Reconsidering the Costs of Business Cycles with Incomplete Markets

We measure the potential welfare gains from countercyclical policy in an economy with incomplete markets. In the course of conducting this measurement, we focus on two questions as central to the determination of those potential gains: (1) What is the likely effect of countercyclical policy on the nature of the income risk faced by individuals in the economy, (2) What are the likely general equilibrium effects brought about as asset prices change due to the implementation of countercyclical policies? In taking up the first question, we see it as critical to distinguish whether the main effect of countercyclical policy is to reduce directly the income risk faced by each individual or is simply to reduce the correlation across individuals in the income risk that they face. We present a model of the wage and employment risk faced by individuals over the cycle in which the levels of those risks are chosen endogenously. On the basis of that model, we argue that the main effect of countercyclical policy aimed at reducing aggregate fluctuations may be simply to remove the correlation across individuals in the unemployment risk that they face. We then use asset price data to argue that in an incomplete markets framework, the potential welfare gains from countercyclical policy are close to zero.