To Banegas and Rosa, another key issue may be at play: the poor performance of the momentum signal itself.
Factor models set forth by academics Eugene Fama and Kenneth French are meant to have both long and short positions. Historically, about half of the momentum style’s alpha stems from the long side, and the other half from the short side. That’s the case over the period from 1927 to 2014.
Since then, however, momentum’s performance has been almost entirely driven by the short side of the portfolio, the authors found. That means chasing winners has stopped working. That happened partly because so much money has piled into the same stocks, like technology shares. And when these equities dominate benchmarks, it’s hard to gain an edge by going after them.
The distortion doesn’t bode well for funds that are restricted from making bets against stocks. Nevertheless Rosa admits that the long side of the trade’s misfiring could be temporary.
“I only looked at six years data and it may not be long enough to completely lose confidence,” he said.
--With assistance from Justina Lee.
This article was provided by Bloomberg News.