AN INVESTIGATION INTO THE RISK ARGUMENT FOR THE VALUE PREMIUM: A STOCHASTIC DOMINANCE APPROACH

Abstract

A considerable amount of academic literature has been devoted to the study of the value premium, a phenomenon where value stocks, defined by their low price-to-performance ratio, seem to outperform growth stocks, defined by their high price-to-performance ratio. One explanation for this phenomenon is that value stocks are riskier than growth stocks, so investors in value stocks will be compensated for the additional risk. We examine the risk argument by comparing conventional measures of risk between value and growth. Besides conventional measures, we also utilize a stochastic dominance approach which, due to its non-parametric nature, avoids the model misspecification problem of other factor models. We examine the dominance relationships in good and bad states of the market, identified by National Bureau of Economic Research data and a regime- switching model, to investigate if the superior performance of value strategies persists in times where investors are more risk-averse. Our findings seem to imply that value strategies gain higher returns than growth stocks throughout the sample period from 1963 to 2004, but the returns are not accompanied by higher risk. Even in bad times, value strategies are preferred to growth under stochastic dominance rules. These findings still hold after controlling for size and the January effect.