Shock-Independent Business Cycle Theories

Abstract

The common feature of all models in Chapter 3 is that they can generate cyclical movements only for special parameter ranges. This may have two consequences for business cycle theory. Since in most models (at least in those of the linear accelerator-type) permanent oscillations occur only for exactly one parameter constellation, these models should be treated as having a classroom character because usually no empirical evidence for the existence of these magnitudes can be found.¹ On the other hand, it is, of course, possible to concentrate on damped oscillations only and to make further assumptions on the occurrence of exogenous shocks. Aside from the argument that this does not provide a complete and consistent theory of the cycle, it should be noted that, abstracting from Hicks’ nonlinear accelerator, the solutions of all models of Chapter 3 behave in a sinusoidal manner with equal frequency and amplitude, which definitely does not fit the empirical facts. Thus a series of exogenous shocks which keeps the cycle alive is either timed in such a way that every single oscillation can fade away, or the shocks appear in such a way that the oscillations overlap in such a manner that it is impossible to trace the time path of an isolated oscillation. While this may be attractive from an empirical point of view, it is theoretically rather unsatisfactory because it leaves too much room for the arbitrary occurrence of exogenous shocks. Though it cannot be ignored that shock-dependent models of the cycle may be useful tools to describe actual cycles from time to time, it is theoretically desirable to construct models that are capable of generating persistent endogenous cycles without restrictive assumptions on singular parameter values.