Characteristic Scaled Betas

Abstract

Recent asset pricing research has suggested that expected returns are determined not only by systematic risk factors as proposed by equilibrium pricing theories but also by non-risk characteristics such as firm size and book-to-market. In a recent study, Gomes, Kogan, and Zhang (2001) reconcile the ability of such characteristics to predict returns within a dynamic pricing paradigm. Firm characteristics can appear to predict stock returns because they may be correlated with the true conditional factor loadings, thereby motivating the scaling of betas by firm specific variables. We test whether such a scaling procedure improves the performance of the theoretically motivated CAPM and consumption CAPM. The evidence shows that equity characteristics often enter beta significantly. However, 'characteristic scaled factor models' do not outperform their unscaled counterparts. The results are robust to various specifications including different proxies for the market portfolio and using both time-series and cross-sectional regressions.