 4,683 3,432 Granted by U.S 1,122 1,424 United Kingdom: Balance 11,440 10,877 Granted to US 14,117 13,676 Granted by US 2,677 2,799 Otfier E.E.C. countries' Balance 5,700 5,439 Granted to U.S 6,483 6,253 Granted by U.S 783 814 U.S.S.R : Balance -63 -115 Granted to U.S 3 Granted by US 66 115 1968 1969 1970 1971 1972 1973 16,686 17,583 897 18,153 19,147 994 17,598 18,663 1,065 16,665 17,992 1,327 16,045 17,289 1,244 11,619 12,964 1,345 362 3,804 3,442 -40 4,483 4,523 -1,552 2,882 4,434 -1,128 4,393 5,521 -1,153 4,575 5,728 -639 4,949 5,588 3,439 4,903 1,464 2,505 4,657 2,152 2,149 4,774 2,625 1,667 5,700 4,033 794 5,948 5,154 546 5,485 4,939 10,107 12,588 2,481 9,503 12,678 3,175 9,776 12,728 2,952 9,226 12,682 3,456 9,837 13,001 3,164 8,866 11,717 2,851 5,481 6,225 744 5,842 6,777 935 5,743 6,670 927 5,143 6,346 1,203 5,093 6,287 1,194 4,914 6,071 1,157 -30 65 95 -76 83 159 -17 201 218 -128 198 326 -96 259 355 -177 205 382 ' Other European Economic Community (EEC) countries inclucje Belgium, Denmark, IrelancJ, Luxembourg, and the Netherlands Data from France are not reliable tor use in this indicator SOURCE World Intellectual Property Organization. Industrial Property. Geneva 1966-73 (December Issues). 164 Table 1-12. Major technological innovations, by selected countries, 1953-73 United United West Period States Kingdom Germany Japan France Total Percentage of total 1953-55 1956-58 1959-61 1962-64 1965-67 1968-70 1971-73 1953-55 1956-58 1959-61 1962-64 1965-67 1968-70 1971-73 NOTE Detail may not add to totals because o1 rounding SOURCE: Gellman Research Associates. Inc., Indicators ol International Trends in Technological Innovation, 1975 (A study commissioned specifically for ttiis report). 75 14 6 5 100 82 9 5 5 100 68 21 2 2 7 100 66 17 5 12 100 55 23 12 8 3 100 57 19 8 13 4 100 58 16 9 10 8 100 Number of innovations 63 12 5 4 84 36 4 2 2 44 38 12 1 1 4 56 55 14 4 10 83 36 15 8 5 2 66 45 15 6 10 3 79 46 13 7 8 6 80 Table 1-13. Mean time in years between invention and innovation, for selected countries, 1953-73' Period United States Japan West Germany France United Kingdom 1953-62 8.4 3.6 5.5 5.6 7.5 7.3 5.1 1963-73 64 7.5 ' Refers to the date of the innovation. ^ Sample size does not allow calculation of the time interval SOURCE Gellman Research Associates, Inc.. Indicators of International Trends in Technological Innovation. 1975 (A study commissioned specifically for this report). 165 Table 1-14. "Radicalness" of innovations by selected countries, 1953-73 Country Imi exist provement of ing technology Major technological advance Radical breakthrough Pe rcentag( i of each country's innovations United States United Kingdom . . France 41 4 12 36 38 31 40 65 50 54 27 56 24 West Germany , . . . Japan 14 8 N lumber of innovations United States United Kingdom . . France 98 2 2 8 10 74 18 11 11 14 65 25 4 West Germany . . . . Japan 3 2 NOTE Detail may not add to 100 percent because ol rounding SOURCE Gellman Research Associates, Inc . Indicators of International Trends in Technological Innovation. 1975 (A study commissioned specifically for this report). 166 Table 1-15. U.S. receipts and payments for patents, manufacturing rights, licenses, etc., by country, 1960-74' [Dollars in millions] Year 1960 Balance Receipts Payments . . . . 1961 Balance Receipts Payments . . . . 1962 Balance Receipts Payments . . . . 1963 Balance Receipts Payments . . . . 1964 Balance Receipts Payments . . . . 1965 Balance Receipts Payments . . . . 1966 Balance Receipts Payments . . . . 1967 Balance Receipts Payments . . . . 1968 Balance Receipts Payments . .. . 1969 Balance Receipts Payments . . . . 1970 Balance Receipts Payments . . . . 1971 Balance Receipts Payments . .. . 1972 Balance Receipts Payments .... 1973 Balance Receipts Payments .... 1974 (preliminary) Balance Receipts Payments .... "otal Europe $210 248 38 $105 140 35 201 244 43 94 132 38 212 256 44 95 133 38 222 273 51 98 144 46 241 301 60 106 162 56 268 335 67 128 189 61 277 353 76 119 186 67 289 393 104 97 190 93 331 437 106 102 196 94 366 486 120 115 222 107 459 573 114 148 247 99 495 618 123 158 268 110 516 655 139 150 270 120 549 725 176 160 306 146 601 781 180 200 348 148 Japan Developing nations Other $48 48 50 52 2 51 53 2 57 58 1 65 66 1 65 66 1 67 70 3 91 95 4 126 130 4 151 155 4 198 202 4 219 223 4 234 240 6 261 274 13 241 249 8 $25 26 1 25 26 1 29 30 1 30 31 1 33 34 1 35 37 2 46 50 4 48 51 3 59 63 4 56 61 5 61 68 7 67 71 4 74 80 6 71 81 10 91 107 16 $31 33 2 32 34 2 38 40 2 37 39 2 35 38 3 40 43 3 46 48 2 54 57 3 45 49 4 44 49 5 52 56 4 51 56 5 58 65 7 58 65 7 69 77 8 ' Represents U S receipts and payments arising out of agreements by U S residents with residents or governments of foreign countries to sell or buy outrigtit or provide or be provided with the use of intangible assets or rights such as patents, techniques, processes, formulae, designs, trademarl^s. copyrights, franchises, manufacturing rights, and other similar intangible property or rights Excludes fees and royalties connected with U.S. and foreign direct investments and excludes film rentals. NOTE Detail may not add to totals because of rounding. SOURCE: Department of Commerce. Bureau of Economic Analysis. Survey oi Current Business, June 1975 167 Table 1-16. Real Gross Domestic Product per employed civilian, for selected countries compared witli the IJnited States, 1960-74 (Indexes, United States = 100) United West United Year States France Germany Japan Kingdom 1960 1965 1967 1970 1971 1972 1973 1974 SOURCE U S Department of Labor, Bureau of Labor Statistics. Office of Productivity and Technology. Comparative Real Gross Domestic Product. Real GDP per Capita, and Real GDP per Employed Civilian for Six Countries, July 1975. 100 55.1 52.0 24.4 50.7 100 60.2 55.7 31.7 48.6 100 62.9 56.4 36.3 49.3 100 71.4 67.0 48.7 52.6 100 72.9 67.0 50.4 53.5 100 74.1 67.6 53.3 53.3 100 75.7 69.2 55.9 53.4 100 81.1 73.8 57.4 55.6 Table 1-17. Productivity' in manufacturing industries, by selected countries, 1960-74 (Index, 1960= 100) Year United States Japan France West Germany United Kingdom 1 960 1 00 100.0 113.1 118.1 127.6 144.6 150.7 165.9 190.5 214.5 247.6 279.0 289.0 312.2 368.8 380.8 100.0 104.7 109.5 116.0 121.8 1288 137.8 145.6 162.2 168.0 176.4 185.6 198.1 209.6 221.1 100.0 105.4 112.2 118.1 127.3 136.1 141.6 150.6 162.0 171.4 175.8 184.2 195.9 209.9 215.1 100.0 1961 102.5 100.8 1962 108.3 103.3 1963 112.7 108.9 1964 118.0 116.8 1965 122.7 120.3 1966 124.3 124.6 1967 124.2 130.2 1968 130.2 138.9 1 969 133.3 140 8 1970 134.0 142.1 1971 143.1 148.7 1972 151.2 154.8 1973 159.4 165.6 1974 (est) 160.5 165.8 ' Output per man-hour SOURCE P Capdevielle and A Neef. "Productivity and Unit Labor Costs in the United States and Abroad". Monthly Labor Review. July 1975 168 Table 1-18. Unit labor cost' in manufacturing industries, by selected countries, 1960-74 (Index, 1960 = 100) Year United States Japan France West Germany United Kingdom I960 1000 100.0 1028 112.5 116.3 115.3 124.7 124,8 121.8 125.7 128.9 135.8 151.8 162.6 171.3 220.1 100.0 105.2 110.7 115.5 118.2 120.4 119.6 122.7 124.9 127.5 136.1 144.8 151.7 162.8 180.7 100.0 105.9 112.6 114.2 114.2 117.3 123.0 122.4 120.6 124.2 139.5 151.7 159.2 168.7 187.5 100.0 1961 100.6 106.9 1 962 99.1 109.8 1 963 . . 98 4 108.9 1 964 98.2 108.8 1 965 97.0 115.6 1 966 100.0 121.0 1967 105 119.0 1968 107.5 119.5 1969 111.7 127.4 1970 118.9 144.9 1971 118.9 158.0 1972 1973 118.8 120 6 171.5 181.7 1974 (est.) 131.2 216.9 ' In national currency unadjusted for inflation. SOURCE P Capdevlelle and A Neef, ■Productivity and Unit Labor Costs in tfie United States and Abroad". Monthly Labor Review. July 1975. Table 1-19. U.S. trade balance in R&D-intensive and non-R&D-lntensive manufactured products. 1960-74 (Dollars in millions) 1974 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 R&D-intensive Balance 5,891 6,237 6,720 6,958 7,970 8,177 8,020 8.817 9,755 10,471 11,722 11,727 11,012 15,101 23,612 Export 7,597 8.018 8,715 8,975 10,267 11,107 12,203 13,407 15,312 16,955 19,274 20,228 22.003 29,088 41,115 Import 1,706 1,781 1,995 2.017 2,297 2,930 4183 4,590 5,537 6,484 7,552 8,501 10,991 13,987 17,503 Non R&D-intensive Balance -179 -12 -691 -765 -678-2,027-3,325-3,729-6,581 -6,698 -8,285 -11,698 -15,039 -15,370 -16,296 Export 4,962 4730 4,940 5,284 6,121 6,281 6,913 7,437 8,506 9,830 10,069 10,215 11,737 15,643 22412 Import 5,141 4,742 5,631 6,049 6,799 8,308 10,238 11,166 15,087 16,528 18.354 21,913 26,776 31,013 38,708 SOURCE U.S. Department of Commerce, Domestic and International Business Administration, Oi/ersess Business Reports, April 1975 and April 1972 169 Table i-20. U.S. trade balance in R&D-lntensive manufactured products, by product group, 1960-74 (Dollars in millions) Product 1960 Chemicals Balance 955 Export 1,776 Import 821 Machinery, nonelectrical Balance 2,948 Export 3,386 Import 438 Electncal machinery Balance 804 Export 1,090 Import 286 Aircraft Balance 970 Export 1,024 Import 54 Professional and scientific instruments Balance 214 Export 321 Import 107 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1,051 1,789 738 1,104 1,876 772 1,294 2.009 715 1,662 2,364 702 1.633 2.402 769 1,719 2,676 957 1,844 2,802 958 2,158 3,287 1,129 2,155 3,383 1,228 2,376 3,826 1,450 2,224 3,836 1.612 2,118 4,133 2,015 3,286 5,749 2,463 4,831 8,822 3,991 3,288 3,743 455 3.547 4,087 540 3,574 4,209 635 3,989 4,860 871 4,114 5.274 1.160 4,102 5,779 1,677 4,218 6,181 1,963 4,280 6,560 2,280 4,838 7,460 2,622 5,583 8,686 3,103 5.268 8,772 3,504 5,325 9,864 4,539 6,904 12,556 5,652 10,633 17,299 6,666 891 1,225 334 946 1,361 415 1,074 1,493 419 1,222 1.665 443 1,020 1,660 640 883 1,899 1,016 962 2,098 1,136 792 2,284 1,492 729 2,677 1,948 728 2,999 2,271 512 3,067 2,555 321 3,698 3,377 533 5,032 4,499 1,602 7,019 5,417 766 903 137 857 980 123 726 817 91 791 874 83 997 1,137 140 828 1,101 273 1,271 1,519 248 2,015 2,309 294 2,140 2,423 283 2.382 2,656 274 3,049 3,387 338 2,580 2,995 415 3,556 4,119 563 5,256 5,766 510 241 358 117 266 411 145 290 447 157 306 504 198 413 634 221 488 748 260 522 807 285 530 872 342 609 1,012 403 653 1,107 454 674 1,166 492 668 1.313 645 822 1.632 810 1,290 2,209 919 SOURCL U S Department of Commerce. Domestic and International Business Administration. Overseas Business Reports. April 1975 and April 1972. Table 1-21. U.S. trade balance with selected nations In R&D-lntensive manufactured products, 1966-73 (Dollars in millions) Nations 1966 Developing nations Balance 4,053 Export 4,316 Import 263 Western Europe Balance 1,890 Export 3.865 Import 1,975 Canada Balance 1,800 Export 2,838 Import 1,038 Japan Balance -133 Export 661 Import 794 1967 1968 1969 1970 1971 1972 1973 4,033 4,430 4,445 4,928 5,087 5,277 6,675 4,332 4.822 5,002 5,679 5,996 6,765 8,968 299 392 547 751 909 1,488 2,293 2,283 2,566 2,986 3,942 3,599 3,089 4,165 4,359 5,020 5,655 6,927 6.861 7,345 9,597 2,076 2.454 2,669 2,985 3,262 4,256 5,432 1,760 1,719 1,914 1,684 1,865 2,333 3,011 2,983 3,142 3,478 3,513 3,914 4,678 5,741 1,223 1,423 1,564 1,829 2,049 2,345 2,730 -115 -200 -324 -224 -516 -971 -839 772 930 1,180 1,536 1.520 1,639 2,216 887 1,130 1,504 1,760 2,036 2,610 3,055 SOURCE: U.S. Department of Commerce, Domestic and International Business Administration, Overseas Business Reports, June 1974 and August 1973. 170 Table 2-1. National R&D expenditures, 1960-74 [Dollars In billions] Year 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 (est.) Current dollars Constant dollars' $13.6 $15.4 14.3 16.1 15.4 17.1 17.1 18.8 18.9 20.4 20.1 21.3 21.9 22.6 23.2 23.2 24.7 23.7 26.7 23.6 26.0 22.6 25.7 22.2 28.4 22.9 30.4 23.2 32.0 22.1 ^ GNP Implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE: National Science Foundation, National Patterns of R&D Resources. (953-75 (NSF 75-307) Table 2-2. Scientists and engineers' employed in R&D, by sector, 1961-74 [In thousands] Yearly average Total 1961 425.7 1965 494.1 1968 550.4 1969 558.2 1970 549.5 1971 529.7 1972 521.5 1973 523.1 1974 (est.) ... 527.8 Federal Universities Government Industry and colleges FFRDC's^ Other nonprofit institutions 51.1 312.0 42.4 9.1 11.1 61.8 348.4 53.4 11.1 19.4 68.1 381.9 66.0 11.2 23.2 69.9 385.6 6C.3 11.6 22.8 69.8 375.4 68.5 11.5 24.3 66.5 358.3 68.4 11.5 25.0 65.2 352.6 66.5 12.0 25.2 62.3 359.2 64.6 12.4 24.6 65.0 359.5 66.8 12.1 24.4 ' Full-lime-equivalent basis, excluding those employed in State and local agencies, calculated as the yearly average. Graduate students are Included ^ Federally Funded Research and Development Centers administered by universities. SOURCE: National Science Foundation. National Patterns of R&D Resources. 1953-75 (NSF 75-307). 171 Table 2-3. National R&D expenditures as a percent of GNP by source, 1960-74 [Current dollars in billions) Year 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 (est.) Gross National Product ^GNP) All sources Federal sources All other sources Total R&D R&D as a percent of GNP Total R&D R&D as a percent of GNP Total R&D R&D as a percent of GNP $503.7 520 1 560.3 590.5 632.4 684.9 749.9 793 9 864.2 930.3 977.1 1,054.9 1,158.0 1,294.9 1,396.7 513.6 14.3 15.4 17.1 18.9 20.1 21.9 23.2 24.7 25.7 26.0 26.7 28.4 30.4 32.0 2.70 2.75 2.75 2.90 299 2.93 2.92 2.92 2.86 2.76 2.66 2.53 2.45 2.35 2.29 9.3 9.9 11.2 12.6 13.0 14.0 14.4 15.0 14.9 14.8 15.0 15.9 16.5 17.0 1.75 1.79 1.77 1.90 1.99 1.90 1.87 1.81 1.74 1.60 1.51 1.42 1.37 1.27 1.22 $4.8 5.1 5.5 5.9 6.3 7.1 7.9 8.8 9.7 10.8 11.3 11.8 12.5 14.0 15.1 0.95 0.98 0.98 1.00 1.00 1.04 1.05 1.11 1.12 1.16 1.16 1.12 1.08 1.08 1.08 NOTE: Detail may not add to totals because ot rounding SOURCE National Science Foundation. National Patterns ot R&D Resources. 1953-75 (NSF 75-307) and The Budget ol the United States Government, Fiscal Year 1976. 172 Table 2-4. National expenditures for R&D by source, 1960-74 [Dollars in millions] Other Federal Universities nonprofit Year Total government Industry and colleges institutions Current dollars 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 (est.) 1960 $15,427 1961 16,124 1962 17,147 1963 18,756 1964 20,410 1965 21,310 1966 22,594 1967 23,205 1968 23,717 1969 23,560 1970 22,648 1971 22,249 1972 22,857 1973 23,186 1974 (est.) 22,143 $13,551 $8,752 $4,508 $149 $142 14,346 9,264 4,749 165 168 15,426 9,926 5,114 185 201 17,093 11,219 5,449 207 218 18,894 12,553 5,880 235 226 20,091 13,033 6,539 267 252 21,894 13,990 7,317 303 284 23,205 14,420 8,134 345 306 24,669 14,952 8,997 391 329 25,686 14,914 9,998 420 354 26,047 14,764 10,434 461 388 26,745 14,982 10,817 529 417 28,402 15,875 11,508 576 443 30,427 16,472 12,880 604 471 32,045 16,955 13,916 683 491 Constant 1967 dolla rs' $9,964 10,412 11,034 12,310 13,561 13,824 14,438 14,420 14,376 13,680 12,837 12,463 12,776 12,552 11,716 $5,132 5,338 5,685 5,979 6,352 6,936 7,551 8,134 8,650 9,171 9.072 8.998 9,261 9.815 9,616 $170 185 206 227 254 283 313 345 376 385 401 440 464 460 472 $162 189 223 239 244 267 293 306 316 325 337 347 357 359 339 ^ GNP implicit price deflators used to convert current dollars to constant 1967 dollars. NOTE Detail may not add to totals because of rounding. SOURCE National Science Foundation, National Patterns ol R&D Resources (NSF 75-307). 173 Table 2-5. National expenditures for R&D by performer, 1960-74 [Dollars in millions] Year Total 1 Federal government Industry Universities and colleges FFRDC's' Other nonprofit institutions Current dollars 1 960 .. $13,551 $1,726 1,874 2,098 2,279 2,838 3,093 3,220 3,396 3,493 3,503 3,855 4,156 4,482 4,619 4,900 $10,509 10,908 11,464 12,630 13,512 14,185 15,548 16,385 17,429 18,308 18,062 18,311 19,371 20,937 22,020 $646 763 904 1,081 1,275 1.474 1,715 1,921 2,149 2,220 2,335 2.500 2.675 2.934 3,008 $360 410 470 530 629 629 630 673 719 725 737 716 764 817 865 $310 1961 14,346 391 1 962 .. 15.426 490 1 963 .. 17,093 573 1 964 . . , . .. 18,894 640 1 965 .. 20,091 710 1966 .. 21,894 781 1967 . . 23,205 830 1 968 . . 24,669 879 1 969 . . 25,686 930 1970 . . 26,047 1.058 1971 . . 26,745 1,062 1972 . . 28,402 1,110 1973 . . 30,427 1,120 1974 (est.) .... . . 32,045 1,252 Constant 1967 dollars^ I960 .. $15,427 $1,965 2,106 2,332 2,501 3,066 3,281 3,323 3,396 3,358 3,213 3,352 3,457 3,607 3,520 3,386 $11,964 12,260 12,743 13,858 14,597 15,046 16,045 16,385 16,757 16,793 15,705 15,233 15,589 15,954 15,216 $735 858 1,005 1,186 1,377 1,563 1,770 1,921 2,066 2,036 2,030 2,080 2,153 2.236 2.078 $410 461 522 582 679 667 650 673 691 665 641 596 615 623 598 $353 1961 16 124 439 1 962 .. 17,147 545 1 963 . . . . 18 756 629 1 964 .. 20,410 691 1 965 .. 21,310 753 1966 . . 22,594 806 1967 . . 23,205 830 1968 .. 23,717 845 1 969 . . 23,560 853 1 970 . . 22,648 920 1971 , .. 22,249 883 1972 , . . 22 857 893 1973 , .. 23,186 853 1974 (est) .... , ., 22,143 865 ' Federally Funded Research and Development Centers administered by universities. ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. NOTE Detail may not add to totals because of rounding SOURCE National Science Foundation. National Patterns ot RSD Resources (NSF 75-307)- 174 Table 2-6. National R&D expenditures, by ctiaracter of work, 1960-74 [Dollars in millions] Current dollars Constant 1967 dollars' Basic Year research 1960 $1,183 1961 1.378 1962 1,695 1963 1,974 1964 2.301 1965 2,572 1966 2,825 1967 3,029 1968 3,286 1969 3,378 1970 3,548 1971 3.544 1972 3.705 1973 3,800 1974 (est.) 3,991 Applied research Develop- ment Basic research Applied research Develop- ment $3,057 $9,311 $1,347 $3,480 $10,600 3,115 9.853 1,549 3.501 11,075 3,727 10.004 1,884 4.143 11,120 3.825 11.294 2,166 4,197 12.392 4.238 12.355 2,486 4.578 13.347 4.470 13.049 2-728 4,741 13,841 4.747 14.322 2,915 4,899 14,780 4.968 15,208 3,029 4.968 15.208 5.356 16,027 3,159 5.150 15.409 5,533 16,775 3,099 5.075 15.387 5,892 16.607 3.085 5.123 14,440 6,047 17.154 2.948 5.030 14,270 6.272 18.425 2.982 5,047 14,828 6,839 19.788 2.896 5,211 15,079 7,460 20.594 2,758 5.155 14,230 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE: National Science Foundation, National Patterns ol R&D Resources. 1953-75 (NSF 75-307). 175 Table 2-7a. Basic research expenditures by source, 1960-74 [Dollars in millions] Year Total Federal Government U Industry Iniversities and colleges Other nonprofit institutions Current dollars I960 , .. $1,183 $693 841 1.091 1.310 1.595 1.817 1.986 2.173 2.327 2.386 2.469 2.379 2,528 2,605 2,724 $331 350 382 414 424 448 496 477 518 519 536 556 528 561 594 $72 85 102 121 144 164 196 223 276 298 350 400 428 416 434 $87 1961 1.378 102 1962 1 .695 120 1 963 1.974 129 1 964 2,301 138 1 965 2.572 143 1 966 2.825 147 1 967 3,029 156 1 968 3,286 165 1969 3.378 175 1970 3.548 193 1971 3,544 209 1972 3,705 221 1973 3,800 218 1974 (est.) 3,991 239 Constant 1967 dollars' 1 960 . .. $1,347 $789 945 1,213 1,437 1,723 1.927 2.050 2.173 2.237 2,189 2,147 1,979 2,034 1,985 1.882 $377 393 425 454 458 475 512 477 498 476 466 463 425 427 410 $82 96 113 133 156 174 202 223 265 273 304 333 344 317 300 $99 1961 1,549 115 1962 1,884 133 1963 2,166 142 1 964 2.486 149 1965 2.728 152 1 966 2.916 152 1967 3,029 156 1 968 3.159 159 1969 3.099 161 1 970 3,085 168 1971 2.949 174 1972 2.981 178 1973 2.895 166 1974 (est ) 2.757 165 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE. National Science Foundation, National Patterns ol R&D Resources: 1953-75 (NSF 75-307) 176 Table 2-7b. Applied research expenditures by source, 1960-74 [Dollars in millions] Universities Federal and Year Total Government Industry colleges Current dollars 1960 $3,057 $1,725 $1,228 $66 1961 3,115 1,804 1.197 69 1962 3.727 2,127 1.473 70 1963 3.825 2,205 1,487 72 1964 4,238 2,503 1,596 77 1965 4,470 2,653 1,658 88 1966 4,747 2,729 1,844 89 1967 4,968 2,874 1,895 102 1968 5,356 3,020 2,132 97 1969 5,533 2,982 2,327 105 1970 5,892 3,258 2,406 98 1971 6,047 3,313 2,476 115 1972 6,272 3,387 2,601 132 1973 6,839 3,670 2,835 158 1974(est)i 7,460 3,992 3,080 214 Constant 1967 dollars' 1960 $3,480 $1,964 $1,398 $75 1961 3,502 2,028 1,345 78 1962 4,142 2,364 1,637 78 1963 4,197 2,419 1,632 79 1964 4,578 2,704 1,724 83 1965 4,741 2,814 1,759 93 1966 4,899 2,816 1,903 92 1967 4,968 2,874 1,895 102 1968 5,150 2.904 2,050 93 1969 5,074 2,735 2,134 96 1970 5,123 2,833 2,092 85 1971 5,031 2,756 2,060 96 1972 5,047 2,726 2,093 106 1973 5,211 2,797 2,160 120 1974 (est.) 5,154 2,758 2,128 148 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE National Science Foundation, National Patterns ol RiD Resources. (NSF 75-307). Other nonprofit institutions $38 45 57 61 62 71 85 97 107 119 130 143 152 176 174 $43 51 63 67 67 75 88 97 103 109 113 119 122 134 120 177 Table 2-7c. Development expenditures by source, 1960-74 [Dollars in millions] Year Total Federal Government Industry Universities and colleges Other nonprofit institutions Current dollars 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 (est.) 1960 $10,600 1961 11,075 1962 11,121 1963 12,392 1964 13,347 1965 13,841 1966 14,781 1967 15,208 1968 15,409 1969 15,387 1970 14,440 1971 14,270 1972 14,827 1973 15,079 1974 (est.) 14,230 $9,311 $6,334 $2,949 $11 $17 9,853 6,619 3,202 11 21 10,004 6,708 3,259 13 24 11,294 7,704 3,548 14 28 12,355 8,455 3,860 14 26 13,049 8,563 4,433 15 38 14,322 9,275 4,977 18 52 15,208 9,373 5,762 20 53 16,027 9,606 6,347 17 57 16,775 9,546 7,152 17 60 16,607 9,037 7,492 13 65 17,154 9,290 7,785 14 65 18,425 9,960 8,379 16 70 19,788 10,197 9,484 30 77 20,594 10,239 10,242 35 78 Constant 1967 dollars' $7,211 7,440 7,457 8,453 9,134 9,083 9,572 9,373 9,236 8,756 7,858 7,728 8,015 7,770 7,075 $3,357 3,599 3,623 3,893 4,170 4,702 5,136 5,762 6,102 6,560 6,514 6,476 6.743 7,227 7,077 $13 12 14 15 15 16 19 20 16 16 11 12 13 23 24 $19 24 27 31 28 40 54 53 55 55 57 54 56 59 54 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE National Science Foundation. National Patterns ol R&D Resources. 1953-75 (NSF 75-307). 178 Table 2-8. Federal expenditures' for research, development and R&D plant, as a percent of total Federal outlays, and as a percent of tfie relatively controllable portion of the Federal outlays, 1960-74 Year Federal R&D expenditures^ Total Federal outlays^ Expenditures as a percent of total Federal outlays Expenditures as a percent of controllable outlays 1960 $ 7.7 9.3 10.4 12.0 14.7 14.9 16.0 16.9 17.0 16.4 15.7 16.0 16.7 17.5 18.6 $ 92.2 97.8 106.8 111.3 118.6 118.4 134.7 158.3 178.8 184.5 196.6 211.4 231.9 246.5 274.7 8.4 9.5 9.7 10.8 12.4 12.6 11.9 10.7 9.5 8.9 8.0 7.6 7.2 7.1 6.8 NA 1961 NA 1962 NA 1963 NA 1964 NA 1965 NA 1966 NA 1967 16.4 1968 14.7 1969 14.6 1970 13.7 1971 14.0 1972 13.9 1973 15.1 1974 (Rst ) 14.8 ' Reported by Federal agencies- ' tn billions of current dollars NOTE: NA = not available. SOURCE National Science Foundation, Federal Funds lor Research. Development, and Other Scienlillc Activities. Vol. XXlll (NSF 74-320) and earlier volumes Table 2-9. Federal obligations for R&D, by major function, 1969-74 [Dollars in millions] 1974 Function 1969 1970 1971 1972 1973 (est.) Current dollars Total $15,641 $15,340 $15,564 $16,512 $16,821 $17,743 National defense 8,354 7,976 8,106 8,898 8,998 9,180 Space 3,732 3,510 2,893 2,714 2,601 2,510 Total civilian R&D 3,556 3,855 4,564 4,900 5,222 6,055 Constant 1967 dollars' Total $14,347 $13,338 $12,947 $13,288 $12,818 $12,260 National defense 7,663 6,935 6,743 7,161 6,857 6,343 Space 3,423 3,052 2,407 2,184 1,982 1.734 Total civilian R&D 3,262 3,352 3,797 3,943 3,979 4.184 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars NOTE. Detail may not add to totals because of rounding SOURCE: National Science Foundation, An Analysis ol Federal R&D Funding by Function, 1969-75 (NSF 74-313). 179 Table 2-10. Federal obligations for R&D by function, 1969 and 1974 [Dollars in millions] Function 1969 1970 1971 1972 1973 1974 (est.) National defense Space Health Environment Transportation and communication 461 Science and technology base Natural resources Energy development and conversion Education Income security & social services Area & community development & housing Economic grow/th & productivity Crime prevention & control International cooperation & development National defense $7,663 Space Health Environment Transportation and communication 423 Science and technology base 475 Natural resources 378 Energy development and conversion 301 Education 145 Income security & social services 85 Area & community development & housing ... 45 Economic grow^th & productivity 67 Crime prevention & control 5 International cooperation & development 25 Current dollars $8,354 $7,976 $8,106 $8,898 $8,998 $9,180 3,732 3,510 2,893 2,714 2,601 2,510 1,111 1,111 1,319 1,564 1,592 2,085 321 359 475 547 678 738 593 782 617 625 516 651 497 476 689 518 529 531 606 610 648 412 462 553 625 618 633 328 317 324 383 442 574 158 151 198 208 231 227 93 102 123 115 151 131 49 91 108 102 117 127 73 99 109 78 90 117 5 9 10 25 35 52 27 32 32 30 33 34 Constant 1967 dollars' $7,663 $6,935 $6,743 $7,161 $6,857 $6,343 3,423 3,052 2,407 2,184 1,982 1,734 1,019 966 1,097 1,259 1,213 1,441 294 312 395 440 517 510 476 460 442 488 465 448 402 460 503 471 437 276 270 308 337 397 131 165 167 176 157 89 102 93 115 91 79 90 82 89 88 86 91 63 69 81 8 8 20 27 36 28 27 24 25 23 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE: National Science Foundation. An Analysis ol Federal flSD Funding by Function, 1969-75 (NSF 74-313) 180 Table 2-11. Federal obligations for civilian R&D, by character ot work, 1970 and 1974 (Dollars in millions] Table 2-13. NSF obligations lor permanent laboratory equipment, 1966-74 [Dollars in millions] Character 1970 1974 (est.) Current dollars Total R&D $3,845.6 $6,040.9 Basic research 1,200.1 1,636.8 Applied research 1,671.4 2,733.5 Development 974.1 1,670.6 Constant 1967 dollars' Total R&D $3,343.8 $4,174,2 Basic research 1,043.5 1,131.0 Applied research 1,453.3 1,888.8 Development 847.0 1,154.4 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE National Science Foundation, special tabulations. Table 2-12. Proportion of NSF and NIH' research project grant funds allocated for permanent laboratory equipment, fiscal years 1966-74 [Percent] Year NSF NIH 1966 11.2 11.7 1967 86 11.8 1968 7.5 9.5 1969 7.0 7.5 1970 6.1 5.9 1971 6.3 6.2 1972 5.6 6.6 1973 5.5 4.9 1974 5.4 5.7 ' Includes the National Cancer Institute, the National Institute of General Medical Sciences, and the National Heart and Lung Institute, SOURCE: National Science Foundation. Databook, annual series, and National Institutes of Health, unpublished data Year Current dollars Constant 1967 dollars' 1966 1967 1968 1969 1970 1971 1972 1973 1974 $17.6 $18.2 14.4 14,4 12.8 12.3 12.3 11.3 9.9 8.6 11.0 9.2 13.6 10.9 14.7 11.2 15.2 10.5 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE National Science Foundation, Databook, annual series. Table 2-14. Federal expenditures for R&D plant, 1960-74 [Dollars in millions] Year 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 (est.) .... Current Constant 1967 dollars dollars' $528.3 $596.9 548.5 609,4 779.1 857,1 1,168.3 1,269,9 1,098.5 1,177,4 1,077,4 1,134,1 1,047,8 1,081,3 786.1 786,1 715,9 691.7 6522 603.9 578,9 508.3 612.7 511.4 564,4 454,4 638.0 494,6 894,1 640,9 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars. SOURCE: National Science Foundation, Federal Funds for Research. Development and Other Scientific Activities, Vol XXIII (NSF 74-320-A) and earlier volumes 181 Table 2-15. Federal obligations for R&D plant, by performer. 1962-74 [Dollars in millionsl Performer 1974 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 (est.) Current dollars Total 777.6 1,186.0 1,177.5 1,131.6 8583 620.1 6038 669.0 524.4 611.2 602.1 774.3 971.9 Federal intramural 619.9 974.4 974.5 913.0 629,0 239.0 294.2 260.4 166.0 200.0 246.6 323.8 409.5 Industry NA NA NA NA NA NA 81.7 141.7 102.3 167.4 142.4 221.8 339.1 Universities & colleges NA NA 97.5 141.6 162.9 111,7 98.1 61.9 56.1 49.2 45.3 42.6 49.2 FFRDC's (administered by universities) NA NA 36.0 50.2 311 138.8 101.7 176.6 169.0 178,7 130.4 162.3 134.1 Nonprofit institutions NA NA NA NA NA NA 20.9 25.8 28.8 5.8 30.0 18.8 36.2 Constant 1967 dollars' Total 864.4 1,301.3 1,272.0 1,200.3 885.8 620.1 580.5 613.6 456.0 508.4 484.5 590.0 6716 Federal intramural 689.1 1,069.1 1,052,7 968,4 649,1 239,0 282.9 238.0 144.3 166.4 198.5 246.7 283.0 Industry NA NA NA NA NA NA 78.6 130,0 88.9 139.3 114.6 169.0 234.3 Universities & colleges NA NA 105.3 150.2 168.1 111.7 94.3 56.8 48.8 40.9 36.5 32.5 34.0 FFRDC's (administered by universities) NA NA 38,9 53.2 32.1 138,8 97,8 162,0 146.9 148.7 104.9 123.7 92.7 Nonprofit institutions NA NA NA NA NA NA 20.1 23.7 25.0 4.8 24.1 14.3 25.0 • GNP implicit price deriators used to convert current dollars to constant 1967 dollars NOTE NA = not available SOURCE National Science Foundation. Federal funds for Research. Development and Other Scientific Achvities, Vol, XXIII {NSF 74.320-A) and earlier volumes. Table 2-16. Federal obligations for R&D plant as a percent of Federal obligations for total R&D, by performer, 1962-74 Percent Performer 1974 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 (est.) Total 8 9 8 Federal intramural 30 43 34 Industry NA NA NA Universities and colleges NA NA 9 FFRDC's (administered by universities) NA NA 6 Nonprofit institutions NA NA NA 30 NA 12 NA 6 20 NA 12 5 NA 4 7 NA 21 NA 14 3 24 4 23 4 25 1 17 4 22 2 17 4 NOTE. NA = not available SOURCE: National Science Foundation. Federal Funds lor Research. Development and Other Scientihc Activities. Vol, XXIII (NSF 74-320-A), and earlier volumes. 182 Table 2-17. Federal obligations for scientific and technical information activities compared with total Federal R&D obligations, 1960-74 Obligations for scientific and technical information activities (in millions) Ratio of tfiese obligations Current Constant1967 to total Federal R&D Year dollars dollars' obligations 1960 W6 $~87 Ho 1961 92 103 .010 1962 129 143 .013 1963 165 181 .013 1964 203 219 .014 1965 225 239 .015 1966 278 287 .018 1967 324 324 .020 1968 359 345 .023 1969 362 332 .023 1970 387 336 .025 1971 398 331 .026 1972 419 337 .025 1973 438 334 .026 1974(est.) 468 323 .026 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE National Science Foundation, Federal Funds lor Research. Deyeloprr\enl and Other Scienlilic Aclivilies, Fiscal Years 1973, 1974 and 1975. Vol. XXIII (NSF 75-320-A) and earlier volumes 183 Table 2-18. Federal obligations for scientific and technical information activities, by agency, 1960-74 (Dollars in millions] 1974 Agency 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 (est.) Current dollars Total $76 $92 $129 $165 $203 $225 $278 $324 $359 $362 $387 $398 $419 $438 $468 Dept. of Defense 16 23 38 53 84 99 119 139 155 147 145 141 150 161 158 Dept. of Health, Education and Welfare 10 12 24 27 24 24 37 53 60 65 66 73 68 67 82 Dept of Commerce 23 27 28 31 33 37 42 46 47 52 60 69 78 85 93 Library of Congress 5 6 6 8 9 10 13 13 17 20 22 25 30 32 35 National Aeronautics and Space Administration ... 1 3 7 14 20 19 23 24 27 28 27 27 27 25 24 Dept of the Interior 4 4 5 7 8 9 10 12 14 13 13 14 14 16 21 Nahonal Science Foundation 7 7 10 10 12 13 16 12 16 12 15 14 12 11 10 Dept of Agriculture 4 4 4 4 5 6 6 14 8 9 10 10 11 13 13 Other agencies 6 6 7 11 8 8 12 11 15 16 29 25 29 28 32 Constant 1967 dollars' Total $87 $102 $144 $181 $221 $238 $286 $324 $344 $333 $335 $331 $338 $333 $324 Dept of Defense 18 26 42 58 91 105 123 139 149 135 126 117 121 123 109 Dept. of Health, Education and Welfare 11 13 27 30 16 25 38 53 58 60 57 61 55 51 57 Dept. of Commerce 26 30 31 34 36 39 43 46 45 48 52 57 63 65 64 Library of Congress 6 7 7 9 10 11 13 13 16 18 19 21 24 24 24 National Aeronautics and Space Administration ... 1 3 8 15 22 20 24 24 26 26 23 22 22 19 17 Dept. of the Interior 5 4 6 8 9 10 10 12 13 12 11 12 11 12 15 National Science Foundation 8 8 11 11 13 14 17 12 15 11 13 12 10 8 7 Dept of Agriculture 5 4 4 4 5 6 6 14 8 8 9 8 9 10 9 Other agencies 7 7 8 12 9 8 12 11 14 15 25 21 23 21 22 ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars NOTE Detail may not add to totals because of rounding. SOURCE- National Science Foundation, FedersI Funds for Research. Development snd Other Saentiltc Activities. Fiscal Years 1973. 1974 and 1975. Vol XXIII (NSF 74-320-A| and earlier volumes. 184 Table 3-1. Basic research expenditures, 1960-74 [Dollars in millions] Year 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974(est.) ' GNP implicit price deflators used to convert current dollars to constant 1967 dollars SOURCE: National Science Foundation. National Patterns of R&D Re- sources. 1953-75 (NSF 75-307) Constant Current 1967 dollars dollars' $1,183 $1,347 1,378 1,549 1,695 1.884 1,974 2.166 2,301 2,486 2.572 2,728 2,825 2.915 3.029 3,029 3,286 3,159 3,378 3,099 3,548 3,085 3,544 2,948 3,705 2.982 3,800 2,896 3.991 2.758 185 Table 3-2. Basic research expenditures, by performer, 1960-74 [Dollars in millions] Year Total Universities and colleges Federal Government Industry FFRDC's' Nonprofit institutions Current dollars 1960 .... 1961 .... 1962 .... 1963 .... 1964 .... 1965 .... 1966 .... 1967 .... 1968 .... 1969 .... 1970 .... 1971 .... 1972 .... 1973 .... 1974(est.) 1960 .... 1961 .... 1962 .... 1963 .... 1964 .... 1965 .... 1966 .... 1967 .... 1968 .... 1969 .... 1970 .... 1971 .... 1972 .... 1973 .... 1974{est.) $1,183 $ 433 $160 $376 $ 97 $117 1,378 536 206 395 115 126 1,695 659 251 488 136 161 1,974 814 299 522 159 180 2,301 1,003 364 549 191 194 2,572 1,138 424 592 208 210 2,825 1,303 445 624 227 226 3,029 1,457 472 629 250 221 3,286 1,649 502 642 276 217 3,378 1,707 565 618 275 213 3.548 1,796 646 629 269 208 3,544 1,914 535 610 260 225 3.705 2,024 607 579 250 245 3,800 2,058 585 605 297 255 3,991 2,151 635 640 291 274 Constant 1967 dollars^ $1,347 1,549 1,884 2.165 2,486 2,729 2,915 3,029 3,159 3,098 3,086 2,947 2.981 2,895 2,757 $ 492 602 733 893 1,084 1,207 1,345 1,457 1,586 1,566 1.562 1,592 1,629 1,568 1,486 $182 232 279 328 393 450 459 472 483 518 562 445 488 446 439 $428 444 542 573 593 628 644 629 617 567 547 507 466 461 442 $110 129 151 174 206 221 234 250 265 252 234 216 201 226 201 $133 142 179 197 210 223 233 221 209 195 181 187 197 194 189 ' Federally Funded Research and Development Centers administered by universities ^ GNP implicit price deflators used to convert current dollars to constant 1967 dollars NOTE Detail may not add to totals because of rounding SOURCE National Science Foundation, National Patterns of R&D Resources. 1953-75 {NSF 75-307) 186 Table 3-3. Basic research expenditure, by source 1960-74 [Dollars in millions] Federal Universities Nonprofit Year Total Government Industry and colleges' institutions Current dollars 1960 $1,183 $ 693 $331 $ 72 $ 87 1961 1,378 841 350 85 102 1962 1,695 1,091 382 102 120 1963 1,974 1,310 414 121 129 1964 2,301 1,595 424 144 138 1965 2,572 1,817 448 164 143 1966 2,825 1,986 496 196 147 1967 3,029 2,173 477 223 156 1968 3,286 2,327 518 276 165 1969 3,378 2,386 519 298 175 1970 3,548 2,469 536 350 193 1971 3,544 2,379 556 400 209 1972 3,705 2,528 528 428 221 1973 3,800 2,605 561 416 218 1974{est.) ... 3,991 2,724 594 434 239 Constant 1967 dollars^ 1960 $1,347 $ 789 $377 $ 82 $ 99 1961 1,549 945 393 96 115 1962 1,884 1,213 425 113 133 1963 2,166 1,437 454 133 142 1964 2,486 1,723 458 156 149 1965 2,728 1,927 475 174 152 1966 2,915 2,050 512 202 152 1967 3,029 2,173 477 223 156 1968 3,159 2,237 498 265 159 1969 3,098 2,189 476 273 161 1970 3,086 2,147 466 304 168 1971 2,947 1,979 463 333 174 1972 2,981 2,034 425 344 178 1973 2,895 1,985 427 317 166 1974(est.) ... 2,757 1,882 410 300 182 ' Includes State and local government sources ^ GNP implicit price deflators used to convert current dollars to constant 1967 dollars. NOTE Detail may not add to totals because of rounding SOURCE National Science Foundation, National Patterns of R&D Resources, 1953-75 (NSF 75-307). 187 Table 3-4. Federal obligations for basic research as a percent of eacfi agency's R&D obligations, by agency, 1960-74 Year All agencies USDA DOD HEW AEC NASA' NSF All other agencies 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974(est.) .... 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974(est.) .... 1960 $ 7,552 1961 9.059 1962 10,290 1963 12,495 1964 14,225 1965 14,614 1966 15,320 1967 16,529 1968 15,921 1969 15,641 1970 15,340 1971 15,564 1972 16,512 1973 16,821 1974(est.) 17,743 Basic research as a percent of all R&D obligations 8 27 3 32 14 26 91 19 9 29 3 32 20 24 92 19 11 32 3 33 19 22 91 20 11 33 3 36 20 16 92 20 11 36 3 35 19 12 91 22 12 40 4 35 21 11 91 22 12 40 4 32 23 11 91 18 12 40 4 32 24 12 91 15 13 39 3 32 21 15 89 17 13 41 4 29 20 17 91 15 13 41 3 32 21 17 85 12 14 39 3 27 21 21 81 9 15 39 3 26 21 24 81 12 14 39 3 25 20 25 82 10 15 39 3 25 20 24 79 10 Federal obligations for basic research (Ci jrrent dollars in millions) $ 610 $ 34 $168 $103 $104 $ 97 $ 68 $ 35 825 41 173 137 167 190 77 39 1,106 50 204 190 192 316 104 50 1,389 56 231 236 219 447 141 59 1,567 68 241 274 238 524 155 66 1,690 90 263 303 258 528 171 77 1,840 94 262 326 281 559 223 95 2,004 100 284 372 302 603 239 104 2,056 100 263 397 282 656 252 106 2,077 107 276 371 285 678 248 112 2,042 116 247 388 287 637 245 122 2,132 118 262 397 277 680 273 125 2,411 137 270 461 268 768 368 139 2,420 143 258 458 275 769 392 125 2,569 150 253 588 286 734 421 138 Federal obligations for all R&D (Current dollars in millions) ;i26 $5,712 $ 320 $ 762 $ 369 $ 75 $ 189 143 6,574 429 850 777 84 202 157 6,723 577 1,029 1,439 114 251 168 7,286 656 1,078 2,857 154 295 189 7,262 777 1,236 4,287 170 305 225 6,797 869 1,241 4,952 187 344 235 7,024 1,014 1,212 5,050 244 541 253 8,049 1,147 1,259 4,867 262 694 254 7,709 1,252 1,369 4,429 284 625 260 7,696 1,297 1,406 3,963 274 744 281 7,360 1,221 1,346 3,800 289 1,043 305 7,509 1,476 1,303 3,258 337 1,377 350 8,318 1,751 1,298 3,157 455 1,183 367 8,404 1,838 1,363 3,061 480 1,309 386 8,599 2,347 1,431 3,026 530 1,425 ' The large amounts reported by NASA for basic research are due to the substantial cost of support equipment such as spacecraft and launch vehicles peculiar to space exploration, and the statistical proration ot costs for tracking and data acquisition. NOTE Detail may not add to totals because of rounding SOURCE National Science Foundation, Federal Funds for Research. Development, and Other Scientific Activities, Fiscal Years 1973. 1974 and 1975. Volume XXIIl (NSF 74-320-A) and earlier volumes. 188 >• 00 o (D c o> 0) ai IB >• A r- CD ■c O) U k IS (/) w c g ^ o = m 01 Ifl E ■- IS (A — a O) JO o ■D I I C35 o to 00 CD ■o- T- 00 (O in in CO 00 CO C\J CO in ^- CvJ in CM t- ■* t- o CO CO 00 in o> CM in CO ■<]• in in r^ CD en CM ■* T- CM ■» OJ h- CO T- CM CO 00 CM y- (D T- CM 1- h- CO CM o CO CD CO in t^ CM CM 1- cm" CM s (D h- ^ CM 00 r^ 00 00 CO CM CO CD in CM ■ CO y- CM CO CM o -^ CM CM CO O CO r^ o o T- CM CO CO CD •» CM CD 1- O) CJ> CD CM 00 in CM CO CM in O CO CO CO CO CD o in CM CM CO CM in CO y- CD •^ CM ■* 00 f~ CO CM CM CD T- CD O 1^ in CO CO 1- ■^ CM CM CM Tf O -tt O CM in O (35 O) CM >- T- 00 CO ■^ (D O O CO y- y- -r- (U LU < Q ^ o o o £ „ „ „ ra c c ^ -^ ffi 0) 0) (U E E E§ etc m (0 ro " a Q. Q. 5 Q) 0) a) "^ D QQ c o en c E " c in E o 51 CO CM CM T- CM CD in O CM T- CO o CM 1- co in CM 0> CM •<1- CD CO o o ID r^ -"J- h- O) T- CO CO CD CM O r^ CM CO CM in CM r- CD OO o (3) l-~ cn o CD O) in o o> •t y- 00 o> CJ) CO CM in CM O h- T- CD 00 CD T- r- y- O) T- ^ CM CO CM ! O CO I-- O O CO O T- CM CO CO CD CM t- r^ o CD o r^ O 00 O) r^ CO Ol r^ CO cn CM CO CNJ in C\J in cj> .- ^3- o r- CM 0) CD T^ CM r^ CD 00 OO E c CM CO CM in '— Q.9 CO O (£) T^ r^ CD O) ID CM CM CVJ >- CO CD O o in in CD in in -^ O) in CD CM CM CM •^ r- CD r^ T- CO T- CD CD in CM in 1- in CM CM CM CO r- CO ,-■,-■,- T- m Q) - - - c O) O g (D (D 0) TD O . t= < < CO E.oa.i < z z ■D HI LU cn- c C31 O CD ID C < LU _ E.9 2 ra < z ;; CD a O) ^ (0 189 o 0) u c 0> o 0)