Swedroe: Shooting Down 2-And-20 Fees

In 1997, hedge fund assets amounted to only about $118 billion. By the end of 2014, they had reached roughly $2.5 trillion, having grown at a compound rate of nearly 20 percent per year.

Given this significant expansion in assets under management, the source of hedge fund returns—and whether they can be attributed to systematic risk exposures rather than managerial skills—has been a topic generating much interest in the academic community. Much of the recent literature suggests that hedge fund performance is mostly determined by alternative betas (investing in systematic factors).

If this is the case, the typical “two-and-twenty” fees (2 percent of assets plus 20 percent of any profits) charged by hedge funds should shrink as new competitors (in the form of lower-cost hedge fund “clones” or “replicators”) are created.

As another benefit, investors could then use such clones to help monitor a hedge fund’s performance and determine whether its managers are worth the fees. In other words, are returns truly the result of alpha, or do they come from exposure to common factors?

Hedge Funds And Linear Models

Kay Eichhorn-Schott, Margherita Giuzio, Sandra Paterlini and Vincent Weber, authors of the July 2015 paper “Tracking Hedge Funds Using Linear Models,” contribute to the academic literature on the sources of hedge fund performance. Using a constrained linear regression model, they investigated the exposure of a wide set of hedge fund indexes to 21 liquid asset class factors.

Their returns data covers the period April 2004 to October 2014. The table below provides a list of the asset classes/factors used in the study.

The authors also addressed the question of whether hedge fund returns can be replicated by investments in liquid asset class indexes. Following is a summary of their key findings:

• Style analysis confirms the hypothesis that alternative betas are the main driver of performance for most hedge fund strategies.
• The evidence gathered during the investigation of constrained hedge fund clones justifies an optimistic outlook for the success of synthetic hedge fund products.
• The evidence on most hedge fund indexes suggests that clones are able to capture the risk and return characteristics of hedge funds, providing similar average returns with relatively low tracking error while also exhibiting high return correlations with the respective replicated indexes.
• At least at the index level, most managers are not changing their risk exposures quickly enough to make the factor models’ efforts worthless.

The authors concluded that the statistics support the idea of replication because clones don’t underperform HFRX indexes. In fact, clones tend to achieve superior returns, suggesting that replication products can actually be good substitutes for direct hedge fund investments.

The authors noted: “Linear models perform indeed quite well since a large part of the index return variation is explained for most strategies.” (Note that nonlinear models have payoff structures such as those found in options.)

The authors added: “The empirical analysis conducted in this paper on 25 out of a total of 28 hedge fund indices confirms that alternative beta strategies still largely determine hedge fund returns.”

They further observed that, out of the 28 models, 21 of them “show an adjusted R² statistic of more than 70%. Only three strategy models (i.e. the HFRI Funds of Funds Market Defensive Index, the HFRI Macro Total and the HFRI Macro Systematic Diversified Index) have adjusted R² values smaller than 50%.”

Hedge Funds Aren’t Special

These findings are consistent with those from the August 2015 study by Mikhail Tupitsyn and Paul Lajbcygier, “Passive Hedge Funds.” They found that a large majority of hedge fund managers rely on “passive” linear risk exposures to generate their returns (in other words, many behave like alternative beta portfolios) while only 20 percent of hedge funds exhibit nonlinear exposure to systematic risk factors.

The authors also found that, on average, active (nonlinear) funds are inferior to linear (passive) funds in terms of both raw and risk-adjusted returns (they have lower Sharpe ratios). And they also have greater negative tail risk (meaning they exhibit negative skewness and excess kurtosis).

In short, the evidence that there really isn’t anything special about most hedge funds is piling up. Most of them simply provide relatively passive exposures to common systematic factors. And exposure to these common factors can now be obtained using lower-cost, passively managed vehicles.

While the securities industry often refers to these types of vehicles as “smart beta,” they are just products that provide access to alternative betas—alternatives that can provide diversification benefits to a portfolio otherwise limited to traditional long-only stock and bond portfolios.

Larry Swedroe is the director of research for The BAM Alliance, a community of more than 140 independent registered investment advisors throughout the country.