War Discourse and the Cross Section of Expected Stock Returns

Abstract

A war-related  factor model derived from textual analysis of media news reports explains the  cross section of expected stock returns. Using a semi-supervised topic model to extract discourse topics from 7,000,000 New York Times stories spanning 160 years, the war factor  predicts the cross section of returns  across  test assets derived from  both traditional and machine learning construction techniques, and spanning 138 anomalies. Our  findings are consistent with assets that are good hedges for  war risk receiving lower risk premia, or with assets that are more positively sensitive  to war prospects being more overvalued.  The return premium on the war  factor is incremental to standard effects.