Information Aggregation and the False Consensus Effect

Abstract

Social psychologists have documented the false consensus effect, which refers to the tendency for people to overestimate their similarity to others. I model the false consensus effect by assuming that agents overestimate the correlation of their private signal errors. By modifying a simple rational expectations equilibrium model to incorporate a false consensus, I show that a false consensus causes the market to underreact to news - a prediction consistent with two of the biggest anomalies in asset pricing: momentum and post-earnings announcement drift. To generate new testable implications, I develop predictions in the setting of analysts' earnings forecasts. My model predicts that the likelihood of an analyst underreacting in his forecast revision is increasing in the number of analysts who issue a forecast between the analyst's earlier forecast and his revised forecast. I empirically confirm this prediction and rule out several alternative explanations for my findings.

Keywords: Information Aggregation, Market Underreaction, Analyst Underreaction, Post-Earnings Announcement Drift, Momentum

JEL Classification: D03, D8, G2