This conference is supported by Fuller and Thaler Asset Management and Bracebridge Capital
Cassella, Golez, Gulen, and Kelly provide empirical evidence that optimism bias increases with the forecasting horizon. They label this empirical regularity the horizon bias. The researchers document significant horizon bias in the macroeconomic expectations of professional forecasters, both in the U.S. and abroad. In their empirical setting, horizon bias is unlikely to be the result of strategic considerations, information rigidities, or common heuristic rules of belief formation. At the same time, Cassella, Golez, Gulen, and Kelly show theoretically that horizon bias can arise in theories of motivated beliefs. Moreover, following the conceptual framework of Benabou(2015), the researchers show that many theory-based drivers of motivated beliefs can help explain time-series variation in horizon bias.
Using survey forecasts, De la O and Myers find that systematic errors in expectations of long-term inflation and short-term nominal earnings growth are the main driver of prices and return puzzles for bonds and stocks. De la O and Myers demonstrate this by deriving and testing a single necessary and sufficient condition based on accounting identities. Errors in expectations of short-term inflation and long-term nominal earnings growth do not play a role in either asset market. Because of these systematic errors, real cash flow expectations closely match aggregate bond and stock prices, leaving little room for time-varying discount rates. These expectations also accurately match key return puzzles for bonds and stocks: the rejection of the expectations hypothesis and stock return predictability. These results are consistent with a simple model in which agents believe the persistences of inflation and nominal earnings growth are magnified versions of the objective persistences.
A fund manager evaluated relative to a benchmark index optimally invests a fraction of the fund's assets under management (AUM) in her benchmark, and such demand is inelastic. Using a dataset of 34 U.S. equity indices, Pavlova and Sikorskaya construct a stock-level measure of benchmarking intensity (BMI), which captures the inelastic component of fund managers' demand. The BMI of a stock is computed as the cumulative weight of the stock in all benchmarks, weighted by AUM following each benchmark. Exploiting a variation in the BMIs of stocks that transition between the Russell 1000 and Russell 2000 indices, Pavlova and Sikorskaya show that the change in BMI resulting from an index reconstitution is positively related to the size of the index effect. Their measure allows us to compute the price elasticity of demand more accurately than in the literature. Furthermore, using fund holdings around the index cutoff, the researchers present evidence of inelastic demand of active managers for stocks in their benchmarks. Finally, Pavlova and Sikorskaya confirm the prediction of their theory that stocks with higher BMIs have lower long-run returns.
Bastianello and Fontanier develop a theory of "Partial Equilibrium Thinking" (PET), whereby agents fail to understand the general equilibrium consequences of their actions when inferring information from endogenous outcomes. PET generates a two-way feedback effect between outcomes and beliefs, which can lead to arbitrarily large deviations from fundamentals. In financial markets, PET equilibrium outcomes exhibit over-reaction, excess volatility, high trading volume, and return predictability. The researchers extend their model to allow for rationality of higher-order beliefs, general forms of model misspecification, and heterogenous agents. Bastianello and Fontanier show that more sophisticated agents may contribute to greater departures from rationality. They also draw a distinction between models of misinference and models with biases in Bayesian updating, and study how these two departures from rationality interact. Misinference from mistakenly assuming the world is rational amplifies biases in Bayesian updating.
Kwon and Tang propose a model of stock price reaction to corporate news in which investors use significant past observations to evaluate new information. The model predicts that event-types with more extreme distributions of fundamental realizations experience greater overreaction to news. Using a comprehensive database of corporate news events, the researchers identify substantial heterogeneity in both reactions to corporate news and extremenesses of fundamental realizations across types of corporate events. Consistent with their model, Kwon and Tang document overreaction to more extreme event-types, such as leadership changes, mergers and acquisitions, and customer-related announcements, and underreaction to less extreme event-types such as earnings announcements. Kwon and Tang also document greater trading volume to corporate events with more extreme fundamental realizations conditional on the magnitude of the return. They calibrate the model to quantitatively fit the variation in investor overreaction and underreaction, as well as trading volume, across different event-types.
Several papers argue that financial economics faces a replication crisis because the majority of studies cannot be replicated or are the result of multiple testing of too many factors. Jensen, Kelly, and Pedersen develop and estimate a Bayesian model of factor replication, which leads to different conclusions. The majority of asset pricing factors: (1) can be replicated, (2) can be clustered into 13 themes, the majority of which are significant parts of the tangency portfolio, (3) work out-of-sample in a new large data set covering 93 countries, and (4) have evidence that is strengthened (not weakened) by the large number of observed factors.