Measuring Tail Risks at High Frequency
61 Pages Posted: 1 Apr 2015 Last revised: 27 Oct 2018
Date Written: October 25, 2018
I exploit information in the cross section of bid-ask spreads to develop a new measure of extreme event risk. Spreads embed tail risk information because liquidity providers require compensation for the possibility of sharp changes in asset values. I show that simple regressions relating spreads and trading volume to factor betas recover this information and deliver high-frequency tail risk estimates for common factors in stock returns. My methodology disentangles financial and aggregate market risks during the 2007-2008 Financial Crisis; quantifies jump risks associated with Federal Open Market Committee announcements; and anticipates an extreme liquidity shock before the 2010 Flash Crash.
Keywords: Tail Risks, High-Frequency Market Making, Bid-Ask Spreads
JEL Classification: C58, G01, G12, G14, G17
Suggested Citation: Suggested Citation