¹ Samuelson, Paul A., “Bogeyman for Japan: A Kondratieff Downswing?” (mimeo, October 1982);Google ScholarWold, Herman, “Basic Innovation and Industrial Growth,” Tenth International Conference on the Unity of Science (Seoul, 1981);Google ScholarDelbeke, Jos, “Towards an Endogenous Macro-Economic Interpretation of the Long Wave: Belgium 1830–1980” (paper presented at the Third Conference of The Council For European Studies: “Cycles and Periods in Europe: Past and Present,” Washington, D.C., 05 1982);Google ScholarMensch, G. and Wold, Herman, “Soft Bi-Equilibrium Modelling of Kondratieff-Mensch Waves” (manuscript in preparation);Google ScholarMensch, G., Stalemate in Technology: Innovations Overcome the Depression (Cambridge, 1979);Google ScholarSuman, James B. and Rosenav, David, The Kondratieff Wave (New York, 1972);Google ScholarMass, Nathaniel J. and Forrester, Jay, Understanding the Changing Basis for Economic Growth (Washington, D.C., 1976); Jay Forrester, “Innovation and the Economic Long Wave” (MIT System Dynamics Group, Paper D-2990-l);Google ScholarGraham, Alan K. and Senge, Peter, “A Long Wave Hypothesis of Innovation,” Technological Forecasting and Social Change, 17 (1980);CrossRefGoogle ScholarSenge, Peter, “The Economic Long Wave: A Survey of the Evidence” (MIT System Dynamics National Model Project, 04 1982);Google ScholarKasper, Wolfgang, “Issues of Long Term Stabilisation,” Australian Political Economy (Melbourne, 1982), pp 78–85;Google ScholarMansfield, Edwin, “Long Waves and Technological Innovation” (mimeo, 12 1982);Google ScholarMandel, Ernest, Long Waves of Capitalist Development (Cambridge, 1980).Google Scholar

² Samuelson, “Bogeyman for Japan.”Google Scholar

³ Roughly speaking, food and raw materials have accounted for about 80–95 percent of Australia's exports. Typically about three-quarters of this produce has been exported unprocessed.Google Scholar Manufactured exports were never higher than 10 percent before the mid-1960s and today stand at their highest level, 20 percent. Imports, by contrast, have mostly comprised what the Commonwealth Statistician (since the 1960s) has described as elaborately transformed; 70–80 percent have fallen into this category, with only 11–12 percent (in 1976) being categorized as crude, the balance being simply transformed. Over the last quarter of a century fuels and lubricants have only constituted 2–3 percent of Australia's imports.Google Scholar

⁴ An excellent review of most of this literature can be found in Soper, John, “The Long Swing in Historical Perspective: An Interpretive Study (Ph.D. diss., University of Massachusetts, 1970). For some recent writings see footnote 1.Google Scholar

⁵ Kondratieff, N. D., “Long Waves in Economic Life,” Review of Economic Statistics, 17 (11 1935), 115.CrossRefGoogle Scholar

⁶ Schumpeter, Joseph, Business Cycles (New York, 1939). See in particular vol. 1, pp. 82–86 and 170.Google Scholar

⁷ According to Rostow, Schumpeter's theory does not fit the facts in two respects: first, the experimental phase of innovation does not always require inflationary credit creation (for instance, cotton did not), and second, the maturing of some innovations did not actually coincide with the periods of falling prices. See Rostow, W. W., Why the Poor Get Richer and the Rich Slow Down (Austin, Texas, 1980), p. 6.CrossRefGoogle Scholar

⁸ Kuznets, Simon, Secular Movements in Production and Prices (Boston, 1930).Google Scholar

⁹ Kuznets, Simon, “Long Swings in the Growth of Population and Related Economic Variables,” Proceedings of the American Philosophical Society, 102, no. 1 (1958), 25–52;Google ScholarCairncross, A. K., Home and Foreign Investment, 1870–1913 (Cambridge, 1953);Google ScholarThomas, Brinley, Migration and Economic Growth (Cambridge, 1954).Google Scholar Rostow remains skeptical of Kuznets's treatment of railways as merely “population sensitive” because this blurs the motives behind different railway booms. Nor were the immigration booms systematically correlated with railway expansion (railways were on the decline as a leading sector before 1914, at a time when immigration attained a secular peak). See Why the Poor Get Richer, p. 21.Google Scholar

¹⁰ Rostow, , Why the Poor Get Richer, p. 61.Google Scholar

¹¹ Rostow, W. W., “Cycles and the Irreducible Complexity of History” (paper delivered at the International Economic History Conference, Budapest, 08 1982).Google Scholar

¹² Producers of I goods, the so-called “core” countries of the world economy, do not necessarily experience Kondratieff cycles in industrial production (and their RGDP). The explanation seems to be that, among other things, countries' particular stages of development—and differing leading sectors—yield different national growth rates. Rostow does not provide us with figures of the core's industrial production. But Arthur Lewis does, at least for the United States, Britain, France, and Germany combined, 1870–1913, concluding that the shape of the sum of these (which hints at an association between Kondratieff long-run price rises and rising output, and vice versa) turns out to be accidental: it has no design to which the parts conform (Lewis, W. Arthur, Growth and Fluctuations, 1870–1913 [London, 1978], pp. 64–65).Google Scholar

¹³ Rostow, W. W., “The Terms of Trade and Development” (manuscript), p. 16. As Australia became more industralized, agricultural cum export-led RGDP cycles would, of course, be muted—but not, we think, obliterated.Google Scholar

¹⁴ The current upswing (1972–) makes little sense either: the curve changes from convex to concave. Rostow points out, however that oil exporters (and Australia is not one) have indeed enjoyed a boom in real output. Rostow argues that the world economy—except the oil exporters—are in a recession but argues that the recession is occurring within a Kondratieff upswing defined by “generalized inflation” and “radically shifted terms of trade.” Other writers (for example, Ernest Mandel) define the Kondratieff in terms of real output, and dating the swings differently than Rostow, arrive at the conclusion that the world economy is in its fifth Kondratieff downturnGoogle Scholar

¹⁵ Taking the upswings, the first has a life of 25 as does the second; the third 24, and the fourth 16. The first downswing runs for 33 years, the second 23, and the third runs its course in 15 years.Google Scholar

¹⁶ Slutsky, Eugen, “The Summation of Random Causes as the Source of Cyclic Processes,” Econometrica, 5 (04 1937), 105–45. Other workers have adapted the “Slutsky effect” to a direct critique of the Kuznets cycle.CrossRefGoogle Scholar See Bird, Roger C. et al. , “Kuznets Cycles in Growth Rates: The Meaning,” International Economic Review, 6 (05 1965), 229–39.CrossRefGoogle Scholar Also see Melnyk, M., “Long Fluctuations in Real Series of American Economy” (Bureau of Economic and Business Research, No. 9, Kent State, 1969).Google Scholar

¹⁷ See Chatfield, Christopher, The Analysis of Time Series: An Interpretation, 2nd ed. (London, 1980), pp. 136–65;CrossRefGoogle ScholarBox, George E. and Jenkins, Gwilym M., Time Series Analysis: Forecasting and Control (San Francisco, 1970), pp. 36–45;Google ScholarGranger, C. W. J. in association with Hatanaka, M., Spectral Analysis of Economic Time Series (Princeton, 1964).Google Scholar

¹⁸ Non-stationarity means that it is difficult to interpret the spectrum. It is to be noted that the power spectral estimates do not represent the contribution of each frequency component to the total variance of the series, because some true components will be closed out or overshadowed by the non-stationary elements.Google Scholar

¹⁹ The types of curves we fitted to remove trend do not have the property of attenuating low- frequency cycles as with so-called low-pass filters.Google Scholar

²⁰ See Granger, C. W. J., “The Typical Spectrum Shape of an Economic Variable,” Econometrica, 34 (01 1966), 150–61.CrossRefGoogle Scholar

²¹ Granger, and Hatanaka, , Spectral Analysis, p. 17. The authors specifically mention the futility of measuring Kuznets cycles by spectral methods, concluding “to measure the reality of the Kuznets long wave, data of 140 years or so are required as a minimum. As reliable economic series are usually shorter than this it is seen that there is an insufficient amount of data available to make any test for such cycles.”Google Scholar

²² The estimates fell within our constructed confidence bands indicating the estimates were not statistically different from one another.Google Scholar

²³ Kuznets, Simon “Schumpeter's Business Cycle,” Economic Change: Selected Essays in Business Cycles, National Income and Economic Growth (New York, 1953), pp. 117–18.Google Scholar

²⁴ Pope, David, “Price Expectations and the Australian Price Level, 1900–1930,” Economic Record (12 1982), pp. 328–52, especially pp. 112–13.Google Scholar

²⁵ Nevile, John, “Domestic and Overseas Influences on Inflation in Australia,” Australian Economic Papers, 16 (06 1977), 121–29.CrossRefGoogle Scholar

²⁶ Pope, David, “Wage Regulation and Unemployment 1900–1930,” Australian Economic History Review, 22, no. 2 (1982), 103–26.CrossRefGoogle Scholar

²⁷ Kelley, A. C., “International Migration and Economic Growth: Australia, 1865–1935,” this Journal, 25 (09 1965).Google Scholar

²⁸ Richardson, H. W., “British Emigration and Overseas Investment, 1880–1914”, Economic History Review (02 1972).Google Scholar

²⁹ Pope, David, “Modelling the Peopling of Australia: 1900–1930,” Australian Economic Papers (12 1981).Google Scholar

³⁰ Rostow, , “The Terms of Trade and Development,” p. 17.Google Scholar

³¹ (italics added).

³² Wilson, Roland, Capital Imports and the Terms of Trade (Melbourne, 1931).Google Scholar

³³ Hall, A. R., The London Capital Market and Australia: 1870–1914 (Canberra, 1962), pp. 123, 125.Google Scholar

³⁴ Rostow does not, of course, ignore residential building. It will be remembered that immigration (triggered initially by the real income effect of favorable movements in the terms of trade) is seen to raise the demand for housing. But as Hall and later Kelley convincingly show, even as regards immigration, contemporaneous immigration was at most only half the story. What mattered in housing demand was the skewed age distribution of the Australian population, a legacy of the rapid growth of births following on the heels of the 1850s gold rushes. This made for a high proportion of the population in the age bracket customarily demanding housing in the 1880s, their children then creating their own demands in subsequent years. See Hall, A. R., “Some Long Period Effects of the Kinked Age Distribution of the Population of Australia, 1861–1961,” Economic Record (03 1963).Google Scholar Hall's hypothesis was further examined and elegantly extended by Kelley, A. C., “Demographic Change and Economic Growth: Australia, 1861–1911,” Explorations in Economic History (12 1969).Google Scholar

³⁵ Macfarlane, I. J., “The Balance of Payments in the Seventies” (Conference in Applied Economic Research, Reserve Bank of Australia, Sydney, 1979).Google Scholar

³⁶ In 1913 producers' goods accounted for about 65 percent of imports. In 1928 the figure was close to 70 percent and thereafter has varied between 70 and 80 percent.Google Scholar

³⁷ Pope, R., “Revaluation: Help or Hindrance to Australian Manufacturing?” (Centre for Applied Economic Research Paper No. 14, Sydney, 1981).Google Scholar

³⁸ Among the list of other influences on industrialization in the early twentieth century must be included the relative abundance of cheap labor, changes in tastes, population growth, growth of hire purchase, and new methods of company finance (plus capital inflow). In R. Pope's model of the determinants of manufacturing output, 1950–1980, the following were included alongside overseas prices: domestic output prices, real expenditure, inventory accelerator, exchange rate, import quotas, overseas production capacity, Australian wage rate adjusted by productivity, monetary policy constraint, and producers' adjustment costs.Google Scholar

³⁹ To see this, let, u = u(p^x, p^M, V) where: u = index of industrialization p^x = export prices p^M = import prices V = vector of other determinants Now Rostow emphasizes ∂u/∂p^x <0. If the principal effect on u of a rise in p^M is the cost-saving effect, then ∂u/∂p^M >0 and predictably ∂u/∂(p^x/p^M) < 0. But if the competitive effect overweighted the cost-saving effect (that is, ∂u/∂p^M <0), then knowing the direction of change in the ratio (^x/p^M) does not permit us to say whether the rate of industrialization will rise or fall.0+and+predictably+∂u/∂(px/pM)+<+0.+But+if+the+competitive+effect+overweighted+the+cost-saving+effect+(that+is,+∂u/∂pM+<0),+then+knowing+the+direction+of+change+in+the+ratio+(x/pM)+does+not+permit+us+to+say+whether+the+rate+of+industrialization+will+rise+or+fall.>Google Scholar

⁴⁰ Foley, J. C., Droughts in Australia: Review of Records From Earliest Years (Bureau of Meteorology, Melbourne, 1957).Google Scholar

⁴¹ Warren Musgrave, “Land Degradation, Land Administration and the Economics of Pastoralism” (paper delivered before Centre for Resource and Environmental Studies, March 1983); Lawrence, Denis and McKay, Lloyd, “Inputs, Outputs and Productivity Change in the Australian Sheep Industry,” The Australian Journal of Agricultural Economics, 24 (04 1980), 46–59.CrossRefGoogle Scholar Rostow is well aware of the general point; see his The Process of Economic Growth (Oxford, 1953), p. 172, footnote 1 and p. 197. Nonetheless the point remains: export price signals can be ambiguous and for this reason it is difficult to build generalizations upon them.Google Scholar