¹ Kondratieff, N. D., “The Long Waves in Economic Life,” The Review of Economic Statistics, XVII (Nov. 1935)Google Scholar. See also Garvy, George, “Kondratieff's Theory of Long Cycles,” Review of Economic Statistics, XXV (Nov. 1943)Google Scholar, and the discusinsion of Kondratieff's views in Kuznets, Simon, Secular Movements in Production and Prices (Boston, 1930), pp. 263–265Google Scholar.

² N. D. Kondratieff, “Long Waves,” pp. 112–115. Subsequent troughs in relative prices in the 1930's and in 1972, and the peak about 1951 impart a special contemporary interest to Kondratieff's hypothesis.

³ Schumpeter evidently knew and was respectful of Kuznets' Secular Movements in Production and Prices and Burns', A. F.Production Trends in the United States Since 1870 (New York 1934)Google Scholar. Significant work on sectoral patterns was also done in the late 1930's by Hoffmann, Walther, Stadien und Typen der Industrialisierung (Kiel, 1931)Google Scholar.

⁴ Note, in particular, his peroration, Kuznets, Secular Movements, p. 329.

⁵ Ibid., pp. 3–5.

⁶ Ibid., pp. 71, 74, 83, 90, 91, 103.

⁷ Ibid., pp. 4–5.

⁸ Ibid., pp. 206–59.

⁹ Ibid., p. 258.

¹⁰ Kuznets, S., National Product Since 1869 (New York, 1946), especially pp. 87–90Google Scholar; Quantitative Aspects of the Economic Growth of Nations, I; “Levels and Variability of Rates of Growth,” Economic Development and Cultural Change, V (Oct. 1956), 44–51Google Scholar; and “Long Swings in the Growth of Population and in Related Economic Variables,” Proceedings of the American Philosophical Society, CII: 1 (1958), 25–52Google Scholar. Between Kuznets' “Secular Movements” and his National Product Since 1869, Arthur F. Burns (1954) had moved the analysis of trend cycles in growth in the U.S. from a sectoral to an aggregate basis (see, especially, ch. vi in his Production Trends in the United States Since 1870).

¹¹ Thomas, Brinley, Migration end Economic Growth (Cambridge, 1954), especially pp. 102–113Google Scholar; Cairncross, A. K., Home and Foreign Investment, 1870–1913 (Cambridge, 1953), especially pp. 187–221Google Scholar. It should be noted that, still earlier, Walter Isard and Norman J. Silberling had demonstrated an association between migration and construction in the United States, as well as with transport cycles; Walter Isard, “A Neglected Cycle: The Transport-Building Cycle,” Review of Economic Statistics (Nov. 1942), pp. 149–158; Silberling, Norman J., The Dynamics of Business (New York and London, 1943) especially pp. 175–238Google Scholar. See also Thomas', Brinley recent challenge to what he regards as excessively “American-centered” views of the long cycle, Migration and Urban Development (London, 1972), especially chs. i-ivGoogle Scholar.

¹² S. Kuznets, “Long Swings,” p. 33.

¹³ For reasons set out in Appendix B (pp. 223–230) of the second (1971) edition of The Stages of Economic Growth, I do not regard the well-known works of Robert Fogel and Albert Fishlow as altering the judgment that railroadization, in all its multiple consequences, was the leading sector in American economic growth from the 1840's through the 1880's.

¹⁴ S. Kuznets, Secular Trends, p. 191; and Ulmer, Melville J., Capital in Transportation, Communication and Public Utilities: Its Formation and Financing (Princeton, 1960), p. 112Google Scholar.

¹⁵ S. Kuznets, “Long Swings,” p. 36: “One could also probably find long swings in price structures, that is, in the relations of prices of various factors of production or of various groups of goods.” In his various expositions of the Kuznets long cycle Moses Abramovitz has sometimes listed price movements as related to them; but he did not establish a firm and lucid linkage. See, for example, Abramovitz' colloquy on this subject before the Joint Economic Committee,Employment, Growth, and Price Levels, 86th Congress, 1st Session, Part II(Washington,1959), pp. 458–59Google Scholar.

¹⁶ In addition to Abramovitz' 1959 statement before the Joint Economic Committee, see also his “The Nature and Significance of Kuznets' Cycles,” Economic Development and Cultural Change, IX (April 1961), 225–48Google Scholar; and “The Passing of the Kuznets Cycle,” Economica (Nov. 1968), pp. 349–67.

¹⁷ Kelley's widest ranging paper in this field is “Demographic Cycles and Economic Growth: The Long Swing Reconsidered,” Journal ofEconomicHistory, XXIX (Dec. 1969), pp. 633–56Google Scholar.

¹⁸ Easterlin, Richard A., Population, Labor Force, and Long Swings in Economic Growth (New York, 1968)Google Scholar.

¹⁹ Bloomfield, Arthur I., Patterns of Fluctuation in International Investment before 1914, Princeton Studies in International Finance No. 21 (Princeton, 1968), p. 5Google Scholar. Conversely, P. J. O'Leary and W. Arthur Lewis conclude that British capital movements were perverse: motivated merely by dissatisfaction with low rates of return in domestic capital markets and supplying excessive capital to agricultural regions at a time when their export prices were falling, in the period 1883–1896; “Secular Swings in Production and Trade, 1870–1913,” The Manchester School, XXIII (May 1955), 145–46Google Scholar. This judgment is based, I believe, on a somewhat superficial view of the motives for pre-1896 capital exports (see below, pp. 735–742). The opening of highly fertile areas could be profitable even in a general environment of falling prices. But, on the face of it, the build-up of Argentine, Canadian, and Australian infrastructure in the 1880's was a fortunate preparation for the period of rising prices and production that followed, roughly, 1896. Neverthless, the O'Leary-Lewis analysis must be accounted a major effort to take into account price and international capital movements in assessing long cycles.

²⁰ Williamson, Jeffrey G., American Growth and the Balance of Payments, 1820–1913 (Chapel Hill, 1964)Google Scholar; Ford, A. G., “British Investment in Argentina and Long Swings, 1880–1914,” Journal ofEconomicHistory, XXXI (Sept. 1971), 650–663Google Scholar.

²¹ A. I. Bloomfield, Patterns of Fluctuation, p. 5.

²² S. Kuznets, Secular Movements, pp. 74, 84, 91, and 104.

²³ The lag of capital imports behind series reflecting domestic expansion can be traced out in the economic history of particular foodstuff and raw material producing regions, but it is caught statistically in general terms by A. I. Bloomfield, Patterns of Fluctuation, p. 34.

²⁴ Brown, E. H. Phelps with Hopkins, Sheila V., “The Course of Wage-Rates in Five Countries, 1860–1939,” Oxford Economic Papers, II (June 1950), especially pp. 238–39Google Scholar; Brown, E. H. Phelps with Handfield-Jones, S. J., “The Climacteric of the 1890's: A Study in the Expanding Economy,” Oxford Economic Papers, IV (Oct. 1952)Google Scholar; Brown, E. H. Phelps and Weber, B., “Accumulation, Productivity, and Distribution in the British Economy, 1870–1938,” Economic Journal, LXIII (June 1953)Google Scholar; Weber, Bernard and Handfield-Jones, S. J., “Variations in the Rate of Economic Growth in the U.S.A., 1869–1939,” Oxford Economic Papers, VI (June 1954)Google Scholar. For earlier speculation on retardation in older sectors incorporating new technology, see A. F. Burns, Production Trends, especially pp. 276–81; also, Jones, G. T., Increasing Return (Cambridge, 1933)Google Scholar.

²⁵ Compare, in particular, Moses Abramovitz in “The Nature and Significance of Kuznets Cycles,” p. 226, with “The Passing of the Kuznets Cycle,” p. 351. On this point, see also P. J. O'Leary and W. Arthur Lewis, “Secular Swings,” pp. 116–18: “One may be tempted to deny that there is fundamentally a Kuznets cycle, and may prefer to say that all that happens is that once every twenty years one of the Juglar depressions gets out of hand, and lasts for 6 to 8 years, instead of lasting 1 and 2 years only.” This point relates to Arthur F. Burns' emphasis on the “sharp divergence of production trends” and “the strain and loss of industrial balance” during the upward phase of a trend-cycle movement, yielding a protracted downward phase of a re-adjustment and recovery of balance; Bums, Production Trends, pp. 248–49. It is also connected to Kuznets' finding, cited above, p. 723, that the amplitude of secondary cycles is associated positively with the momentum of a sector, as measured by its primary trend.

²⁶ It is impossible to chart the transition in leading sectors with precision for two reasons. First, conceptually, what we need are data embracing the whole leading sector complex; that is, the sector incorporating the new technology plus its backward and lateral linkages to the rest of the economy. Such data are not available. Second, there are narrower data problems. The traditional industrial classifications derive from institutional rather than technological history, embracing sub-sectors with neither uniform price nor income elasticities of demand nor with uniform technological influences, nor, even, uniform short-run elasticities of supply. Without further disaggregation, these categories leave us often in the position of having no analytic grip on the statistical movements they describe. Thus, Kuznets' despairing conclusion: “Since the high and accelerated rate of technological change is a major source of the high rates of growth of per capita product and productivity in modern times and is also responsible for striking shifts in production structure, it is frustrating that the available sectoral classifications fail to separate new industries from old, and distinguish those affected by technological innovations”; Kuznets, S., Economic Growth of Nations: Total Output and Production Structure (Cambridge, 1971), p. 315CrossRefGoogle Scholar. Nevertheless, some useful propositions about leading sectors are possible short of satisfactory measurement. And, as I have argued on other occasions, I believe it wiser to get at leading sectors with such data as we can muster than to confine our study to over-aggregated data which conceal from view variables critical to growth, trend, and cyclical analysis; see, for example, Rostow, W. W., Stages of Economic Growth, 2d ed. (Cambridge, 1971), especially pp. 183–89Google Scholar.

²⁷ E. H. Phelps Brown with Sheila V. Hopkins, “The Course of Wage Rates,” p. 291. It is useful to compare their data with Kuznets' calculations of deviations of per capita product from straight line trends in growth rates, arrayed in overlapping averages. The American cyclical depression of the 1890's was longer and deeper than elsewhere. The average level of unemployment in the U.S. for the years 1893–1898 was 14.2 percent; in Britain it was 3.7 percent. Britain moved rapidly out of its trough of 1893; the U.S. wallowed in a trough from 1893 to 1898. This differential experience affected, of course, the contours of the subsequent expansions which, in the British case, were affected also by the Boer War, a more substantial affair economically than the Spanish-American War. Nevertheless, after its rebound from the disproportionately severe depression of the 1890's, retardation emerges in the U.S. as in the British and German data.

²⁸ Ibid., p. 287. The sharp decline in French real wages late in the period derives from a particularly rapid increase in the cost of living. In a chart of product wage rates, E. H. Phelps Brown with S. J. Handfield-Jones shows Belgium sharing the relative stagnation of real wages after 1900; Phelps Brown and Handfield-Jones, “The Climacteric of the 1890's,” p. 268. Paul Douglas' classic study of real wages in the United States showed, of course, no increase in real full time earnings between the period 1890/99 and 1914; Douglas, Paul, Real Wages in the United States, 1890–1926 (Boston, 1930), p. 582Google Scholar. The data chosen by Phelps Brown and Hopkins are hourly wage rates.

²⁹ See, for example, Jörberg, Lennart, Growth and Fluctuations of Swedish Industry (Stockholm, 1961)Google Scholar.

³⁰ The trend rate of growth of Japanese GNP apparently rises from a trough in 1902 to a peak in 1916; see Kazushi Ohkawa and Henry Rosovsky, ch. i in Lawrence Klein and Kazushi Ohkawa, editors, Economic Growth, The Japanese Experience Since the Meiji Era (Homewood, 111., 1968), p. 8. On the diversification of Japanese industry in this period, based, as in Sweden, on the absorption of more sophisticated technologies, see Yuichi Shinoya, ch. iii, Ibid., especially pp. 74–77. On the sharp rise of real consumption expenditure per capita, after a brief setback during the period 1904–1909 (embracing the Russo-Japanese War), see Simon Kuznets, ch. vii, Ibid., especially p. 198.

³¹ Jan Marczewski, “Some Aspects of the Economic Growth of France, 1660–1958,” Economic Development and Cultural Change, IX (April 1961), especially pp. 382 and 386, as well as Tables 5, 7, and 9. François Crouzet “Essai de construction d'un indice annuel de la production industrielle françhise au XIX^e siècle,” Annales, Economies, Sociétés, Civilisations, XXV (Jan.-Feb. 1970), especially pp. 71–73, 78–81, and Tables 6, 10b, 11a, and 11b in the statistical annex.

³² Arthur Burns, Production Trends, especially ch. v. For retardation in productivity after 1899, see Kendrick, John, Productivity Trends in the United States (Princeton, 1961), ch. iiiGoogle Scholar, especially the five-year moving average chart of total factor productivity, p. 67.

³³ Solomon Fabricant with the assistance of Shiskin, Julius, The Output of Manufacturing Industries, 1899–1937 (New York: National Bureau of Economic Research, 1940)Google Scholar. Shaw, William H., Value of Commodity Output Since 1869 (New York: National Bureau of Economic Research, 1947)Google Scholar, also supplies highly disaggregated data permitting isolation of rapidly growing new sectors.

³⁴ S. Fabricant p. 296.

³⁵ Ibid., p. 102. This is a significant underestimate, since the impact of the automobile on steel, glass, battery, and other component production is not included.

³⁶ Wagenfuhr, R., “Die Industriewirtschaft Entwicklungstendenzen der deutschen und internationalen Industrieproduktion 1860–1932,” Sonderheft 31 Verteljahreschafte zur Konjunktorforschung (Berlin, 1933), p. 18Google Scholar; Hoffmann, Walther G., Des Wachstum der Deutschen Wirtschaft sert der Mitte des 19. Jahrhunderts (Berlin, 1965), pp. 451–52Google Scholar. The two indexes of industrial production move as follows, by decades (1913=100):

³⁷ W. G. Hoffmann, Das Wachstum der Deutschen Wirtschaft, pp. 392–93.

³⁸ Ibid, pp. 361–62.

³⁹ See, for example, Landes, David, The Unbound Prometheus (Cambridge, 1969), especially pp. 287–90Google Scholar.

⁴⁰ W. G. Hoffmann, Das Wachstum der Deutschen Wirtschaft, pp. 388.

⁴¹ Ibid., p. 358.

⁴² Phelps Brown and Handfield-Jones, “The Climateric of the 1890's,” p. 283.

⁴³ In sulphuric acid, for example, the British loss of leadership is suggested by the following table exhibiting sulphuric acid production:

⁴⁴ The increases in steel production were: U.K., 45 percent; Germany, 142 percent; France, 108 percent; Ibid., p. 261. Those for coal production were: U.K. 28 percent; Germany, 74 percent; France, 25 percent; Ibid., p. 252.

⁴⁵ David Landes marshals the case for a substantial entrepreneurial role in Britain's pre-1914 falling behind the German (and American) pace; D. Landes, The Unbound Prometheus, pp. 326–58. Six papers deal with facets of this question in McCloskey, Donald N., editor, Essays on a Mature Economy: Britain after 1840 (Princeton, 1971)Google Scholar. Five of the papers argue the rationality of British industrial practice. For a general defense or the British entrepreneurs as efficient maximizers in the Victorian era, see McCloskey, Donald N., “Did Victorian Britain Fail?,” Economic History Review XXIII (Dec. 1970)Google Scholar. McCloskey's argument, however, is not germane to the present discussion, since he does not extend his defense to the Edwardian period.

⁴⁶ For successive five-year periods starting with 1885–1889, W. Malenbaum's data show wheat yield per acre moving as follows: 13 percent, 3 percent, —2 percent, 9 percent, —2 percent; Malenbaum, W., The World Wheat Economy, 1885–1939 (Cambridge, Mass., 1953), pp. 236–39CrossRefGoogle Scholar. See also Malenbaum's discussion of U.S. regional wheat yield trends, pp. 97–99. For trend cycles in U.S. production and yields for individual agricultural commodities, see Arthur Burns, Production Trends, pp. 207–15, 323, for stagnation of output per unit of labor input and decline per unit of capital input in the U.S.

⁴⁷ W. Malenbaum, The World Wheat Economy, pp. 238–39.

⁴⁸ For various wheat price movements over these years, see, for example, Ibid., p. 118.

⁴⁹ Butlin, N. G., Investment in Australian Economic Development, 1861–1900 (Cambridge, 1964), pp. 320–22, 29Google Scholar.

⁵⁰ See the contemporary comment quoted in W. Malenbaum, The World Wheat Economy, p. 141.

⁵¹ See, especially, Hall, A. R., The London Capital Market and Australia, 1870–1914 (Canberra, 1963)Google Scholar, page opposite 88, and Appendixes II and IV.

⁵² See Ford, A. G., The Gold Standard, 1880–1914, Britain and Argentina (Oxford, 1962)Google Scholar. For general economic background and data, see pp. 81–89. For capital import statistics, in the form of annual U.K. issue for Argentina, p. 195. See, also, Williams, John H., Argentine International Trade Under Inconvertible Paper Money, 1880–1900 (Cambridge, Mass., 1920)Google Scholar; and Ferns, H. S., Britain and Argentina in the Nineteenth Century (Oxford, 1960)Google Scholar.

⁵³ See, especially, John H. Williams, Argentine International Trade, pp. 223–26.

⁵⁴ Easterbrook, W. T. and Aitken, Hugh G. J., Canadian Economic History (Toronto, 1956), p. 483Google Scholar.

⁵⁵ Buckley, Kenneth, Capital Formation in Canada, 1896–1930 (Toronto, 1955), p. 14Google Scholar.

⁵⁶ Ibid., p. 64.

⁵⁷ Feis, Herbert, Europe: The World's Banker, 1870–1914 (New Haven, 1930), pp. 210–11Google Scholar. Also, Cameron, Rondo, France and the Economic Development of Europe, 1800–1914 (2d ed., Chicago, 1961), pp. 300–1Google Scholar.

⁵⁸ Cairncross, A. K., “Investment in Canada, 1900–1913,” reprinted in Hall, A. R., editor, The Export of Capital from Britain, 1870–1914 (London, 1968), p. 157Google Scholar, from Caimcross', Home and Foreign Investment, 1870–1913 (Cambridge, 1953)Google Scholar. Also see, Wilson, Roland, Capital Imports and the Terms of Trade (Melbourne, 1931)Google Scholar; and for the U.S. in the 1830's and 1850's, Smith, W. B. and Cole, A. H., Fluctuations in American Business, 1790–1860 (Cambridge, Mass., 1935), pp. 66 and 100, and discussion on pp. 66–69 and 93–101Google Scholar.

⁵⁹ For a fuller exposition of the balance sheet for Britain, see the author's British Economy of the Nineteenth Century (Oxford, 1948), pp. 26–28Google Scholar. For similarities and differences in European net barter terms of trade movements over this period, see, especially, Kindleberger, C. P., The Terms of Trade, A European Case Study (New York, 1956), pp. 26–27Google Scholar.

⁶⁰ See, for example, A. K. Cairncross, Home and Foreign Investment, 1870–1913, p. 41.

⁶¹ See, for example, John H. Williams, Argentine International Trade, p. 207.

⁶² See, for example, the annual data in Imre Ferenczi and Willcox, W. F., International Migrations, I (New York: National Bureau of Economic Research, 1929), p. 947Google Scholar.

⁶³ S. Kuznets, “Long Swings in the Growth of Population and in Related Economic Variables,” pp. 35 and 51. In the population-association mood of this paper, Kuznets states, “… quickening of population growth would involve similar movements in the rate of expansion ts new lands.”

⁶⁴ Historical Statists of the United States, p. 12.

⁶⁵ Cairncross underlines this point: “… it would be a mistake to lay stress exclusively on agricultural development. The rise in export prices was not the sole factor at work. Between 1900 and 1913 rich deposits of mineral ores were found and developed in the Cobalt, Porcupine and other districts. Canada became the leading nickel producer in the world, and copper, silver and gold were exported in large quantities. These minerals sold at low prices, but mining costs were also low and production was highly profitable.” Caimcross, Home and Foreign Investment, 1870–1913, p. 42.

⁶⁶ The period 1897–1905 was a phase of modest but regular net capital export for the United States; but from 1906 to 1914 (excepting 1908 and 1913) the United States was once again a net capital importer. These were the years of highest immigration. Over the whole span 1897–1914, the U.S. exported net $1.2 billion. See, for example, J. G. Williamson, American Growth, p. 151.

⁶⁷ Robert S. Gallman and Edward S. Howie, “Trends in the Structure of the American Economy since 1840,” ch. oil in Fogel, Robert W. and Engerman, Stanley L., editors, The Reinterpretation of American Economic History (New York, 1971), p. 26Google Scholar.

⁶⁸ Lebergott, Stanley, Manpower in Economic Growth: The American Record since 1800 (New York, 1964), p. 510Google Scholar.

⁶⁹ For unemployment and immigration data, 1890–1914, see Ibid., p. 43.

⁷⁰ Ibid., p. 189.

⁷¹ I deal with these matters in The Process of Economic Growth, and more currently in “Technology and the Price System,” a chapter in a volume in tribute to Clarence Ayres, edited by William Breit and W. P. Culbertson, Jr., to be published in 1976 by the University of Texas Press.

⁷² Hollis Chenery and Lance Taylor have associated the coming in of different leading sector complexes with the rise of GNP per capita (in effect, with stages of growth) in the contemporary world in “Development Patterns: Among Countries and over Time,” The Review of Economics and Statistics, L (Nov. 1968), especially pp. 405 ff.

⁷³ S. Kuznets, “Long Swings in the Growth of Population,” p. 33. This proposition does not, of course, exclude capital imports, the typical rise in the investment proportion during the early phase of modem growth, or the possibilities of changes in the proportion, as during the railway age or in post-1945 Western Europe and Japan.

⁷⁴ If we remain loyal to Kondratieff's challenge, we are left with gold and wars. As has often been noted, gold discoveries cannot be regarded as wholly exogenous events. A falling price trend in an international regime dominated by a gold standard increased progressively the incentive to find and mine gold as well as to develop more efficient mining technologies. The two major periods of gold discovery in the century after 1815 come after protracted periods of falling prices. Once found, the exploitation of gold yielded in the mining areas a version of the “new country” dynamics; that is, flows of internal and international migration, capital flows, broadbased booms transcending mining itself, and local wage and price increases greater than those elsewhere. See, for example, Gilbert, Donald Wood, “The Economic Effects of the Gold Discoveries Upon South Africa: 1886–1910,” Quarterly Journal of Economics, XLVII (Aug. 1933), pp. 553–97CrossRefGoogle Scholar.

As for war, it strongly reinforced, if, indeed, it did not largely create, the first Kondratieff upswing in prices and the abnormal expansion in British agriculture. In the second, we have the Crimean and American Civil Wars as well as Bismarck's three ventures. In the third, the Spanish-American and Boer Wars, the Russo-Japanese and Balkan Wars, climaxed by that of 1914–1918. In the fourth, the Second World War and the Korean War occur. All these conflicts reinforce the inflationary tendency at work in the Kondratieff upswings; but the price increases of the early 1850's, 1890's, and late 1930's, preceded the relevant wars.

Kondratieff implied not that wars caused or contributed to long-cycle price increases, but that, somehow, long-cycle upswings led to wars. Only the American Civil War can be linked in any coherent way to the expansion process decreed by the rising prices and increasing demands for food which are the mark of the Kondratieff upswing. These asserted themselves in the 1850's and are linked to the railroad expansion of that decade. The westward march of the railroad in the 1850's and the foreseeable absorption into the Union of all that lay between the Mississippi and California did, indeed, force on to the agenda the constitutional issues that led to war, although those issues were rooted, of course, in slavery. That is all one can rationally make of the causal linkage Kondratieff implied between long-cycle upswings and wars. But the tensions that exist or might develop over scarce or prospectively scarce resources in the period which began in 1972 suggest that we might take Kondratieff's observation as a warning against the dangerous potentialities of the neo-mercantilism which is one possible outcome of the fifth Kondratieff upswing.