Research

Small information imperfections can fundamentally distort financial market outcomes. Building on psychological foundations, I develop a behavioral model in which investors universally believe in the Efficient Market Hypothesis, termed the “Believer Expectations Equilibrium” (BEE). The model provides a unified explanation for three central asset pricing anomalies: inelastic asset demand, excess price volatility, and the idiosyncratic volatility (IVOL) anomaly (i.e., stocks with higher idiosyncratic volatility earn lower subsequent returns). It further generates novel predictions, including a U-shaped relationship between price volatility and market risk aversion, and the emergence of noise-driven instability when market risk aversion is low, which amplifies noise trader risk to arbitrarily large levels and provides an explanation for tremendous market fluctuations in the absence of fundamental shocks. These outcomes under BEE arise from a distorted relative scale between two forces: the “market learning echo” and the “learning-absent market confidence.” More generally, the relative strength of these two components critically determines the stability of any market in which agents learn from prices. Finally, the model delivers policy implications regarding optimal market transparency in the presence of mislearning.

We investigate the impacts of information characteristics on the patterns of reaction to information in equity markets. Employing earnings call transcript data, we find the asymmetric impacts of complexity and noise on belief reaction: To good news, complexity leads to overreaction and noise leads to underreaction while to bad news, complexity leads to underreaction and noise leads to overreaction. Inspired by the empirical findings, we develop a three-stage unified model of belief formation basing on Ba, Bohren and Imas (2024). By incorporating the novel attraction effect and the shared cognitive costs, our model precisely captures the asymmetric impacts of complexity and noise on reaction to information. Furthermore, our model predicts the systematic overreaction to good news and underreaction to bad news, offering a new explanation for the longstanding puzzle of asymmetric market responses to good and bad news.