Benchmarking Accessible Hedge Funds: Morningstar Broad Hedge Fund Index and Morningstar Nexus Hedge Fund Replication Index

Transcription

1 Benchmarking Accessible Hedge Funds: Morningstar Broad Hedge Fund Index and Morningstar Nexus Hedge Fund Replication Index Morningstar White Paper June 29, 2011

2 Introduction Hedge funds as an asset class have demonstrated their potential to diversify portfolios and mitigate downside risk. These investment products achieve these results because of relatively unrestricted investment mandates. While traditional investments are typically purchases of equities or bonds, alternative strategy products, including hedge funds, generally invest both long and short in a wide variety of markets including: equity, fixed income, currencies, and commodities. Hedge funds obtain long or short market exposure both directly, through cash instruments (stocks and bonds), as well as indirectly, through derivatives (both exchange-traded and over-the-counter). Most hedge funds earn returns through trading strategies using liquid instruments, but others invest in securities whose returns are tied to liquidity premiums more commonly associated with private equity. Many have restrictive terms or minimum investment requirements that limit all but the largest investors without liquidity concerns. Morningstar has developed the Morningstar Broad Hedge Fund Index (MBHFI) as a benchmark of hedge funds with more-permissive terms that are more realistic for high-net-worth individuals as well as small and moderate-size institutions. Additional research into the factors that drive the returns of this index has been used to create a replication index, the Morningstar Nexus Hedge Fund Replication Index. The replication index is composed of a basket of securities, which in aggregate perform in line with the hedge fund index, and it is a proxy for the daily returns of the accessible hedge funds that compose the MBHFI. Hedge Fund Benchmarks Having a reliable benchmarking system is one of the biggest challenges that investors face when selecting and evaluating hedge fund managers. Investors are interested in analyzing how hedge funds strategies correlate with both broad market strategies for portfolio construction, optimization and asset-allocation purposes. Hedge fund indexes come in two basic varieties. Broad hedge fund indexes are considered noninvestable and are calculated based upon hundreds or thousands of fund returns. Investable hedge fund indexes are based upon relatively small samples of funds. Broad hedge fund indexes typically include a wide spectrum of the investment strategies of varying availability to investors. Broad indexes are derived from the constituents in a hedge fund database, which rarely include the returns of the largest or most prominent funds. These indexes generally apply currency unhedging or conversion to present results in a single currency. Some broad indexes have no asset minimum or operating history requirements, allowing small and untested funds to drive index performance. Most of these indexes include funds that are closed to new investors as well as funds that still available. Both the Morningstar 1000 Index Series and HFRI Fund-Weighted Composite Index include both open and closed funds as well as funds denominated in multiple currencies. Investable hedge fund indexes are generally limited to funds that remain open to new investments. They are also restricted to funds on one investment platform. Investable indexes use individual platforms because they offer larger investors the ability to purchase the entire index customized for their investment currency (assuming they meet the platform minimums for each of the underlying funds). The construction and rebalancing rules of many investable hedge fund indexes result in returns that are more similar to that of funds of hedge funds (albeit with reduced fees) than those of broad market measures. Investable indexes are thus more subject to risks from individual funds. For example, the Credit Suisse/Tremont Market Neutral Investable Index lost approximately 40 in December 2008, primarily due to Madoff exposure through a single feeder fund. 2

3 Morningstar Broad Hedge Fund Index The MBHFI has been created as a benchmark for investors with constrained budgets and high liquidity requirements for their hedge fund investments. Research, including Morningstar and Barron s Annual Alternative Investment Survey, indicates that lack of liquidity is the number-one reason that both advisors and institutions hesitate to invest in alternative investments. Additionally, many advisors and institutions indicated that the length of lock-ups 1 and size of investment minimums prevent their firms from allocating to hedge funds. This index is considered appropriate for investors with limited budgets for direct hedge fund investments and the need for regular rebalancing. The index rules have been created to reflect the needs of these investors. The starting point for the MBHFI is the Morningstar hedge fund database, which contains approximately 4,500 direct hedge funds (as opposed to funds of hedge funds). Operational and liquidity filters are applied to identify those funds that are most available to investors. The index requires a minimum performance history of one year and at least $100 million within the strategy to ensure that the constituent funds are not in incubation and have sustainable businesses. To ensure that underlying funds would be eligible for an institutional rebalancing program, funds must offer redemptions and subscriptions no less than quarterly, with a limited lock-up and advance notice period. 2 Funds also must specify investment minimums of no more than $500,000 so that smaller institutions and high-net-worth individuals could use these funds as a direct investment program. Lastly, funds must be open to new investments and denominated in U.S. dollars. The U.S.-dollar-denomination requirement is applied to remove the effects of currency hedging or translation from the index returns. The choice by a manager to hedge currency risk is often a discretionary, active decision, and, by excluding funds with return streams in other currencies, the index is able to isolate these decisions to a single reference currency. This rule ensures that any currency hedging and returns are part of the investment process rather than a function of index calculation. MBHFI Constituency Characteristics The Morningstar Broad Hedge Fund Index has been calculated back to January Morningstar and acquired databases have been collecting hedge fund data throughout this time. Funds that were operating historically but have ceased operations or ceased reporting to the database are included in the calculated returns. The MBHFI includes hedge funds that use a wide variety of trading strategies. The following table compares the May 2011 constituents of the MBHFI to the full Morningstar database. In May 2011, the index contained more than 500 funds and the database had more than 4,500 funds. 1. Lock-ups forbid investors from redeeming for a set period following the initial investment; a lock-up is considered soft if investors can redeem subject to an additional redemption fee, and hard if there is no provision for early withdrawal. These terms make direct hedge fund investments illiquid and more difficult for investors to use in a dynamic allocation. 2. Advance Notice Periods require investors provide managers redemption requests prior to the dealing day. For many hedge funds these notice periods require redemption planning by investors months in advance. 3

4 Figure 1: Morningstar Broad Hedge Fund Index Constituent Comparisons Morningstar Broad Hedge Fund Index Fund Breakdown () Database Fund Breakdown () Difference () Asia/Pacific Long/Short Equity Bear Market Equity China Long/Short Equity Convertible Arbitrage Currency Debt Arbitrage Distressed Securities Diversified Arbitrage Emerging Markets Long/Short Equity Equity Market Neutral Europe Long/Short Equity Event Driven Global Long/Short Equity Global Macro Long/Short Debt Merger Arbitrage Multistrategy Systematic Futures U.S. Long/Short Equity U.S. Small Cap Long/Short Equity Volatility Source: Morningstar After applying the construction rules, the MBHFI tilts toward strategies based upon trading and security selection, rather than strategies that are based upon liquidity premiums. While most hedge fund strategies focus on liquid investments and earn trading profits, a small but significant portion of the industry earns returns associated with liquidity discounts on the underlying investments. Many of the excluded funds earn returns from liquidity premiums, which are most commonly associated with private equity. This is in contrast to most hedge fund strategies, which earn returns from trading hedged positions. Funds using strategies described as activist or distressed securities often have restrictive liquidity terms. This allows time for portfolio managers to engage with corporate management and to allow corporate restructurings to play out. The MBHFI rules will intentionally limit exposure to these funds. Because of the illiquid nature of the underlying investments, these funds often have longer lock-ups, less frequent liquidity, and high investment minimums. In addition to the exclusion of longer-term strategies, other notable characteristics of the index constituency are the exclusion of Morningstar s China long/short equity category, which only includes renminbidenominated funds. The MBHFI also has fewer funds trading in Asia/Pacific long/short equity and Europe long/short equity than the database does because of the large numbers of these funds that are denominated in a currency other than U.S. dollars. On the other hand, the largest category of hedge funds with an overweighting is the systematic futures category. These funds generally trade liquid futures, and few have restrictive terms. Other fund categories with an overweighting include the U.S. small cap long/short equity and U.S. equity categories, which unsurprisingly are dominated by U.S.-dollardenominated funds. Morningstar will regularly publish the category weights both by assets and number of funds. 4

5 Index Return Characteristics The returns of hedge funds demonstrate low systematic exposures to equity and fixed-income indexes. From January 2003 through April 2011, the Morningstar Broad Hedge Fund Index has a beta of 0.26 and R-squared of 49 to the S&P 500 and beta of 0.14 and R-squared of less than 1 to the BarCap US Aggregate Bond Index. With less than 50 of fund returns statistically explained by movements in the S&P 500 and a de minimis amount of returns correlated with the total returns of the BarCap index, the diversification benefits are strong. Diversification benefits cannot be evaluated in a vacuum. While low beta and R-squared show diversification, these only help a portfolio if the return for risk assumed is adequate. Since 2003, the MBHFI demonstrates higher returns than the S&P 500, BarCap U.S. Aggregate Bond Index, and the S&P Goldman Sachs Commodity Index. Despite the highest returns, only bonds demonstrate a lower standard deviation than does the MBHFI. The distribution of returns for the MBHFI aren t quite as positive a stand-alone feature, with the largest negative skewness and peaked returns (larger kurtosis). The MBHFI has shown relatively larger single-month declines than stocks, bonds, and commodities. Despite this weakness, the index has recorded the highest number of positive months and a maximum drawdown significantly lower than equities and commodities. The Sharpe ratio for the index is 1.01 and the Sortino ratio is 1.52, both the highest among the four asset classes. Figure 2: Index Comparisons Annual Return Ann. Standard Deviation Skewness Kurtosis Maximum Drawdown Up Months Morningstar Broad Hedge Fund Index S&P BarCap U.S. Agg Bond S&P GSCI MSCI World Willshire Source: Morningstar Over the same time period, from 2003 through the present, a $10,000 investment in the Morningstar Broad Hedge Fund Index would have grown to more than $18,700, a higher value than an equivalent investment in stocks, bonds, commodities, or two possible balanced portfolios. The total return for the period doesn t tell the whole story. Depending on your entry and exit points, each of the six investments shown could have provided the highest relative returns. Figure 3: Annual Return on a $10,000 Investment j Morningstar Broad Hedge Fund Index 30k j S&P GSCI TR j S&P 500 TR j BarCap U.S. Agg Bond TR j 45 S&P 35 BarCap U.S. Agg 20 MBHFI j 60 S&P 40 BarCap U.S. Agg 25k 20k 15k 10k Source: Morningstar 5

6 Inclusion in a Portfolio While the return characteristics of the MBHFI look quite attractive as a stand-alone investment since 2003, this period s turbulence has provided a tailwind to the relative returns of hedge funds as a strategy as compared with long-only investments. Prudent investors avoid having all their eggs in one basket and diversify across asset classes and strategies. Morningstar s recent survey of institutional investors indicated that approximately half expect an allocation of more than 20 in alternatives five years from now in their standard diversified portfolios. The MBHFI has return and correlation characteristics that should provide diversification benefits for a portfolio without the individual manager risk associated with a direct hedge fund investment program. For simplicity sake, let s compare an investor with a 60/40 portfolio split between U.S. stocks and bonds with an investor allocated 45/35/20 across stocks, bonds, and hedge funds, respectively. Over the period reviewed, the portfolio with a 20 allocation to accessible hedge funds: added 24 basis points of annual return; reduced sensitivity to large-capitalization equities by 16; lowered portfolio standard deviation by more than 15; and had a maximum drawdown more than 5 below the traditional portfolio. Additionally, a $10,000 investment would have grown to $17,620 versus $17,290 throughout the full period. Figure 4: Example Investor Portfolio Comparisons Annual Return () Beta () Standard Deviation () Maximum Drawdown () 60 S&P, 40 BarCap US Agg S&P, 35 BarCap US Agg, 20 MBHFI Source: Morningstar Direct Passively Investing in Hedge Funds There are two primary approaches to passively investing in hedge funds. Investments linked to investable hedge fund indexes offer returns of the specific underlying hedge funds composing the index. Investable hedge fund indexes and the products linked to these indexes are subject to selection biases that result in returns that are quite sensitive to the individual index constituents. This results in returns more akin to a fund of hedge funds rather than the full hedge fund marketplace. The second approach is known as hedge fund replication. Replication products seek to identify systematic factors driving hedge fund returns, otherwise known as alternative beta. Replication products are typically based upon broad noninvestable indexes. These products generally use long and short positions in exchange-traded funds, futures, and other derivative instruments to gain both long and short exposure to individual market risks that drive hedge fund returns. Morningstar believes that the aggregate returns of many hedge fund strategies are largely driven by common factors that can be replicated through investments in index investment products. The MBHFI has been used as the basis for the Morningstar Nexus Hedge Fund Replication Index (Nexus). This index is a daily return index currently composed of approximately 20 liquid financial instruments including equity index futures, volatility derivatives, sovereign debt futures, currency contracts, and commodity futures. The commodity futures include positions in precious metals, energies, agricultural commodities, and industrial metals. There are two primary methods to determine factors and weights when replicating a hedge fund index. Regression-based models often use fixed, rolling time periods and inherently make assumptions about the distributions of returns of all factors. The assumptions frequently assume constant correlation rela- 6

7 tionships and other assumptions that sometimes differ from reality. Filtering techniques employ advanced statistics that incorporate data for the longest common periods. Filtering makes few assumptions related to the distribution of returns and attempts to reflect all available data. Morningstar s research has determined that replication models based upon filtering most accurately reflect the realities of the dynamic investment market. Nexus is a filter-based replication index. Rather than basing its replication on the unadjusted MBHFI, Nexus adjusts its target to reflect the expected management costs within funds. This is appropriate as the underlying returns of the MBHFI are reported to the Morningstar Hedge Fund Database net of all management and performance fees. After adjusting the MBHFI returns, the filter-based model provides the constituent positions and weights for the index each month. The back-history of Nexus has been calculated since January The following table outlines the performance of Nexus compared with MBHFI from January 2004 through April Figure 5: Morningstar Nexus Hedge Fund Replication Index Performance Compared to Morningstar Broad Hedge Fund Index Annual Return Standard Deviation Skewness Kurtosis Maximum Drawdown Tracking Error Information Ratio Morningstar Broad Hedge Fund Index NA NA Nexus Source: Morningstar The Morningstar Nexus Hedge Fund Replication Index does a good job tracking the MBHFI with a small tracking error of 5.41, along with a moderately positive information ratio. The annualized returns are 1.48 higher than the MBHFI. This result is intentional in order to cover the costs of any products created using the replication basket. These higher returns do come at a cost: the standard deviation of Nexus is higher than that of the MBHFI, but only moderately so and still lower than a balanced portfolio without hedge fund exposure. Another difference with the replication index is the improvement in maximum drawdown. While this deviates from the performance of the reference index, most investors are less concerned with minor positive outperformance. Conclusion The Morningstar Broad Hedge Fund Index offers a unique benchmark for the most available hedge funds for U.S. dollar investors. The underlying funds offer frequent liquidity and minimal lock-ups, characteristics that are in increasing demand by investors. The index construction rules are aimed to capture the returns available to the majority of hedge fund investors, who desire regular access to rebalance their holdings. The Morningstar Nexus Hedge Fund Replication Index provides investors a proxy for the daily returns of the MBHFI. The securities that are used are highly liquid and quite available for investors to create a replication portfolio. This index offers a transparent method to passively invest in the hedge funds available to liquidity-conscious investors. 7

8 References Fung, William and Hsieh, David A., Asset-Based Hedge-Fund Styles and Portfolio Diversification (October, 2001). Duke University Fuqua School of Business Working Paper. Available at SSRN: or doi: /ssrn Amenc, N. Ge`hin,L. Martellini, and J.-C. Mayfredi., Passive Hedge Fund Replication: A Critical Assessment of Existing Techniques. The Journal of Alternative Investments. Fall 2008, Vol. 11, No. 2, pp Roncalli, Thierry and Weisang, Guillaume, Tracking Problems, Hedge Fund Replication and Alternative Beta (January 9, 2009). Available at SSRN: Li, Daniel, Markov, Michael and Wermers, Russ R., Monitoring Daily Hedge Fund Performance When Only Monthly Data is Available (October 1, 2009). Available at SSRN: Freimann, Eckhard, The Diversification Properties of Hedge Fund Investments. Available at SSRN: Bali, Turan G., Brown, Stephen J. and Caglayan, Mustafa O., Do Hedge Funds Exposures to Risk Factors Predict Their Future Returns? (October 26, 2010). Available at SSRN: Bekkers, Niels, Doeswijk, Ronald Q. and Lam, Trevin W., Strategic Asset Allocation: Determining the Optimal Portfolio with Ten Asset Classes (October 2009). Available at SSRN: Jaeger, Lars A., Factor Modelling and Benchmarking of Hedge Funds: Can Passive Investments in Hedge Fund Strategies Deliver? (November 7, 2005). Available at SSRN: Hsieh, David A. and Fung, William, The Risk in Hedge Fund Strategies: Theory and Evidence from Trend Followers. The Review of Financial Studies, Vol. 14, No. 2, Summer