I present a dynamic equilibrium model of investor sentiment in which investors form beliefs by overly extrapolating past returns. The key contribution of the paper is that I connect mispricing with investor sentiment by the market impact of extrapolators, and provide novel insights into the predictability of returns in the market. When their wealth level is high, extrapolators are driving the asset prices. In this case, high investor sentiment makes the current asset price overvalued, and the future asset price will decline because high investor sentiment will cool down over time. Therefore, investor sentiment negatively predicts future market returns. When the wealth level is low, investor sentiment positively predicts future returns since the market is under a price correction. I empirically test the model implications and find strong support for my results. My model also matches investor sentiment in surveys, and captures many documented patterns of boom-bust cycles in the stock market.
" Dark Matter of Finance in the Survey."
Using time series from investor expectation surveys and the Shiller tail risks survey, I show that different types of investor belief surveys contain consistent and mutually supportive information. The variations in investors' perceived tail event probability are well explained by changes in investor expectations in the surveys as much as 76%. Moreover, I show that variations in both investors' perceived expectations and left-tail probabilities can be well interpreted under one unified belief formation framework based on return extrapolation. This evidence not only reinforces the validity of investor belief survey data but also points to the usefulness of return extrapolation for understanding a broader sense of investor belief formation patterns. In addition, I also document close connections between investors' perceived tail risks in the survey and the time-varying rare disaster probabilities in the literature.
Presented at: Finance Down Under Conference 2018 (scheduled), NBER Behavioral Finance Working Group Meeting 2017; Caltech; Maastricht University; Tilburg University; University of California, Irvine; the Young Economists Symposium at Yale; the Caltech Junior Faculty Behavioral Finance Conference.
We develop a representative agent general equilibrium model with return extrapolation and recursive preferences. Our model is the first return extrapolation model that can be taken to the data in a serious way; it allows for direct comparisons with leading rational expectations models about the stock market in the literature. The model matches investors' extrapolative expectations disciplined by survey evidence. It also generates a large equity premium, low interest rate volatility, strong excess volatility and predictability for equity returns, persistence of price-dividend ratios, as well as low correlations between consumption growth and stock returns. Extrapolative beliefs generate perceived persistence in dividend and consumption growths that, under recursive preferences, serves as an important source for discount rate variation and helps explain our model predictions.