Sharpe to Fama & French: Don't Bury Beta Just Yet

Fama and French groupies beware! in his new book "Investors & Markets," William Sharpe has a warning for you: the expected out performance of value & small-cap stocks may be a pipe dream. On page 200 of his book, Sharpe concludes that "the announcement of the death of beta appears to be highly premature."

Explains Sharpe:

In recent years, some researchers have concluded that the expected returns on some factors may not depend wholly on their beta values. Prominently, Fama and French (1992) have studied the performance of portfolios of securities grouped on the basis of market capitalization and price-to-book ratios… Fama and French find that relative to their beta values, small stocks seem to have performed better than large stocks and that (value) stocks seemed to have performed better than (growth) stocks. (page 196)

Interestingly, long/short equity hedge fund managers are often accused, en masse, of also being long small cap names and short large cap names. Many long/short managers acknowledge such a bias, but point to the need for high liquidity in the short book so they can bail out in a short squeeze (or so a short squeeze can be avoided in the first place). But in effect, if not by intent, the long/short equity hedge fund community seems to be implicitly aligned with Fama and French.

In any case, Sharpe challenges Fama and French on a number of fronts. First, he suggests that the high costs of establishing and maintaining a Fama and French strategy likely mitigate any benefit:

It is possible that the costs associated with implementing the investment strategies required to obtain the returns on the Fama/French factors (”Small Minus Big” & “High Minus Low” book-to-price ratios) could easily be greater than any associated advantages. (page 198)

Second, he suggests that small cap companies, like short put positions, might just outperform in the short term because they are more likely to mess up in a future market calamity:

It could be that returns from value stocks and small stocks would be particularly poor in very poor markets. Perhaps small and downtrodden companies are more likely to fail in a serious depression than are large and profitable companies. (page 199)

The empirical record may indicate that markets are more complex than posited by the simple CAPM. But it seems highly unlikely that expected returns are unrelated to the risks of doing badly in bad times. (page 200)

Again, this argument will sound very familiar to those in the hedge fund industry since it is a key weapon in the hedge fund skeptic’s arsenal. Hedge funds, the argument goes, are essentially “short vol” (i.e. they “sell volatility”). They do this by (in effect) writing options and earning a steady premium from them. As long as volatility remains low, hedge fund managers look like heroes, generating miraculously steady returns that by definition are uncorrelated with market returns. This works, say the skeptics, until a “100 year flood” when the hedge funds have to pay the piper.

Third, Sharpe says that even if small cap stocks once had a higher risk-adjusted return than large cap stocks, this phenomenon has disappeared. And the same thing is destined to happen to value stocks, according to Sharpe.

…the superiority of small stock returns diminished substantially after 1980 following widespread attention to the phenomenon. More recently, the superiority of value stocks has been broadly publicized. If this truly reflected a market inefficiency, some future diminution might be anticipated. Methods for beating the market carry the seeds of their own destruction. (page 200)

For their part, Fama and French aren’t fans of Sharpe’s primary contribution to finance [CAPM] either. In the abstract to a 2003 paper titled "The Capital Asset Pricing Model: Theory and Evidence," the two authors delivered a somewhat backhanded compliment, followed by a firm rebuke of CAPM:

The attraction of the CAPM is its powerfully simple logic and intuitively pleasing predictions about how to measure risk and about the relation between expected return and risk. Unfortunately, perhaps because of its simplicity, the empirical record of the model is poor - poor enough to invalidate the way it is used in applications. The model’s empirical problems may reflect true failings.

Touche!