This is the authors' abstract. We don't add key points for this paper.
This paper investigates the underlying behavioral mechanisms driving the belief heterogeneity puzzle in an emerging market, contrasting the competing hypotheses of short-term overreaction-reversal (H1) and gradual underreaction-momentum (H2). Using the sorting approach and Fama-Macbeth regression over the period from July 2007 to December 2023, our results show that the divergence in beliefs is conditional, with its magnitude varying according to market regimes and informational conditions. Our results suggest that momentum captures some of the information contained in the dispersion of beliefs, indicating that the effect of disagreement on returns operates primarily indirectly through a gradual price adjustment, consistent with the investor underreaction hypothesis. Furthermore, we show that limited investor attention acts as the primary cognitive friction enabling this phenomenon; the disagreement-momentum dynamic operates heavily in low-attention environments but completely vanishes when attention is high. More interestingly, we identify a novel, asymmetric transmission channel: the negative relationship between belief heterogeneity and subsequent returns is concentrated almost entirely within past losing stocks, representing a persistent downward trend driven by slow information integration. Finally, our temporal analysis shows that the puzzle is highly state-dependent and countercyclical, proving significant during political and macroeconomic crises but becoming entirely insignificant during stable periods or the COVID-19 pandemic.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Boujlida et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: