Alpha­the outperformance from manager skill­is hard to find. So when you've got it in one asset class, why not move it where you want?

By Miriam Bensman

A litany of troubles pushed Don Nesbitt, director of investments at the $14 billion Teachers' Retirement System of the State of Illinois, to engage in a bit of radical thinking. The plan's $1.67 billion international equity portfolio wasn't performing well enough. It was expensive to run, took up too much of his time in manager searches and, worst, it wasn't making good enough returns.

So Nesbitt tried a bold experiment with a small chunk of the fund. The plan would start a do-it-yourself index replication program that would add value through clever cash management. It took a year to work out the methodology, but in September 1995 the Illinois Teachers plan started buying the seven large, liquid futures contracts that make up 98 percent of the return of its benchmark, Morgan Stanley Capital International's Europe­Africa­

Far East (EAFE) index. Illinois Teachers de-leverages the futures with cash instruments and seeks to beat the index return by getting a higher cash return than is embedded in the futures, and by trading options and futures skillfully.

The program outperformed the EAFE return by 2.1 percent in its first six months and saved the fund about $600,000 in management fees, Nesbitt says. His board was so pleased with the program's performance that in May it more than doubled the program's size, to $115 million from $50 million.

Count Illinois Teachers among the happy converts to alpha transport strategies, which have been slowly gathering recruits and gaining theoretical stature in the last few years. The strategies range from enhanced cash programs and market-neutral strategies overlaid with futures, to equity index­bond index swaps on actively managed portfolios. One way or another, these strategies systematically shift alpha-the outperformance that comes from manager skill-from one asset class to another, primarily through the use of derivatives.

Taken to the extreme, alpha transport posits a radical new approach to investing in which derivatives would play a huge part. "In my ideal world," says William Sharpe, the Nobel-winning economist, "pension fund managers would have all long/short managers [that have shown they can add value], and then do their asset allocation with index funds or derivatives, or both. That's much cleaner and simpler."

Sharpe himself admits, however, that such an extreme application may not be practical or cost-effective, and even those advocates who say the practice can maximize alpha production list a raft of safety precautions.

Unbundling the return

Alpha transport starts with the basic insight that active managers provide two basic types of return: beta, the return from a given market or asset class, and alpha, the variation from the market return that comes from skill. Normally, plan sponsors hire managers to achieve a bundled return. Alpha transport, by contrast, involves unbundling the components of an active portfolio and then rebundling them for an optimal portfolio return.

The search for alpha, of course, has been the obsession of the money management business for years-and for good reason. An additional 1 percent return on $1 billion over a ten-year period (assuming reinvestment at 7 percent) provides almost $200 million in additional assets, note Amoco Corp.'s top investment pros, Marvin Damsma and Greg Williamson, in the first chapter of Alpha-The Positive Side of Risk: Daring to be Different (Investor's Press, 1996). Alpha, however, is all too hard to find. Few money managers outperform, and finding a high-performing manager for each slot in the asset-allocation matrix is often nearly impossible.

Some plan sponsors simply give up. They hire passive managers to provide exposures to efficient markets, such as U.S. equities, where relatively little value can be added, and shift their search for alpha to less efficient markets, such as venture capital, hedge funds or international equity. Thomas K. Philips, a consultant with RogersCasey & Associates, advocates just this approach, coupled with a clear focus on optimal asset allocation, in another chapter of Alpha-The Positive Side of Risk.

Rather than give up on adding value-on that extra 1 percent return that could save taxpayers or the corporate sponsor hundreds of millions of dollars down the road-alpha transport adherents look for alpha wherever it can be found and then "attach" it to the asset mix they want.

Index plus

Illinois Teachers' international program, for example, is a variant of Pacific Investment Management Co.'s (Pimco) Stock Plus strategy, in which Illinois Teachers has been investing since August 1991. In Stock Plus, the Newport Beach, California­based Pimco buys the S&P 500 futures contract and de-leverages with cash. It seeks to beat the S&P 500 by getting a higher return on cash than the short Eurodollar rate built into the futures. Pimco does this primarily by extending duration out as long as one year and, to a lesser extent, by taking a little credit risk with high-grade commercial paper.

"It's done great," Nesbitt says. In less than five years the program has delivered a cumulative excess return of 7.1 percent. Fees are minuscule because Illinois Teachers only pays 15­20 percent of the excess return. A typical equity manager would charge a flat fee of 10­20 basis points of the assets under management plus 10 percent of the outperformance. One result: Illinois Teachers has boosted the allocation to the program from $300 million to $800 million, about one seventh of its $5.5 billion domestic equity portfolio.

Pimco's program is far from unique. Los Angeles­based bond manager Paydon & Riegel offers a similar program, as does Mellon Capital Management and JP Morgan Investment Management.

Market-neutral

Another principal form of alpha transfer strategies involves market-neutral programs, which transport alpha from trading in almost any market into a cash-plus return. Although market-neutral has become a dirty word since David Askin's mortgage fund blew up in 1994, it can be considered the pure form of active management.

Consider, for example, the long/short equity manager who plays only within the S&P 500. Conceptually, the longs and shorts are similar to the overweight and underweight positions in a tilt fund run by a long-only active manager; putting on a S&P 500 futures overlay almost turns the long/short portfolio into a tilt fund. (The difference will be explained later.)

Not all market-neutral managers, of course, stay within the S&P 500. (Neither do all active managers.) Some long/short equity players, most notably Jacobs Levy Equity Management, delve into the small-cap arena. Others, such as First Quadrant, focus on sector rotation strategies. Still others, such as Camden Asset Management, run convertible arbitrage programs. JP Morgan Investment Management, First Quadrant and Advanced Portfolio Technologies run long/short programs in the Japanese equity market.

Regardless of what market pool the manager fishes in to generate alpha, the plan sponsor can keep the market-neutral portfolio as a cash alternative or lay on a market exposure. Asea Brown Boveri (ABB), for example, has used market-neutral programs for more than five years; they now represent 15 percent of its $900 million U.S. pension plan. BZW Barclays Global Advisors runs a U.S. equity long/short portfolio; First Quadrant runs a U.S. long/short, which will soon become global; and JP Morgan Investment Management provides a Japanese market-neutral strategy known as Nippon Neutral. All three put on a S&P 500 futures overlay at the plans' request.

ABB is happy with its market-neutral program, because the current and past managers have, as a group, outperformed the S&P 500 by 250 basis points on average for the last five years, with about 5 points of tracking error. "This is a pretty efficient part of the market where it's tough to find someone who can pick stocks well," notes Eric Wood, ABB's director of pension and thrift management.

Some investors equitize long/short portfolios themselves; others ask their money managers to do it. David Leinwebber, managing director at First Quadrant, says that investors in his firm's U.K. market-neutral product usually put on a U.K. equity overlay; those

in the U.S. product put on a U.S. overlay. JP Morgan Investment Management, however, has some clients in its Nippon Neutral program that ask the firm to put on a international bond futures overlay, says vice president Rhonda Kerschner. And William Jacques, chief investment officer at Martingale Asset Management in Boston, says he has a few clients that ask his firm to equitize their long/short portfolios with the S&P 500, but the clients then swap the S&P 500 return for their benchmark, the Russell 3000. "When I first heard of it," Jacques adds, "I thought, 'that's a clever client.'"

Long-only managers swapped

More often, however, swaps are used to shift the market exposure from a long-only portfolio to some other exposure. "We've seen it used in both directions," says David Duebendorfer, managing director and head of global equity derivatives for Deutsche Morgan Grenfell in New York. "It used to be mostly S&P 500 versus a LIBOR floating rate, but now people are getting more sophisticated, and use it for indices such as the Russell 2000, or a bond index."

Such swaps are frequently used to rebalance a portfolio. Even pension plans well-satisfied with their equity managers probably found themselves forced to take money away early this year, because last year's fabulous equity returns had pushed their equity allocation too high. A swap from the S&P 500 to, say, the Lehman Government Corporate Bond index would save the manager from incurring the market impact and transaction costs for selling the cash securities.

Pushed further, such swaps become a form of tactical asset allocation. "If we have a manager taking a big exposure in the U.K. where he has skill, but we don't like the exposure to U.K. equities now, we may hedge it out with futures and leave behind their stock-picking skills," notes Michael de Marco, director of risk management at GTE Investment Management Co. That lets GTE keep the alpha and maintain its relationship with a valued manager.

Swaps or equivalent futures trades can also prevent a plan's alpha production from being driven by asset allocation, Amoco's Damsma notes. If an equity portfolio is generating 200 basis points of alpha and a bond portfolio is generating 100 basis points, a tactical shift from equities to bonds would sacrifice the larger alpha from equities, unless the shift was done through an overlay.

The next logical step, of course, is to reverse the normal procedure. Rather than set the asset allocation first, and then struggle to find managers to fill in the matrix, a plan sponsor could assemble a portfolio of the best performing long-only or market-neutral managers, regardless of their asset mix, and then use futures or swaps to get to the target asset allocation. That's more or less what Damsma and Williamson advocate in their article, at least conceptually, albeit with many warnings on proper risk control.

Amoco's story

In their article Damsma and Williamson describe a conceptual approach to pension management that, says Damsma today, Amoco does not yet fully implement. The fund began using a portfolio of long/short managers five years ago to generate pure alpha, using futures and swaps to lay on the equity or bond exposure its strategic allocation required, Damsma says. "Because that entailed significant new risks, it was necessary for us as fiduciaries to build better information and risk control systems to support those strategies," he explains.

For the last two years Amoco has been constructing a daily accounting and analytic system for risk monitoring and control. Such systems, Damsma argues, are sorely needed at a time when even the best custodian bank's accounting and analytic systems are taxed by coping with long/short strategies and derivatives. They are also well worth the estimated $400,000­600,000 cost, he adds, if they can help a $1 billion fund to generate an extra $200 million over 10 years by extracting more alpha.

When it completes its new systems sometime this summer, Damsma says, Amoco will start evaluating the existing alpha systems within the portfolio, most of which are with traditional long-only managers, to see if they can be enhanced. To reduce risk, he will also check the correlations between alpha strategies. As it seeks new strategies, Amoco will remain flexible in choosing between long-only and market-neutral strategies, adds Damsma, depending on their expected risk and reward payoff.

The fact is, alpha transport isn't always cost-effective. "If you are going to have a permanent exposure to the U.S. equity market of, say, $5 billion for the next 20 years, you can't get that more efficiently with derivatives," argues GTE's de Marco. "The best way to get it is with cash securities. It only becomes more expensive to use cash securities when you're switching, because of the market impact and transaction costs." GTE uses derivatives to make short-term adjustments in its asset mix.

"You have to get much higher alpha from market-neutral strategies to offset the higher cost of rolling futures over a long period," Damsma agrees. The point is to make the cost comparison, adds Lisa Polsky, managing director and chief derivatives strategist at Morgan Stanley & Co. "Once you separate alpha from the index return, it helps you to see what you pay for what-how much for index replication and how much for alpha." Polsky advocates comparing the alpha returns, risk and fees of long-only managers to those of long/short managers that pick up alpha using similar strategies plus a market overlay.

Confronting the risks

An alpha transport strategy, of course, entails specific risks from using derivatives. Illinois Teachers incurs basis risk from combining seven futures to get EAFE. "Anyone using equity futures to hedge this year has had to confront circuit breakers that suspend trading" and expose traders to market gaps, adds GTE's de Marco. Swaps, of course, may eliminate basis risk and gap risk, but they entail counterparty risk instead.

Then there's the risk inherent in alpha production. Sometimes it's fairly simple to understand. Pimco generates return by extending duration out to one year in the cash instruments behind its Stock Plus strategy. Mellon Capital Management aims for a more consistent, smaller excess return by investing primarily in floating-rate notes, which pay a small liquidity premium, says Tom Hazuka, chief investment officer at Mellon. A big index fund provider, Mellon also switches between futures and cash securities, depending where the price is best.

Market-neutral strategies typically entail more complex risks. For one thing, even the simplest-long/short equity strategies-are not simply tilt funds minus the index exposure. Tilt fund managers are much more constrained in their negative bets. If a stock is only 1 percent of the index, the tilt fund manager can only be underweight by 1 percent; the market-neutral manager could short far more. That increases the chances to make (or lose) money, but exposes the client to the risks of shorting.

Also, long/short managers are inherently double-leveraged. Think of a $100 portfolio with two stocks and an equal-weighted benchmark. The tilt fund manager can only sell off the less attractive one-half the portfolio-to shift it into the more attractive stock; it still only has $100 invested. The market-neutral manager can short $100 of the less attractive stock and go long $100 of the more attractive stock. The "double alpha" of which long/short equity managers boast is the result of that leverage, which amplifies both gains and losses.

Other risks come specifically from alpha transport. Consider the practice of using a market-neutral strategy to fill the cash portion of asset allocation. It exposes the plan sponsor to far more volatility than cash investments normally provide-as some learned to their distress in 1992 and 1993. That's when Jacobs Levy Equity Management's market-neutral strategy returned 3.6 percent and 3.7 percent below Treasury bills, after outperforming by well over 10 percent in its first two years.

"A cash manager's alpha is usually consistent, predictable, real and small," declares Eric Sorensen, managing director and head of equity derivatives research at Salomon Brothers. "As you move up the asset spectrum, alpha gets less consistent and more risky." Although it varies considerably by strategy, alpha from playing in the equity markets will typically be more volatile than alpha from playing with cash instruments.

Then there's the ambiguity about what alpha really is. Its not simply the difference between the manager's return and the relevant benchmark, notes Maarten Nederlof, a consultant with Capital Market Risk Advisors, because there is often residual risk from luck or accident that cannot be easily disentangled from alpha.

Also, Sorensen says, "Alpha is supposed to be return to skill, not to asset class volatility. Often, the higher return from a manager isn't from insight, but from moving to higher beta areas." The manager of a corporate/government bond portfolio who outperforms his benchmark index by buying high-yield bonds isn't delivering alpha, he is taking more risk, Sorensen notes.

But what Sorensen sees as a problem for alpha transport strategies, Morgan Stanley's Polsky sees as an argument for them. Long-only managers, Polsky says, often take positions which create cross-asset risk, such as buying Telmex in a large-cap, U.S. equity fund. Others take significant concentration risk, for instance by putting 40 percent of a U.S. equity fund into technology stocks. Such large deviations from the benchmark make it questionable whether the sponsor is getting the right market exposure for the manager's slot in the matrix.

"When funds do asset allocation," Polsky argues, "they assume the assets are mutually exclusive, well-diversified and exhaustive, and have different returns and low correlations to each other. Increasingly, that hasn't been the case." She adds: "Separating alpha from the index return helps you key in on the risk you take from active management. That's the risk-management reason for alpha transport strategies."

Plan sponsors, of course, have long been expected to understand the source of their manager's returns and to diversify their manager and strategy risk. As so often is the case with derivatives-based applications, alpha transport's conceptual framework simply focuses sharply on plan sponsors' and consultants' traditional tasks-asset allocation and manager selection and supervision-and in so doing perhaps providing a clearer picture.

Alpha­The Positive Side Of Risk by Marvin Damsma and Greg Williamson is available for $45 from Investor's Press. Phone: 860-868-9411; Fax: 860-868-0715.