Identification of Non-Rational Risk Shocks

Abstract

This paper studies how non-rational risk shocks affect the macroeconomy. Using a novel identification design which exploits survey data on expectations of financial executives in the US, I identify non-rational risk shocks via distortions in beliefs. Belief distortions are measured through surprises in beliefs of credit spreads, defined as the difference between subjective and objective forecasts. They are then used as a proxy for exogenous variation in the risk premium. Belief distortions elicit due to overreaction of credit spreads, eventually leading to exaggerated beliefs on financial markets. Results indicate that the constructed shocks have statistically and economically meaningful effects. This has sizeable consequences for the U.S. economy: A positive non-rational risk shock moves credit spreads remarkably while real activity and the stock market decline.