Superstars or Average Joes? A Replication-Based Performance Evaluation Of 1917 Individual Hedge Funds

By: Harry Kat & Helder Palaro, Cass Business School, City University London
Published: February 3, 2006

This paper is the third and final installment of our “Harry Kat Trilogy” this week.  In it, Kat & Palaro apply their methodology to individual hedge funds (as opposed to funds of funds as they did in earlier research).

It is quite apparent from their conclusions why they subsequently embarked on a quest to replicate hedge funds:

“…we find that no more than 17.7% of the hedge funds in our sample beat the benchmark. In other words, the majority of hedge funds have not provided their investors with returns, which they could not have generated themselves by mechanically trading S&P 500, T-bond and Eurodollar futures. Over time, we observe a substantial deterioration in overall hedge fund performance. In addition, we find a tendency for the performance of successful funds to deteriorate over time, which supports the hypothesis that increasing assets under management endanger future performance.”

The paper covers a lot of ground from “Who Needs Hedge Funds?”, the duo’s earlier treatise, and then launches into the results of applying the “KP” methodology to nearly 2,000 individual hedge funds.

Lacking any intuitive rationale for the success of KP’s approach, the lower-order intellects at AllAboutAlpha.com have had to trust Kat & Palaro so far when they say:

“From the theory of dynamic trading it is well known that in the standard theoretical model with complete markets any payoff function can be hedged perfectly.”

But this paper offers us a glimmer of hope.  Buried in the detail is the following statement:

“The idea behind the KP measure is that in the longer run investors receive a return that is fair compensation for the bottom-line risk that they have taken, irrespective of how that risk profile is obtained.”

And this really sums up the KP approach for us.  While the media seems to interpret the approach as a way of cheaply replicating the “outperformance” of hedge funds, KP are actually arguing that hedge fund investors are compensated (only) appropriately for the risk they assume.  In other words, the market (e.g. the futures market) already prices the return profiles of hedge funds in an efficient manner.  So why buy these return distributions in expensive hedge fund packaging?

The authors use their “KP Measure” (over 100 = hedge fund is efficient, under 100 = hedge fund can be replicated more efficiently by a dynamic trading strategy) to determine if there has been a recent deterioration of hedge fund results.  They conclude:

“Over all funds, the average over the first period (i.e. the first half of their lives) was 100.56, while over the second, more recent, period the average was only 98.28. This clearly indicates a substantial deterioration in hedge fund performance over time.”

The duo admits that their pessimistic conclusions are markedly more dire than those obtained via factor analysis.  They chalk this up to the fact that it is impossible to truly know which factors to use in a multiple regression.

But as this paper points out in conclusion, neither methodology can overcome the problem of small samples sizes given the short history of the hedge fund industry.

“A fund manager may have been taking the most horrific risks, but if so far he has been lucky, the premium collected for taking on those risks will show from his track record, but the risk itself won’t.”

Read Full Paper

Be Sociable, Share!

One Comment

Leave A Reply

← Alpha as a Net Zero Sum Game: How Serious a Constraint? Who Needs Hedge Funds? A Copula-Based Technique for Hedge Fund Replication →