“Overlay hedging” in funds of funds improves alpha: Edhec

Okay.  So the truth is out, hedge funds have long equity exposure.  Our back-of-the-envelope analysis of the HFRI Index last week showed that all strategies – particularly “Equity Hedge” had a positive correlation with equity markets.

So what can an investor seeking truly uncorrelated returns do about this?  After all, it’s quite possible that a hedge fund could produce alpha, but deliver it to investors with a side helping of over-priced beta.  Short bias managers, for example, are often said to produce a positive alpha even though they lose their shorts year after year.  It’s cases like this that make the term “absolute returns” a misnomer (see related post).

A new paper by the Edhec Risk and Asset Management Research Centre illustrates the ways that a fund of hedge funds can mitigate itself from these not-so-hidden factor exposures.

To begin with there’s that pesky equity exposure.  The following graph from the report shows the contribution of the MSCI World (blue line) to the volatility of the HFRI Fund of Funds Index:

Obviously, buying an off-setting futures contract could dramatically reduce this exposure.  But the exposure of a particular fund of funds to the MSCI could also fluctuate from time to time depending on the makeup of managers in the fund.  Thus, by dynamically managing this off-setting position, a funds of funds is able to mitigate for sudden changes resulting from a manger being fired or a new manager being added to the portfolio.  If the usage of passive positions to plug these temporary holes or keep things balanced during manager transitions sounds vaguely familiar, it’s because Andrew Lo suggested using hedge fund replication products for essentially the same thing (see related post).

As Edhec also points out, shoring up factor imbalances using futures can also prevent a fund of funds from having to burn any bridges with underlying managers as a result of redeeming its investment.

While this sounds like a great way to create purely uncorrelated returns, Edhec points to several drawbacks.  Chief among them is what they call “negative fee arbitrage” – paying performance fees for beta, only to hedge out that beta with off-setting futures positions.  Another is that the futures contracts would require collateral – reducing the capital available to actually invest in the underlying managers.  Then, of course, there’s the issue of the futures contract not being a perfect hedge against the equity exposure in the portfolio.  Finally, there’s the problem of buying the off-setting futures contract in a different currency than the portfolio itself.  In this case, another hedge would be required to protect against currency fluctuations.

But funds of funds aren’t just correlated with equities.  Edhec points out that many hedge funds are also “short volatility”.  In other words, they make easy money in times of stability, but can take a bath when volatility strikes.  So funds of funds also need to consider using options to balance out this annoying propensity.  The paper explains how to use variance swaps, the VIX, or everyone’s favorites, credit derivatives to create off-setting exposures that can mitigate a portfolio’s implicit “short volatility” positioning.

And it doesn’t even end there.  Edhec also suggests funds of funds consider hedging out a variety of factor exposures that are inherent in most portfolios of hedge funds.  The resulting off-set would be a multi-factor monster that simultaneously inoculates the portfolio against various alternative betas.  In other words, creating what is essentially a short position in a customized “hedge fund replication” product.   Here’s a hypothetical example Edhec proposed to make the point…

The general idea of creating a customized portfolio of futures to hedge out various exposures may sound vaguely familiar to those who have studied “distributional hedge fund replication”.  As Professor Harry Kat, the original developer of that concept has often stated, it’s true value is that it can be used to manage the unique risks in a given portfolio, not simply to “replicate” a hedge fund (see related post).  While the mechanics are obviously very different, we were struck by how hedge fund factor replication is also now being presented as a risk management tool by this paper.

The bottom line, according to Edhec:

“Overall, a hedging capability removes a significant constraint from funds of funds, thus rendering them much more flexible in terms of the types of portfolios they are able to assemble. This should have the net result of improving alpha, allowing for more unique and idiosyncratic portfolios, and for more
creative structured products.”

Be Sociable, Share!


  1. Jerome Abernathy
    October 31, 2008 at 8:48 am

    We’ve been approached by a number of groups looking to hedge their FoF portfolios. The problem is that the hedge exposures are far larger than most FoF are comfortable with taking. It’s like giving a 50% notional allocation to a single manager.

    For FoF, I think hedging can be used for incremental performance gains (100 to 300bps.) but complete hedging is more of a hybrid product.

  2. Harry M. Kat
    October 31, 2008 at 4:20 pm

    For many years, this hedging out standard risks has been used by hedge funds and funds of funds as well as sophisticated investors to create uncorrelated returns. Glad to see that news about this technique has now also reached France. An alternative approach would be to simply eliminate the funds that cause the unwanted correlation, such as equity long/short. Instead of first buying a bit of everything and then implement clumsy corrections, one might as well just buy what is needed. Isn’t that a lot easier?

  3. Lars Jaeger
    November 3, 2008 at 7:40 am

    Nice comment, Harry! For once I agree with you at least partly 🙂 The idea about hedging out beta (or alternative beta) is by no means new. Some have attached the name “portable alpha” to it. I have mentioned this idea in my 2004 book (p.150), but I am by no means the originator of that idea, either. I agree to just get rid of pure beta (or alternative beta) players is a good – and not necessarily a very innovative – idea. However, some or most HF managers are a mix of the two. So we will unformtunately not just fix the problem by pouring out the baby with the bath.

  4. Peter Urbani
    November 3, 2008 at 10:13 pm

    Hi Prof. Kat

    Its getting a little difficult to find funds that are sufficiently uncorrelated hence the need to ‘re-hedge’. We are doing something similar just using ETF’s which, although they do require the equivalent of a 50% notional allocation to a single L/S Overlay portfolio, overcome some of the weaknesses of a futures based approach. See http://content.infiniti-capital.com/docs/Infiniti%20Pure%20Alpha%20Fund%20(%20Sep%202008%20).pdf
    for further details ( Back-tested out-of-sample results. )

  5. Jean-René Giraud
    November 4, 2008 at 1:48 pm

    Interesting views on the state of research in France by Dr Harry M. Kat.

    Unfortunately for the accuracy of his remarks, not only are the authors of the paper in question not French but they are also, in addition to their role as research associate with EDHEC-Risk, operating in an internationally-renowned firm based in the US.

    So all we can say at this stage is that the hedging techniques presented in the aforementioned paper are indeed popular in the US. When it comes to the issue of dissemination of the concept in France, we’ll leave Harry the responsibility for his comments, as we have not seen so far any appropriately documented study on the question …

  6. Harry M. Kat
    November 5, 2008 at 11:52 am

    Hi JR,

    I knew that French comment would get your juices flowing! Apologies for assuming that research was done by yourselves. Didn’t realize you hire external people to do your research and then simply slap the Edhec label on. Maybe you can now reinstate my password to the Edhec website, which mysteriously got blocked shortly after I posted my earlier comment?

Leave A Reply

← Alternative Viewpoints: When the "100 year flood" really is a 100 year flood... Liquidity and Emotion in Beantown →