Investors may be underestimating the risks of factor investing, smart beta firm Research Affiliates has warned.
Although investment factors like value and momentum theoretically provide uncorrelated, positive sources of return, the reality is that even multi-factor portfolios are at risk for “severe” drawdowns, according to a new paper from Vitali Kalesnik, Research Affiliates’ director of European research, and Juhani Linnainmaa, an advisor focusing on strategic research and product development.
In the paper, the pair argued that the forecasted behavior of single- and multi-factor strategies differed sharply from reality over the last 54 years, with factors displaying more volatility and worse losses than simulations would have suggested.
“The usual factor-investing sales presentation leaves an impression that investing in factors means almost guaranteed excess returns and that investing in a number of factors eliminates most of the risks of underperformance as a result of diversification,” Kalesnik and Linnainmaa wrote.
In reality, they argue, most factor returns are “substantially” more variable than they are projected to be, with extreme gains or losses occurring “far more frequently than investors might expect.” In addition, some factors’ future returns may be related to their past performance — meaning poor performance could be followed by more bad returns.
“Such continuation can make periods of underperformance excruciatingly painful,” they added.
Then, of course, there is the risk of data mining. “A factor may look good because it is good, or because the historical record is randomly good,” Kalesnik and Linnainmaa wrote.
[II Deep Dive: Smart Beta Is Making This Strategist Sick]
To demonstrate their argument, the Research Affiliates authors compared real and projected returns for six common factors — value, momentum, low beta, size, investment, and profitability — over the period between 1963 and 2017. They also examined the performance of a multi-factor portfolio which equally weighted those six factors.
Specifically, they compared expectations for the three worst drawdowns of a given factor or multi-factor portfolio with the three worst losses realized by that factor or factor portfolio during the 54-year period. For U.S. equities, drawdowns were more severe than expected 83 percent of the time.
Kalesnik and Linnainmaa found that while diversification through exposure to multiple factors did reduce risk, it was “far from completely” eliminated, with the multi-factor portfolio still suffering sharper-than-expected losses during its three worst drawdowns.
“If we assume correlations are constant, and that we can diversify away almost all factor risk by investing in multiple factors, we may be in for an unpleasant surprise,” the authors concluded. “Investors in the last quant crash, who were assuming low correlations among investment assets, can bear witness to the cost of such an assumption.”