Assessing Macroeconomic Tail Risk

Abstract

Real GDP and industrial production in the US feature substantial tail risk. While this fact is well documented, several questions remain unanswered. Is this asymmetry driven by a specific structural shock? No. We show that the 10th percentile of the predictive growth distributions responds about three times more than the median to \emph{both} monetary policy and financial shocks. What mechanism can generate this asymmetry in the data? We discuss nonlinear VAR models and a nonlinear equilibrium model that are capable of matching our empirical findings. Furthermore, we provide empirical evidence that allows us to differentiate between two competing theories.

Keywords: Macroeconomic Risk, Shocks, Local Projections

JEL Classification: C21, C53, E30, E37