Tail Risk and Investors' Concerns: Evidence from Brazil
38 Pages Posted: 2 Aug 2021
Date Written: July 28, 2021
I estimate tail risk for Brazil from January 2001 to July 2020 and investigate the origins of tail risk variation. The tail risk measure peaks at stock market crashes, financial crises, political shocks and disaster events such as the coronavirus pandemic. Moreover, I find that tail risk is countercyclical, has strong predictive power for market returns and negatively predicts real economic activity. In order to identify the investors' concerns associated with tail risk, I extract daily news from the largest financial newspaper in Brazil. The co-movement between news and tail risk indicates that tail risk variation is mainly driven by disaster concerns, followed by economic and government uncertainty. While economic uncertainty explains the countercyclical property of tail risk, investors only require compensation for bearing tail risk implied by disaster concerns. Similarly, tail risk negatively impacts real outcomes because of the disaster concerns that it identifies. These findings support recent models explaining asset pricing puzzles with time-varying disaster risk.
Keywords: Tail Risk, Risk-Neutral Measure, Return Predictability, Machine Learning, Text-based Analysis
JEL Classification: C14, C58, G12
Suggested Citation: Suggested Citation