War Discourse and Disaster Premia: 160 Years of Evidence from the Stock Market
The Review of Financial Studies, Volume 38, Issue 2, Feb 2025, Pages 457-506
163 Pages Posted: 18 Feb 2024 Last revised: 30 Jan 2025
There are 3 versions of this paper
War Discourse and Disaster Premia: 160 Years of Evidence from the Stock Market
War Discourse and Disaster Premia: 160 Years of Evidence from Stock and Bond Markets
War Discourse and Disaster Premia: 160 Years of Evidence from Stock and Bond Markets
Date Written: February 1, 2024
Abstract
Using a semi-supervised topic model on 7,000,000 New York Times articles spanning 160 years, we test whether topics of media discourse predict future stock market excess returns to test rational and behavioral hypotheses about market valuation of disaster risk. Media discourse data address the challenge of sample size even when disasters are rare. Our methodology avoids look-ahead bias and addresses semantic shifts. Our discourse topics positively predict market excess returns, with War having an out-of-sample R2 of 1.35%. We call this effect the war return premium. The war return premium has increased in more recent time periods.
Keywords: War, Topic Modelling, Machine Learning, Predictability, Disaster Risk
JEL Classification: G0, G1, G2
Suggested Citation: Suggested Citation