ICAPM and the Accruals Anomaly

Abstract

We propose new multifactor models to explain the accruals anomaly. Our baseline model represents an application of Merton's ICAPM in which the key factors represent (innovations on) the term and small-value spreads. The model shows large explanatory power for cross-sectional risking premia associated with three accruals portfolio groups. A scaled version of the model shows better performance, suggesting that accruals risk premia are related with the business cycle. Both models compare favorably with popular multifactor models used in the literature, and also perform well in pricing other important anomalies. The risk price estimates of the hedging factors are consistent with the ICAPM framework.