By SCOTT PATTERSON
THE WALL STREET JOURNAL ONLINE
January 27, 2005 9:45 a.m.
Burton Malkiel, the Princeton economist who popularized the belief that a blindfolded monkey throwing darts could pick stocks as surely as the experts, has become the target of some dart throwing himself.
Hedge-fund industry leaders are lining up against Mr. Malkiel, author of the widely read investment guide "A Random Walk Down Wall Street," and a study he co-authored that argues hedge-fund indexes, which track the returns of select groups of the funds, are artificially inflated by several percentage points.
The brewing battle, which has gone mostly unnoticed outside of academia, raises legitimate questions over how the largely unregulated hedge-fund industry uses databases and indexes to track and publicize its performance.
At stake: enormous fees -- as much as 20% of profits, in some cases -- these turbo-charged investment pools command for the promise of outsized returns.
As hedge funds become a bigger cog in the global market -- roughly $1.5 trillion is under management now, up from $400 billion in 2001 -- their performance is drawing more scrutiny. That's particularly true as the funds, once the playground of the wealthy, increasingly attract more conventional dollars from pension funds, endowments and charities.
The Securities and Exchange Commission late last year said most of the 8,000 hedge-fund firms in existence will have to register as advisers by February 2006 and open their books for periodic inspection. This will provide regulators, but not the public, an unadulterated look at performance.
Catching a Nerve
In October, Mr. Malkiel and Atanu Saha of the Analysis Group, a Boston research outfit, submitted a report to the Journal of Finance, later obtained by the press, titled "Hedge Funds: Risk and Return." The authors combed through the TASS Research database, a massive collection of more than 5,500 hedge funds and managers dating back to the early 1990s. The database is owned by Tremont Capital Management Inc., a Rye, N.Y., hedge-fund firm offering research, advice and other services.
The authors concluded that hedge-fund indexes, such as those compiled by Tremont and Van Hedge Fund Advisors International, are skewed upward by built-in "biases" -- nothing short of a shot across the bow.
Industry defenders are firing back. Tremont and Van's chief complaint is that the study confuses indexes with databases. Roughly two-thirds of all hedge funds report to one or more databases such as Tremont's, according to a recent study by Strategic Financial Solutions, a Memphis-based developer of software. But while similar, databases in no way are used to compile indexes.
Robert Shulman, co-chief executive at Tremont, called Mr. Malkiel "an academician who does not fully understand the business he's writing about."
Here's how it works: Historical results in a database are frequently tinkered with to reflect the turnover of funds and managers that report to it. A fund launched in 1998 that begins reporting to a database in 2000 will include performance going back to the fund's inception. That's not true of an index. The historical returns of the Dow Jones Industrial Average, for example, weren't altered when Intel Corp. was added to the list of 30 components. The impact of Intel's stock performance on the Dow average is tracked only going forward.
William Goetzmann, a business professor at the Yale School of Management and co-author of several papers on hedge-fund performance, said the Malkiel-Saha paper's real flaw is its tendency to make broad charges about the industry after studying only a single database. The authors should have sought to "understand the methodology used by specific vendors before suggesting that industry measures of hedge-fund performance are flawed," he said.
ON MALKIEL
Malkiel-Saha: We conclude that hedge funds are far riskier and provide much lower returns than is commonly supposed.
Van Rebuttal: Evidence shows the mutual fund industry is becoming increasingly concerned about the rise of [hedge funds]. Malkiel also is a proponent of the proposition that you can't beat general market indexes -- which [hedge funds] have. In fact, he has built his career on this proposition.
Mr. Saha said this road has already been traveled. The TASS database is one of the most comprehensive sets of data on the industry and has been used in numerous studies of hedge-fund performance in the past, he countered.
The animosity has been gathering steam. In December, Van Hedge Fund, based in Nashville, Tenn., published a 37-page rebuttal accusing Mr. Malkiel, a Wall Street icon, of ideological biases. The rebuttal charged, "As a professional associated with the mutual-fund industry … it comes as no surprise that [Mr. Malkiel's] latest paper portrays" hedge-fund indexes in a negative light.
Mr. Malkiel is "tarring and feathering the whole industry" because of his alliance to mutual-fund providers such as Vanguard, George Van, chairman and founder of Van Hedge Fund, said in an interview. "They are extrapolating their logic from that database that is not representative of the industry."
Mr. Malkiel, a director at Vanguard since 1977 who was paid $111,000 in 2003, the latest data available, scoffed at the charge. "I'm an academic," he said. "I'm not doing this because I happen to be a director of Vanguard."
At the heart of the dispute is the claim, made by some hedge-fund defenders, that active managers consistently beat the Standard & Poor's 500-stock index. The Van Hedge Fund Web site boasts that hedge funds provide "both superior returns and lower statistical risk than the S&P 500 or mutual funds."
The claim flies in the face of the "efficient markets" theory, championed by Mr. Malkiel, that stock prices fully reflect all public information, making it well-nigh impossible for a trader or actively managed fund to consistently beat a broad index such as the S&P 500 over the long term.
If hedge funds don't beat the broader market over time, then they have little justification for the high fees they demand, critics say. Hedge funds typically charge 2% for assets under management, in addition to taking 20% of the profits. Some mangers charge much more. Mutual funds, on the other hand, charge on average much less -- a 2% fee is at the high end of the range. Mutual-fund managers book none of the fund's profits, though their compensation often is based on performance.
"That's one of the big knocks on hedge funds; they're very expensive," said David Twibell, president of Flagship Capital Management, a portfolio-management firm in Colorado Springs, Colo., which invests in hedge funds.
That's why the study "catches a real nerve," Mr. Malkiel said. "This is a very profitable business."
Hedge-fund advocates say their product is worth the high fees because of strong historical returns, as illustrated by the indexes. From 1988 through 2004 the Van U.S. Hedge Fund Index rose 17.3%, compounded annually, compared with a 12.3% gain for the S&P 500. To be sure, those results slowed in 2004. The Van index rose just 8.4% last year versus a 9% gain for the S&P 500.
Backfills and Death Spirals
There's the rub. The Malkiel-Saha study, which is still under peer review, claims those returns are inflated, citing several flaws they found in the TASS database. Although the opponents say databases aren't used to compile indexes, Messrs. Malkiel and Saha say the only way to match reported index figures is to factor in the flaws.
One flaw, they say, is "end-of-life reporting" -- when hedge funds in a death spiral stop reporting results months before closing shop. Then there's "survivorship bias," when dead funds' results are pulled from databases. Some funds never report, such as Long Term Capital Management, which wiped out in 1998. It's as if the funds never existed, though investors still feel the impact.
Another allegation is "backfill bias." When a fund begins reporting to a database, historical results are added to provide a more comprehensive view of performance. Messrs. Malkiel and Saha found that "backfilled returns tend to be substantially higher than the contemporaneously reported ones" because funds tend to begin reporting only after a strong run. The study found that backfilled returns are on average 5% higher annually than normal returns.
The result of such distortions, among others, is a misrepresentation of the success of the entire industry, the authors argue. They conclude "hedge funds are far riskier and provide much lower returns than is commonly supposed."
Van, in its rebuttal, said it includes all funds in its database, "dead, dying or alive." That's not an issue, though, since Van keeps its database indoors. Messrs. Malkiel and Saha studied databases only available to the public.
Critics argue the study not only is off kilter because it looks at the wrong instrument, but is doubly unfair because it only focuses on factors that inflate numbers. Large, heavily capitalized hedge funds that perform extremely well often stop reporting results, too, because they no longer wish to attract clients. That can skew database and index results downward. The absence of such funds has a bigger net impact on the indexes than that of dead funds, Mr. Shulman said.
"There's a whole universe of large funds that do very well that do not report, and our research tells us that that's a significantly larger dollar volume than the funds that stop reporting because they're fading," he said. Funds that stop reporting because they are failing, on the other hand, are rarely successful enough to have gathered much capital, he said.
Messrs. Malkiel and Saha concede that they didn't look at the Van Hedge Fund method of data collection, or that of any other index or database. They argue that, using results contained in the TASS database, the only way they can produce figures that match the historical returns of the Van Hedge Fund Index, as well as several other industry indexes, is by including the discovered biases.
"If you take out the biases, you get a rate of return that is considerably lower than the returns of the various indexes," Mr. Saha said. "If you put those biases back in, then the return is very comparable to the returns contained in the various [hedge fund] indexes."
Hedge funds, including two biases found in the TASS database -- dead funds and backfill -- returned 13.14% for 1995 to 2003, according to the study. That number is close to the 13.61% return sported by the Van Hedge Fund Index for the same time period, the researchers note. Take out the biases: Compounded returns are 9.29% -- much closer to the S&P 500's return of 9.38%.
"All we are saying … is that the only way we can come up with the numbers that Van has is to not correct" for the biases, said Mr. Malkiel. That argument doesn't hold water with Mr. Van. "Sounds like academic mumbo jumbo to me," he says. "They haven't seen the inside of our database."
Write to Scott Patterson at scott.patterson@wsj.com
ON 'DEAD' FUNDS
Malkiel-Saha: Data bases available at any point in time tend to reflect the returns earned by currently existing hedge funds. They do not include the returns from hedge funds that existed at some time in the past but are presently not in existence or do still exist but no longer report their results. ... Hence, unsuccessful funds tend to close, leaving only the more successful funds in the data base.
Van: Van retains all funds in its database, dead, dying or alive, as well as in the index. We also record all reported losses in the database (and in the index should that fund be in the index). Should a fund become terminal, we seek to obtain the magnitude of the loss.
ON 'BACKFILL' BIAS
Malkiel-Saha: Managers often will establish a hedge fund with seed capital and begin reporting their results at some later date and only if the initial results are favorable. Moreover, the most favorable of the early results are then 'backfilled' into the data base along with reports of contemporaneous results.
Van: The authors appear confused between the returns of the database (which does not report returns), and the returns of the index which does report returns but does not, in a serious index like Van's, accept backfilling.
ON 'END-OF-LIFE' BIAS
Malkiel-Saha: "Hedge funds generally stop reporting their results during the last several months of their lives. For example, Long-Term Capital Management lost 92 percent of its capital between October 1997 and October 1998.
Van: In reality, this does occur, but it is so inconsequential compared to other factors that it isn't worth the effort to study it.
ON OVERALL RETURNS
Excerpts from the Malkiel-Saha study and the Van rebuttal:
Malkiel-Saha Paper: In this paper, we … carefully examine the data bases that have been used to measure hedge fund performance and estimate the magnitude of two substantial biases in the data series. We shall see that these biases are far greater than has been estimated in previous studies. … We conclude that despite the claims that they are bias free, the popular hedge fund return indexes are substantially biased upward.
Van Advisors Rebuttal: The study appears to betray a lack of understanding of the basic differences between [hedge-fund] databases and [hedge-fund] indices and their uses. It also seems to point out a general lack of knowledge of the procedures [hedge-fund] index providers use in maintaining their indices.
-- Excerpts compiled by Worth Civils