[We develop a parsimonious model in which frictions in the labor market may turn small, continuous labor productivity declines into large drops in employment, endogenously causing disasters. Assuming one state variable and CRRA agents, we solve for prices in closed form, calibrate the model using labor market data, and show that this simple setting captures the high, countercyclical volatility and equity premium observed in the United States. Moreover, returns in our model are conditionally predicted by dividend yields. Finally, as in the data, in our setting the disasters are larger when the capital's share of income is higher.]

We develop a parsimonious model in which frictions in the labor market may turn small, continuous labor productivity declines into large drops in employment, endogenously causing disasters. Assuming one state variable and CRRA agents, we solve for prices in closed form, calibrate the model using labor market data, and show that this simple setting captures the high, countercyclical volatility and equity premium observed in the United States. Moreover, returns in our model are conditionally predicted by dividend yields. Finally, as in the data, in our setting the disasters are larger when the capital's share of income is higher.

We propose and test a simple explanation for institutional investors’ tendency to chase return trends. When investors face uncertainty about the precision of their private information, they wait for subsequent confirming news before establishing stock positions. While such news impact the stock price, at the same time they increase investors’ estimates of the precision of their information. With low information quality the latter effect dominates and causes investors to purchase the stock after confirming good news. We formalize these ideas in a simple model and test the model’s predictions on mutual funds’ stock holdings data. Using mutual funds’ past return experiences with individual stocks as a proxy for their stock-specific information quality, we find evidence for the prediction that trend chasing is more likely when information quality is low.