Different Opinion or Information Asymmetry: Machine-based Measure and Consequences

Abstract

We leverage machine learning to introduce belief dispersion measures to distinguish different opinion (DO) and information asymmetry (IA). Our measures align with the human-based measure and relate to economic outcomes in a manner consistent with theoretical prediction: DO negatively relates to illiquidity and volatility, while IA positively does. Moreover, IA negatively predicts the cross-section of stock returns, while DO predicts returns positively for underpriced stocks and negatively for overpriced ones.  Our findings reconcile conflicting disagree-return relations in the literature and are consistent with Atmaz and Basak (2018) model. We also show that the return predictability of DO and IA stems from their unique economic rationales, underscoring that components of disagreement can influence market equilibrium via distinct mechanisms.

Keywords: Belief Dispersion, Machine Learning, Cross-Section of Stock Returns, Mispricing, liquidity