¹ A recent exploration of this argument may befound in Pavitt, K.L.R., “Introduction and Summary,” in Technical Innovation and British Economic Performance, ed. Pavitt, K. L. R. (London, 1980).Google Scholar

² Innovative activity at the firm and industry level—measured in terms of either inputs (e.g., expenditures or employment in research and development) or outputs (e.g., patents)— has been shown in several studies to influence productivity growth and export performance. Mansfield, E., The Economics of Technological Change (New York, 1968)Google Scholar and Terleckyj, N., Effects of R&D on Productivity: An Exploratory Inquiry (Washington, D.C., 1974)Google Scholar present empirical evidence on the links between industry and firm productivity growth and R&D activity. Keesing, D. B., “The Impact of Research and Development on United States Trade,” Journal of Political Economy 75 (Jan./Feb. 1967): 38–48CrossRefGoogle Scholar, and Gruber, W., Mehta, D., and Vernon, R., “The R&D Factor in International Trade and International Investment of United States Industries,” Journal ofPolitical Economy 75(Jan./Feb. 1967): 20–37CrossRefGoogle Scholar, both found a significant statistical linkage between export performance and R&D investment in a number of industrialized nations during the postwar period. While their findings cover only the period after 1950 (because of data limitations), similar links between research activity and trade performance probably also were present earlier. Statements such asthat of Arthur Lewis in note 3 below are consistent with this view.

³ See Matthews, R. C. O., Feinstein, C. H., and Odling-Smee, J. C., British Economic Growth 1856–1973 (Stanford, Calif., 1982), esp. 436–37.CrossRefGoogle ScholarLewis, W. A., Economic Survey, 1919–1939 (London, 1949), 79Google Scholar, noted that “the largest category of British exports was in those commodities expanding least in world trade. The leader par excellence in world trade was the United States; only 17.1 per cent of her manufactures exports were in the lowest category (in terms of growth in world trade during 1913–29) compared with Britain's 42.1 per cent, and 28.6 per cent were in the highest category, compared with Britain's 4.3 per cent. Germany also was well ahead.”

⁴ See Rostas, L., Comparative Productivity in British and American Industry (Cambridge, 1948)Google Scholar and Brown, E. H. Phelps, “Levels and Movements of Industrial Productivity and Real Wages Internationally Compared, 1860–1970,” Economic Journal 83 (March 1973): 58–71.CrossRefGoogle Scholar

⁵ Chandler, A. D. Jr, The Visible Hand (Cambridge, Mass., 1977).Google Scholar

⁶ Allen, G. C., British industry and Economic Policy (London, 1979), 74CrossRefGoogle Scholar, argued that “the measures taken to deal with this condition of excessive capacity must be classed among the noteworthy failures of economic policy of the interwar period.”

⁷ In Germany, antitrust policy was also weak, while industrial research activity in many industries was substantial. But Germany differed from Britain in other respects, notably the educational system, the involvement of banks in the organization and financing of mergers, and the resulting pattern of firm structure. The interactions among these three factors are of great importance in explaining international differences in the level of R&D activity. Moreover, the impact of industrial research on performance may depend as much on its organization as on the level of R&D investment. German firms appear generally to have been more successful than British enterprises in organizing and managing industrial research.

⁸ For example, a number of proposals have beenmade for the establishment within the United States of publicly funded cooperative industrial research organizations. For a discussion of the theoretical basis for and results of some of these programs, see Mowery, D. C., “Economic Theory and Government Technology Policy,” Policy Sciences 16 (winter 1983): 27–43.CrossRefGoogle Scholar

⁹ See Peck, M. J., “Science and Technology,” in Britain's Economic Prospects, ed. Caves, R. E. (Washington, D.C. 1968).Google Scholar

¹⁰ Freeman, C. H., “Research and Development: A Comparison between British and American Industry,” National Institute Economic Review 20 (Feb. 1962): 21–38Google Scholar, stated that U.S. industry's R&D expenditures per employee were three times as great as British industry's. Industrial research employed 0.13 percent of the U.S. work force, compared with 0.05 percent in Britain in 1961. See also Sanderson, M., “Research and the Firm in British Industry,” Science Studies 2 (April 1972): 107–51.CrossRefGoogle Scholar Both Freeman and Sanderson found electrical machinery, chemicals, and petroleum products to be among the most research-intensive industries in the United States and Great Britain during the 1930s and 1950s. Freeman concluded in addition that the most research-intensive industries in his analysis (including aircraft, chemicals, electrical and nonelectrical machinery, and instruments) accounted for a larger percentage of net manufacturing output in Great Britain (32 percent in 1935 and 45 percent in 1958) than in the United States (28 percent in 1935 and 40 percent in 1958).

¹¹ See the data in National Research Council, Industrial Research Laboratories of the United States, bulletins 91, 104, and 113 (Washington, D. C., 1933, 1940, 1946)Google Scholar, reported and analyzed in Mowery, D. C., “The Emergence and Growth of Industrial Research in American Manufacturing, 1899–1946” (Ph. D.diss., Stanford University, 1981).Google Scholar

¹² The American data for 1946 reflects the impact of high levels of wartime federal support of industrial research. Nevertheless, the prewar data suggests that the gap between British and American manufacturing in in-house research activity was somewhat greater then than in the postwar period.

¹³ See Bernal, J.D., The Social Function of Science (London, 1939)Google Scholar; Perazich, G. and Field, P. M., Industrial Research and Changing Technology, Works Progress Administration National Research Project Report M-4 (Philadelphia, 1940)Google Scholar; and Senate Military Affairs Committee, Subcommittee on War Mobilization, The Government's Wartime Research and Development (Washington, D.C., 1945), 8.Google Scholar

¹⁴ Data on the identity of the 200 largest British manufacturing firms was compiled from various sources by Alfred Chandler and associates, while data on the existence of in-house research facilities was taken from a 1936 survey conducted by the British Association of Scientific Workers, Industrial Research Laboratories (London, 1936)Google Scholar, and da C. Andrade, E. N., Industrial Research: 1946 (London, 1946).Google Scholar Similar data for the 200 largest American manufacturing firms in 1930 and 1948 was assembled by Chandler, and data on research employment in these firms was extracted from the National Research Council surveys of industrial research for 1933 and 1946 (see note 11).

¹⁵ For a more detailed discussion of the data, assumptions, and statistical techniques involved in this analysis, see Mowery, D. C., “Industrial Research and Firm Size, Survival, and Growth in American Manufacturing, 1921–46: An Assessment,” Journal of Economic History 43 (Dec. 1983): 953–80CrossRefGoogle Scholar, and idem., “British and American Industrial Research: A Comparison, 1900–1950,” presented at the Anglo-American Conference on the Decline of the British Economy, Boston, 30 September to 1 October 1983.

¹⁶ See Mowery, “Industrial Research and Firm Size, Survival, and Growth.”

¹⁷ See Edwards, R. C., “Stages in Corporate Stability and the Risks of Corporate Failure,” Journal of Economic History 35 (June 1975): 428–57CrossRefGoogle Scholar; Collins, N. R. and Preston, L. E., “The Size Structure of the Largest Industrial Firms, 1909–1938,” American Economic Review 51 (Dec. 1961): 986–1011Google Scholar, or Kaplan, A.D.H., Big Business in a Competitive System (Washington, D.C., 1964).Google Scholar

¹⁸ For further discussion, see Reich, L. S., “Research, Patents, and the Struggle to Control Radio: A Study of the Big Business and the Uses of Industrial Research,” Business History Review 51 (summer 1977): 208–35CrossRefGoogle Scholar; idem., “Industrial Research and the Pursuit of Corporate Security: The Early Years of Bell Labs,” Business History Review 54 (winter 1980): 504–29; Mueller, W. F., “Du Pont: A Study in Firm Growth” (Ph.D. diss., Vanderbilt University, 1955)Google Scholar; and Brock, G. W., The Telecommunications Industry (Cambridge, Mass., 1981).Google Scholar

¹⁹ See Hannah, L., The Rise of the Corporate Economy (London, 1976), 117, 167.Google Scholar

²⁰ See Terleckyj, N., “Sources of Productivity Advance: A Pilot Study of Manufacturing Industries, 1899–1953” (Ph. D. diss., Columbia University, 1961)Google Scholar, as well asthe results presented in Mowery, “British and American Industrial Research.”

²¹ See Rostas, Comparative Productivity in British and American Industry. Kendrick, J. W., Productivity Trends in the United States (Princeton, 1961), 136Google Scholar, concludes that total factor productivity in U.S. manufacturing grew at an average annual rate of 5.3 percent from 1919 to 1929, and 1.9 percent from 1929 to 1937. Matthews et al., British Economic Growth, 229, estimate that total factor productivity in British manufacturing grew at an average annual rate of 1.9 percent from 1924 to 1937. While some tendency for British productivity growth to catch up with that of American manufacturing thus is indicated for the 1930s (possibly because of the severe impact of the Great Depression on output growth in the United States), these estimates are broadlyconsistent with those of Rostas in pointing to a large gap in the 1920s.

²² For further discussion, see Mowery, D. C., “The Relationship between the Contractual and Intra-firm Forms of Industrial Research in American Manufacturing, 1900–1940,” Explorations in Economic History 20 (Oct. 1983): 351–74CrossRefGoogle Scholar; or Cohen, W. M. and Mowery, D. C., “Firm Heterogeneity and R&D: An Agenda for Research,” in Strategic Management of Industrial R&D, ed. Crow, M., Bozeman, B., and Link, A. N. (Lexington, Mass., 1984).Google Scholar

²³ In-house research grew rapidly within American manufacturing between 1900 and 1940, and independent contract research organizations' share of total industrial research employment declined from 15 percent in 1921 to 7 percent in 1946; see Mowery, “The Relationship between the Contractual and Intrafirm Forms of Industrial Research.” By 1940, over 60 percent of the clients of such independent research organizations as the Mellon Institute had in-house research facilities, and contract research was operating primarily as a complement to, rather than as a substitute for, in-house research. An example of the importance for technical change of the incorporation of several functions within the firm is Courtaulds' exploitation of the technology for rayon production in the early 1900s. As Coleman, D. C., Courtaulds: An Economic and Social History (Oxford, 1969), 2:66Google Scholar, points out, the fact that the firm was a textile, as well as an aspiring rayon yarn, producer, was of considerable help in refining and exploiting the rayon production technology.

²⁴ For additional discussion see Mowery, “Industrial Research and Firm Size, Survival and Growth.”

²⁵ Ibid. Note, however, that while the largest firms were the leaders in the early years, they were no more research-intensive, relative to their size, during the 1921–46 period than during the postwar period.

²⁶ See Hannah, L., “Mergers in British Manufacturing Industry, 1880–1919.” Oxford Economic Papers, new ser., 26 (March 1974): 1–20, esp. 10.CrossRefGoogle Scholar

²⁷ See Salter, M. S. and Weinhold, W. A., Merger Trends and Prospects, report for the Office of Policy, U.S. Department of Commerce (Washington, D.C., 1980), esp. 4.Google Scholar

²⁸ See Chandler, A. D. Jr, “The Growth of the Transnational Industrial Firm in the United States and the United Kingdom: A Comparative Analysis,” Economic History Review, 2d ser.,33 (Aug. 1980): 396–410CrossRefGoogle Scholar; Payne, P. L., “The Emergence of the Large-Scale Company in Great Britain, 1870–1914,” Economic History Review, 2d ser., 20 (winter 1967): 519–42CrossRefGoogle Scholar; and Hannah, L., “Managerial Innovation and the Rise of the Large-Scale Companvin Great Britain,” Economic History Review, 2d ser., 28 (May 1974): 252–71.CrossRefGoogle Scholar

²⁹ Macrosty, H. W., The Trust Movement in British Industry (London, 1907), 147.Google Scholar

³⁰ Quoted in ibid., 152.

³¹ “Within his own business he [Lever] salved his Victorian economic conscience bv the application of competitive principles; all associated companies should 'compete strongly with Lever Brothers in every line ….‘” Wilson, C. M., The History of Lever (London, 1954), 1:273.Google Scholar

³² “United Steel was not intended to be a very centralized business…. There is no record of the founders of the business having as their object the achievement of increased productive efficiency horizontally as between the businesses which were to become branches of the combine through what was later to be called ‘rationalization,’ and an enforced specialization upo different parts of their joint product.” Andrews, P. W. S. and Brunner, E., Capital Development in Iron and Steel (Oxford, 1951), 121.Google Scholar

³³ Reader, W. J., Imperial Chemical Industries: A History (Oxford, 1975), 2:93.Google Scholar Reader argues further that ICI, in sharp contrast to Du Pont, failed to master the integration of product diversification with its internal structure until the late 1950s, because of an inability to reorganize existing divisions within the firm according to new product opportunities and strategies (ibid., 461).

³⁴ Jones, R. and Marriott, O., Anatomy of a Merger: A History of GEC, AEI, and English Electric (London, 1970), 155.Google Scholar

³⁵ See Baker, W. J., A History of the Marconi Company (London, 1970), 332.Google Scholar Yet another large firm formed in the 1920s, Metal Box, also was characterized by a lack of internal cohesion (see Reader, W. J., Metal Box: A History [London, 1976], 40–41).Google Scholar

³⁶ See Reader, Imperial Chemical Industries:A History, 2:414.

³⁷ Mathias, P., “Visible and Invisible Hands in Great Britain,” in Managerial Hierarchies, ed. Chandler, A. D. Jr, and Daems, H. (Cambridge, Mass., 1980), 67Google Scholar, states that “in important respects—particularly in coordinating sales policies and securing monopolistic control over prices—cartels were an alternative to merger that was open to European entrepreneurs but closed to their counterparts in the United States. Firms that preferred to maintain a single-unit structure in Britain were therefore free to do so while, at the same time, reducing competition by joining a cartel; in the 1930s and during World War II, in fact, these arrangements were actively encouraged by the government. United States industrialists like Gerald [sic] Swope, the dynamic president of the American General Electric Company, accustomed to the antitrust tradition, were advised by bankers in Britain that they need not create large, centralized corporations through mergers for their European operations; market competition could be regulated through agreement with other firms, and there was thus no need to acquire them.” Macrosty, The Trust Movement in British Industry, and Lucas, A.F., Industrial Reconstruction and the Control of Competition (London, 1937)Google Scholar, among others, provide detailed accounts of price and output agreements among firms in a wide range of industries.

³⁸ See Galambos, L., Competition and Cooperation (Baltimore, 1966), 35.Google Scholar

³⁹ This argument is stressed in Stigler, G. J., “Monopoly and Oligopoly by Merger,” in The Organization of Industry, ed. Stigler, G.J. (Homewood, Ill., 1968)Google Scholar. The Supreme Court ruled in the Trans-Missouri Association case in 1898 and the Addyston Pipe case in 1899 that the Sherman Act outlawed all agreements among firms on prices or market sharing. According to Thorelli, H. B., Federal Antitrust Policy (Baltimore, 1954)Google Scholar, a total of 84 mergers occurred from 1890 to 1897, followed by 24 in 1898, 105 in 1899, and a totalof 83 from 1900 to 1902. The evidence is circumstantial, but still strong, in support of the argument that the Supreme Court decisions had an impact on merger activity. See also Chandler, The Visible Hand, 331–36.

⁴⁰ Reader, W. J., “The Chemicals Industry,” in British Industry between the Wars, ed. Buxton, N. K. and Aldcroft, D. H. (London, 1979), 174Google Scholar, notes that “it was settled policy in ICI (Imperial Chemical Industries) to avoid competition with customers or suppliers which meant, in effect, avoiding competition with virtually every manufacturing company of any importance in the United Kingdom. For many years, accordingly, ICI kept clear of the range of chemical products, associated with industrial alcohol, which were supplied by the Distillers’ Company Ltd and which were of considerable importance in the fields of organic activity opening up in the late 1920s and early 1930s. Similarly, in order to avoid giving offence to Courtaulds, important customers for caustic soda, ICI, unlike Du Pont, kept well clear of rayon and cellophane.”

⁴¹ The Committee on Finance and Industry's Report (London, 1931), 165, remarked on the absence of such close links.

⁴² See Lazonick, W. A., “Factor Costs and the Diffusion of Ring Spinning in Britain prior to World War I,” Quarterly Journal of Economics 96 (Feb. 1981): 89–109CrossRefGoogle Scholar; and idem., “Industrial Organization and Technological Change: The Decline of the British Cotton Industry,” Business History Review 57 (summer 1983): 195–236. The share of industry capacity accounted for by “combined” firms (i.e., firms that carried out bothweaving and spinning operations) declined substantially between 1884 and 1914, according to data in Tyson, R. E., “The Cotton Industry,” in The Development of British Industry and Foreign Competition, ed. Aldcroft, D. H., (Glasgow, 1968).Google Scholar See the Committee on Commercial and Industrial Policy after the War, Final Report (London, 1918)Google Scholar; S. B. Saul, “The Engineering Industry,” in The Development of British Industry and Foreign Competition; K. Warren, “Iron and Steel,” in British Industry between the Wars; and Tolliday, S., “Industry, Finance and the State: An Analysis of the British Steel Industry during the Inter-War Years” (D.Phil, thesis, University of Cambridge, 1979).Google Scholar

⁴³ Saul, “The Engineering Industry.” Byatt, I. C. R., The British Electrical Industry 1875–1914 (Oxford, 1979), 177–78Google Scholar, argues that the reliance on consulting engineers in the British electrical machinery industry reduced the flow of incremental innovations and adaptations resulting from the actual operation of the machinery. On the northeast coast, where consulting engineers were not widely used, this feedback channel was not obstructed, and important innovations resulted.

⁴⁴ For a discussion, see A. Thackray, “University-Industry Connections and Chemical Research: An Historical Perspective,” in U. S. National Science Board, The University-Industry Connection: A Collection of Papers (Washington, D.C., 1982)Google Scholar. M. B. W. Graham, “Industrial Research in the Age of Big Science,” presented at the meetings of the Society for the History of Technology, Washington, D.C., 22 October 1983, provides a description of the relationship between academic and industrial research during and after the inter-war period in the United States.

⁴⁵ Sanderson, M. W., The universities and British Industry, 1850–1970 (London, 1972), 117Google Scholar, argues that the demand on the part of industry for technically trained personnel was weak, especially before 1914.

⁴⁶ See Briggs, A., “Social History 1900–1945,” in The Economic History of Britain Since 1700, ed. Floud, R. and McCloskey, D. W., vol. 2 (Cambridge, 1981)Google Scholar; U. S. Bureau of the Census, Historical Statistics of the United States: Colonial Times to 1970 (Washington, D.C., 1975), 1:386.Google Scholar The size of the higher educational system was an important “supply-side” influence on the growth of industrial research; Beer, J. H., The Emergence of the German Dye Industry (Urbana, Ill., 1959)Google Scholar cites the high rate of production of chemistry Ph.D. s by German higher education in the late nineteenth century as an important influence on the growth of industrial research in the German chemicals industry. As the supply of professional chemists exceeded available academic employment opportunities, emigration or industrial research were the only alternatives open to the German graduate chemist.

⁴⁷ Bernal, The Social Function of Science; Carnegie Foundation for the Advancement of Teaching, The States and Higher Education (San Francisco, 1976), 35.Google Scholar

⁴⁸ Peck, “Science and Technology,” 450.

⁴⁹ Miles, F. D., A History of Research in the Nobel Division of I.C.I. (London, 1955), 58.Google Scholar

⁵⁰ See Wickenden, W. E., A Comparative Study of Engineering Education in the United States and in Europe, Bulletin 16 of the Investigation of Engineering Education (New York, 1929), esp. 140-42Google Scholar. R. Floud, “Technical Education and Economic Performance: Engineering in the Late Nineteenth Century,” unpublished ms, 1978, focuses largely on the ability of this sytem of engineering education to respond to increasing industrial demand during the late nineteenth century, rather than considering the content and the implications for innovation of this form of engineering education, in arguing that British technical education during this period wasnot deficient.

⁵¹ The importance of this linkage is underlined in Reich, L. S., “Irving Langmuir and the Pursuit of Science and Technology in the Corporate Enviroment,” Technology and Culture 24 (April 1983): 200.CrossRefGoogle Scholar He observes that at General Electrics central research facility in the early years, “Many industrial researchers were both scientists and engineers by training. Five of the best people at Bell in the early years had taken undergraduate degrees in engineering, then proceeded to Ph.D.’s in science. Irving Langmuir received a B.S. in metallurgical engineering before acquiring his Ph.D. in physical chemistry, and William Collidge, his well-known contemporary at GE, began with a B.S. in electrical engineering, then went on to a physics Ph.D.”

⁵² See Chandler, A. D. Jr, Strategy and Structure (Cambridge, 1962), 317Google Scholar; Wickenden, “A Comparative Study of Engineering Education,” 258.

⁵³ See for additional discussion, Thackray, “University-Industry Connections and Chemical Industry Research.” Servos, J. W., “The Industrial Relations of Science: Chemistry at M.I.T.,” Isis 71 (Dec. 1980): 531–49Google Scholar, discusses the rise and decline of a similar organization, the Research Laboratory for Applied Chemistry, at the Massachusetts Institute of Technology during the period 1910–35, arguing that advocates of pure academic research eventually eclipsed such industrially oriented research activity at MIT. While such conflicts undoubtedly took place elsewhere, several points should be noted. Even after the reduction in influence of the applied chemistry laboratory, links between MIT and industry remained strong, and certainly were far stronger than those between British industry and elite institutions of higher education. The laboratory's decline was also a reflection of the fact that such applied research facilities were much more widespread in American industry in the 1930s than before World War I, as noted above in the discussion of private independent research organizations in the United States.

⁵⁴ See Erickson, C., British Industrialists. Steel and Hosiery 1850–1950 (Cambridge, 1959)CrossRefGoogle Scholar, and Chandler, Visible Hand.

⁵⁵ See Coleman, D. C. “Gentlemen and Players,” Economic History Review, 2d ser., 26 (May 1973): 92–116.CrossRefGoogle Scholar

⁵⁶ Chandler, Visible Hand.

⁵⁷ Advisory Council to the Committee for Scientific and Industrial Research, Report (London, 1916), 8.

⁵⁸ Reader, W. J., Imperial Chemical Industries: A History (Oxford, 1970), 1:275.Google Scholar

⁵⁹ Johnson, P. S., Co-operative Research in Industry (New York, 1973), 19.Google Scholar

⁶⁰ Hill, D. W., Co-operative Research in Industry (London, 1947), 54.Google Scholar

⁶¹ Committee on Industry and Trade, Final Report (London, 1929), 215, 217.

⁶² A. Albu's description of the research association in marine engineering is illustrative: “By the time of the last war Parsons [marine engine and turbine] designs had become conservative…. The anxiety of the Admiralty led to the setting up of the Parsons and Marine Turbine Research and Development Association (PAMATRADA) in 1944; but although it had a well-staffed research department the marine engine builders, most of whom were adjacent to the shipyards, were too small to be able to support it, and lacked the level of technical management to be interested in the results of research. The Association declined into a licensing organization for Parsons designs.” A. Albu, “Merchant Shipbuilding and Marine Engineering,” in Technical Innovation and British Economic Performance, 175.

⁶³ Johnson, Co-operative Research in Industry, 51–52.

⁶⁴ See Tolliday, “Industry, Finance, and the State, “276.

⁶⁵ Reader, W. J., Professional Men (London, 1966), 115Google Scholar, argues, in terms similar to those of Coleman, “Gentlemen and Players,” that the slow pace of reform reflected the English middle and upper middle classes' lack of interest in such issues during the late nineteenth and early twentieth centuries.

⁶⁶ British public policy toward R&D since World War II bears some striking similarities to that of the interwar period. A high level of government spending on R&D has been overwhelmingly concentrated (or overcommitted—see Peck, “Science and Technology“) in two sectors that have important military applications but are isolated from the rest of the British economy: nuclear energy and aircraft. This policy has reduced the resources (both financial and human) available for R&D elsewhere in the economy.