Why is Cliff Asness, hedge fund manager, raising so many questions about the value of hedge funds? Ask Cliff Asness, academic.
By Hal Lux
Institutional Investor Magazine
Why is Cliff Asness, hedge fund manager, raising so many questions about the value of hedge funds? Ask Cliff Asness, academic.
ate last year a draft of an academic study began making the rounds among finance professors and investment practitioners. Titled “Do Hedge Funds Hedge?,” the paper was written in the dense verbiage of academic finance, and its conclusions were startling.
Analyzing seven years of returns from the benchmark CSFB/Tremont hedge fund index, the authors concluded that the hedge fund industry has been dramatically underestimating the correlation of its returns to the overall stock market and thus remarkably overestimating the value of these funds for investors. The bottom line: During those seven years, an off-the-shelf stock index would have been a better risk-adjusted investment, after expenses, than the index of glamorous hedge funds.
“The broad index of hedge funds and most hedge fund subcategories do not add value over this period,” the authors summed up. “Many hedge funds make bold claims. . . . In light of the results of this paper, we believe that careful scrutiny of these claims is important.”
The hedge fund industry might have been able to brush aside these conclusions as the half-baked theories of egghead academics. Except that the authors themselves run a hedge fund and boast sterling academic credentials. Clifford Asness, Robert Krail and John Liew run AQR Capital Management, a New York hedge fund with a quantitative bent whose investors include Stanford University and quantitative money manager Robert Arnott. The AQR partners met as Ph.D. students in finance at the University of Chicago in the late 1980s, then moved into the real world of money management at famed trading shop Trout Trading Co. and Goldman Sachs Asset Management. Three years ago the trio left Goldman to launch AQR, which made waves as the prototypical flashy hedge fund start-up that was able to raise a billion dollars overnight despite its limited track record.
Why would Asness and company take on their own industry? Asness pleads guilty to a kind of compulsive truth-telling. “We’re not on a crusade against hedge funds,” he says. “I think a smart investor can put together a great hedge fund portfolio. But it’s not easy. Why would it be easy? Hedge funds have been oversold as an asset class that’s a miracle.”
The AQR study, which is still being finalized for submission to financial journals, is stirring up a storm of controversy - and for good reason. Oversold or not, hedge funds, as a class, have never been more popular or more avidly sold as a panacea for investing ills. Just a decade ago there were about 2,000 hedge funds, with some $67 billion under management, according to consulting firm VAN Hedge Fund Advisors. Today more than 6,000 funds oversee $500 billion-plus in assets, and VAN Hedge estimates that assets under management, which grew by $75 billion last year, could reach $1 trillion in the next five years. Already, they have spawned a multitude of funds-of-funds, as well as a Wall Street cottage industry that caters to them with prime brokerage services such as stock lending and leverage financing.
With investment styles ranging from global macroeconomic trend plays to convertible bond arbitrage, from long-short equity plays to mortgage-backed securities trading, hedge funds are generally tax-inefficient and notoriously expensive - managers routinely pocket 1 percent of assets and 20 percent of returns. But they have become the leading edge of active money management, catering largely to the wealthy - and increasingly to huge institutional investors like the California Public Employees’ Retirement System, which plans to invest $1 billion in hedge funds over the next five years.
Their appeal is based not just on the lure of outsize returns delivered by, ostensibly, the best portfolio managers in the world, but on the additional promise that these returns will not be correlated to the overall stock market - and hence will be much less risky. In a recent study titled “Why Hedge Funds Make Sense,” Morgan Stanley’s quantitative strategies group calculates that hedge fund indexes have outperformed stock indexes on a risk-adjusted basis for the past decade. And a bulky 1998 Goldman, Sachs & Co. study showed that hedge funds, blended with other holdings in a large institutional portfolio, could boost risk-adjusted performance. But if Asness’ judgment is right, and hedge fund managers are getting most of their returns from just being in the market, it undercuts much of their appeal - to say nothing of their fee structure. By AQR’s analysis, many hedge fund managers are nothing more than extremely expensive indexers.
“It’s a very, very strong study,” says Massachusetts Institute of Technology finance professor Kenneth French, a former dissertation adviser to Asness who himself works in the money management business as a consultant to Santa Monica, California-based Dimensional Fund Advisors. “It just smells right. You already hear a lot of academics saying, ‘I wish I had written the paper.’ The only people I hear questioning it are those with a vested interest.”
But the vested interest charge cuts two ways: Some hedge fund managers and consultants insist that AQR’s partners have their own ax to grind. Critics, who think the study is flawed, argue that AQR’s founders are trying to overcome the poor returns they posted in their first year and a half and want to position themselves as a trusted voice in a crowded field.
“We don’t believe it,” says the appropriately named James Hedges IV, managing director of LJH Global Investments, which advises on $3 billion in hedge fund holdings for wealthy individuals and institutions. “People do studies to get credentials for marketing purposes. And in this case, I think it’s obvious that they are selling a quantitative strategy that claims to be superior to traditional long-short equity strategies.” Still, even managers skeptical of AQR’s motivations say objective research on hedge fund returns is sorely needed. “Hard research is long overdue,” says another fund chief.
Whatever the study does for AQR, it won’t win Asness and his colleagues many friends. “We have indirectly felt some hostility - people saying, ‘You’re whistle-blowers,’” says Asness.
But those who know him well say it’s not surprising that he would go ahead and publish something clever. “Part of him is still very much an academic. He’s intellectually honest,” says Larry Kohn, formerly head of institutional marketing at Goldman Sachs Asset Management, now a partner at executive search firm Heidrick & Struggles International. “He doesn’t care if his fellow hedge fund managers don’t like him.”
The AQR partners emphasize that their study does not discount the value of select hedge funds. They don’t have much choice on this point. After all, they’re hedge fund managers, with no intention of giving back their $875 million in assets. Some hedge funds are sound, even exceptional, investments, they say, but most are merely average, like the rest of actively managed funds. Says AQR partner Krail, “No one freaks out when people do studies that say most mutual funds don’t add value.”
AQR was in no position to critique anyone’s performance for the first year and a half of its existence.
Launched in August 1998 in the midst of the raging bull market, AQR had a miserable start, losing 35 percent of the assets in its flagship hedge fund product over the next 20 months. It was an embarrassing debut for an organization that had been touted on Wall Street as the next great quant shop - trading stocks, bonds and futures around the globe - and that had turned away money from eager investors. Asness blames an historic stock valuation bubble and some of his own mistakes for AQR’s poor start. “It was the worst time ever for a value-based stock selection system, and we were taking too much risk on the value side,” he says. “Nothing would have turned that into a good period for us, but our fund today is a lot better than it was two years ago.”
Asness, 34, had reason to take the losses personally. The trading models that run AQR are partly built on investment ideas that he first hatched as a Ph.D. candidate. A native of Roslyn Heights, Long Island, Asness is the son of a relatively innumerate Manhattan assistant-district-attorney-turned-defense-lawyer father and a mother who was a star math student at Queens College. “I get the geek skills from her,” says Asness. A mediocre high school student, he scored well enough on his SATs to enroll in the University of Pennsylvania, where he earned undergraduate degrees in economics and computer engineering. He entered the graduate program at the University of Chicago’s business school in 1988. Even at an institution that practically invented modern finance, Asness’ acumen stood out. “He’s extremely quick,” says MIT professor French. “He was one of my best, if not my best ever, at Chicago.”
Working under French and renowned finance theorist Eugene Fama, Asness researched value and momentum investing and uncovered the benefits of combining two investment styles that most people consider almost contradictory. His dedication to pure academics, however, didn’t last long. Nearly from the start of his studies at Chicago, he was working part-time at Goldman Sachs for legendary finance innovator Fischer Black. In 1992, his dissertation still unfinished, Asness took a one-year leave to trade mortgages for GSAM, where he had spent the previous summer. When the year ended, he decided to stay on Wall Street but moved to a job as head of quantitative research for GSAM. (He completed his dissertation, “Variables that Explain Stock Returns,” and received his doctorate in 1994.)
Staffing his group with fellow Chicago alums, including future AQR partners Liew and Krail, Asness again focused on finding investment strategies that combined elements of value and momentum investing, this time to help Goldman portfolio managers allocate assets among countries. His quant group also began running as much as $7 billion of institutional money, including a $560 million internal hedge fund called Global Alpha that compiled impressive gross annual returns of 74 percent in its first two years of trading stocks, currencies and bonds.
Like Asness, Liew and Krail mixed academics and trading. After a stint as a corporate finance analyst for Dean Witter Reynolds, Krail entered Chicago’s Ph.D. program in finance but dropped out to work for Trout Trading. In 1995 he joined Goldman to work with Asness on research and money management. Liew received his Ph.D. in 1995 and also went to work for Trout, developing global quantitative market-neutral stock selection strategies. In 1994 he joined Asness’ group.
Asness displayed a knack not only for racking up impressive investment returns, but also for pitching business. For example, he played a major role in winning money management mandates from GTE Corp. and Shell Oil for Goldman, recalls Kohn. Asness’ group is said to have generated $100 million in profits his final year; as a result, he reportedly was among the highest-paid nonpartners at the firm.
Witty, gregarious and extremely self-confident, Asness found punch lines everywhere. Eating at the posh Manhattan steak house Sparks just before he resigned from Goldman, Asness began to choke on a jumbo shrimp. “We thought he was kidding, because he’s a joker,” recalls Kohn. “Then he started turning blue.” After a Spark’s employee used the Heimlich maneuver to dislodge the shrimp, says Kohn, Asness “put down a $100 bill and said, ‘I hope you don’t mind if I don’t finish the rest of my meal.’”
As any Chicago Ph.D. knows, two years of even jumbo returns may be the result of nothing more than luck. This statistical verity didn’t prevent Asness from using his Goldman returns to market himself as a trading prodigy. In 1997 Asness, Liew and Krail (and a fourth, nonquant, AQR partner, David Kabiller, who runs marketing) resigned from Goldman, took office space in Rockefeller Center (coincidentally, next door to a University of Chicago fundraising office) and touted their Global Alpha performance numbers. In general, AQR seeks to combine elements of value and momentum trading. In practice, that can mean buying battered-down securities that are starting to improve or stocks that have been bid up but are now dropping. “We buy things that over the long term look cheap but are starting to get better,” says Asness. “I’m a big believer that people are either too optimistic or too pessimistic.”
As investors bid up Nasdaq stocks regardless of their valuations in 1998 and 1999, value funds famously underperformed. So did AQR. In hindsight, Asness says AQR, which like most quant funds holds very broad portfolios, should have been able to limit some of its losses. One problem, he believes, was the firm’s rigid adherence to value positions, even as investors continued to abandon traditional valuation models. In some respects, the fund was too correlated - not to the market, but to a philosophy. “We took too much risk on the value side,” says Asness. “When I look at the market, I really should not be able to figure out squat about how our portfolio did.” The losses prompted AQR to revise its models so that the firm can more quickly abandon pure value positions.
AQR runs about ten different trading strategies as part of several hedge fund and traditional institutional money management products. More than half of the firm’s assets are in its flagship AQR Absolute Return multistrategy hedge fund. The pain peaked for AQR in March 2000 when the fund hit a drawdown of 35 percent from its August 1998 launch. Then the Nasdaq bubble burst, high-valuation stocks began falling, and AQR came roaring back. AQR Absolute Return ended last year up 19 percent, compared with a 39.3 percent loss for the Nasdaq composite index and a 10.1 percent loss for the S&P 500. This year, with stock markets continuing to struggle, AQR Absolute Return is up 14 percent through April, giving it a gain of 60 percent since March 2000. But those who invested from its inception are just about flat, a disappointing return for a fund that aims to generate 20 to 25 percent returns per year.
Not surprisingly for a company that has kicked up all this commotion about stock market correlation, AQR asserts that its returns have virtually no connection to the overall market. Says Asness, “We run very close to a zero beta.”
The AQR partners left academics, but they
never stopped publishing. Even as they grew rich from trading, AQR’s founders churned out research. In the past five years, Asness and his partners have published more than a dozen academic papers.
Their new study on hedge funds seems certain to be their most controversial offering. In it, Asness and company sought to determine the percentage of hedge fund returns that could be attributed to the skill of managers and the percentage that was simply the result of owning stocks during a historic bull market. To do so, they used the CSFB/Tremont index, which tracks the monthly returns of 337 hedge funds and is widely regarded as representative of all hedge funds. Using regression analysis of monthly returns for the years 1994 through 2000, AQR initially calculated that the CSFB/Tremont index had a relatively low beta of 0.37. Beta measures how much an individual security or index moves in relation to the overall market - in this case, a 1,000-point rise in the overall market would have resulted in a comparatively modest 370-point gain for the hedge fund index. The results were especially impressive because a pure stock market index fund with a beta of 0.37 (the stock market has a beta of 1.0) would have posted annual returns 2.6 percentage points below those of the 13 percent achieved by the hedge fund index.
Long suspicious about anecdotal reports of these kinds of hedge fund returns, AQR began studying whether the reported performance of hedge funds was being inflated by their holdings of illiquid securities. When securities don’t trade much, they can make a portfolio seem less volatile and correlated. After performing a statistical study that turned up evidence that hedge fund returns were inexplicably tracking stock market returns from the previous three-month period - suggesting the presence of illiquid securities - AQR recalculated the hedge fund index beta using a standard finance technique called a “lagged beta regression.” The technique, first popularized in the 1970s by Nobel laureate and former Long-Term Capital Management partner Myron Scholes, has long been used to clean data in money management studies.
After AQR adjusted for these illiquid securities, the correlation of the hedge fund index soared, and its relative risk-adjusted performance plummeted. The revised hedge fund index beta came out to a high 0.84, meaning that hedge funds mirror most of the movement of the market. The AQR partners went on to calculate that instead of making outsize returns by buying a hedge fund index, investors would have gotten returns 4.5 percentage points better each year in a plain-vanilla market index. “When we estimate betas that take into account stale pricing, the hurdle for hedge fund performance goes way up,” says Liew. “It now looks like hedge funds lost to passive index funds.” Even some hedge fund investors acknowledge that it’s a good idea to question the pricing of many hedge fund returns. “I know billion-dollar hedge funds that take a very unrigorous approach to pricing their portfolio on an intraquarterly basis,” says one money manager who oversees several major hedge fund accounts.
Still, a number of hedge fund managers and investors who have read the study believe that AQR is overrepresenting the effect of illiquid securities - particularly because no one, including AQR, can determine exactly the level of illiquid securities held by hedge funds. “What he is saying could be true for a minority of funds,” says Ezra Mager, a partner with the fund-of-funds operation Torrey Funds who has read the AQR study. “But it is not true in the majority of funds.” Convertible bond arbitrage and private equity are two areas where interim pricing marks should be scrutinized, says Mager, but he believes that most other hedge funds hold portfolios of fairly liquid securities.
To be sure, by mixing business and academics, Asness inevitably leaves himself and his motives open to criticism. “I think his fund had a terrible start,” says a hedge fund investor. “And this study means that some people, who might not [otherwise], are going to take a second look at him.” An executive at one risk arbitrage hedge fund also notes that Asness and a couple of Harvard Business School professors are planning to launch an index fund for risk arbitrage that would offer a cheap alternative to traditional takeover funds. The study, in some respects, is marketing ammunition for such a product. Says the hedge fund official, “I think I know what game he’s playing.” (Actually, Asness believes that risk arb funds, which are not a distinct category in the study, have performed well.)
Of course, it’s implausible to think that many hedge fund managers would happily accept Asness’ conclusions. After all, he’s suggesting that, on average, most of what they do is worthless.