By: Tuomo Vuolteenaho, Arrowstreet Capital
Published: January 2006
Dresdner Kleinwort economist James Montier refered to this article in a recent piece. As Montier pointed out, Vuolteenaho argues that low beta stocks outperform higher beta stocks on a risk-adjusted basis (according to Montier’s own research, they may even outperform on an absolute basis). Vuolteenaho makes the logical step from this conclusion to the construction of a beta-neutral portfolio consisting of long positions in low beta names and short positions in high beta names. If this portfolio has a positive gross return, then “CAPM is C.R.A.P.” (to use Montier’s technical jargon). As this chart copied from the report shows, the CAPM seems especially stinky recently (the “late sample” = 1963-2001 while the “early sample” = 1927-1963).
What gives? Aren’t investors supposed to be compensated for taking on more risk? What causes higher beta stocks to produce essentially the same return as low beta stocks since the early 60’s? One explanation, says Vuolteenaho, is that some investors want octane, but get a rash when they have to take on actual leverage. So they clamor to buy higher beta stocks (taking on implicit leverage) and push down the expected returns from these names. The irony, of course, is that these highly volatile stocks are essentially levered to the market anyway – whether it be through financial leverage on the balance sheet or simply leverage to a certain exogenous business factor.
Vuolteenaho’s second explanation is equally as intriguing and rests on a peculiarity of mutual fund marketing. Since retail mutual funds can’t take on much leverage (by law), their managers have no choice but to buy high beta stocks if they want to take on more risk. Sure, they might also lose money when high-beta names go back down, but Vuolteenaho points out that mutual fund inflows are asymmetric – they go way up when you hit the lights out and then don’t go down as much if returns are negative (especially if the rest of the market is down anyway).
But isn’t this just a way of re-discovering Fama & French’s value and small-cap bias? No, says Vuolteenaho. In fact, he proves it by showing that his beta-neutral strategy has a positive return even when the Fama/French variables are neutralized. In addition, Vuolteenaho concludes that the playing field is only tilted toward low beta stocks for part of the time. Sometimes, it actually pays to invest in high beta stocks and short low beta stocks. He refers to this toggling as “tactical beta arbitrage”.
Tactical beta arbitrage involves going long low beta names and short high beta names (or vice versa). Since the resulting portfolio is beta neutral, it will involve a minimal amount of market risk. But the low beta stocks will outperform the high beta stocks on a risk-adjusted basis and the portfolio should appreciate, at least according to this research.
Curiously, long/short equity managers are often accused of simply being long value, short growth. But according to this research, that would actually be a very prudent strategy. Says Vuolteenaho:
“…our estimates suggest that high-beta stocks will earn a negative premium over low-beta stocks, making short positions in high-beta stocks a particularly attractive way to hedge equity risk.”