Financial frictions and shocks

This paper aims to quantify the extent to which sources of economic uctuations generate in the nancial markets. First, a novel identi cation method is introduced into a Bayesian VAR model in order to identify a nancial type shock which we refer to as a `risk news' shock. We identify the risk news shock in macroeconomic time series for the US, while simultaneously identifying other standard macroeconomic shocks with the use of sign restrictions. Motivated by Barsky and Sims (2011), the identi ed risk news shock is constructed to be uncorrelated with today's volatility, but to contribute maximally to future volatility. Our VAR evidence suggests that the risk news shock is small, but yet contributes signi cantly to movements in the business cycle. Second, we estimate a standard New Keynesian DSGE model with nancial frictions, modi ed as in Christiano, Motto, and Rostagno (2010) to incorporate a risk news shock, and compare the impulse responses to those implied by the VAR. We use simulation methods to check for the suitability of our identi cation strategy. The main message of the paper is that it provides evidence that a small shock in the nancial system can cause substantial e ects on real variables. ∗University of Cambridge †Bank of England ‡Bank of England We are grateful to seminar participants at the Bank of England and the LSE for helpful discussions.