Books / Digital Text
16. Schumpeter's Theory of Business Cycles
Joseph Schumpeter's business cycle theory is one of the very few that attempts to integrate an explanation of the business cycle with an analysis of the entire economic system. The theory was presented in essence in his Theory of Economic Development, published in 1912. This analysis formed the basis for the “first approximation” of his more elaborate doctrine, presented in the two-volume Business Cycles, published in 1939.65 The latter volume, however, was a distinct retrogression from the former, for it attempted to explain the business cycle by postulating three superimposed cycles (each of which was explainable according to his “first approximation”). Each of these cycles is supposed to be roughly periodic in length. They are alleged by Schumpeter to be the three-year “Kitchin” cycle; the nine-year “Juglar”; and the very long (50-year) “Kondratieff.” These cycles are conceived as independent entities, combining in various ways to yield the aggregate cyclical pattern.66 Any such “multicyclic” approach must be set down as a mystical adoption of the fallacy of conceptual realism. There is no reality or meaning to the allegedly independent sets of “cycles.” The market is one interdependent unit, and the more developed it is, the greater the interrelations among market elements. It is therefore impossible for several or numerous independent cycles to coexist as self-contained units. It is precisely the characteristic of a business cycle that it permeates all market activities.
Many theorists have assumed the existence of periodic cycles, where the length of each successive cycle is uniform, even down to the precise number of months. The quest for periodicity is a chimerical hankering after the laws of physics; in human action there are no quantitative constants. Praxeological laws can be only qualitative in nature. Therefore, there will be no periodicity in the length of business cycles.
It is best, then, to discard Schumpeter's multicyclical schema entirely and to consider his more interesting one-cycle “approximation” (as presented in his earlier book), which he attempts to derive from his general economic analysis. Schumpeter begins his study with the economy in a state of “circular flow” equilibrium, i.e., what amounts to a picture of an evenly rotating economy. This is proper, since it is only by hypothetically investigating the disturbances of an imaginary state of equilibrium that we can mentally isolate the causal factors of the business cycle. First, Schumpeter describes the ERE, where all anticipations are fulfilled, every individual and economic element is in equilibrium, profits and losses are zero—all based on given values and resources. Then, asks Schumpeter, what can impel changes in this setup? First, there are possible changes in consumer tastes and demands. This is cavalierly dismissed by Schumpeter as unimportant.67 There are possible changes in population and therefore in the labor supply; but these are gradual, and entrepreneurs can readily adapt to them. Third, there can be new saving and investment. Wisely, Schumpeter sees that changes in saving-investment rates imply no business cycle; new saving will cause continuous growth. Sudden changes in the rate of saving, when unanticipated by the market, can cause dislocations, of course, as may any sudden, unanticipated change. But there is nothing cyclic or mysterious about these effects. Instead of concluding from this survey, as he should have done, that there can be no business cycle on the free market, Schumpeter turned to a fourth element, which for him was the generator of all growth as well as of business cycles—innovation in productive techniques.
We have seen above that innovations cannot be considered the prime mover of the economy, since innovations can work their effects only through saving and investment and since there are always a great many investments that could improve techniques within the corpus of existing knowledge, but which are not made for lack of adequate savings. This consideration alone is enough to invalidate Schumpeter's business-cycle theory.
A further consideration is that Schumpeter's own theory relies specifically for the financing of innovations on newly expanded bank credit, on new money issued by the banks. Without delving into Schumpeter's theory of bank credit and its consequences, it is clear that Schumpeter assumes a hampered market, for we have seen that there could not be any monetary credit expansion on a free market. Schumpeter therefore cannot establish a business-cycle theory for a purely unhampered market.
Finally, Schumpeter's explanation of innovations as the trigger for the business cycle necessarily assumes that there is a recurrent cluster of innovations that takes place in each boom period. Why should there be such a cluster of innovations? Why are innovations not more or less continuous, as we would expect? Schumpeter cannot answer this question satisfactorily. The fact that a bold few begin innovating and that they are followed by imitators does not yield a cluster, for this process could be continuous, with new innovators arriving on the scene. Schumpeter offers two explanations for the slackening of innovatory activity toward the end of the boom (a slackening essential to his theory). On the one hand, the release of new products yielded by the new investments creates difficulties for old producers and leads to a period of uncertainty and need for “rest.” In contrast, in equilibrium periods, the risk of failure and uncertainty is less than in other periods. But here Schumpeter mistakes the auxiliary construction of the ERE for the real world. There is never in existence any actual period of certainty; all periods are uncertain, and there is no reason why increased production should cause more uncertainty to develop or any vague needs for rest. Entrepreneurs are always seeking profit-making opportunities, and there is no reason for any periods of “waiting” or of “gathering the harvest” to develop suddenly in the economic system.
Schumpeter's second explanation is that innovations cluster in only one or a few industries and that these innovation opportunities are therefore limited. After a while they become exhausted, and the cluster of innovations ceases. This is obviously related to the Hansen stagnation thesis, in the sense that there are alleged to be a certain limited number of “investment opportunities”—here innovation opportunities—at any time, and that once these are exhausted there is temporarily no further room for investments or innovations. The whole concept of “opportunity” in this connection, however, is meaningless. There is no limit on “opportunity” as long as wants remain unfulfilled. The only other limit on investment or innovation is saved capital available to embark on the projects. But this has nothing to do with vaguely available opportunities which become “exhausted”; the existence of saved capital is a continuing factor. As for innovations, there is no reason why innovations cannot be continuous or take place in many industries, or why the innovatory pace has to slacken.
As Kuznets has shown, a cluster of innovation must assume a cluster of entrepreneurial ability as well, and this is clearly unwarranted. Clemence and Doody, Schumpeterian disciples, countered that entrepreneurial ability is exhausted in the act of founding a new firm.68 But to view entrepreneurship as simply the founding of new firms is completely invalid. Entrepreneurship is not just the founding of new firms, it is not merely innovation; it is adjustment: adjustment to the uncertain, changing conditions of the future.69 This adjustment takes place, perforce, all the time and is not exhausted in any single act of investment.
We must conclude that Schumpeter's praiseworthy attempt to derive a business cycle theory from general economic analysis is a failure. Schumpeter almost hit on the right explanation when he stated that the only other explanation that could be found for the business cycle would be a cluster of errors by entrepreneurs, and he saw no reason, no objective cause, why there should be such a cluster of errors. That is perfectly true—for the free, unhampered market!
- 65. Joseph A. Schumpeter, The Theory of Economic Development (Cambridge: Harvard University Press, 1936), and idem, Business Cycles (New York: McGraw-Hill, 1939).
- 66. Warren and Pearson, as well as Dewey and Dakin, conceive of the business cycle as made up of superimposed, independent, periodic cycles from each field of production activity. See George F. Warren and Frank A. Pearson, Prices (New York: John Wiley and Sons, 1933); E.R. Dewey and E.F. Dakin, Cycles: The Science of Prediction (New York: Holt, 1949).
- 67. On the tendency to neglect the consumer's role in innovation, cf. Ernst W. Swanson, “The Economic Stagnation Thesis, Once More,” The Southern Economic Journal, January, 1956, pp. 287–304.
- 68. S.S. Kuznets, “Schumpeter's Business Cycles,” American Economic Review, June, 1940, pp. 262–63; and Richard V. Clemence and Francis S. Doody, The Schumpeterian System (Cambridge: Addison-Wesley Press, 1950), pp. 52 ff.
- 69. In so far as innovation is a regularized business procedure of research and development, rents from innovations will accrue to the research and development workers in firms, rather than to entrepreneurial profits. Cf. Carolyn Shaw Solo, “Innovation in the Capitalist Process: A Critique of the Schumpeterian Theory,” Quarterly Journal of Economics, August, 1951, pp. 417–28.