We are about to defend, and in some cases praise, the use of leverage and derivatives in managing a pension plan. Although that may seem like a fool’s errand given the devastating financial crisis we have just lived through, we do realize that we are not in the Kansas of 2006 anymore. In fact, let’s paraphrase some inspiring comments that have become commonplace in recent times:
• Derivatives are financial weapons of mass destruction.
• There has been no valuable financial innovation for decades.
• Short-selling is un-American.
• Excessive leverage caused all our problems.
There is, of course, some truth behind these quotes. But there is also much more exaggeration and overreaction behind them. Leverage and derivatives are financial tools that have many legitimate and helpful applications. Like any tool, they can be used for good or for bad. A few years ago, as would be likely before any crisis, many of us were undervigilant to their dangers. Today, as should be expected after a crisis, the conventional wisdom is overvigilant, stifling many of their benefits.
We could offer a broad defense of derivatives and leverage. We could argue that the financial crisis was rooted in bad credit, not in the actions of those trying to protect themselves from it, and not created by a bogeyman inhabiting a web of derivatives. We could mention that Greece is in trouble not because of short-selling speculators’ using derivatives like credit default swaps, but because the country spent way more than all of its money. We could point out that some leveraged institutions need taxpayer support now, but they are banks and financial companies, not investors like hedge funds or tax-exempt institutions. We could note that in the recent events in Greece, during the tech bubble and through the corporate scandals of Enron Corp. and WorldCom — and even in this last credit bubble — it’s been short-sellers, often using derivatives and leverage, who have done the hard work of first discovering our problems. Earlier is better. No matter how big a problem turns out to be, letting it go on longer is always worse. And while we’re on the subject of the bursting of the tech bubble, with its trillions of dollars of losses, just in case anyone thinks to blame it on leverage or derivatives, please remember that it was about just old-fashioned, grossly overpriced stocks.
We could do all those things, and, as you probably noticed, we did mention them very quickly (and insufficiently) above, but that’s not our purpose here. Our purpose is to highlight how these tools can be used for better, and for worse, but mostly the former. Our emphasis will be on how institutional investors can use derivatives and leverage to improve portfolio performance — that is, to increase returns, improve diversification and reduce risk.
Going forward, institutional investors are certain to have high required return targets, but they are likely to be operating in a world of modest risk premiums. They will face a choice: Limit their investment options and continue to concentrate their risks in traditional equities or diversify and build more stable portfolios more likely to meet their difficult benefit and spending requirements. Choosing the latter course of action has a lot of appeal, but it will require, you guessed it, the prudent use of leverage and derivatives.
Institutions that choose this path will generally have aggressive goals, and most will have important constraints, but it’s in the interaction between the two where all of the fireworks occur. The problem is that hardly any investor’s goals can be met without taking risk, so careers are spent trying to build better portfolios, develop strategies to expand the “efficient frontier,” seek managers that can deliver elusive “alpha” and reduce drags on performance, all while working within some explicit or implicit idea of how much risk is permissible. The effect of constraints on this process depends on whether they make economic sense. When you layer on irrational constraints, born of fear disguised as prudence, managers have too few levers to pull to meet their challenges. This usually forces risk taking to get very concentrated in the few allowable instruments that offer any hope of delivering the required returns.
Leverage and derivatives are particularly dangerous when used to concentrate rather than spread risk. For example, a number of years ago, a multi-employer pension plan with no active benefit accruals found itself in an awkward situation. By its accounting it was fully funded but required a high rate of return to stay that way — one that far exceeded the low Treasury bond yields then available in the market. As a benefit plan for an industry that had long since moved offshore, it had no deep-pocketed sponsors. Understandably, the trustees had an intense desire to avoid risk taking, lest any losses obligate plan sponsors to make contributions that they could ill afford. So risk taking was necessary (because of the high required return) but severely constrained; equities were shunned and meaningful credit risk not permitted. Pretty much the only allowable investment choices were fixed-income securities backed by the U.S. government.
You may already see where this story is headed. Keeping within its narrow range of permissible investments, the fund began to make unleveraged investments in certain government-backed securities. Eventually, the portfolio was filled with them. They looked good as line items on a portfolio listing. But those investments, agency-backed mortgage bonds — specifically, “inverse floaters” — were mortgage derivatives that had plenty of leverage embedded in them (even when purchased without explicit leverage). As long as interest rates stayed stable — or, even better, fell — they generated a very high return that helped the fund meet its objectives. Of course, they helped until rates started to rise, at which point they proceeded to cause a disaster. Those superbullish bond bets fell super fast, and the plan was mortally wounded.
What happened here was simple: Out of an abundance of caution, the institution had set investment constraints that allowed it few ways to meet its essential return goals. In a long-shot attempt to meet the investment objective, the manager had deployed nearly all fund capital in leverage and derivatives packaged into traditional-looking investments in government-backed securities. The entire fund had become one big, undiversified and leveraged bet. It didn’t work. An untenable combination of objectives and constraints had encouraged a particularly toxic use of leverage and derivatives.
This unfortunate case arose because the plan placed restrictions on allowable investment types. Another common way to create risk by trying to avoid it is to allow a broad range of asset classes but forbid explicit leverage. This strategy may seem judicious, but it forces many institutions into portfolios inefficiently concentrated in equities. These portfolios rely too heavily on equity risk and its juiced-up and very highly correlated cousin, private equity risk. The seemingly “prudent” stand, avoiding things that sound risky, like explicit leverage, drives portfolios toward even larger risks — namely, concentration in one asset class, implicit leverage and illiquidity.
We argue that derivatives and leverage can be used toward positive ends, particularly when they are used to diversify risk. And the good news is that some institutional constraints are easing in ways that encourage just that. We will examine two examples, both of which use derivatives and leverage, that allow institutions to reduce their heavy focus on equities while targeting higher expected returns. The first, risk-parity investing, is a method for allocating assets to a more diversified portfolio that has a superior risk-reward trade-off than that of a traditional allocation. Leverage is used to increase return expectation on a lower-risk portfolio, and derivatives are used as a means to attain most safely and inexpensively the desired market exposures. The second example, investing in a broad portfolio of hedge fund strategies, is about adding diversifying risk premiums and expanding the efficient frontier. Implementing these strategies requires some leverage and derivatives.
We believe risk-parity investing is an example of a good use of derivatives and leverage. More traditional portfolios (those without risk parity) are often allocated among a few asset classes and weighted by capital deployed, with a tilt toward equities. These portfolios may seem diversified, but really they are not. Asset class exposures appear to be somewhat spread around, but because equities are so much riskier than most other holdings, portfolio returns are typically 90 percent driven by the stock market. This would not be so bothersome if the equity expected return premium were so large that it compensated investors for holding all that equity risk. But both empirical data and theory say investors are not being paid enough. (For more than 80 years, U.S. equities have not paid investors enough to be such a big part of their risk budgets.) Instead, the data show that portfolios much less reliant on equity risk and much more broadly reliant on everything else provide a better risk-reward trade-off.
Risk-parity portfolios try to improve the risk-reward ratio by taking their risk, and gathering their expected return, in a balanced way across the spectrum of assets, which includes not just global stocks and bonds, but also emerging-markets assets, commodities, Treasury inflation-protected securities (TIPS), credit and mortgage exposure. The key word here is “balanced,” because risk-parity portfolios try to do much more than just get exposure to the broad range of assets. What’s different about them is that they seek meaningful diversification by diversifying risk, not just exposure (capital). To get that benefit, to allocate roughly equal risk to each asset class, these portfolios need to invest more dollars in the lower-volatility assets, like bonds, and fewer dollars in higher-volatility assets, like stocks.
So what’s the problem? A well-diversified (in risk, not just dollar terms), unlevered risk-parity portfolio offers high expected return for its level of total portfolio risk but is not aggressive enough for most investors. Unlike equities, where the companies themselves are levered, other assets don’t come with much built-in leverage. So the risk-parity portfolio may be a great portfolio, but it’s too low-octane in a world of aggressive return requirements. That’s where leverage comes in. To satisfy typical return objectives, you need to use some leverage to get the total risk level of this very diversified portfolio to about the risk level of traditional, less-diversified portfolios. And derivatives come in handy, too. The best way to construct the risk-parity portfolio turns out to be by using derivatives, like futures.
Why derivatives? Derivatives are used because for this particular application they are the most cash efficient, and so the safest, way to get the leverage. They are also the cheapest to trade and finance and the easiest to manage. The portfolio that results usually leaves about 90 percent of investor capital in cash that stands ready to support exposures should global markets decline. Some might think it’s contradictory to use leverage but have a lot of cash. Actually, when you use leverage, it is essential that you have cash; the more, the better.
Risk-parity investing has been around for several years now and lived through the fall of 2008, the historical ground zero for exposing the flaws in leverage- and derivatives-based strategies. It turns out that risk parity generally did better than traditional allocations; it avoided concentration in equities on the one hand and proved a safe use of leverage on the other. We believe risk-parity investing creates a superior unlevered portfolio and then uses a prudent amount of leverage to put it on an equal risk footing with traditional portfolios, which translates risk parity’s advantage from lower risk into better long-term returns (see Figure 1, page 51). Risk-parity investing absolutely needs prudently managed leverage and derivatives to accomplish these things. And it is consistent with institutional constraints when constraints focus on the economic substance of asset allocation and implementation rather than only on their form.
Another example of using derivatives and leverage in a positive way is expanding the efficient frontier — the highest return for a given amount of risk — with direct investing in hedge fund strategies. Creating a better asset allocation is partly about getting your chosen mix of assets right and partly about including as many worthwhile assets as you can find. Hedge fund strategies, particularly those that provide returns largely uncorrelated to traditional assets, can be great additions since they offer exposure to several alternative risk premiums that expand the efficient frontier. And, of course, their implementation typically requires leverage and frequently uses derivatives, too.
The importance of hedge funds is not about their “genius” managers — there aren’t enough of those to go around. It’s that they can improve your portfolio by adding new attractive exposures. Academic and industry research of the past few years has shown that many hedge fund strategies can be thought of as capturing some systemic, albeit nontraditional, risk premium (let’s call them “hedge fund betas,” as they are systematic strategies on which nobody has a monopoly). By identifying the sources of return for a good portion of the hedge fund world, this work has served to both demystify alternative investments and give us a road map for improving our analysis of how hedge fund strategies expand the efficient frontier.
Some examples (out of many) of hedge fund strategy types include merger arbitrage, convertible-bond arbitrage and managed futures. For merger and convertible arbitrage, persuasive academic research has shown that disciplined, highly diversified portfolios of merger targets and acquirers, as well as portfolios of fully hedged convertible bonds, can offer attractive long-term returns. Furthermore, research has shown that these diversified returns are comparable to those of similar, but much less diversified, hedge fund portfolios focused on selecting individual merger deals or specific convertible bonds. In other words, there is a “beta” return you get just for implementing the strategy well (which is easier than adding alpha but a heck of a lot harder than say market-cap indexing of equities).
To construct these strategies in useful form, you once again have to rely on leverage and, for best construction, derivatives. For convertible-bond arbitrage you need to short equities to capture the premium embedded in the bonds without layering on unnecessary equity risk. You need leverage because without it the hedged position has a good Sharpe ratio but its risk and expected return are very low. You can further refine your strategy to capture the premium in convertible arbitrage by using derivatives (bond futures) to hedge your interest rate exposure and possibly even credit default swaps to hedge residual credit risk. Merger arbitrage is little different, buying acquirees, short-selling acquirers and using leverage to generate enough return to make this generally low-risk portfolio adequately aggressive.
Managed futures focuses on the tendency of major assets and asset classes to trend on their own (that is, not a relative-value trend but an absolute-value trend). A wealth of research has shown that the return to a portfolio of futures constructed to follow those trends tends to be positive. And, arguably more importantly, this return pattern has, on average, a very low correlation with traditional markets. Furthermore, in a rare exception to most successful strategies, it has had its greatest success when movements of traditional markets have been extreme. All in all, a strategy that makes money, on average, does not suffer, on average, when equities decline and succeeds when the world is going mad (in either direction) is a valuable one, indeed.
But, at the risk of asking a question with a now obvious answer, to implement such a strategy, what do you need? Certainly derivatives (it’s managed futures, after all!) — and usually leverage, depending on the level of risk required. Again, a strategy that we believe belongs in an institutional portfolio, and that can help achieve return hurdles while possibly providing some protection against the inevitable vicissitudes of markets (which presumably did not end when the stock market hit bottom in March 2009), is only possible if one is to accept a modest and prudent use of our bad boys.
The first step has been done. We have identified the underlying return source of these leveraged, derivatives-using hedge fund styles — that is, their “betas.” Now we can use traditional methods to see how they improve your overall portfolio choices by expanding the efficient frontier. And improve it they do. Particularly when you make sure they are hedged strategies, not delivered with the embedded market exposure too often found in hedge funds. Our studies, using traditional asset classes plus a diversified set of hedge fund betas, shows that, at the same level of portfolio risk, an economically meaningful increase in expected return is possible.
Note that these hedge fund betas aren’t valuable because they use our two investment tools, derivatives and leverage. They are valuable because they represent sources of return (risk premiums) not commonly present in traditional portfolios. You just need leverage and derivatives to turn them into a useful investable form.
Hedge fund betas don’t need traditional long-only stock market exposure. In fact, it’s far more useful to your portfolio to invest in hedge fund betas when traditional stock market exposure is fully hedged (this is in contrast to the broad universe of actual hedge funds, which is far from fully hedged). The graph on page 68 (Figure 2) provides some real-life evidence during a time of market extremes — first down, then up. Note the cumulative return since the worst of the financial crisis (when, in some odd timing, we began trading this in a live portfolio). Hedge fund betas did not suffer. However, most real-world hedge funds suffered, for they were (and generally still are) long the stock and credit markets. In other words, the average real-world hedge fund is not very hedged. The fascinating implication is that although hedge funds had a rough run during the fall of 2008, it was mostly because of their market betas. Over this volatile period the underlying sources of hedge fund beta, if fully hedged for stock market exposure, performed fine!
Of course, despite our defense of, and advocacy for, derivatives and leverage when used to diversify and expand opportunities, there are real dangers if they are used imprudently. Leverage and derivatives are used badly when they are deployed for the wrong purposes — like increasing risks along one dimension because you are constrained from taking risk in others. But beyond that, there is also bad implementation of good intentions.
To start, leverage and derivatives should not be used to deceive. Leverage should not be implicit and unstated. Derivatives should not be vehicles for repackaging disallowed risks into allowable risks or, even worse, into hidden ones (ask Greece).
In practice both tools rely to some extent on market liquidity. Leverage magnifies exposures and so magnifies transaction costs as well as returns, making portfolio rebalancing more expensive when bid-ask spreads are wide (when liquidity is poor). More importantly, when leveraged investors with inadequate cash reserves and illiquid portfolios find the bid prices for their assets falling so low that lender collateral requirements cannot be met with available cash, they may be forced to unwind some leverage, which will be expensive. In the futures markets margin requirements may be raised on a position too large to trade down efficiently. When inadequate cash reserves are held to support this position, portfolio sales may be forced, quite possibly at a time when markets are disrupted and trading is expensive. This is one case in which the long-term leverage embedded in equities is less dangerous than financial leverage taken on by a portfolio manager. Note, it’s not much of a practical problem for things like risk-parity investing, which as we discussed above, can have a cash cushion as high as 90 percent, enabling it to withstand a very wide range of market disruptions without forcibly deleveraging. As a general rule, be wary of the combination of leverage and illiquidity, especially without lots of free cash and stable financing terms.
Over-the-counter derivatives also introduce counterparty credit risk. Daily mark-to-market collateral payments in bilateral credit agreements mitigate this as long as the instrument is reasonably liquid and counterparty credit is reasonably good. When counterparty credit deteriorates, or before that if you hold a position that cannot be unwound and worry that your counterparty’s credit may eventually deteriorate, this can be a meaningful risk factor. Exchange-traded derivatives represent an important alternative. They have far less counterparty credit risk and worked remarkably well during the financial crisis. We like their use for risk-parity implementation, to the extent possible.
In sum, even though there are mechanisms to keep counterparty risk small, and for risk parity it was quite manageable during the worst of the credit crisis, we still note it is a risk not faced by unleveraged investors in cash markets. But fear of illiquidity and counterparty credit are not reasons to avoid these tools. Anyone using them just needs to respect these challenges. That means following a portfolio strategy that limits illiquid positions, monitoring portfolio liquidity and cash availability and paying careful attention to financing terms and counterparty exposures.
During the credit bubble, of course, many market participants used leverage to increase their total risk, not to diversify it. Leverage was employed to fund risky projects that would not have existed without it, and synthetic instruments were created so financial risk exceeded physical economic risk. Interlocking leverage and derivatives counterparty risk were fatal to some firms, and near-fatal to others, when liquidity evaporated in a wild market. This experience should be kept in the front of our minds, even though it was mostly a phenomenon for financial institutions. The proper lesson is to be prudent, not to stop using these legitimate tools for appropriate purposes.
In conclusion, we should not always be fighting the last war, either in what to avoid or what to extol. But it’s worth mentioning again that both investment examples (risk parity and hedge fund betas), using leverage and derivatives, existed before the credit crisis — and came through it quite well. Even as each strategy is near and dear to our hearts, and admittedly to our wallets as well, they are far from the only examples of the judicious usage of leverage and derivatives that garnered no undue headlines during the past few years’ turmoil. They are simply the ones with which we are most familiar.
Leverage and derivatives are just tools. At their best they are used consciously to improve asset allocation, to make shifts efficiently and cheaply, to implement intended bets, to close market inefficiencies that would be left open (and thus improve markets) and to transfer risk between parties.
There have been volumes written recently that make using leverage and derivatives sound like you are taking your pension plan to the casino. In contrast, we believe that sensible uses of the two instruments help you avoid some of the bad gambles that many have been all too willing to take — holding portfolios with risk concentrated in one asset class, equities, or holding a lot of illiquidity risk in private investments, like loans to casino developers, which are highly correlated to these same equities. Histrionic articles that overemphasize some of the lessons of the last crisis are more fun to write than sober analytical ones, but investors don’t have the luxury of forsaking leverage because certain types of leveraged institutions struggled mightily (that is, banks and broker-dealers, generally not leveraged investors or institutions). Neither do investors lower their risk by avoiding simple derivatives just because some insurers overloaded on ones that bet against the left tail.
Rather than worrying about the labels put on the techniques and investment products they use in their portfolios, investors would do better to ask themselves the following questions:
• Do these things help me reach my return goals?
• Do these things add to my total portfolio risk (not risk viewed investment by investment)?
• Am I using them to help create a better portfolio, or am I using them to concentrate risks because of institutional constraints?
• Does the manager using these techniques, or offering these or similar products, have respect for their power? Is their use prudent?
If the answers to the above are yes, for the ideas we highlight or others, investors should not let labels blind them to opportunities to improve their portfolios. Particularly in a world of narrow risk premiums, growing liabilities and lofty return goals, improvement over tradition is necessary. The way forward is to use everything we have learned from experience and theory to invest wisely the great pools of assets we manage. Retreating toward 19th century notions of good and evil isn’t going to get us there, nor will overaggressive use of new investment technologies. Instead, we should be guided by our time-tested understanding of the benefits of diversification, coupled with our more modern understanding of how best to realize them.
Clifford Asness is managing and founding principal of AQR Capital Management, a Greenwich, Connecticut–based investment management firm employing a disciplined, multi-asset, research process. Previously, he was a managing director at Goldman, Sachs & Co., where he was director of quantitative research for its asset management division.
David Kabiller is a founding principal and head of client strategies at AQR. In that role he manages new and existing client relationships, new products and strategic initiatives. Previously, he worked at Goldman Sachs in the pension services group, overseeing relationships with the CIOs of many of the largest institutional investors in North America.
Michael Mendelson is a principal at AQR. He is portfolio manager of the firm’s statistical arbitrage fund and co–portfolio manager of its global risk premium funds. Prior to joining the firm, he worked at Goldman Sachs, where he was a managing director and head of quantitative trading.
