An earlier version of this article was presented to seminars at the Center for Political Studies, University of Michigan; Economics Program, Claremont Graduate School; International Political Economy Program, University of California, Los Angeles; and Department of Political Science, University of California, Riverside. In addition to the useful feedback from these sources, the comments of Terry Boswell, George Modelski, Karen Rasler, Suzanne Frederick, Stephen Krasner, and two anonymous reviewers were helpful.

¹ See, for example, Maddison, Angus, Phases of Capitalist Development (New York: Oxford University Press, 1982)Google Scholar; Mansfield, Edwin, “Long Waves and Technological Innovation,” American Economic Review 73 (05 1983), pp. 141–45Google Scholar; Rosenberg, Nathan and Fishtak, Claudio R., “Long Waves and Economic Growth: A Critical Appraisal,” American Economic Review 73 (05 1983), pp. 146–51Google Scholar; Beenstock, Michael, The Economy in Transition, 2d ed. (London: Allen & Unwin, 1984)Google Scholar; Rosecrance, Richard, “Long Cycle Theory and International Relations,” International Organization 41 (Spring 1987), pp. 283–301CrossRefGoogle Scholar; and Gilpin, Robert, The Political Economy of International Relations (Princeton, N.J.: Princeton University Press, 1987)CrossRefGoogle Scholar.

² See Eklund, Klas, “Long Waves in the Development of Capitalism?” Kyklos 30 (Summar 1980), pp. 383–419CrossRefGoogle Scholar. See also Garvey, George, “Kondratieff's Theory of Long Cycles,” Review of Economic Statistics 25 (11 1943), pp. 203–20CrossRefGoogle Scholar; Tinbergen, Jan, “Kondratiev Cycles and So-Called Long Waves: The Early Research,” Futures 13 (08 1981), pp. 258–63CrossRefGoogle Scholar; and Goldstein, Joshua, Long Cycles: Prosperity and War in the Modern Age (New Haven, Conn.: Yale University Press, 1988)Google Scholar.

³ See Goldstein, Long Cycles.

⁴ For representative works, see Schumpeter, Joseph A., Business Cycles: A Theoretical Historical and Statistical Analysis of the Capitalist Process (New York: McGraw-Hill, 1939)Google Scholar; Mensch, Gerhard, Stalemate in Technology: Innovations Overcome the Depression (Cambridge, Mass.: Ballinger, 1979)Google Scholar; Hartman, Raymond S. and Wheeler, David R., “Schumpeterian Waves of Innovation and Infrastructure Development in Great Britain and the United States: The Kondratieff Cycle Revisited,” Research in Economic History, vol. 4, 1979, pp. 37–85Google Scholar; Bousquet, Nicole, “From Hegemony to Competition: Cycles of the Core?” in Hopkins, Terence K. and Wallerstein, Immanuel, eds., Processes of the World System (Beverly Hills, Calif.: Sage, 1980), pp. 46–83Google Scholar; Kleinknecht, Alfred, “Innovation, Accumulation, and Crisis: Waves in Economic Development,” Review 4 (Spring 1981), pp. 687–711Google Scholar; Freeman, Christopher, Clark, John, and Soete, Luc, Unemployment and Technical Innovation (London: Frances Pinter, 1982)Google Scholar; and Kogane, Yoshiro, “Long Waves of Economic Growth: Past and Future,” Futures 20 (10 1988), pp. 532–50CrossRefGoogle Scholar. Note that, for Schumpeter, the term “innovation” was not restricted to a narrow type of technological change. Innovation could take five forms: new goods, new production methods, new markets, new sources of raw materials, and new organizational formats in industries.

⁵ See Mandel, Ernest, Late Capitalism (London: New Left Books, 1975)Google Scholar; Mandel, Ernest, Long Waves of Capitalist Development: The Marxist Interpretation (Cambridge: Cambridge University Press, 1980)Google Scholar; David M. Gordon, “Stages of Accumulation and Long Economic Cycles,” in Hopkins, and Wallerstein, , Processes of the World System, pp. 9–45Google Scholar; and Boswell, Terry, “Accumulation Innovations in the American Economy: The Affinity for Japanese Solutions to the Current Crisis,” in Boswell, Terry and Bergesen, Albert, eds., America's Changing Role in the World-System (New York: Praeger, 1987), pp. 95–126Google Scholar.

⁶ See, for example, Kondratieff, Nicolai D., The Long Wave Cycle (New York: Richardson & Snyder, 1928)Google Scholar; Graham, Alan K. and Senge, Peter M., “A Long-Wave Hypothesis of Innovation,” Technological Forecasting and Social Change 17 (08 1980), pp. 283–311CrossRefGoogle Scholar; Forrester, Jay W., “Innovation and Economic Change,” Futures 13 (08 1980), pp. 323–31CrossRefGoogle Scholar; and Sterman, John D., “An Integrated Theory of the Economic Long Wave,” Futures 17 (04 1985), pp. 104–31CrossRefGoogle Scholar.

⁷ See Rostow, Walt W., The World Economy: History and Prospect (Austin: University of Texas Press, 1978)CrossRefGoogle Scholar; Rostow, Walt W., “The World Economy Since 1945: A Stylized Historical Analysis,” Economic History Review 38 (05 1985), pp. 252–75Google Scholar; Lewis, W. Arthur, Growth and Fluctuations, 1870–1913 (Boston: Allen & Unwin, 1978)Google Scholar; Marchetti, Cesare, “Society as a Learning System: Discovery, Invention and Innovation Cycles Revisited,” Technological Forecasting and Social Change 18 (12 1980), pp. 267–82CrossRefGoogle Scholar; Modelski, George, “Long Cycles, Kondratieffs and Alternating Innovations: Implications for U.S. Foreign Policy,” in Kegley, Charles W. Jr, and McGowan, Patrick J., eds., The Political Economy of Foreign Policy Behavior (Beverly Hills, Calif.: Sage, 1981), pp. 63–83Google Scholar; Modelski, George, “Long Cycles and the Strategy of United States International Political Economy,” in Avery, William and David, Rapkin, P., eds., America in a Changing World Political Economy (New York: Longman, 1982), pp. 97–118Google Scholar; Shinohara, Miyohei, Industrial Growth, Trade and Dynamic Patterns in the Japanese Economy (Tokyo: University of Tokyo Press, 1982)Google Scholar; and Volland, Craig S., “A Comprehensive Theory of Long Wave Cycles,” Technological Forecasting and Social Change 32 (09 1987), pp. 123–45CrossRefGoogle Scholar.

⁸ For examples of this type of interpretation, see Rose, Arnold, “Wars, Innovations and Long Cycles: A Brief Comment,” American Economic Review 31 (03 1941), pp. 105–7Google Scholar; Thompson, William R. and Zuk, Gary, “War, Inflation and Kondratieff's Long Waves,” Journal of Conflict Resolution 26 (12 1982), pp. 621–44CrossRefGoogle Scholar; and Thompson, William R., On Global War: Historical-Structural Approaches to World Politics (Columbia: University of South Carolina Press, 1988)Google Scholar.

⁹ See Kondratieff, Nicolai D., “The Long Waves in Economic Life,” Review of Economic Statistics 17 (11 1935), pp. 105–15CrossRefGoogle Scholar, an article reprinted in Review 2 (Spring 1979), pp. 519–62; Imbert, Gaston, Des mouvements de longue duree Kondratieff (The movements of Kondratieff long waves) (Aix-en-Province: La Pensee Universitaire, 1959)Google Scholar; Vayrynen, Raimo,“Economic Cycles, Power Transitions, Political Management and Wars Between Major Powers,” International Studies Quarterly 27 (12 1983), pp. 389–418CrossRefGoogle Scholar; Craig, Paul P. and Watt, Kenneth E. F., “The Kondratieff Cycle and War,” The Futurist, 19 04 1985, pp. 25–27Google Scholar; and Goldstein, Long Cycles.

¹⁰ There are, of course, a number of other topics which are relevant to long waves but which have not been mentioned. Kurth and Gourevitch deal with sectoral politics in a way that could easily be related to long waves. See Kurth, James R., “The Political Consequences of the Product Cycle,” International Organization 33 (Winter 1979), pp. 1–34CrossRefGoogle Scholar; and Gourevitch, Peter, Politics in Hard Times (Ithaca, N.Y.: Cornell University Press, 1986)Google Scholar. Vayrynen explores long waves in relation to arms races and military-industrial complexes. See Vayrynen, Raimo, “Economic Fluctuations, Technological Innovations and the Arms Race in an Historical Perspective,” Cooperation and Conflict 18 (Summar 1983), pp. 135–59CrossRefGoogle Scholar. Bergesen, Albert examines corporate behavior in two of his works: “Long Economic Cycles and Size of Industrial Enterprise,” in Rubinson, Richard, ed., Dynamics of World Development (Beverly Hills, Calif.: Sage, 1981)Google Scholar; and “Economic Crises and Merger Movements: 1880s Britain and 1980s United States,” in Friedman, Edward, ed., Ascent and Decline in the World-System (Beverly Hills, Calif.: Sage, 1982)Google Scholar. Debt waves are investigated by several authors. See Kowalewski, David, ”Debt Waves: Global Financial Crises in Structural-Cyclical Perspective, 1791–1984,” unpub-lished manuscript, University of Texas at San Antonio, n.d.Google Scholar; Pfister, Ulrich and Suter, Christian,“International Financial Relations as Part of the World-System,” International Studies Quarterly 31 (09 1987), pp. 239–72CrossRefGoogle Scholar; and Marichal, Carlos, A Century of Debt Crises in Latin America (Princeton, N.J.: Princeton University Press, 1989)Google Scholar. A number of scholars have loooked at temporal rhythms in attitudes toward foreign policies and other subjects. See Klingberg, Frank L., “The Historical Alternation of Moods in American Foreign Policy,” World Politics 4 (01 1952), pp. 239–73CrossRefGoogle Scholar; Elder, Robert E. and Holmes, Jack E., “International Economic Long Cycles and American Policy Moods,” in Johnson, Paul M. and Thompson, William R., eds., Rhythms in Politics and Economics (New York: Praeger, 1985), pp. 239–64Google Scholar; Schlesinger, Arthur M. Jr., The Cycles of American History (Boston: Houghton Mifflin, 1986)Google Scholar; and Namenwirth, J.Zvi and Weber, Robert P., Dynamics of Culture (Boston: Allen & Unwin, 1987)Google Scholar.

¹¹ See Goldstein, , Long Cycles,pp. 40–98Google Scholar.

¹² Resorting to more sophisticated statistical examinations, such as spectral analysis, does not address this phase inconsistency problem. Instead, it introduces more evaluation problems both in terms of its demands for time series that are longer than those nominally available for production indicators and in terms of its insistence on finding strict regularities in long wave periodicities. Spectral analysis is not the only dubious technique for long wave analysis. See the exchange over the utility of stepwise polynomial regression in Taylor, James B., “Long Waves in Six Nations: Results and Speculations for a New Methodology,” Review 11 (Summar 1988), pp. 373–92Google Scholar; Brill, Howard, “Stepwise Polynomial Regression: Royal Road or Detour?” Review 11 (Summar 1988), pp. 393–411Google Scholar; and Taylor, James B., “The Trouble with Cycles: A Reply to Brill,” Review 11 (Summar 1988), pp. 413–32Google Scholar.

¹³ See Solomou, Solomos, Phases of Economic Growth, 1850–1973: Kondratieff Waves and Kuznet Swings (Cambridge: Cambridge University Press, 1987)Google Scholar; and Goldstein, Long Cycles.

¹⁴ Solomou devotes an unusually complicated chapter to a discussion of how to test long wave hypotheses. He proposes that the data first be partitioned into a number of short-term phases of roughly eight- to eleven-year lengths (Juglar cycles) to create comparable peak-topeak bases for calculating average growth rates. If each long wave encompasses four to six Juglar cycles, as he assumes, the question then becomes one of ascertaining, with the aid of statistical significance tests, the extent to which periods of high growth and low growth alternate along predictable lines. While it is possible to follow Solomou's procedures and reasoning with careful reading, each series examined begins in a different year. The lengths of the subsequent Juglar cycles thus vary by series. After several series are worked through, it becomes difficult to escape the feeling that the data have been overly sliced and diced in order, ironically, to reduce the potential influence of statistical artifact. See Solomou, , Phases of Economic Growth, pp. 14–26Google Scholar.

¹⁵ Ibid., p. 61.

¹⁶ See Goldstein, Long Cycles. See also Goldstein's, Joshua earlier works: “War and the Kondratieff Upswing,” International Studies Quarterly 29 (12 1985), pp. 411–41Google Scholar; and “Long Waves in Production, War and Inflation: New Empirical Evidence,” Journal of Conflict Resolution 31 (12 1987), pp. 573–600Google Scholar.

¹⁷ See Goldstein, , Long Cycles,p. 217Google Scholar. Whether it is a good idea to compute aggregate t-tests for rather heterogeneous series (in terms of varying lengths at the very least) is debatable. Another way of looking at Goldstein's evidence is to count the number of times an upswing or downswing was predicted incorrectly or missed. In the first, unlagged test, fourteen mispredictions in thirty-four trials were recorded. In the second, lagged test, there were still ten misses in thirty-seven trials. When lags were introduced, the proportional number of misses declined (from 41.2 to 27 percent), but the number of misses remained high. Moreover, only one series (U.S. GNP) in the lagged test emerged without a mispredicted phase. Britain's GNP series fared better in the unlagged test (zero misses) than in the lagged test (three misses). Along these lines, only two series, world and British industrial production, improved substantially with the imposition of the lags. The two series had nine misses in the unlagged test and only two in the lagged test.

¹⁸ Rostow, , The World Economy,pp. 106–7 and 365–72Google Scholar.

¹⁹ Rostow does not elaborate on why this should be the case. There are several possibilities. First, an aging sector has had time to become critical to a large proportion of the economy. Second, large and mature industries make it difficult for new, more innovative firms to emerge, since suppliers and the general economic infrastructure are geared to the older ways of doing things. And, third, venture capital may also be difficult to obtain within a decaying environment. For an application of these arguments to American steel decline in the Pittsburgh area, see Hoerr, John P., And the Wolf Finally Came: The Decline of the American Steel Industry (Pittsburgh: University of Pittsburgh Press, 1988)Google Scholar.

²⁰ Freeman, , Clark, , and Soete prefer to use the term “new technology systems.” In Unemployment and Technical Innovation, p. 64Google Scholar, their point is that emphasis should be placed on the interrelations between the families of innovation that occur in several new industries and their subsequent diffusion throughout the economy to other industries. As long as we keep in mind that the specific leading sector indicators are only indicators, there should be no conflict with the broader interpretation preferred by these authors.

²¹ For example, Landes and Gilpin draw attention to the same sectors as Rostow does. See Landes, David S., The Unbound Prometheus (Cambridge: Cambridge University Press, 1969)Google Scholar; Gilpin, Robert, U.S. Power and the Multinational Corporation (New York: Basic Books, 1975)CrossRefGoogle Scholar; Gilpin, Robert, War and Change in World Politics (Cambridge: Cambridge University Press, 1981)CrossRefGoogle Scholar; and Gilpin, The Political Economy of International Relations. Economic historians are somewhat more likely to disagree over the impact of a specific sector, as in the case of railroads. For railroad examples, compare the following: Fogel, Robert W., Railroads and American Economic Growth: Essays in Econometric History (Baltimore, Md.: Johns Hopkins University Press, 1964)Google Scholar; Fishlow, Albert, American Railroads and the Transformation of the Ante-Bellum Economy (Cambridge, Mass.: Harvard University Press, 1965)Google Scholar; Mitchell, Brian R., “The Coming of the Railway and United Kingdom Economic Growth,” in Reed, M. C., ed., Railways in the Victorian Economy (New York: Augustus M. Kelley, 1969), pp. 13–32Google Scholar; Hawke, Gary P., Railways and Economic Growth in England and Wales, 1840–1870 (Oxford: Oxford University Press, 1970)Google Scholar; and Fremdling, Rainer, “Railroads and German Economic Growth: A Leading Sector Analysis with a Comparison of the United States and Great Britain,” Journal of Economic History 37 (09 1977), pp. 583–604CrossRefGoogle Scholar.

²² Evidence on the time periods during which growth rates of the indicators surpassed those of industrial production can be found in Rostow, , The World Economy, pp. 373–408 and 419–25Google Scholar.

²³ Data sources for the leading sector information are discussed in Thompson, , On Global War, p. 286Google Scholar. While the jet airframe data proved useful in calculating relative leading sector shares (the purpose for which the data were collected originally), the often negative growth rates performed erratically and appeared to depress or elevate unduly the average growth rates. Their analytic utility was also compromised to some extent by the large, continuing American production share. Consequently, the aerospace indicator was not used in computing the average leading sector scores reported in this article. Other possible indicators, such as sales measured in constant values, do not appear to offer much in the way of improvement.

²⁴ A consolidated schedule of leading sector tenures for the major power reference group can be found in Thompson, , On Global War, p. 137Google Scholar.

²⁵ The problems with this approach also bear some resemblance to those found with the alternative focus on the annual frequency of innovations as a measure of technological change. With the latter, more subjective indicator, the basic problem is which innovations should be counted as major developments. Too restrictive a focus will yield a rather sketchy series. Invariably, the available series mix such innovations as the zipper or the ballpoint pen with jet engines and steel-making processes. In this respect, the leading sector focus is more discriminating. However, the multiple innovation series do tend to overlap a great deal. They also tend to peak in the leading sector valleys. See Mensch, Stalemate in Technology; Kleinknecht, “Innovation, Accumulation, and Crisis”; Freeman, Clark, and Soete, Unemployment and Technical Innovation; Haustein, Heinz-Dieter and Neuwirth, Erich, “Long Waves in World Industrial Production, Energy Consumption, Innovations, Inventions, and Patents and Their Identification by Spectral Analysis,” Technological Forecasting and Social Change 22 (09 1982), pp. 53–89CrossRefGoogle Scholar; and Duijn, J. J. Van, The Long Wave in Economic Life (Boston: Allen & Unwin, 1983)Google Scholar.

²⁶ See Kuznets, Simon, “Schumpeter's Business Cycles,” American Economic Review 30 (06 1940), pp. 257–71Google Scholar; and Schumpeter, Business Cycles.

²⁷ See Duijn, Van, The Long Wave in Economic Life, p. 155Google Scholar. Van Duijn's comparable recovery phase begins in 1937.

²⁸ Ibid., pp. 147–58.

²⁹ See Modelski, , “Long Cycles and the Strategy of United States International Political Economy,” p. 104Google Scholar. The “active zone” concept is from Perroux, Francois, “An Outline of a Theory of the Dominant Economy,” in Modelski, George, ed., Transnational Corporations and World Order (San Francisco: W. H. Freeman, 1979), pp. 135–54Google Scholar. See also Thompson, , On Global War, pp. 113–124Google Scholar, for a discussion of the system leader definitions articulated in the following works: Modelski, George, Long Cycles in World Politics (Seattle: University of Washington Press, 1987)CrossRefGoogle Scholar; Keohane, Robert O., After Hegemony: Cooperation and Discord in the World Political Economy (Princeton, N.J.: Princeton University Press, 1984)Google Scholar; and Wallerstein, Immanuel, The Capitalist World-Economy (Cambridge: Cambridge University Press, 1984)Google Scholar. One of the arguments of this comparison is that leading sectors lie at the heart of all three definitions as well.

³⁰ Relative share scores are given in Thompson, , On Global War, p. 140Google Scholar.

³¹ Data for the industrial production series are derived from Mitchell, Brian R., European Historical Statistics, 1750–1975, 2d rev. ed. (New York: Facts on File, 1980)Google Scholar; U.S. Department of Commerce, Historical Statistics of the United States: Colonial Times to 1970 (Washington, D.C.: Government Printing Office, 1971)Google Scholar; and U.S. Office of the President, Economic Report of the President (Washington, D.C.: Government Printing Office, 1987)Google Scholar.

³² A more detailed discussion of the prerequisites for systemic leadership can be found in Modelski, , Long Cycles in World Politics, pp. 217–33Google Scholar. Five hundred years of concentration and deconcentration in naval global reach capabilities are explored in Modelski, George and Thompson, William R., Seapower and Global Politics, 1494–1993 (Seattle: University of Washington Press, 1988)CrossRefGoogle Scholar. For two quite different theoretical slants on the rise of a globally oriented ruling coalition, see Modelski, , Long Cycles in World Politics, pp. 161–93Google Scholar; and Frieden, Jeff, “Sectoral Conflict and Foreign Economic Policy, 1914–1940,” International Organization 42 (Winter 1988), pp. 59–90CrossRefGoogle Scholar.

³³ Olson's thesis on distributional coalitions raises the interesting question of whether vested interests are a significant cause of decline or are simply more likely to increase their political activities in the context of relative decline. See Olson, Mancur, The Rise and Decline of Nations (New Haven, Conn.: Yale University Press, 1982)Google Scholar.

³⁴ By 1914, the United States was producing 95 percent or more of the world's motor vehicles. The proportion remained above 50 percent until 1958.

³⁵ The growth-slowing processes and practices that are internal to aging industrial sectors are no less complicated than the other facets of systemic leadership decline. For useful discussions of the nineteenth-century British case, see Levine, Aaron L., Industrial Retardation in Britain, 1880–1914 (New York: Basic Books, 1967)Google Scholar; and Elbaum, Bernard and Lazonick, William, eds., The Decline of the British Economy (Oxford: Oxford University Press, 1987)Google Scholar.

³⁶ See Huntington, Samuel P., “The U.S.: Decline or Renewal?” Foreign Affairs 67 (Spring 1989), pp. 76–96CrossRefGoogle Scholar; and Vogel, Ezra F., “Pax Nipponica?” Foreign Affairs 64 (Spring 1986), pp. 752–67CrossRefGoogle Scholar.

37. See Albert Bergesen, Robert M. Fernandez, and Chintamani Sahoo, “America and the Changing Structure of Hegemonic Production,” in Boswell, and Bergesen, , America's Changing Role in the World-System, p. 173Google Scholar.