No. 140: Heterogeneous Priors and Information Choices: Evidence from Sell-Side Financial Analysts
Abstract
We examine the effect of heterogeneous priors on economic agents’ information choices and their posterior beliefs variance. Drawing upon a Bayesian updating framework, we propose that agents acquire information consistent with their biased priors, which in turn manifests in their posterior beliefs. Exploiting location- and time-specific variation in priors stemming from mood shifts after the Super Bowl, we find that positive mood increases information acquisition, whereas negative mood leads to negatively skewed information choices and less receptiveness of new information. We further show that heterogeneity in priors weighting mechanism amplifies the contrary bias in analysts’ posterior beliefs dependent on the outcome of the Super Bowl. Overall, our study provides new evidence on the mechanism of how behavioral biases lead to disagreement among economic agents.