Now more than ever, it’s time to think outside the Fama-French factor box

The factors in the widely-used Fama-French five-factor model¹ experienced a lost decade. Over the 2010-2019 period, these equity factors – namely: value, size, profitability and investment – delivered a negative return on average, while the return on each individual factor was well below its long-term average. However, dismissing factor investing altogether based solely on this outcome would be short-sighted.

As it turns out, the dismal performance of the Fama-French factors between 2010 and 2019 is not unprecedented. New research² by Robeco shows that, in fact, the returns of the past decade were remarkably similar to those seen previously for these four factors during the 1990-1999 period. That did not prevent the same factors from making a strong comeback during the following decade.

Moreover, we find that many time-tested alternative equity factors which are not considered in the Fama-French model did deliver a positive performance over the 2010-2019 decade. These factors include payout yield, accruals, intangibles, price momentum, analyst revisions, earnings momentum, seasonals, short-term reversal, and low risk.

From these findings, a clear dichotomy emerges: while the most commonly accepted factors experienced a lost decade from 2010 to 2019, many other factors, which the academic community often considers to be inferior or redundant, were actually the ones that delivered the highest returns during that same period.

Only time will tell if the Fama-French factors will again be able to make another comeback in the next decades

Altogether, the 2010-2019 decade turned out to be like a mirror image of the 2000-2009 decade, during which the Fama-French factors had an exceptionally strong performance and left most other factors in their wake. Only time will tell if the Fama-French factors will again be able to make another comeback in the next decades.

In the meantime, their weak recent performance will have implications for asset pricing research. For one, the five-factor model will generally have a hard time explaining strong CAPM alphas over the 2010-2019 period, as positive loadings on the Fama-French factors will not help to explain returns if the Fama-French factors themselves have no premium to begin with.

Our findings also challenge the ambition to reduce the entire ‘factor zoo’ of the hundreds of alleged equity factors reported in the academic literature to just a handful of truly relevant factors, such as the four Fama-French factors, that should explain the whole cross-section of stock returns.

Although the Fama-French factors still show a strong long-term performance, they have now experienced two lost decades during which various other factors were able to deliver. Therefore, it seems that more factors are needed for an accurate and comprehensive description of the cross-section of stock returns.