AAA Exclusive: An interview with Prof. Harry Kat about his newest project, “super-diversification”

On October 21, we commented that Professor Harry Kat, developer of a “hedge fund replication” technique commonly called “distributional replication”, seemed to downplay his tool’s usage in replicating hedge fund returns and emphasized how it can instead be used as a risk management technique.

“He’s apparently trying to move away from a direct attack on hedge funds and is instead proposing that his dynamic trading method can and should be used to create custom return distributions to complement existing long-only portfolios.”

Well today German-based Aquila Capital, a 1.5 billion Euro manager of alternative investments, did just this.  They announced the launch of a market neutral fund that uses Kat’s technique as a risk management overlay.  According to the firm’s press release, its Statistical Value Market Neutral (SVMN) fund uses “a combination of multi-asset investing and a behaviorally driven tactical asset allocation overlay” to actually generate alpha.  Then it uses Kat & Palaro’s “FundCreator” software “to ensure a stable and predictable risk profile over time”.  Specifically, this means the fund is designed to have a volatility of 7%, a skew and kurtosis of zero and a correlation to the S&P 500 of zero.  In addition, the fund aims to deliver a maximum monthly drawdown of 4%.  The zero correlation to the S&P 500 provides what Aquila calls “super-diversification”.

An existing version of the fund (sans Kat’s risk management overlay) has been running since 2004 and has produced a Sharpe ratio of 1.0 with a low (0.1) correlation to S&P 500.  So we were curious about how FundCreator was expected to improve on (or maintain) these results.

“Who better to ask than Kat himself?” we thought.  So what follows is our exclusive interview with the man behind the new risk management approach that has caused so much debate over the past year.

AllAboutAlpha: In your words, what is the “SVMN” fund about?

Harry Kat: The basic idea behind the fund is to provide investors with a high quality, efficient diversifier. That means an investment that has low correlation with everything else, but which at the same time offers a good expected return. In the fund this is accomplished by combining the alpha generating skills of a real-life manager, with a 4-year real-life track record, with the risk management skills of FundCreator. In that way, investors get the best of two worlds: solid risk management topped up with a substantial alpha bonus.

AAA: So how exactly does Aquila Capital create this “alpha bonus”?

HK: They have a very interesting approach, which one might refer to as “behavioral harvesting of risk premia”. No fancy high-frequency trading strategies. Instead, it is based on the occurrence of cycles in risk premia.  They figure out when risk premiums are high and when they are not, and only invest in those markets where risk is properly compensated. Put simply, they only eat in restaurants that offer good value for money.  If a place gets too expensive or the quality of the food goes down, they go somewhere else and only return when prices have come back in line with what’s on offer.

The starting point is an equally weighted mix of a number of different asset classes. This is similar to the portfolio that we have always used in our FundCreator studies; a well diversified portfolio that doesn’t make too many assumptions. The manager overlays this basic portfolio with a tactical overlay based on a number of proprietary behavioral indicators.  Based on his indicators he decides whether he wants to be in an asset class or not.  He only decides whether or not he wants to be invested in an asset.  He doesn’t go short.  That would be going against the market and we all know how that works out in the longer run.  He has been at it for 4 years now and has produced a very good track record. Close to zero correlation with stocks, bonds and other hedge funds, 9% volatility and 13% average return.  He made +2.1% in January. Given the low correlations, that’s an awful lot of alpha.

AAA: But haven’t you always said that there is no such thing as alpha?

HK: What I have been saying – and I’m sticking to it – is that there is a lot less alpha around than people like to think, especially on an after-fee basis. Last year we did an update of our FundCreator-based hedge fund evaluation study, which confirmed the results of our earlier study: only 20% of hedge funds add value after fees.  Of course, we also looked at the SVMN fund manager. He came out very well and is in the top 10% of the 2000+ funds that we looked at.

AAA: Okay, so how does FundCreator fit into this picture?

HK: FundCreator is in there to provide investors with a stable, predictable risk profile.  Although so far the manager has generated a very attractive risk-return profile, there was nothing in place to actually ensure that he produced 9% vol, zero correlation, etc.  With FundCreator integrated in the fund’s daily risk management that has changed.  Now there are over 12,000 lines of C++ computer code aimed at nothing else than making sure the fund delivers the risk profile that investors are promised. That’s a big step forward from the usual case where you are promised 8% vol and 0.4 correlation with the MSCI World in the prospectus and then later find out that what you actually got was 12% vol with 0.7 correlation.

People always talk about risk budgeting, but if the risk profile of funds is completely unstable, then risk budgeting is like building a castle on quick sand, isn’t it?  With FundCreator fully integrated, the fund offers investors a clear-cut and stable risk profile through time, which means that investors can do some real risk budgeting for a change, as they now know what to expect.

AAA: Why did you choose this particular fund to partner with?

HK: Apart from the use of FundCreator, the fund has some other interesting features – all aimed at solving some of the usual problems surrounding hedge funds. First, there is no lock-up or notice period. It offers its investors daily valuation and liquidity. Investors always underestimate the importance of liquidity as it is something you don’t need every day. However, as anybody involved in one of the major hedge fund blow-ups knows too well, when you do need it, it better be there, or else.

Second, what I think is another often overlooked point, the fund is not a fresh start-up. The manager has a 4-year track record and Aquila Capital has been around for a while as well, employing 60+ people and having $2.2 billion under management.  Third, the fund is regulated under UCITS III. That puts restrictions on the strategy, but it also means that the fund can be sold to retail investors throughout Europe. The fund is one of only a handful of funds that can claim that privilege.

AAA: What is your personal involvement with this fund?  How are you balancing your academic work with your new responsibilities?

HK: Of course, I’m really proud that some of my ideas are now being put into practice. Eventually, that’s what every academic dreams of.  I’m not involved in the day-to-day management of the fund though, but purely act as an academic consultant. This fund didn’t come about just like that. We’ve done many in-depth studies, for example, on how to best merge the fund’s original strategy with FundCreator, as well as the behavior of risk premia over time. Now the fund is there, this won’t stop. There are always more alternatives to develop and test. There are always new ways to look at old problems. That’s what makes finance such an exiting area to be in and that’s why we’re ahead of the game.

AAA: Professor Kat, thanks very much for your time.

Be Sociable, Share!


  1. Bill aka NO DooDahs!
    February 12, 2008 at 7:24 am

    I think I read someone suggest, some time ago in your comments section, that the best use of these “replication” tools is probably the targeting of stable, constant return streams, rather than trying to mimic an index of hedge fund returns.

    Go Kat!

Leave A Reply

← Replicating Hedge Funds: Traditional beta or alternative beta? Hedge Fund Replication: Delicacy or a bunch of baloney? →