The Effects of Uncertainty on Macroeconomic Performance: The Importance of the Conditional Covariance Model.

We study the effects of growth volatility and inflation volatility on average rates of output growth and inflation for postwar U.S. data in a multivariate asymmetric GARCH-M model. Our statistical model differs from other work in that we allow the conditional covariance of inflation and growth to be both nondiagonal and asymmetric. We show that the data reject diagonality and symmetry restrictions frequently imposed in the literature. Our results on the macroeconomic effects of uncertainty also differ from those in other recent studies using a more restrictive covariance model. Specifically, we find that increased growth uncertainty is associated with significantly higher average growth, and that higher inflation uncertainty is significantly negatively correlated with lower output growth and lower average inflation.