Assessing Asset Pricing Models using Revealed Preference

We propose a new method of testing asset pricing models that does not rely on prices and returns but on quantities instead. We use the capital flows into and out of mutual funds to infer which risk model mutual fund investors use. Using this metric, we find that the Capital Asset Pricing Model outperforms all other models. Over longer horizons we do find some evidence in support of the recently proposed dynamic equilibrium models. We find no evidence that investors use any of the reduced-form multi-factor models that have been proposed. The field of asset pricing is primarily concerned with the question of how the riskiness of financial assets affects their price. Despite half a century of research on the topic, the field is far from a consensus view on how to adjust for risk. The Capital Asset Pricing Model (CAPM), originally derived by Sharpe (1964), Lint-ner (1965) and Mossin (1966), remains controversial largely because beta does not appear to explain asset returns. As a result, in the years since the model was first proposed, fi-nancial economists have derived numerous extensions in an attempt to bring the model’