Print this article | Return to Article | Return to

A Hedge-Fund Mystery

Why did a number of equity hedge funds suffer big losses in August 2007 — and what do those losses say about systemic risk?
Edward Teach, CFO Magazine
January 1, 2008

Last August was a bad month for the hedge-fund industry. According to fund tracker Barclay Hedge Ltd., about 75 percent of the 2,600 or so hedge funds that reported results for the month showed a loss, of about 1.4 percent on average. The week of August 6 was particularly cruel to quantitative equity hedge funds. A number of prominent "quants," managed by the likes of Goldman Sachs and Renaissance Technologies, suffered losses reportedly ranging from 5 percent to more than 30 percent.

What happened to the quants? The question forms the title of a recent article by two researchers from the Massachusetts Institute of Technology — Amir E. Khandani, a graduate student; and Andrew W. Lo, a well-known professor of finance at MIT's Sloan School of Management and founder and chief scientific officer of AlphaSimplex Group, a hedge fund. Like a pair of detectives, working with indirect clues (hedge funds are famously secretive) and an investment simulation, Khandani and Lo set out to discover why the quants lost so much money.

Their conclusions, they stress, must be regarded as speculative, since the people who know what happened — the hedge-fund managers — aren't talking. Still, Khandani and Lo's diagnosis is entirely plausible, and raises disturbing questions about systemic risk in the hedge-fund industry.

Suspecting a Liquidation
Systemic risk here refers to the possibility that huge, correlated losses can rapidly occur among funds, as in a banking panic. Such correlated losses may be triggered by a single event — such as the near-collapse in August 1998 of the hedge fund Long-Term Capital Management (LTCM), which briefly menaced the stability of the global financial system.

Did a similar event spark the hedge-fund losses last August? Khandani and Lo believe so. Sometime on August 7 or 8, they theorize, one or more sizable quantitative equity market-neutral portfolios were suddenly unwound. The liquidation had a significant, if temporary, impact on equity prices. That caused other quants — many of them equity long/short funds, which hold stocks in both long and short positions — to cut their own losses and sell. What Khandani and Lo call the "perfect financial storm" ended on August 10, when the liquidations ceased and returns climbed sharply.

The steep drop and rebound, they note, were consistent with a "liquidity event" such as a portfolio sell-off. To test their theory, Khandani and Lo simulated an equity long/short strategy for the week of August 6, leveraged approximately 8 to 1. Sure enough, a fund using the test strategy would have lost more than 27 percent of its assets between August 7 and 9, and gained about 24 percent on August 10.

As for why the losses were so large, the researchers point out that investors have poured money into equity long/short funds over the past decade, making them the most popular hedge-fund category. At the same time, the profitability of the strategy has declined, due to increasing competition and technological advances. In order to maintain their accustomed level of returns, therefore, long/short funds have had to increase leverage.

The Subprime Factor
What triggered the sudden sell-off that started the loss parade? After all, as Khandani and Lo observe, "there were virtually no signs of market turmoil outside the world of quantitative equity market-neutral funds on August 7th and 8th" — no major movements in market indexes for stocks, bonds, currencies, or commodities.

The researchers discuss two possibilities. One, a fund manager or a number of managers, worried by conditions in the markets, voluntarily decided to deleverage at the beginning of August. The second possibility, which Khandani and Lo favor, is that a multistrategy fund or proprietary trading desk was forced to meet a margin call or reduce risk because of its exposure on the credit side to the subprime-mortgage market. To do so it sold its most liquid assets, stocks.

The likelihood that problems in the credit markets spilled over to a completely unrelated sector is new, say Khandani and Lo. (By contrast, in August 1998, equity long/short funds were unaffected by LTCM's fall, which similarly roiled credit markets.) A primary reason is the rise of multistrategy funds, now the second-most popular kind of hedge fund. There is so much money now in multistrategy funds that "the correlations [among hedge funds] are that much greater," Lo tells CFO. "When some kind of adverse event occurs for one kind of strategy, a multistrategy fund may well decide to decrease its exposure to other strategies in response."

As for the quants' misfortune in early August 2007, "the important conclusion is that the cascade of losses in the absence of major market news suggests that there is less liquidity, and more commonality, in a liquid category [of hedge fund]," write Khandani and Lo. If equity long/short funds use the same standard factors in their quantitative models, then they are more prone to move in lockstep.

The possibility that a large number of hedge funds in a given sector could suddenly liquidate their positions poses a significant risk to the global financial system, say Khandani and Lo.

Looking at the Black Box
If the systemic risk of the hedge fund industry is rising, and more losses lie ahead, what can be done about it?

"I think financial-market crises are an unavoidable aspect of the degree of capital-market growth and flexibility that we've enjoyed for many years," says Lo. "We shouldn't fool ourselves into thinking we can avoid them altogether." Instead, he says, "we should try to understand the nature of these crises, so we can be better prepared for them."

To that end, Lo advocates the creation of a "capital markets safety board," analogous to the National Transportation Safety Board. Just as the NTSB sends in experts to determine the cause of airplane crashes, so the Securities and Exchange Commission could dispatch a team of forensic accountants, attorneys, and financial engineers to analyze hedge-fund liquidations. "We ought to try to disseminate as best we can the underlying mechanisms of each of these hedge-fund disasters in exactly the same way that the NTSB is charged with collecting and reviewing the black boxes for airplane crashes," argues Lo.

Of course, unlike the black boxes on airplanes, the investing black boxes of hedge funds are not subject to outside review. Nevertheless, "to the extent that the hedge-fund industry becomes an important part of the U.S. economy — I argue that it's already there — it will be important to impose certain kinds of regulatory oversight on hedge funds," says Lo. (Estimates of the size of the global hedge-fund industry vary; the McKinsey Global Institute recently pegged it at $1.5 trillion in 2006. By comparison, mutual funds had assets under management of $19.3 trillion.)

Lo would like to empower the SEC to create and maintain a database of detailed information about hedge funds, such as assets under management, instruments traded, fees, monthly returns, and leverage. Moreover, he would also have the SEC collect similar information from hedge-fund counterparties, including banks. The banks, he points out, would probably be the first casualties of a systemic event involving hedge funds.

Although Lo predicts further hedge-fund losses stemming from the subprime mess, with more costly unwinding of long/short positions, he maintains an optimistic outlook for the industry and financial markets. "We have an enormous set of resources to draw on to deal with these issues," he explains. "I have a great deal of confidence in the President's Working Group on Financial Markets, in Fed chairman [Ben] Bernanke, and Treasury Secretary [Henry] Paulson."

But investors will probably lose a lot more money in coming months, says Lo: "It's an inevitable aspect of financial markets and risk-taking."

Edward Teach is articles editor of CFO.

CFO Publishing Corporation 2009. All rights reserved.