Today’s high risk, low interest rate environment presents a dilemma for prudent trustees seeking to meet the needs of both income and remainder beneficiaries. A portfolio dominated by bonds is likely to feel much more comfortable, but deliver little income, as well as meager growth for the remaindermen. On the other hand, adding to equities doesn’t feel appropriate in today’s uncertain environment.
This makes hedge funds seem appealing to many trustees. Can this asset class really add to the expected returns of a trust with limited stock market exposure? If so, how would it affect the income the portfolio generates? We’ll explore these investment considerations, without addressing contractual, fiduciary or legal considerations.1
Our research has found that hedge funds have historically generated superior returns to stocks, with considerably less volatility, even when performance databases are cleaned of the biases that come with self-reported data. Just as important, hedge fund returns have had a fairly low correlation to stock and bond returns over time, which makes them valuable diversifiers.
Although hedge funds have other risks that volatility doesn’t capture, and their diversification benefit is inconsistent, our research suggests that a modest allocation to hedge funds can improve the remainder value of a trust without significantly affecting income or risk levels.
But first, let’s review the basics.
What Are Hedge Funds?
Strictly speaking, hedge funds aren’t an asset class: They’re actively managed investment pools that can invest in any asset class. While hedge funds vary widely (see “The Many Faces of Hedge Funds,” p. 44), they have some common traits. Typically, hedge funds aim to deliver positive absolute returns in all market environments and are subject to far looser constraints than traditional long-only portfolios. Most importantly, hedge funds typically use leverage; they sell short investments that they expect to lose value, as well as take long positions in investments they expect to gain value. And, they can invest in a wider array of instruments than traditional long-only portfolios can.
These additional freedoms allow hedge funds to reduce their sensitivity to broad market movements, capture some security mispricings more effectively than most long-only managers can and pursue some risk premiums not available to most long-only managers. Among the latter are the risk premiums from currency-carry and commodity-roll strategies.2 As a result, manager skill drives a larger share of the returns of hedge funds than of traditional investment portfolios.
Over the 16 years for which we have reasonably reliable data, 40 percent of the variability of a typical hedge fund’s return after fees came from exposure to the markets, or beta. The remaining 60 percent of its return variability came from manager decisions, or alpha (see “Sources of Return Variability,” p. 45). By contrast,
87 percent of the return variability of the average long-only, active equity portfolio came from beta and 13 percent from alpha.
Hedge fund sensitivity to market movements varies widely by category. Hedge funds that aim to be market-neutral are the least sensitive to the market, but still receive some of their return variability from equity market exposure, because the success of some of their alpha strategies depends, in part, on favorable markets.
Although hedge funds retain some market exposure, the higher portion of their return from alpha makes them useful diversifiers. From 1996 to 2011, the median hedge fund manager generated alpha of 3.3 percent per year, compared with 0.4 percent for long-only equity managers and 0.2 percent for long-only bond managers.
Hedge Fund Return and Risk Data
To understand whether to consider hedge funds for inclusion in a trust’s asset allocation, it’s important to understand their performance characteristics and correlations with other potential investments.
We took a skeptical look at hedge funds’ reported results, because the track record for the category is much shorter than for stocks or bonds, and the funds offer much less transparency. In addition, hedge fund performance databases contain embedded biases, because managers can submit results—or not—for however long they choose.
We analyzed the data available and adjusted for the biases that arise from self-reported results. We arrived at an adjusted annual hedge fund return of 7.3 percent from 1996 to 2011, significantly below the unadjusted return of 9.8 percent, but still attractive relative to stock and bond returns over the same period (see left side of “Performance Data,” p. 45).
We also found that hedge funds have been far less volatile than equities over the same 16-year period. The annualized volatility of the asset-weighted index of all hedge funds in the database was 8.2 percent, about half the 16.5 percent volatility of stocks, as represented by the MSCI World Index, but higher than the 3.6 percent volatility of bonds, as represented by the Barclays Capital U.S. Aggregate Bond Index (see right side of “Performance Data,” p. 45). Hedge funds seem to offer a risk and return trade-off superior to that of traditional asset classes, even after adjusting for biases.
Lastly, we found that hedge funds diversified stocks effectively over this 16-year period, but not as well as bonds did. Hedge funds’ correlation to stocks was about 0.6; bonds’ correlation to stocks, about 0.2.
With higher returns, lower volatility and good diversification benefits, hedge funds seem to be a slam-dunk addition to a portfolio. No wonder many investors have been tempted to throw out the stock/bond paradigm and put all—or a very large chunk—of their money in hedge funds!
Reasons for Caution
Of course, no one should take such drastic action based on 16 years of self-reported data. There are also several reasons to be cautious about embracing hedge funds.
1. Manager skill is hard to identify. Good performance may reflect luck, rather than skill. And even a skilled manager’s strategy may be out of favor for years at a stretch. That’s also true for long-only managers, but for hedge funds, return from skill (alpha) is a much larger part of total return.
2. Manager results vary widely. Hedge fund performance varies enormously from manager to manager in any one year and over time (see “Dispersion Among Managers,” p. 46). While extremely successful long-only equity managers (those ranking at the 10th percentile) returned 9.2 percent per year on average and extremely unsuccessful long-only equity managers (those at the 90th percentile) returned –1.1 percent from 1996 to 2011, extremely successful hedge fund managers returned 19.5 percent and extremely unsuccessful hedge fund managers returned –8.9 percent. We also found a wide dispersion of returns within hedge fund categories. Since there’s no investable hedge fund index, the very wide dispersion of hedge fund managers’ results means that effective diversification of managers is critical to success in this investment category.
We believe the uncertainty of alpha is crucial to understanding how rewarding hedge fund investments are likely to be. This factor should weigh as heavily as volatility and potential returns in investor deliberations about whether, how and how much to invest in hedge funds. The uncertainty of alpha is a central element in our conclusions about the need for rigorous due diligence, broad diversification and hedge funds’ place in an overall asset allocation.
3. The diversification benefit is conditional. Over the 16 years for which we have data, hedge funds had a relatively low correlation to equities, but it wasn’t stable. In months when the equity markets rose, hedge funds’ correlation to equities was almost as low as the correlation of bonds to equities (see left side of “Correlation Between Hedge Funds and Stocks and Bonds,” this page). In months when the equity markets fell, hedge funds’ correlation to equities was much higher (see right side of “Correlation Between Hedge Funds and Stocks and Bonds,” this page).
It makes sense that high quality bonds tend to protect portfolios better than hedge funds do during equity bear markets. During periods of economic and equity-market stress, interest rates tend to fall, which boosts bond prices. By contrast, the aggregate performance of hedge funds is, at best, independent of economic and equity-market conditions.
Until the 2008 credit crisis began, many investors believed that they could use certain types of hedge funds to completely replace a bond allocation. Our findings on hedge fund performance when the equity market falls suggest that this use of hedge funds would be imprudent.
4. Leverage amplifies losses as well as gains. This makes risk and liquidity management crucial.
5. Beware of funds that finance illiquid investments with short-term debt. They may be difficult to roll over in a market crisis and may force a fund to sell at distressed prices; in the presence of leverage, such losses may be large. As a result, such funds are particularly likely to restrict redemptions in a market crisis.
6. Even typically liquid instruments can become illiquid in a market crisis. As a result, simply matching the expected liquidity or duration of assets to funding may not adequately protect a fund from a liquidity squeeze.
Executing a Strategy
Many of the enduring principles that govern traditional investments govern hedge funds, too. In particular, due diligence, diversification and rebalancing are critical.
Due diligence. In the world of long-only investing, investors and their consultants perform extensive due diligence that far exceeds checking for a history of positive alpha. Among other things, they seek to understand whether a manager has an experienced team, investment philosophy and strategy that exploit a known pricing anomaly or a risk premium that’s likely to persist. They also check for sound risk and liquidity management, operating processes and fair treatment of all investors.
The same considerations apply to hedge funds, although their managers offer far less transparency, which makes due diligence more difficult. Nonetheless, we believe that with appropriate due diligence you can reduce the chances of fraud and improve the likelihood that you select managers who will deliver alpha over time. Checking for sound risk and liquidity management and good operating controls is particularly important, given the complications that can arise from short positions, leverage and complex relationships with prime brokers.
Diversification. Investors in long-only portfolios typically make strategic allocations to diverse strategies. They diversify by asset class (among stocks, bonds and real assets), by geography (among developed and emerging markets) and by style (such as value and growth). Our research shows that hedge fund investors would benefit from having strategic allocations to diverse categories of hedge funds and diversifying within each of these broad categories.
We found that the median return on risk (Sharpe ratio) for a single long/short equity strategy over the last 16 years was 0.14, while a portfolio of three randomly chosen long/short equity strategies would have a return on risk of 0.28. Why? One of these strategies might be long/short equity within the U.S. market, another within the Japanese market and the third within the global equity market. Even two U.S. long/short equity strategies might perform quite differently if one focused on pairs of stocks within a given industry, while another took broad bets across industries or went long on large-cap stocks while shorting small-cap stocks. Furthermore, one manager might retain far more market exposure than the other.
We also found the improved return on risk from diversifying across hedge fund categories is substantial. Diversifying from a single hedge fund to one fund in each of 10 categories would have increased the return on risk from 0.14 to 0.49, while diversifying further to three managers in each of the 10 categories would have increased the return on risk to 0.65. Thus, we recommend investing with at least 10 hedge fund managers; investing with 30 or more hedge fund managers would maximize risk-adjusted return. Our asset allocation recommendations assume a well-diversified hedge fund portfolio.
An investor who wants to diversify hedge fund exposures can select a group of funds he believes to be first-rate or can opt for the ready-made format, called a “fund of funds.”
Direct investment offers two principal advantages: control over manager selection and lower fees. However, most hedge funds require minimum investments of $500,000 or more; that puts direct investment in 30 or more hedge funds out of reach for smaller institutions and for all but the wealthiest individuals.
A fund-of-funds pools investor capital, collecting enough to give each investor access to multiple funds—often 40 or more. The manager of the fund-of-funds selects the funds and may provide access to funds closed to all but the most well-connected investors; he also provides due diligence and risk management. Because funds-of-funds provide these services, they charge an additional layer of fees. Our research found that the median fund-of-funds delivered attractive risk-adjusted returns after fees. However, the wide dispersion of results makes due diligence in choosing a fund-of-funds critical.
Our approach to asset allocation starts by assessing the investors’ circumstances, objectives and risk tolerance, as shown by the stock/bond mix they would choose if they invested in no other asset classes.
To determine the optimal allocation for long-term investors who require distributions along the way, we took the risk, return and correlation outputs of our Wealth Forecasting System3 and applied a mean-variance optimization to identify the number of hedge funds that would maximize the portfolio return on risk. We took into account the diminishing benefits of low correlations as portfolio weights increase.
We found that the optimal allocation to hedge funds rises with the overall risk tolerance of the investor, up to an allocation of about 25 percent (see “How Much to Invest,” this page). In each case, the long-term expected return would be about the same with or without hedge funds, if the hedge fund allocation was sourced pro rata from stocks and bonds. The addition of the hedge fund allocations, however, reduces the likelihood of a large temporary loss, because of the diversification benefit that hedge funds provide.
Now, let’s apply these findings to a trust with a different set of goals: one that seeks to add to expected returns without increasing risk substantially, as outlined in the introduction. Our research suggests that a well-diversified allocation to hedge funds is likely to improve the remainder for income trusts and unitrusts,4 while taking little from the income beneficiary and only modestly increasing the risk of a large drop in portfolio assets. Here’s how we reached that conclusion.
Most investors, we’ve found, can tolerate a 10 percent peak-to-trough drop in wealth, but find a 20 percent loss excruciatingly painful. So, we looked at the odds of a 20 percent peak-to-trough drop in wealth for various asset mixes. In today’s volatile markets, our Capital Markets Engine, which projects 10,000 plausible market outcomes based on initial conditions and proprietary economic models, projects that the odds of a 20 percent peak-to-trough decline at some points over the course of the next 20 years is just 6 percent for a portfolio with 40 percent in stocks and 60 percent in bonds.
If a trustee sought to improve expected returns by raising the equity weight to 60 percent, the odds of such a big drop would rise substantially, to 29 percent. However, if the trustee sought to add to expected returns by shifting 20 percent of the portfolio assets from bonds to hedge funds (so that the portfolio had 40 percent in stocks, 40 percent in bonds and 20 percent in hedge funds) the odds of a 20 percent peak-to-trough loss would increase only slightly, to 9 percent (see “Adding Return With Limited Increase to Risk,” p. 49).
Next, we looked at the range of outcomes for both types of trusts, with and without hedge funds.
We project that in the median case, a $10 million trust, of either type, that invested 40 percent in bonds and 60 percent in stocks would generate $12 million in inflation-adjusted distributions over 20 years, including both the annual income and the remainder. The total distributions would rise to $12.5 million if the trust shifted the asset mix to 40 percent stocks, 40 percent bonds and 20 percent hedge funds.
In the current low-interest rate environment, the income trust’s cash flows would be skewed in favor of the remainder beneficiary, whether or not the trust invested in hedge funds. Without hedge funds, $4.2 million of the projected $12 million in total distributions would go to the income beneficiary, and $7.8 million would go to the remainder beneficiary, in the median case. With hedge funds, the distributions to the income beneficiary would fall modestly, to $4.0 million, but the distributions to the remainder beneficiary would rise materially, to $8.5 million (see left side of “Projected Trust Outcomes,” p. 50).
A unitrust’s distributions would more evenly split the wealth between the beneficiaries, in the current environment. But, as with the income-only trust, adding hedge funds would increase expected wealth to the remainder beneficiary and detract modestly from the current beneficiary. For a unitrust with a 4 percent payout and no hedge fund allocation, $5.7 million in after-tax distributions would go to the income beneficiary in the median case, leaving $6.3 million for the remainder beneficiary. With hedge funds, the unitrust’s distributions to the remainder beneficiary would rise significantly, to $6.8 million, while distributions to the income beneficiaries would fall modestly, to $5.6 million (see right side of “Projected Trust Outcomes,” p. 50). A trustee could also adjust the unitrust payout to split the incremental value of adding hedge funds.
Note that these are the median outcomes. We also stress-tested the results for favorable markets (top 10th percentile) and poor markets (bottom 10th percentile) and found that the division of the distributions between the beneficiaries would remain comparable: Adding hedge funds to the mix would provide significantly more to the remainder beneficiary, while taking only a little from the income beneficiary.
In sum, our research suggests that a well-diversified, moderate allocation to hedge funds is likely to improve the total expected returns of a somewhat conservative
trust asset allocation, while adding less risk than a comparable increase in the equity weight.
—Bernstein Global Wealth Management does not offer tax, legal, or accounting advice. In considering this material, you should discuss your individual circumstances with professionals in those areas before making any decisions.
1. Charitable remainder trusts shouldn’t invest in hedge funds, since the leverage used makes the gains and losses taxable.
2. Currency-carry strategies buy high interest rate currencies and short low interest rate currencies; commodity-roll strategies buy some commodity futures at a discount to spot and sell others at a premium to spot.
3. The Bernstein Wealth Forecasting System, driven by the Capital Markets Engine, uses a Monte Carlo model that simulates 10,000 plausible paths of return for each asset class and inflation and produces a probability distribution of outcomes. The model doesn’t draw randomly from a set of historical returns to produce estimates for the future. Instead, the forecasts: (1) are based on the building blocks of asset returns, such as inflation, yields, yield spreads, stock earnings and price multiples; (2) incorporate the linkages that exist among the returns of various asset classes; (3) take into account current market conditions at the beginning of the analysis; and (4) factor in a reasonable degree of randomness and unpredictability.
4. Rules on how hedge fund income and tax liability are allocated to income and remainder beneficiaries can vary based on the trust document and state law. For the purpose of our analysis, we assumed that even if the hedge fund doesn’t distribute income, the trustee can use his “power to adjust” to distribute the income reported on the K-1 (interest, dividends and short-term capital gains) and that the taxes on the income would be borne by the income beneficiary. Long-term capital gains (and the taxes due on them) would remain in the trust.