Common factor portfolios in the academic literature, such as Fama–French-style equity factor portfolios, typically take long positions in stocks with desirable characteristics, combined with short positions in stocks with undesirable characteristics. They assume that both the long and short positions contain relevant information for understanding asset prices.

The authors of this study look at the long and short dimensions of factor premiums separately to investigate each one’s contribution to portfolio performance.

The authors examine commonly studied equity factor premiums from 1963 to 2018. Specifically, they consider monthly returns for five Fama–French-style portfolios over this time period: value, momentum, profitability, investment, and low risk.

First, they break down the performance of the long legs and short legs of the five factors separately, evaluated by Sharpe ratios (i.e., volatility-adjusted returns). They then test how Sharpe ratios are affected when the factors are combined in different ways.

Next, the authors assess the extent to which equity size plays a role in their findings, noting that small caps generally have higher limits to arbitrage and that many studies have found that the factor premiums tend to be larger in the in the small-cap universe.

The authors then evaluate the findings of recent studies that argue that the value and low-risk factors are subsumed by new Fama–French factors—specifically, profitability and investment. Finally, they discuss whether costs and frictions might lead to different conclusions from the ones they have reached.

The authors find that factor premiums originate in both the long and short legs of the portfolios they study. However, they discover that most added value tends to come from the long legs. For example, the authors determine that the alpha of the long leg of each factor is positive and mostly significant at the 5% level. Once regressed against the other factors, the alphas of short legs range from zero to significantly negative, suggesting that the short legs add no additional value over and above the long legs.

Furthermore, long legs are found to offer more diversification than the short legs. When all five factors are combined, the long side’s Sharpe ratio (1.10) is clearly superior to the short leg’s (0.69).

The authors also discover that the performance of the short legs is largely subsumed by that of the long legs. The short legs offer essentially the same exposure but with lower rewards. For example, on the one hand, the alpha of the long-leg portfolio—its remaining performance after adjusting for the exposures to the short-leg portfolio—is +1.09% and statistically significant. On the other hand, the short-leg portfolio has a statistically significant negative alpha of –1.00%.

The authors find that these results apply across both the large- and small-cap universes. In addition, their conclusions apply across various regional equity markets, are found to be robust over time, and cannot be attributed to differences in tail risks.

The authors also show that the findings of recent studies that argue that the value and low-risk factors are subsumed by new profitability and investment factors are entirely driven by short-leg considerations. By contrast, the long legs of both value and low-risk factors offer premiums that cannot be explained by the long legs of other factors.

This study suggests that decomposing common factors into their long and short legs is an important part of understanding factor premiums and building efficient factor portfolios. Its key finding is that factor premiums are most attractive on the long side, especially in smaller caps. The authors state that investors may therefore efficiently capture these premiums by focusing on the long legs of factors and by using highly liquid market index futures to hedge out the market exposure.