Alternative Viewpoints: Sustainable Hedge Fund Performance

Every year, pure random chance dictates that exactly half of all investors will outperform the median and half underperform the median.  The Holy Grail of alpha generation, of course, is to outperform more than pure random chance should allow.  In other words, to produce persistent alpha.

In our monthly column featuring the thoughts of a member of the Chartered Alternative Investment Analyst (CAIA) Association, we feature one academic who may have identified a way to uncover such non-random outperformance.  Daniel Capocci, Ph.D., CAIA, is a senior portfolio manager at KBL European Private Bankers, a lecturer at the Luxembourg School of Finance and a Research Associate at the Edhec Risk & Asset Management Center.

Alternative Viewpoints, powered by CAIA

Special to AllAboutAlpha.com by: Dr. Daniel Capocci, CAIA, KBL European Private Bankers

Three fields exist that examine hedge fund performance. The first includes studies that compare the performance of hedge funds with equity and other indices (some authors conclude that hedge funds are able to outperform these indices, whereas others are more cautious in their conclusions).

The second field of hedge fund performance analysis compares the performance of hedge funds with that of mutual funds (where some have found that hedge funds constantly obtain superior performance to mutual funds, although lower and more volatile returns than the reference market indices considered.)

Finally, the third group of hedge fund performance analysis examines the persistence of hedge fund returns.  Persistence is particularly important in the case of hedge funds because the hedge fund industry has a higher attrition rate than mutual funds.

The results have been mixed.  Some researchers have found that offshore hedge funds have positive risk-adjusted returns but attribute this result to style effect and conclude that there is no evidence of particular skill.  Others find evidence of persistence in hedge fund performance, particularly for poor performing funds that continue to underperform.

In a study I co-authored for the Journal of Financial Management last year (earlier version here), we adapted a multi-factor model to identify hedge funds that consistently beat traditional markets.  We tested various ways of classifying funds based not only on past performance, but also past volatility, the Sharpe ratio, alpha, beta, skewness, kurtosis and an “adapted” Sharpe ratio in order to identify funds that consistently outperform the classical markets.

Our multi-factor model included 10 linear factors (US equities, MSCI World, Fama/French size and value factors, a momentum factor, high yield debt, global bond, emerging market debt, a commodity factor and a currency factor).  We also added four non-linear factors to a second version of the model (at-the-money and out-of-the-money puts and calls).

What we found: Factors driving hedge fund returns

Overall, we found that every strategy but macro funds, option strategy funds, no strategy funds and funds of funds created significant alpha at the 5% significance level.  Still, almost all strategies were significantly exposed to the US equity market (inversely for short sellers) with a beta ranging from -0.25 for short sellers to 0.66 for sector funds.

We found that the Fama/French size factor was significantly positive for every strategy except option arbitrage funds and short sellers – suggesting that many hedge funds profit from the small companies’ out-performance (over the period under analysis) by simply being small cap biased.

Option arbitrage strategies, on the other hand, were biased towards large companies.  This is perfectly understandable when you consider that this strategy tends to focus on the S&P 500.

Interestingly, the momentum factor was significantly positive for most strategies -indicating that many hedge funds tend to be momentum players.

For most strategies the alpha and the exposures to the factors remained the same when we added the non-linear (option) factors.  But when we added these factors to the global emerging markets funds, the exposure to the US equity market disappears and the exposure to the at-the-money call option became significant.  Interestingly, this result indicates that the returns offered by emerging market funds offer at-the-money call option return features.  In other words, these managers offer emerging market-like returns only when the market is going up (like a call option on developed markets).

While options factors were helpful in analyzing hedge fund returns for some specific strategies like emerging markets, but they introduced serious statistical risks into the analysis (namely, “multi-colinearity”).  So we focused our study mostly on the results of the model that did not include these non-linear factors.

What we found: Return persistence in hedge funds

In order to determine if some funds consistently and significantly create alpha over time, we ranked all funds based on their total return of the prior year.  Every January, we place all funds into 10 equally weighted portfolios and ordered from highest to lowest past returns.

Interestingly, we found that the alphas were significantly positive for deciles 2 to 8 in the previous year’s performance rankings, but not for the previous year’s best and worst performing funds.

It appears that the middle-of-the-pack alpha creators were long equities, had a small cap biased and were value oriented.  Conversely, the previous year’s best performing funds were strongly exposed to small companies, and pure momentum trades with a short exposure to the USD – leading to high returns, but a lower alpha.

Volatility and persistence

We divided hedge funds into low, medium and high volatility categories, and found that low volatility funds tended to consistently and significantly outperform over time.  Indeed, low volatile funds that could not produce consistent performance tended to become more volatile and actually switch to being “medium” or “high” volatility.

We also ranked hedge funds based on an adapted Sharpe ratio to see if that measure could predict alpha persistence.  Interestingly, we found that the alpha of funds with the lower adapted Sharpe ratio tended to significantly and consistently outperform over time.  These funds tended to have a relatively low exposure to the equity market, had no momentum bets and had little or no emerging market bond exposure.

Since lower volatility funds do significantly and consistently outperform the indices we studied, volatility and returns should be considered together to identify funds that should outperform classical markets in the future.

Real world frictions

While hedge funds that can offer persistence over time are a very attractive proposition for investors, it is, unfortunately, not very easy for them to profit from this opportunity due to real world frictions.

In the best case, hedge funds offer monthly liquidity with a 30-day notice period meaning that it takes a minimum of one month to get out of a fund.  In this best case, if investors decide to close a position based on ranking early in the year, it will take them at least one month before being able to sell and maybe 2 or 3 weeks to get the cash, meaning that they cannot be invested before the end of March.

Conclusion: alpha persistence can be identified using the right measures

Previous studies have all been focused on past performance as the unique tool to analyze alpha creation.  We went one step further in decomposing hedge funds returns and analyzing the persistence in hedge fund returns.  In doing so we found a consistent, systematic way of creating pure alpha using a simple classification methodology based on basic statistics.  Funds offering the highest Sharpe score, funds with a limited volatility and/or funds with a limited exposure to the equity market consistently and significantly outperform equity and bond markets. This analysis is of particular interest because it clearly proves that some funds consistently and significantly outperform classical markets.

The opinions expressed in this guest posting are those of the author and not necessarily those of AllAboutAlpha.com.

Be Sociable, Share!

3 Comments

  1. Javi
    April 1, 2008 at 10:26 am

    “funds with a limited exposure to the equity market consistently and significantly outperform equity and bond markets”

    I don’t think that if I have a fund exploring betas in non equity/bond asset classes, I am creating alpha…


  2. Shelly Jacobs
    May 5, 2008 at 2:58 pm

    Been trying to “break” our models that have been back & forwarded tested ad nauseam, and thus far am unable to do so. IMHO it’s our risk control & adaptive money management, coupled with a periodic reoptimization followed by daily total rescreening of our 30+ algorithms that has created the robustness of our risk:reward ratio and non-leveraged returns. Just damn sustainable performance!

    Saying that, perhaps our log normal return bell curve will begin to skew if we take on too much investment, i.e., in excess of $3B, non-leveraged? Difficult to say, in practice, as it is typically the manager’s goal to accept a larger asset pool if expected lower returns, making up with the larger pool of assets to earn the management & performance incentive fee. So do we stick with $3B as a limit and take our 50% incentive fee above the benchmark, or do we simply accept unlimited client assets at the orthodox 1%/20% and make it up on the volume of assets?

    I suppose if I am able to break our adaptive models, then I’ll have my answer.


Leave A Reply

← Skeptics to hedge fund managers: Your alpha has been faked! New Study: No hedge fund bubble...but a potentially serious capacity constraint →