Equilibrium theory and business cycle theory seem, at the outset, incompatible. Equilibrium theory, after all, speaks of things tending to a certain "static" setting which, when achieved, repeats itself indefinitely. As a consequence, when speaking of cycles, we must be speaking of some aberration. Other than the handful of economists we have covered, most other Neoclassical economists treated business cycles merely as "crises" which interrupt the normal activities of the economy.
"For, just as a lake is, at times, stirred to its very depths by a storm, so also the market is sometimes thrown into violent confusion by crises, which are sudden and general disturbances of equilibrium."
(L. Walras, Elements of Pure Economics 1874: p.381)
However, there are two problems that are not adequately dealt with in the bulk of pre-Keynesian business cycle theory: firstly, why are these aberrations so lengthy (and, so, what propagates them) and, secondly, what causes these aberrations to begin with.
The issue of the "push" was the point of contention of early theory. As Haberler (1937) and Mitchell (1927) document, many imaginative exogenous "push" theories are indeed available: e.g. from sunspots (Jevons), animal spirited expectations (Pigou), institutional changes (Vogel), technological change (Tugan-Baranovsky, Spiethoff), or the financial sector (Hawtrey, Hansen).
Of what consequence is any of this? Why ascribe business cycles to a single, particular cause? One must note that, by doing so, one is in effect arguing that this non-economic cause must itself occur periodically. Therefore, it is also necessary to develop a "theory of sunspots" which is itself regular. Or a theory which explains the periodicity of expectations breakdown. Therefore, if one avoids explaining business cycles endogenously by appealing to outside influences, one is, at the same time, necessarily implying some endogenous cyclicalism of these outside causes. Unfortunately, most "push" theories rarely offer such an supporting theory (Schumpeter is a notable exception).
As the century advanced, the issue of "push" was gradually superseded by the issue of persistence. This was particularly important since it revealed an incompatibility between static equilibrium economics and business cycle theory. After all, whatever the cause, if there is indeed a "crisis", what makes it persist? Why do the regular stability forces fail to quickly bring the economy back into equilibrium?
Emil Lederer (1926) and Adolph Lowe (1926) argued that the study of cycles should transcend simple static equilibrium settings. Instead, they should be referred to in the context of dynamic aggregates where fluctuations are clearly indentifiable. In essence, they argued, any study of business cycles has to be macroeconomic. This was the call Hayek and Keynes responded to. In spite of their efforts, equilibrium economics ignored the German arguments. As such, then, by the 1930s, it seemed as if business cycles were forever to be ignored by equilibrium economics.
This state of affairs was fundamentally altered when Ragnar Frisch (1933) and Eugen Slutsky (1937) presented their theory of stochastic cycles. There was no need in the Frisch-Slutsky approach to appeal to specific determinate causes. All that was argued is that many phenomena existed which could precipitate a real shock in the economy's equilibrium path. These shocks, Frisch and Slutsky argued, were common but entirely random and distributed normally (standard variance with a mean of zero). This implies, then, that most shocks were relatively small and approximately half of them were negative and another half positive. Large shocks, in either direction, however, were rare. However, should a particularly large negative shock appear, it is unlikely that it would be followed immediately by a similarly large positive shock. Consequently, they argued, that single, large negative shock would be sufficient to draw the average output away from equilibrium output for a sufficiently long time to be considered a downswing. Over time, of course, the influence of this single shock would be gradually whittled away as more positive shocks arise.
Of course, it is not necessary that a large shock happen for the average to move. As Slutsky, in particular, pointed out, there could be "clusters" of small negative shocks (i.e. 3, 4 or 5 successive small negative shocks without a positive shock in between) which would also move the economy below its average. Thus, whether by single large shocks or clusters of small shocks, the economy could experience a negative average which was sufficiently persistent to be classified as part of a cycle. Thus, business cycle theory reentered Neoclassical equilibrium theory as a theory of stochastic shocks. Instead of attempting to root out periodic causes of such shocks, the Frisch-Slutsky approach had to assume them as random variations from an equilibrium trend.
The question immediately posed was that of "time". How a large shock or cluster of shocks may move an average down is understandable, but what prevents a large negative stochastic shock from being instantaneously corrected by a policy-engineered large positive shock? Are the stability forces of equilibrium economics not strong or fast enough so that the effects of a large shock last longer? The answer is simply yes. As the economy is thrown off course and adjustments are immediately called upon in prices, output, employment or whatever, there might be differing lags in response from different sectors of the economy and consequently a failure to coordinate a countercyclical shock. These coordination failures may prolong the negative average. Institutional rigidities in wages, prices, interest rates and associated elements may also contribute to the prevention of a swift response.
The Frisch-Slutsky approach to business cycles, then, responds to the previous treatment of cycles by ignoring specific causes. Any strong random shock or cluster of small shocks in the same direction may, propagated by these rigidities and response failures, result in a far larger cyclical swing than the shocks themselves might warrant.
The Frisch-Slutsky theory of business cycles may not be very satisfying since they are no proper "cyclical" phenomenon as such but rather just moving patterns as a result of random events. They are all deviations from an underlying equilibrium. Consequently, most theorists which subscibe to this approach take empirical proof of business cycles proper with a large dose of skepticism.