Analysts' Underreaction and Momentum Strategies

Abstract

I estimate a theory-based behavioral momentum using analysts' predictable (errors driven by) underreaction (APU) as a proxy for newswatchers underreaction. The results show that APU strongly predicts analysts' errors and, more importantly, stock returns. A long-short strategy based on APU generates a value-weighted Fama-French six-factor alpha of 0.85% per month (t-stat = 3.48). Furthermore, I propose an underreaction factor that subsumes the momentum factor in spanning tests and provide higher explanatory power for a wide range of return predictors. The results support behavioral explanations of the momentum effect and show that APU can better capture newswatchers' underreaction than traditional estimates.

Keywords: Analysts' predictable error, Analysts' forecasts, Momentum, Post-earnings announcement drift, Stock market anomaly

JEL Classification: G12, G29, M41