Hedge funds are supposed to thrive in a down market — and 2008 certainly provided a chance for the well-paid people who run them to work some upside magic. Instead, most of these managers made huge sums of money disappear. So much for skill-based alpha. Put to the test during last fall’s sell-off, hedge funds looked almost as hapless as long-only money managers. By December many former top funds were down 40 to 50 percent on the year.
Then along came the Bernard Madoff scandal — a stark reminder that transparency actually does matter. Investors braced for billions more in losses, courtesy of fund-of-hedge-fund firms that fell prey to the disgraced financier’s alleged Ponzi scheme.
Before all this pain transpired, Michael Powell — head of alternative assets at Liverpool-based Universities Superannuation Scheme, the U.K.’s second-biggest pension fund — had been telling people that the average hedge fund manager was no wizard, an assertion he has had plenty of opportunity to repeat of late. “The number of funds that can consistently generate alpha at the net return is relatively small,” Powell says. “Where we can identify alpha, we are willing to pay an appropriate fee for it, but we do not want to pay alpha-type fees for beta-derived returns.”
Powell is quick to cite academic research that supports the view that most hedge fund returns come from beta or from “systematic alpha” — rules-based trading strategies that can be replicated in a process otherwise known as hedge fund cloning. Even when hedge funds yield alpha, he argues, their fees usually end up swallowing it.
Last March, USS acted on these convictions by announcing a $200 million commitment to State Street Global Advisors’ fledgling hedge fund replication product, Premia. SSgA, the $1.7 trillion investment management wing of Boston-based State Street Corp., launched Premia in July 2007. Along with a host of other replication offerings to hit the market during the past few years, Premia — like most clones — purports to offer hedge-fund-like returns with lower fees and less risk.
The USS-SSgA deal was the second such venture for USS, which has $44.6 billion in assets. In 2007 the pension giant put $200 million in a hedge fund clone run by Zug, Switzerland–based alternative asset manager Partners Group. But USS isn’t abandoning traditional hedge funds. Powell says the pension scheme is developing an integrated approach that combines replication strategies and single-manager funds.
Out with the two and 20, then, and in with more-reasonable charges? Maybe, but hedge fund clones are still a mostly nascent and untried sector. Total assets aren’t known, but the number is tiny compared with the $1.5 trillion hedge fund industry.
As devastating as 2008 was, it may have thrown open the door for replication because, even if hedge funds are having a tough time generating alpha, the beta they provide is still a valuable commodity. Proponents say clones can deliver it more cheaply and with greater liquidity than can hedge funds (notorious for limiting withdrawals). And some people see clones as a way to offer mom-and-pop investors a back door into hedge fund strategies.
“We think hedge fund replication is, in a sense, the democratization of hedge fund investing,” says Anthony Davidow, a senior vice president and head of distribution at IndexIQ, a Rye Brook, New York–based firm that sells a replication mutual fund. “The reason it makes a lot of sense for the retail client is, it gives them access they haven’t had historically.”
Hedge fund replication, in fact, may emerge as a serious threat to the very vehicles it seeks to re-create. By doing away with alpha fees for beta results, clones are positioned to revolutionize the hedge fund industry in the same way that index funds transformed the long-only equity business, allowing more-affordable access to the same underlying assets. But for that to happen, managers of hedge fund clones may first have to fine-tune their methods, which don’t always allow for swift adjustments to changing market conditions. This is simply because replication models are based largely on the past rather than what may be happening in the present (meltdowns, rallies, currency and credit crises). Critics note too that replication comes more from academic than real-world experience.
Although the name suggests otherwise, hedge fund replication isn’t about photocopying individual hedge funds. That would be almost impossible because hedge fund managers are so secretive and idiosyncratic. Many clones strive, in fact, to be average by using mathematical models to mimic the return profile of a benchmark hedge fund index like the HFRI funds of funds composite index, compiled by Chicago-based Hedge Fund Research. Others try to copy a particular hedge fund strategy — long-short equity or merger arbitrage, for instance.
Either way, investors get hedge-fund-like exposure while avoiding single-manager risk. And last year there were much worse places in which to invest than hedge funds, whose decline was about half that of the broad market. After finishing 2007 up 9.7 percent, the HFRI composite index was down 18.7 percent in 2008. But the Standard & Poor’s 500 index plunged 37 percent.
On both sides of the Atlantic, investors in search of clones have lots of choices. Credit Suisse, Deutsche Bank, Goldman, Sachs & Co., JPMorgan Chase & Co., Merrill Lynch & Co. and Morgan Stanley all offer replication products. Other players — in addition to Partners Group and State Street — include Cambridge, Massachusetts–based AlphaSimplex Group, Helsinki-based BlueWhite Alternative Investments and Greenwich, Connecticut–based AQR Capital Management, which recently launched a fund it bills as an improvement over earlier replication strategies.
Still, hedge fund clones remain a niche product. In early 2008 the Nice, France–based Edhec Risk and Asset Management Research Centre surveyed asset managers, institutional investors, private bankers, family offices and finance professionals about hedge fund replication. Just 15 percent of respondents said they used replication products; 30 percent said they would never do so.
Andrew Lo, the Harris & Harris Group professor of finance at the MIT Sloan School of Management and a pioneer in replication (see “Attack of the Clones,” June 2006), says he doesn’t expect resistance to persist, however, and that the replication trend is poised for exponential growth. Lo, who is also chief scientific officer and chairman of $625 million-in-assets quantitative hedge fund firm AlphaSimplex — where he manages two new replication products, ASG Laser Fund and ASG Global Alternatives Fund, with a combined $50 million in assets — says that as recently as a year ago replication was treading water because most investors were still focused on traditional hedge funds and were happy with their returns. (He started his funds in September and October.)
“The dislocation hadn’t really set in,” explains Lo, who in the late-1990s was one of the first academics to study hedge fund replication. “What’s happened over the past several months has awakened individuals and institutions to the fact that there is value to liquidity — and that scalability and diversification are probably more important than absolute return.”
Despite replication’s ties to academia, some professorial types frown on it. Harry Kat, a professor of risk management at City University London’s Cass Business School, makes a distinction between replication and “synthetic hedge funds,” something he helped develop (see related article, “A Third way? “Synthetic” Hedge funds”). Kat, former head of equity derivatives for Europe, at Bank of America’s London office, also points out that most replication products are similar.
“The problem is that they do live up to their promise — that is, they do replicate a well-diversified hedge fund index,” says Kat. “That index, however, is highly correlated with the equity market and therefore doesn’t make a suitable diversifier for most investors.”
Replication, in other words, doesn’t deliver much alpha, Kat argues.
Still, the shakeout rumbling through the hedge fund sector may make replication more attractive. Deepak Gurnani is co-head of hedge funds at Investcorp, an alternative-investment firm whose $17 billion (as of June 30, the most recent figures the firm has released) in assets include some $4 billion in hedge funds and funds of hedge funds. Investcorp had planned to launch its own replicator this year but poor market conditions scotched the idea. This is noteworthy in part because Investcorp is such an influential player, with offices in London, New York and Bahrain. After the hedge fund industry’s performance last year — and as it continues to struggle — Gurnani expects institutional investors to think twice about whether hedge fund managers are earning their fees.
“We believe that recent events are going to increase pressure on managers to show alpha — or not survive,” he says. “In that kind of environment, there should be demand for replication strategies. Why pay two and 20 if you’re not getting alpha?”
When State Street launched Premia, it did so in consultation with three academics: William Fung, visiting research professor of finance at the London Business School’s BNP Paribas Hedge Fund Centre; David Hsieh, Bank of America Professor of Finance at Duke University’s Fuqua School of Business; and finance professor Narayan Naik, director of the BNP Paribas center. Fung and Hsieh — who coined the term alternative beta — were among the first researchers to isolate the market exposures that many hedge funds share.
Fung, Hsieh and Naik run the Premia models and also do some marketing. Hsieh says institutions find that replicators are stepping into a vacuum created by the fact that hedge funds are an increasingly tough sell: “The combination of high fees, low transparency and low liquidity makes it very hard for institutional investors to justify.”
The three professors argue that alpha isn’t as common as it used to be. (Their paper, “Hedge Funds: Performance, Risk and Capital Formation,” published in the August 2008 edition of the Journal of Finance, offers support for their position.)
Premia draws on research by Fung, Hsieh and Naik that shows that several “factor exposures” explain the bulk of the return variation of most hedge fund indexes, the standard industry benchmark. To obtain the right weightings of these exposures — two equity-related, two interest rate ones and three connected to options — Premia, Hsieh says, “regresses” them against historical data (it’s similar to the linear regression analysis that Nobel Prize–winning economist William Sharpe applied to mutual funds in the early 1990s, using historical returns data to study the relationship between well-known risk factors). Premia then gains exposure to hedge fund beta by building a portfolio using liquid instruments like futures, option contracts and swaps.
SSgA vice president Bailey Bishop Jr., a senior portfolio manager in the firm’s global structured products group, says queries about Premia have poured in from Asia, Europe and the U.S. and that American foundations and endowments are interested in using it to reach their hedge fund allocation targets while they hold off on investing in single managers or funds of hedge funds. “They also see potential use for a strategy like this as a permanent allocation to one’s overall hedge fund exposure,” he notes.
Including the March commitment from USS, Premia had about $400 million in assets as of December 31. Bishop won’t discuss performance or fees, except to say that fees are significantly lower than the hedge fund standard. Replicators typically charge a management fee of 0.75 to 1.5 percent and do not charge for performance.
Like its rivals, Premia relies on hedge fund databases to feed its statistical models, and Bishop concedes that a backward-looking method is not ideal. But, he says, managers consider other factors too. “These models are not static,” he explains.
Drawing a parallel with indexed equity products, Bishop says SSgA has been gauging client interest in a “core-satellite” approach to building hedge fund portfolios. The result would be a foundation of core market returns through replication strategies, combined with returns from a select group of actual hedge funds (the satellites). And like indexed equity funds, clones offer institutional investors virtually unlimited capacity — in contrast with a traditional hedge fund, which might not be able to handle a big allocation from a state pension plan.
Merrill Lynch, recently absorbed by Bank of America, has rolled out a clone similar to Premia. The Merrill Lynch factor model — one of several replicators offered by the New York–based firm — aims to track the HFRI composite index. It does so by investing in six popular and highly liquid indexes, including the Russell 2000, the S&P 500 and the U.S. dollar index. Yonathan Epelbaum, managing director for equity derivatives at Merrill, says the underlying model performs a simple linear regression on the past two years of hedge fund data: “It’s a formula we run every month, and the formula puts out numbers, and those numbers get used.”
The Merrill replicator, which launched in June 2006, has attracted institutional and retail customers in the U.S. and abroad. It was down 13.8 percent in 2008, “doing as the hedge fund community has been doing,” Epelbaum says.
Even if clones are performing as advertised, investors may still need some persuading. In last year’s Edhec survey, most respondents agreed that the main benefits of replication are liquidity and comparatively low fees. But almost half said they remained wary of hedge fund clones. Edhec Graduate School of Business finance professor Lionel Martellini has similar misgivings after tracking the performance of many replication products. “Overall, it looks a little bit disappointing,” says Martellini, scientific director of the Edhec Risk and Asset Management Research Centre. “The data that we have tends to suggest that these products underperform their targets.”
Given that hedge funds are not an asset class but a diverse set of strategies, Martellini questions the value of replicating a broad index. “It is better to try to replicate at the strategy level,” he says.
AlphaSimplex chairman Lo concedes that replication products don’t make sense for traditional hedge fund investors who want 20 percent or 30 percent returns with commensurate risk. He says that clones should appeal instead to institutional investors seeking LIBOR-plus-300 basis points, which would come to about 5 percent annually, at current rates.
“If you’re a pension fund looking to get diversification from a portfolio that’s largely concentrated in stocks and bonds, then making a 5 or 10 percent allocation to a hedge fund beta product is a tremendously valuable way of getting exposure to those alternative asset classes,” Lo reasons.
But he cautions that building and running a replication strategy requires good judgment, economic intuition and constant vigilance: “This is one of the reasons hedge fund beta strategies, while they sound simple and appealing, at this point in time still require a fair bit of expertise to undertake.”
AQR Capital Management is nonetheless betting on hedge fund beta — but not under the formal banner of replication. The firm, which manages some $20 billion in alternative and traditional assets, in October launched what it calls its Delta strategy. Adam Berger, who heads the portfolio solutions group at AQR, says the firm avoids calling the new fund a replicator.
But that’s what it is.
Delta — for dynamic economically intuitive liquid transparency alternative — takes a bottom-up approach. Unlike top-down replicators, such as the clones offered by Merrill and Premia, which mirror the funds in a target index, Delta constructs individual strategies from scratch by trying to build diversity; for example, a merger arbitrage strategy would aim to take a position in every merger deal transpiring. AQR then rolls these strategies into a single product. “If you want a strategy like merger arbitrage or convertible arbitrage, for the most part you have to go to a hedge fund to access that risk,” Berger says. “We think we can give it to people outside a hedge fund’s two-and-20 structure.”
Berger says top-down clones hold too few assets to capture the granularity and richness of hedge funds: “In many cases, they tend to be just repackaging risks that investors already have elsewhere in their portfolios.”
Berger anticipates that Delta will have about $150 million in assets as of this month, most of it from institutional investors. The fund has two fee levels, depending on whether investors choose a lower-risk or a higher-volatility strategy. Each offers the choice of a flat rate or a management fee plus a performance fee. Berger, however, declines to disclose what they are, saying only that fees are “much lower than an investor would pay for a comparable hedge fund or certainly a fund-of-funds investment.”
He also declines to reveal returns, but says Delta was up in October and November. Hedge fund losses during that time had less to do with the strategies those funds were using than with market returns almost anybody could have gotten, he notes. “The fact that we exclude that passive market piece — which seems to dominate the hedge fund index returns — has worked in our favor these past few months,” he says.
New York–based Credit Suisse Alternative Capital is also pursuing a strategy-specific approach with its long-short equity replication index, launched last March. The first in a suite of its Alternative Index Replication offerings devoted to individual hedge fund strategies, the index came out of a consulting partnership with professors Fung, Hsieh and Naik. It was down 16.6 percent in 2008. Its benchmark, the Credit Suisse/Tremont long-short equity hedge fund index, fell 19.8 percent.
To show investors how they might fare with short exposure to the common factors behind the long-short equity sector, the firm also publishes the Credit Suisse inverse long-short equity replication index, which was up 16.7 percent in 2008. Jordan Drachman, Credit Suisse’s head of research for alternative beta strategies, says a broader hedge fund index would have required taking into account all the factors one might use to model various subindexes, an exercise that isn’t feasible, he argues. And besides, such a model may outlive its usefulness if, say, global macro displaces long-short equity as the dominant strategy in the hedge fund world.
By comparison, Drachman explains, the risk-return profile of an individual strategy remains fairly constant: “The trades the managers do are different, from time to time, but the strategies don’t change that much.”
Hsieh foresees a second generation of replicators that will adopt specific hedge fund strategies — but with different allocations than those of the average hedge fund. “This will generate returns that are less correlated to the average hedge fund, and therefore less correlated to the stock market,” he says.
Berger, who acknowledges that a firm without AQR’s resources would have trouble launching a strategy like Delta, says recent market conditions will compel investors to reconsider their commitment to their hedge fund portfolios. He thinks hedge fund beta strategies will win out because they have advantages hedge funds can’t necessarily match: better returns and low correlation to the rest of one’s portfolio, along with more transparency and comparatively reasonable fees.
“For many people, this is the wave of the future,” Berger says. “The development of market-cap-weighted stock indexes was transformative for how people invested, and hedge fund beta could prove to be the same thing.”
