The Contribution of Frictions to Expected Returns

Abstract

We derive a model-free formula to estimate the contribution of market frictions to expected returns (CFER) within a formal asset pricing setting. We document that properly scaled deviations from put-call parity estimate CFER accurately. We show theoretically that CFER is non-linearly related to the size of frictions, its return predictability becomes stronger over stocks subject to greater frictions, and the alpha of CFER-sorted spread portfolios can be as big as twice the round-trip transaction costs. We empirically validate these theoretical predictions, thus justifying the sizable CFER and the alphas of CFER-sorted spread portfolios, which we find among optionable stocks. Our analysis provides a friction-based theoretical explanation to previous evidence about the return predictability of deviations from put-call parity. In contrast to CFER, the relation of other measures of deviations from put-call parity with expected returns, the empirically estimated alphas and the strength of predictability cannot be established theoretically.

Keywords: Limits of arbitrage, Market frictions, Put-call parity, Return predictability

JEL Classification: C13, G10, G12, G13