Characteristic-based Returns: Alpha or Smart Beta?

Abstract We propose new methodology to construct arbitrage portfolios by utilizing information contained in firm characteristics for both abnormal returns and betas (and, therefore, smart-beta risk premiums). Our methodology is an extension of Projected Principal Component Analysis (PPCA) and gives maximal weight to risk-based interpretations of characteristics' predictive power before any attribution to abnormal returns. The method allows the explanatory power of a characteristic for both alpha and beta to ebb and flow. This feature is particularly important when we expect that profit opportunities may be arbitraged away by investors. We apply the methodology to a large panel of U.S. stock returns from 1965–2018. Empirically, characteristics have time-varying explanatory power for both factor betas and alpha. We find the arbitrage portfolio has (statistically and economically) significant alpha and annualized Sharpe ratios ranging from 1.31 to 1.66.