This conference is supported by Fuller and Thaler Asset Management and Bracebridge Capital
Persuasive communication involves not only the content but also the delivery. Yet Hu and Ma know little about the latter. This paper studies the impact of non-content delivery features in persuading financial investors, e.g., facial expressions, tone of voices, style of word choices. The reseachers use a setting in which startups pitch venture investors. Using the full pitch videos as the main data input and machine learning (ML)-based algorithms as the processing technology, Hu and Ma quantify the delivery features in three-V dimensions--visual, vocal, and verbal. Venture investors are more likely to invest in startups showing more positivity (i.e., happy, warm, passionate), even though those startups underperform conditional on funding. Investors do not seem to correctly form beliefs about startup quality based on the delivery features. Instead, these features induce biases in investment decision. Using video analysis and an experiment, the researchers show that these biases can be explained by a taste-based channel (18 percent) and inaccurate beliefs (82 percent).
This paper was distributed as Working Paper 29048, where an updated version may be available.
Does the observed relationship between mutual fund flows and recent performance represent irrational "return chasing" or rational learning about unobserved fund manager skill in the presence of decreasing returns to scale? Roussanov, Ruan, and Wei estimate a structural model of investor beliefs implicit in the fund flows and compare it with the rational Bayesian benchmark that estimated from past performance data. Their estimates imply that investors are more optimistic about fund manager's average skill level than warranted by the historical data. They over-weight recent performance in a manner consistent with models based on the representativeness heuristic, yet respond slowly to changes in these beliefs, consistent with limited attention and/or informational frictions. Flows to retail funds imply more strongly biased beliefs than those to institutional funds.
This paper analyzes how limits to the complexity of statistical models used by market participants can shape asset prices. Molavi, Tahbaz-Salehi, and Vedolin consider an economy in which agents can only entertain models with at most k factors, where k may be distinct from the true number of factors that drive the economy's fundamentals. Molavi, Tahbaz-Salehi, and Vedolin first characterize the implications of the resulting departure from rational expectations for return predictability at various horizons. The researchers then apply their framework to two applications in asset pricing: (i) violation of uncovered interest rate parity at different horizons and (ii) momentum and reversal in equity returns.
Gabaix and Koijen develop a framework to theoretically and empirically analyze the fluctuations of the aggregate stock market. Households allocate capital to institutions, which are fairly constrained, for example operating with a mandate to maintain a fixed equity share or with moderate scope for variation. As a result, the price elasticity of demand of the aggregate stock market is small, so flows in and out of the stock market have large impacts on prices. Using the recent method of granular instrumental variables, the researchers find that investing $1 in the stock market increases the market's aggregate value by about $5. Gabaix and Koijen also show that they can trace back the time variation in the market's volatility to flows and demand shocks of different investors. Gabaix and Koijen also analyze how key parts of macro-finance change if markets are inelastic. They show how pricing kernels and general equilibrium models can be generalized to incorporate flows, which makes them amenable to use in more realistic macroeconomic models, and to policy analysis. Government purchases of equities have a large impact on prices. Corporate actions that would be neutral in a rational model, such as share buybacks, have large impacts too. Their framework allows us to give a dynamic economic structure to old and recent datasets comprising holdings and flows in various segments of the market. The mystery of apparently random movements of the stock market, hard to link to fundamentals, is replaced by the more manageable problem of understanding the determinants of flows in inelastic markets. Gabaix and Koijen delineate a research agenda that can explore a number of questions raised by this analysis, and might lead to a more concrete understanding of the origins of financial fluctuations across markets.
This paper was distributed as Working Paper 28967, where an updated version may be available.
Textbook finance theory assumes that investors strategically try to insure themselves against bad future states of the world when forming portfolios. This is a testable assumption, surveys are ideally suited to test it, and Chinco, Hartzmark, and Sussman develop a framework for doing so. Their framework combines survey experiments with field data to test this assumption as it pertains to any candidate risk factor. The researchers study consumption growth to demonstrate the approach. While participants strategically respond to changes in the mean and volatility of stock returns when forming their portfolios, there is no evidence that investors view this canonical risk factor as relevant.
A robust finding in financial markets is that riskier assets earn lower risk-adjusted returns than less risky assets in equilibrium. A number of theories have emerged to explain this phenomenon that focus on market frictions or preferences for lottery-like payoffs. A related phenomenon from betting markets is the Favorite-Longshot Bias, where returns for betting on riskier "long-shots" are lower than betting on "favorites". Moskowitz and Vasudevan synthesize the evidence from the two settings and study their joint implications. Rational theories of risk-averse investors with homogeneous beliefs predict no cross-sectional relationship between diversifiable risk and return, and therefore cannot simultaneously explain these facts. The researchers evaluate preferences versus belief-based explanations for these facts, and conclude that preferences likely play a dominant role. Specifically, their evidence points to Cumulative Prospect Theory preferences as a unifying explanation for the facts in both markets, with quantitative calibrations in the betting data matching those used to explain financial market phenomena.