Business cycles are recurring cycles of economic events involving a period of more rapid than normal or average growth (the expansionary phase) and culminating in a peak, followed by a phase of slower than average growth (a recession), or a period of negative growth (a depression) culminating in a trough. In the post-war literature, business cycles are normally assumed to be forty to sixty months in duration, and they are distinguished from various longer cycles that have been discussed in the economics literature, such as the six-to eight-year Major trade cycle, the fifteen- to twenty-five-year Kuznets or building cycle, and the fifty- to sixty-year Kondratieff wave.
Business cycles have commonly been viewed as evolving around a long-term growth trend, especially in the post-war period, and this has typically led to a divorce of ‘business cycle theory’, which attempts to explain the fluctuations around the trend, from ‘growth theory’, which attempts to explain the trend growth itself In the 1970s, interest in long waves revived, and an alternative view is that business cycles are short-term fluctuations in economic activity around longer cycles or waves. In this case, business cycles are analysed as growth cycles, with alternating rapid growth expansionary phases and slower growth contractionary phases (or recessions) during the upswing of the long wave; while during the downswing of the long wave they will involve periods of positive growth in the expansionary phase followed by periods of zero or negative growth in the contractionary phase (or depression).
There has been some debate about whether business cycles are systematic economic fluctuations, or whether they are instead purely random fluctuations in economic activity. It is certainly true that business cycles are not regular, in the sense of a sine wave with constant period and amplitude. But the weight of evidence, largely due to the accumulated studies produced through the National Bureau of Economic Research, indicates that business cycles are sufficiently uniform to warrant serious study.
Business cycle modelling in the post-war period has usually adopted the approach, suggested by the work of Frisch (1966 [1933]) and Slutsky (1937 [1927]), of regarding the economic system as fundamentally stable but being bombarded by a series of shocks or unexpected events. Economists commonly attempted to devise a ‘propagation model’ of the economy capable of converting shocks, generated by an impulse model, into a cycle. Using this strategy, many different models have been devised; these vary according to the degree of stability assumed in the ‘propagation model’, the form of the series of shocks emanating from the ‘impulse model’, and the sources of the shocks and sectors of the economy described by the propagation model. The various models have commonly involved linear stochastic second order equation systems. The linearity assumption serves both to simplify analysis and to allow an easy separation of business cycle and growth theory, because growth can be represented by a linear or log linear trend. There have been exceptions to this general modelling strategy that have used nonlinear equa-tion systems capable of generating—in the absence of shocks—self-sustaining ‘limit cycles’. These can be stable and self-repeating even in the face of shocks, which merely impart some additional irregularity. Since the late 1970s catastrophe theory and bifurcation theory have been used to develop models of the cycle in which nonlinear discontinuities play a key role. Chaos theory, which can explain the seeming irregularity of business cycles using simple nonlinearities, has also been employed. Such contributions have been relatively rare but may become more common as economists increasingly familiarize themselves with nonlinear techniques of mathematical and statistical analysis. The possibility of modelling the business cycle as a limit cycle, as an alternative to the Frisch-Slutsky approach, raises the general question of whether the business cycle is something that would die out in the absence of shocks, or whether it is endogenous to the economic system.
The nonlinear models have also commonly treated business cycles and growth theory as separable. There is, however, an alternative view, which is that business cycles and growth should be explained together, and that a theory of dynamic economic development is required. This view is most frequently associated with the work of Schumpeter (1989).
In 1975 papers by Nordhaus (1975) and Lucas (1975) discussing the political and the equilibrium theories of the business cycle had a major impact. The political theory of the business cycle argues that business cycles are in fact electoral economic cycles which result from governments manipulating the economy in order to win elections. This contrasts with the broad Keynesian consensus view of the mid-1960s that governments, through anti-cyclical demand management policies, had on the whole been successful in reducing the amplitude of the cycle, although it was accepted that at times they may have aggravated it because of the problems involved in allowing for the lag in the effect of policy interventions. The equilibrium theory of the business cycle assumes that economic agents are endowed with ‘rational expectations’ but must make decisions based on inadequate information about whether price changes are purely inflationary, so that no real response is required, or whether they indicate a profitable opportunity. In models based on this theory, systematic anti-cyclical monetary policy can have no effect, and the only contribution the government can make is to reduce the shocks to the economy by pursuing a systematic monetary policy. The equilibrium theory of the business cycle contrasts with most other theories, which view business cycles as being fundamentally a disequilibrium phenomenon. Although the political and equilibrium theories have many contrasting features, they both raise questions concerning the appropriate treatment of the government in business cycle models. The Keynesian consensus view was that the government could be treated exogenously. In contrast, the political and equilibrium theories of the business cycle indicate that the government should be treated endogenously in business cycle models. A promising new approach to business cycle modelling is the game theoretic analysis of government policy making.
Both the political and equilibrium theories of the business cycle have their origins in much earlier literature. The evidence in their support is rather less than conclusive. Nevertheless, these modern theories revived interest in the nature and causes of business cycles, and spawned a large literature on ‘real business cycles’, which are equilibrium cycles driven by real, rather than monetary, shocks. They also provoked a reconsideration of the appropriate policy responses to business cycles, based on the analysis of policy credibility and reputation, which led to support for central bank independence.
Andy Mullineux
University of Birmingham
References
Frisch, R.A.K. (1966 [1933]) ‘Propagation and impulse problems in dynamic economies’, in R.A.Gordon and L.R.Klein (eds) Readings in Business Cycles.
Lucas, R.E. Jr (1975) An equilibrium model of the business cycle’, Journal of Political Economy 83.
Nordhaus, W.D. (1975) ‘The political business cycle’, Review of Economic Studies 42.
Schumpeter, J.A. (1989) Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process, Philadelphia, PA.
Slutsky, E. (1937 [1927]) ‘The summation of random causes as the source of cyclic processes’ (in Russian), Problems of Economic Conditions 3(1), Moscow. (Revised English version, Econometrica April.)
Further reading
Dore, M.H.Z. (1993) The Macroeconomics of Business Cycles, Oxford.
