Keeping It Simple: The Disappearance of Premia for Standard Non-Market Factors

There is some confusion surrounding premia commanded by standard stock market factors. Some research attributes the factors to risk and some other research considers them to be a manifestation of mispricing. This note tries to resolve this confusion. My position is that the premia for factors that are priced should remain stable over time, as there is no reason for investors to stop pricing risk. Conversely, unstable premia that trend downward over time suggest mispricing being arbitraged away. The paper considers premia for five well-known factors based on accounting and market capitalization data, and two factors based on past returns, together with the excess market return. Amongst these eight standard and popular factors, only the profitability and market factors yield a reliable non-zero premium over the last 25 years. Factors based on value, size, real investment, short- and long-term reversal, and momentum, all fail to command significant non-zero premia. The evidence therefore suggests that the original in-sample premia for these factors were driven by mispricing, rather than genuine risk pricing. As time goes on, we would expect much the same to happen to the profitability anomaly. The evidence is supports adaptive market efficiency, that is, inefficiencies disappear once discovered.