We use machine learning to construct a statistically optimal and unbiased benchmark for firms' earnings expectations. We show that analyst expectations are on average biased upwards, and that this bias exhibits substantial time-series and cross-sectional variation. On average, the bias increases...

Are stock valuation ratios mainly informative about future earnings growth or future returns? Using a variance decomposition, we find that over 70% of cross-sectional variation in price-earnings ratios is reflected in cross-sectional differences in future returns, while less than 30% is...

Using survey forecasts, we find that subjective expectations of earnings growth and price growth account for over 90% of cross-sectional variation in stock price-earnings ratios. This is largely due to expected earnings growth, which is more volatile and more correlated with P/E ratios than...

A large percentage of text in short-sell research reports pertains to accounting fraud and earnings mismanagement. Using survey cash-flow forecasts as a counterfactual, we find that investors underreact to the cash-flow news contained in short-sell reports. On average, target firms earn abnormal...