NEW YORK -- Leverage is not as important as many pension plan sponsors might think.
At least that's one conclusion in a recent report by the Counterparty Risk Management Policy Group, New York, a private sector group formed in 1998 after hedge fund Long-Term Capital Management melted down. The group was formed at the request of Securities and Exchange Commission Chairman Arthur Levitt to look at ways hedge funds and their trading partners could better manage risk, thus reducing the chances of another Long-Term catastrophe.
The group recently submitted the report, called "Improving Counterparty Risk Management Practices," to the SEC chairman. It concludes: "The policy group believes that leverage, while an extremely important concept with broad intuitive appeal, is not an independent risk factor whose measure can provide useful insights for risk managers and supervisors alike. Rather, leverage is best assessed by its effects, which can be observed in the possible amplification of market risk, funding liquidity risk and asset liquidity risk."
In an interview, Michael J. Alix, senior managing director for global credit at Bear Stearns & Co., New York, who was a member of the policy group, said, "If I had a goal for the group, it would be that people understand that leverage by itself isn't useful. Leverage is the most imperfect risk statistic out there. Simplistic views of leverage can be misleading."
The report points out that it is "common wisdom that leverage has the potential to increase market risk." However, it then says that "in a world of active portfolio management, an increase in leverage may be associated with a decrease in market risk . . . By the same token, a reduction in leverage (as traditionally measured) can be associated with a rise in market risks."
"Leverage is a magnifier of risk for the assets being leveraged," said Mr. Alix. "But a leveraged portfolio of non-risky assets could be less risky than an unleveraged portfolio of more risky assets."
The report also examines the use of value-at-risk and stress testing to supplement the use of leverage in determining risk.
"Most financial intermediaries and a growing number of leveraged investors utilize one or more variants of value-at-risk to estimate, monitor and limit their market risk," the report says.
"Since all forms of VAR estimates have limitations relating to assumptions used about market parameters, market normality and liquidity, many users of VAR estimates supplement them with the results from an array of stress tests . . . As progress is made in the art of stress testing, firms will become comfortable supplementing their existing risk limits with ones based on stress tests."
Waite Rawls, chief operating officer of Ferrell Capital Management, Greenwich, Conn., was a member of the Risk Practices Working Group that produced the report with the policy group. In an interview, he said: "VAR looks at the volatility of what's being leveraged. It's a much more precise measure of risk."
"VAR and leverage go hand in hand along with stress tests" to measure risk, he added. "If someone says 'value-at-risk,' they better say stress test in the next breath.
"VAR doesn't do well in abnormal markets. (One has to) combine stress tests with VAR measurement to get the best results."
He explained that when markets are under extreme conditions, such as those in the fall of 1998, "my volatility measure (VAR) doesn't capture concepts like this -- it's true in illiquid markets in times of stress."
There is "not a consensus on how to do stress tests," said Mr. Rawls. "The issue is to know there are parts of your portfolio that can really get screwy and how do you measure it."
The report offers stress testing components: an initial stress loss; potential capital withdrawal; and liquidation-induced losses, both realized and unrealized.
"Leverage is an assumption that goes into stress tests," said Mr. Alix. When assets are stress-tested, "leverage accelerates the rate of decline for a given portfolio of assets," he added.
"You can measure risk (of a portfolio) using stress tests. To the extent (the assets) are leveraged, it will have a greater affect on net asset value," said Mr. Alix.
The report concludes: "Experience with past market crises suggests that the most useful way to evaluate leverage is not as an independent source of risk, but as a factor that can accentuate market, credit and liquidity risks. When those crises occur, these three elements of risk interact."
If companies in the marketplace had been doing all the things that the report recommends, "Long-Term Capital wouldn't have been able to have the effect on the market that it did," said Mr. Alix. "But we would still have market disruptions," he said. "I'm always surprised at the speed with which things can unravel."
"If you do all the stuff in the report," you would diminish the potential for many market problems, according to Mr. Rawls.
But there's one important lesson from the Long-Term Capital situation that the report didn't deal with: "Even a group of guys who are so smart are still capable of screwing up and need watching," he said.