Essays on beliefs, information, and the asset market

Abstract

My dissertation consists of three related essays, aiming to understand how investors' beliefs and information influence the financial market and macroeconomy. In the first chapter, I study how investors form expectations and react to new information in the economy. I show two empirical facts: revisions of consensus forecasts of macroeconomic variables positively predict announcement day forecast errors, whereas stock market responses to revisions negatively predict announcement day returns. I develop a dynamic noisy rational expectations model with periodic macroeconomic announcements that quantitatively accounts for these findings. Under asymmetric information, average beliefs are not Bayesian: they underweight new information and positively predict subsequent errors. In addition, stock prices are partly driven by noise, and therefore negatively predict returns on announcement days when noise is revealed and the market corrects itself.

Abstract In the second chapter, which is a joint work with Kenneth Kasa and Yulei Luo, we study asset pricing implications of heterogeneous beliefs, uncertainties and information frictions. In particular, we incorporate ambiguity and information processing constraints into a model of intermediary asset pricing. Financial intermediaries are assumed to possess greater information processing capacity. Households purchase this capacity, and then delegate their investment decisions to intermediaries. As in (He and Krishnamurthy, 2012), the delegation contract is constrained by a moral hazard problem, which gives rise to a minimum capital requirement. Both households and intermediaries have a preference for robustness, reflecting ambiguity about asset returns (Hansen and Sargent (2008)). We show that ambiguity aversion tightens the capital constraint, and amplifies its effects. Detection error probabilities are used to discipline the degree of ambiguity aversion. The model can explain both the unconditional moments of asset returns and their state dependence, even with DEPs in excess of 20%.

Abstract My third chapter is a joint work with Hengjie Ai, Xuhui Pan and Lai Xu. This paper relates the time-varying information flow, in terms of periodic monetary policy announcements, and the uncertainty in individual stocks to study the cross section of equity returns. We use the expected option-implied variance reduction to measure the sensitivity of stock returns to monetary policy announcement surprises and show that monetary policy announcements require significant risk compensation in the cross section of equity returns. We present evidence that our sensitivity measure captures the exposure of stock returns with respect to growth rate expectations. We then develop a parsimonious equilibrium model in which FOMC announcements reveal the Federal Reserve's interest rate target, which affects the expected growth rate of the economy. Our model accounts for the dynamics of implied variances and the cross section of the monetary policy announcement premium realized around FOMC announcement days.