The restrictions on predictability implied by rational asset pricing models

This article shows how rational asset pricing models restrictthe regression-based criteria commonly used to measure returnpredictability. Specifically it invokes no-arbitrage argumentsto show that the intercept, slope coefficients, and R² in predictiveregressions must take specific values. These restrictions providea way to directly assess whether the predictability uncoveredusing regression analysis is consistent with rational pricing.Empirical tests reveal that the returns on the CRSP size decilesare too predictable to be compatible with a number of well-knownpricing models. However, the overall pattern of predictabilityacross these portfolios is reasonably consistent with what wewould expect under circumstances where predictability is rational.