C H A P T E R
2 6
Hedge
Funds
937
may be no more than compensation for illiquidity may appear to be true alpha, that is, risk-
adjusted abnormal returns.
Aragon demonstrates that hedge funds with lock-up restrictions do tend to hold less
liquid portfolios.
6
Moreover, once he controlled for lock-ups or other share restrictions
(such as redemption notice periods), the apparently positive
average alpha of those
funds turned insignificant. Aragon’s work suggests that the typical “alpha” exhibited by
hedge funds may be interpreted as an equilibrium liquidity premium rather than a sign
of stock-picking ability, in other words a “fair” reward for providing liquidity to other
investors.
One symptom of illiquid assets is serial correlation in returns. Positive serial correla-
tion means that positive returns are more likely to be followed by positive than by negative
returns. Such a pattern is often taken as an indicator of less liquid markets for the following
reason. When prices are not available because an asset is not actively traded, the hedge
fund must estimate its value to calculate net asset value and rates of return. But such proce-
dures are at best imperfect and, as demonstrated by Getmansky, Lo, and Makarov, tend to
result in serial correlation in prices as firms either smooth out their value estimates or only
gradually mark prices to true market values.
7
Positive serial correlation is therefore often
interpreted as
evidence of liquidity problems; in nearly efficient markets with frictionless
trading, we would expect serial correlation or other predictable patterns in prices to be
minimal. Most mutual funds show almost no evidence of such correlation in their returns,
and the serial correlation of the S&P 500 in most periods is just about zero.
8
Hasanhodzic and Lo find that hedge fund returns in fact exhibit significant serial cor-
relation. This suggestion of smoothed prices has two important implications. First, it lends
further support to the hypothesis that hedge funds are holding less liquid assets and that
their apparent alphas may in fact be liquidity premiums. Second, it implies that their risk-
adjusted performance measures are upward-biased, because any smoothing in the estimates
of portfolio value will reduce total volatility (increasing the Sharpe ratio) as well as covari-
ances and therefore betas with systematic factors (increasing risk-adjusted alphas). In fact,
Figure 26.2 shows that both the alphas and Sharpe ratios of the hedge fund indexes in
Table 26.3 increase with the serial correlation of returns. These results are consistent with
the fund-specific results of Hasanhodzic and Lo and suggest that price smoothing may
account for some part of the apparently superior average hedge fund performance.
Whereas Aragon focuses on the average level of liquidity, Sadka addresses the liquidity
risk of hedge funds.
9
He shows that exposure to unexpected declines in market liquidity is
an important determinant of average hedge fund returns, and that the spread in average
returns across the funds with the highest and lowest liquidity exposure may be as much
6
George O. Aragon, “Share Restrictions and Asset Pricing: Evidence from the Hedge Fund Industry,” Journal of
Financial Economics 83 (2007), pp. 33–58.
7
Mila Getmansky, Andrew W. Lo,
and Igor Makarov, “An Econometric Model of Serial Correlation and Illiquid-
ity in Hedge Fund Returns,” Journal of Financial Economics 74 (2004), pp. 529–609.
8
The 2005–2011 period, in which the serial correlation of monthly excess returns for the S&P 500 was 0.218
(see Table 26.3 ), is a striking exception to this general rule. This aberration arises from the period of the financial
crash, when the return on the S&P 500 was strongly negative in sequential months (September through November
2008, and then again in January and February of 2009). These sequences of large, consecutive negative returns
resulted in positive serial correlation over the sample period, a highly unusual outcome for the index. Note, how-
ever, that even in this period, the average serial correlation of the hedge fund indexes is nearly twice that of the
S&P 500.
9
Ronnie Sadka, “Liquidity Risk and the Cross-Section of Hedge-Fund Returns,” Journal of Financial Economics,
98 (October, 2010), 54–71.
bod61671_ch26_926-950.indd 937
bod61671_ch26_926-950.indd 937
7/25/13 2:04 AM
7/25/13 2:04 AM
Final PDF to printer
