Labor supply response in macroeconomic models: Assessing the empirical validity of the intertemporal labor supply response from a stochastic overlapping generations model with incomplete markets

Abstract

We evaluate the labor supply response in a stochastic overlapping generations model with incomplete markets and a non separable utility function in labor and consumption. Using a simulated panel from the model, we calculate the labor supply response to anticipated changes in wages (holding the marginal utility of wealth constant-that is, the Frisch elasticity) and to unanticipated change in wages (which describes the effect of uncertainty in labor supply responses). The model's Frisch elasticity estimate is 0.33, which is slightly higher than the empirical estimates in the earlier literature but somewhat lower than more recent estimates. The paper also shows that the borrowing constraints in the model reduce substantially the estimates of the Frisch elasticity. The labor supply response to an unanticipated change in wages is small because of large wealth effects. Having all the variables required and no measurement error, we calculate the omitted variable bias of not controlling for the level and variance (risk) of the unexpected changes in wages. Omitting both variables biases the estimates of the Frisch elasticity downward by a factor of 8; omitting measures of wage risk alone biases it by a factor of 1.4