Examining the trades of company insiders, I find that a sale of stock at a loss is a much more negative signal about future returns than a sale of stock at a gain. I consider a range of explanations for my results and find that the evidence is most consistent with the idea that investors derive direct disutility from selling a stock at a loss. Since selling a stock at a loss is painful, an investor who sells at a loss must have particularly negative information. This result offers a novel measurement of the strength of the disposition effect.

We propose that extrapolative beliefs about earnings announcement (EA) returns may contribute to our understanding of EA return patterns. We construct a theoretically-motivated measure of extrapolative investors’ expectations based on a stock’s recent history of EA returns. We then show that this measure explains cross-sectional variation in stock returns and investor behavior around EAs. Stocks expected to have high EA returns according to our measure experience predictable increases in prices before EAs and predictable decreases afterwards. These patterns are economically significant: investors that buy (sell) a portfolio that is long firms with high recent EA returns and short firms with low recent EA returns in the pre-EA (post-EA) period earn daily five-factor abnormal returns of 16.1 bps (18.3 bps). Using individual investor trades data and a measure of institutional trading, we find that individual and institutional investors are more likely to purchase stocks with high recent EA returns, consistent with at least a subset of investors forming extrapolative beliefs about EA returns.

Recent work shows that the term structure of optimism has significant time-series variation and is, on average, upward sloping (horizon bias). We examine whether the horizon bias can explain the time-series variation in the equity term structure. We use analyst earnings forecasts to measure the degree of the horizon bias in the stock market. Consistent with the intuition from a stylized present value model, we find that periods of above-average horizon bias are associated with negative term premia, whereas periods of below-average horizon bias are associated with positive term premia. Furthermore, the horizon bias negatively predicts realized equity term premia. 

We find evidence that investors value dividends differently depending on their level of trust. Our tests indicate that investor demand for dividend-paying stocks increases as trust decreases, and that this relationship affects market values. We begin with survey evidence showing that people think accounting fraud is less likely among dividend payers, and that people with low trust are more likely to hold dividend-paying stocks. We then empirically exploit accounting fraud discoveries within a mutual fund’s portfolio as a shock to trust. In response to these shocks, we show that mutual funds tilt their portfolios toward dividend-paying stocks. This result is not explained by a shift in risk preferences, indicating that these institutional investors are seeking dividends in particular rather than stable firms that just happen to pay dividends. Finally, we provide evidence that dividend payers experience a premium in their market values relative to non-payers when their investor base becomes less trusting.

Existing research finds that investors’ returns vary with their wealth and level of sophistication. We bring a new perspective from the supply side by showing that return heterogeneity can be magnified as assets offered by the market become more complex. Using detailed account-level data, we examine the trading of B funds—complex, structured products in the Chinese market. During a three-year market cycle, the return gap between the naive and sophisticated is an order-of-magnitude greater when trading B funds than when trading simple, non-structured funds. In an event study, we confirm that this disparity is driven by differences in investors’ understanding of product complexity. 

Prior beliefs influence what is foremost in our memories and, therefore, what is representative. This can result in biased interpretations of information signals. We provide evidence of this prior-biased inference in the earnings forecasts of sell-side analysts and the macroeconomic forecasts of professional forecasters. By accounting for this bias in our analysis of forecasts, we are better able to predict future asset prices and macroeconomic outcomes. 

A growing body of evidence suggests that FOMC (Federal Open Market Committee) announcements can affect private sector beliefs about near-term macroeconomic conditions. We measure index option trader beliefs about the short-horizon implications of central bank policy using the return of short-term dividend strips around each FOMC announcement (we term this short-term dividend strip return, “SDR”). We find that SDR predicts both future firm-level earnings and firm-level earnings announcement returns. Furthermore, using analyst earnings forecasts, we provide evidence of belief underreaction to FOMC announcements. We develop a stylized framework to show how investor specialization and segmented markets can generate our empirical results. 

We analyze the equity term structure during Trump 2.0’s first 100 days to examine the narrative from the Trump administration that his policies will bring short-term pain in exchange for long-term gain. We find that the price of short-term assets (assets that entitle the owner to the dividends of the aggregate market over the near term) increased in value, whereas the aggregate market itself decreased in value. This differs from the president’s narrative. The evidence from the pre-election period suggests that investors might have initially overestimated the short- and long-term impacts of Trump’s policies.