Uncertainty Shocks in a Model with Mean-Variance Frontiers and Endogenous Technology Choices

Abstract

This paper builds a model to show how increases in aggregate uncertainty - an uncertainty shock - can generate recessions. In the model managers can alter the amount of firm-specific risk they face by choosing what stochastic Total Factor Productivity (TFP) processes to run their firm with. The set of TFP choices available to the manager lie on a positively sloped mean-variance frontier; that is, a frontier where a lower-variance, and thus lower-risk, TFP choice is also a lower-mean TFP choice. In response to an increase in aggregate uncertainty, risk-averse managers in the model act to reduce the amount of firm-specific risk they face by moving along the mean-variance frontier to a lower-risk, and thus lower-mean, TFP choice. Consequently a positive uncertainty shock in the model leads to a drop in mean TFP, which in turn implies lower production - a recession. Moreover, the model matches various stylized facts about time series and cross-sectional variations in TFP data and suggests shortcomings in using TFP data to calculate exogenous TFP shocks.