Is the study of business-cycle fluctuations “scientific?”

Abstract

The study of macroeconomic fluctuations assumes that the behavior of the whole (aggregates) cannot be reduced to the sum of the parts (agents, markets). This is because interdependencies between markets can substantially amplify, or on the contrary dampen, shocks that at any time disturb the equilibrium. The understanding of general-equilibrium effects, on which direct evidence is limited, which are empirically blurred by multiple potential confounding factors, and for which controlled experiments are almost impossible to design, is necessarily more conjectural than the study of individual behavior or of a specific market. However, ignoring these effects because they do not have the same degree of empirical certainty as a directly observed microeconomic effect can lead to serious policy mistakes.