Short-sale constraints and the market portfolio ∗
We study how short-sale constraints on stock lending affect asset prices in an equilibrium model with multiple assets. We endow investors with heterogeneous beliefs in order to generate short selling demand. We obtain a CAPM-like equation that links asset-specific excess returns with the market equity premium. In the presence of short selling constraints in the market, the model gives rise to assetspecific alphas that are explained by both asset-specific and market-wide short-sale constraints; unconstrained stocks have higher risk-adjusted expected returns relative to the market portfolio, whereas the opposite holds for constrained stocks. In the absence of short-sale constraints, the model reduces to the standard CAPM. We test the model using extensive data on short interest and borrow fees. The model is able to empirically explain asset prices for 10 borrow-fee-sorted portfolios, as opposed to CAPM and factor models which produce unexplained alphas that are significantly different from zero for some low and high borrow fee portfolios. ∗We are extremely grateful to Jonathan Berk, Shai Bernstein, Nick Bloom, Svetlana Bryzgalova, and Isaac Sorkin for helpful conversations and suggestions, as well as to Kevin Mak for practical insights on the institutional details of the short selling market. All errors are ours. †Department of Economics, Stanford University, 579 Serra Mall, Stanford, CA 94305. Email: adugalic@stanford.edu. Website: http://www.stanford.edu/∼adugalic. ‡Department of Economics, Stanford University, 579 Serra Mall, Stanford, CA 94305. Email: dtorresp@stanford.edu.