Rocaton Insights. Managed Futures: The Case for a Strategic Allocation. Anton Gorbounov David Morton. January 2011

Transcription

1 Rocaton Insights Managed Futures: The Case for a Strategic Allocation Anton Gorbounov David Morton January 2011 Copyright Rocaton Investment Advisors, LLC

2 Executive Summary Managed futures is a diverse class of alternative investment strategies which primarily employ systematic strategies to trade liquid futures and currency forward markets both long and short. The most common type of managed futures strategy consists of trend followers who seek to identify and capitalize on major directional price movements in financial, commodity and currency markets. Managed futures offers a number of potential benefits including diversification versus most asset classes, the ability to capitalize on major market moves, an attractive asymmetric return profile, and the potential for alpha generation. Among the risks associated with managed futures are leverage, volatility, counterparty risk, poor performance in times when major trends are absent or when trends reverse rapidly, and performance dispersion among individual managers. We believe an allocation to managed futures as part of an overall institutional portfolio may be appropriate. Such allocations should be implemented through the use of fund-of-funds and/or highly diversified strategies Rocaton Investment Advisors, LLC 1

3 Introduction Investors are still sifting through the aftermath of the credit crisis in 2007/2008 when correlations in markets approached one and many active strategies massively underperformed expectations. While enjoying a partial recovery since the market lows in early 2009, investors have been going through a genuine soul-searching in an effort to take lessons learned from the crisis to construct portfolios better positioned for future downturns. One strategy that deserves to be part of this conversation is managed futures. Managed futures was one of the few investments that produced attractive returns during the perfect storm of Managed futures performed well in a number of previous equity market downturns as well. In addition, it has a lengthy track record of generating positive risk-adjusted returns with low correlation to most asset classes and other alternative investment strategies. Despite these potential benefits, the adoption of managed futures by the institutional investor community has been limited. In this paper, we discuss the benefits and risks of managed futures, describe implementation considerations, and present a case for including managed futures as a part of investors strategic asset allocation policy. Managed Futures Defined The term managed futures is used to describe a diverse class of investment strategies that are generally united by their use of liquid futures and currency forwards and the systematic nature of the execution of the investment strategies. These strategies are typically implemented long and short based on technical signals and, in some cases, fundamental signals as well. Managed futures managers are registered with the Commodity Futures Trading Commission as Commodity Trading Advisors (CTAs). They are also often members of the National Futures Association, the industry s self-regulatory body. Managed futures managers can be categorized by the trading style they employ. The largest group, both by number and assets under management, are medium- and long-term trend followers. These managers seek to identify and capitalize on major price changes in the equity, fixed income, commodity and currency markets. Trend followers typically look at various timeframes in multiple markets and their investment horizons can range from a few weeks to several months. While some trend followers have a track record dating back to the 1970s, the use of this strategy has evolved significantly since that time. Trend followers of the 1970s and 1980s generally employed only a handful of investment professionals who utilized a small number of relatively simple models. The institutional trend follower of today may employ dozens of research specialists using numerous, mathematically complex models. Other managed futures programs include: Short-term trading strategies based on trend or non-trend signals that attempt to capitalize on intra-day and intra-week moves Counter-trend strategies that seek to profit from trend reversals A variety of other systematic strategies, such as pattern recognition One of the oldest providers of data for the managed futures industry, BarclayHedge, estimates that the managed futures fund industry had $248 billion in assets under management at the end of September Of this total, $180 billion, or 72%, was managed by systematic traders while $133 billion, or 54%, was invested in diversified programs which traded across multiple market types, such as currency, commodities, interest rates, and equities Rocaton Investment Advisors, LLC 2

4 Like most hedge funds, managed futures managers typically charge a base fee, as well as a fee tied to their performance. A common starting point is a "2 & 20 structure, within which the manager receives 2% per year on the assets they manage and 20% of the profits they generate each year. Fees can often be negotiated for larger accounts. Managed account structures are fairly commonplace in the managed futures industry and full position transparency is common. As a result of the liquidity of the markets in which these managers operate, most offer favorable liquidity terms to their investors, and gates and side pockets are generally not utilized. As a result, managed futures investors have largely avoided the experience of illiquidity, gates and fund restructurings that plagued the hedge fund industry in 2008 and Managed Futures Performance Over the long term, managed futures has delivered performance that was comparable to or exceeded the equity market. In particular, managed futures performance was strong when meaningful trends in a variety of markets were present, such as in and in Not surprisingly, managed futures results were relatively weak when trends were largely absent, as was the case in and Fig. 1: Annualized and Calendar Year Performance As of September 30, YTD 1 Year 3 Years 5 Years 7 Years 10 Years 15 Years 20 Years Managed Futures Indices: BTOP Dow Jones Credit Suisse N/A U.S. Equities: S&P Managed Futures Indices: BTOP Dow Jones Credit Suisse U.S. Equities: S&P Sources: Credit Suisse, BarclayHedge, Rocaton. Performance for periods greater than one year is annualized. BTOP 50 is an equal-weighted composite of the largest managed futures trading programs, designed to capture 50% of the managed futures universe. Dow Jones Credit Suisse is an asset-weighted composite of 32 (currently) large managed futures funds with at least $50mm AUM and a 12 months track record. Managed futures delivered strong results when strong market trends were present Rocaton Investment Advisors, LLC 3

5 12 10 Fig. 2: Risk and Return 10 Years Ending September 30, 2010 EMD Annualized Return (%) Core Fixed Income Broad Hedge Funds Managed Futures (DJCS) High Yield Managed Futures (BTOP) Commodities Non-U.S. Equities 0 U.S. Equities Annualized Standard Deviation (%) Sources: BNY/Mellon Asset Servicing, Credit Suisse, BarclayHedge, Rocaton. For a full list of index proxies, please refer to the appendix. Over the last 10 years, managed futures have outperformed equities, with less volatility. Managed futures also outperformed core fixed income and most other hedge fund strategies, albeit with greater volatility. Seeking to Capitalize on Market Trends A market trend is a general price move in one direction. Markets do not move from peak to trough overnight; rather, they tend to move gradually over time. Trend following strategies attempt to capitalize on this movement by identifying a trend in its early stages and then exiting the position as the trend begins to weaken. As demonstrated in the chart that follows, trends can vary greatly in strength, speed and length. Markets can also trade within a range, which is often referred to as being trendless or range-bound H2008: Strong Upward Trend Fig. 3: Trend Example West Texas Intermediate (WTI) Crude Oil, H2008: Strong Downward Trend 2009: Weak Upward Trend with Multiple Reversals 20 Long-Term Trends Short-Term Trends : Range-Bound Trading Sources: Bloomberg, Rocaton Crude oil provided two highly profitable trends in 2008 as prices rose to $147 per barrel and then fell to $34 per barrel Rocaton Investment Advisors, LLC 4

6 The premise that past price history can be used to predict future results represents a significant departure from some of the most prominent academic theories in finance, such as the Efficient Market Hypothesis. There is a growing body of academic research, however, that both supports and explains the existence of trends. For example, many of the behavioral finance concepts used to explain price momentum in individual securities (e.g. anchoring-and-adjustment, herding, etc.) can be used to explain broad market trends as well. Buying or selling that is motivated by something other than pure economic gain such as a central bank s intervention in the currency markets, risk management-based trades, and forced selling by leveraged entities facing margin calls may add to this effect. Simply put, the behavior of market participants creates identifiable patterns in the markets, such as trends, which managed futures managers attempt to exploit. 1 The Benefits of Managed Futures The use of managed futures as a part of a strategic asset allocation policy offers a number of potential benefits. Historically, managed futures have provided diversification versus most traditional asset classes, as well as other hedge fund strategies. Furthermore, correlations of managed futures tended to be positive, albeit low, during months of strong performance for risk assets and negative during months when risk assets performed poorly. Fig. 4: Diversification Benefits Offered by Managed Futures Strategies As of September 30, 2010 Managed Futures Correlations with Asset Classes All Months Up Months Down Months U.S. Equity Int'l Equity U.S. Fixed Income High Yield Emerging Debt (since 1994) Commodities (since 1991) Managed Futures Correlations with Other Hedge Fund Strategies All Months Up Months Down Months Equity long/short Global macro Distressed Fixed Income Arbitrage Convertible Arbitrage Merger Arbitrage Multi-Strategy Short Bias Sources: BNY/Mellon Asset Servicing, Credit Suisse, BarclayHedge, Bloomberg, Rocaton For a full list of index proxies, please refer to the appendix Managed futures have exhibited low overall correlation to most asset classes, modestly positive upside correlation, and meaningfully negative downside correlation. 1 While a full discussion of academic theory behind trend following and momentum-based strategies is beyond the scope of this paper, for a good introduction to some of the behavioral finance concepts that can be used to explain the existence of trends, see Brian Hurst, Yao Hua Ooi and Lasse Pedersen, Understanding Managed Futures (AQR Capital Management, February 2010) Rocaton Investment Advisors, LLC 5

7 Correlation statistics, however, tell only one part of the story. Managed futures managers posted some of their strongest returns during periods of significant market dislocation. This is logical, as trend following strategies, by their very nature, attempt to capture major moves in the financial markets. Trend followers have often been able to identify and capitalize on these moves earlier than managers who employ fundamentally-oriented strategies. Trend followers tend to perform better when markets experience steep declines versus rising market environments. This is possibly due to the fact that markets often move more dramatically to the downside. As a case in point, during the last five equity market downturns of 15% or more, managed futures managers generated positive absolute returns. The benefits of this return profile are particularly evident when viewed in the context of an overall portfolio. Over the past three-, five- and ten-year periods, adding a managed futures allocation to a 60% equity/40% bond portfolio would have both increased its overall return and reduced risk. Adding a managed futures allocation to a more diversified portfolio (containing U.S., developed international and emerging equities, high yield, emerging markets debt and commodities) produced a similar improvement in results. Fig. 5: Managed Futures in a Portfolio Context As of September 30, 2010 Annualized Return (%) Ann. Standard Deviation (%) 3 Years 5 Years 10 Years 3 Years 5 Years 10 Years "60/40" "60/40" + 10% CTA Improvement Diversified Portfolio Diversified + 10% CTA Improvement Sources: BNY Mellon Asset Servicing, BarclayHedge, Rocaton For a full list of index proxies and portfolio weights, please refer to the appendix. A managed futures allocation would have boosted portfolio returns and tempered risk. Another benefit of managed futures is the profile of returns that a trend-following strategy often delivers. While trend models vary significantly from manager to manager, most include a signal that attempts to identify the trend and a "stop loss" that is designed to cap the losses in the case of a losing position. In doing so, the downside to any one trade is limited, while the upside is not. This can lead to a return profile similar to a long option position, whereby frequent smaller losses, followed by less frequent but larger gains, result in overall positive outcomes. As shown in the chart below, managed futures returns tend to have a return distribution with a limited number of left tail events and a larger number of right-tail events. This profile means, in practice, that managed futures managers can be right less than half the time and still be successful. In contrast, many fundamentally-oriented strategies exhibit return profiles similar to option sellers, with periods of smaller gains followed by infrequent, but larger, losses (this is especially prevalent when the strategy attempts to capture some form of relative value among different securities). In the absence of profitable trends, managed futures return profiles may be dominated by frequent small losses as positions are established, but then eliminated upon reaching stop loss limits Rocaton Investment Advisors, LLC 6

8 Fig. 6: The Distribution of Managed Futures Monthly Returns January 1987 September 2010 # of Observations "Left Tail" "Right Tail" <-3σ -3σ to -2σ -2σ to -1σ -1σ to mean mean to +1σ +1σ to +2σ +2σ to +3σ +3σ to +4σ >4σ Standard Deviation Bands Sources: BarclayHedge, Bloomberg, Rocaton Managed futures returns tend to have a positive skew right-tail events are more prominent than the left-tail events. Expectations for Managed Futures Managed futures managers pursue alpha-based investment strategies and therefore do not constitute an asset class. The profitability of the strategy depends not only upon the identification of a profitable trend, but also on the efficient harvesting of profits and the careful management of losses. Because of these factors, managed futures returns tend to be inconsistent. However, we do expect predictable return patterns from managed futures managers over longer periods of time. Specifically, over the long term we expect trend followers to profit when unexpected events substantially move the markets but to lose money when markets are range-bound or when trends reverse sharply. The more markets move, the more profitable the trend. This dynamic makes the addition of managed futures to a diversified portfolio compelling. What should one expect to earn by investing in managed futures? Academic opinion varies on this question and much of what has been recently written on this subject makes arguments both for and against managed futures. For example, one recent study found that a simple trend following model applied across many different markets from 1985 to 2009 produced annualized returns in the high teens, with a volatility of just under 10%. 2 In contrast, a paper written in 2008 reached a different conclusion. It concluded that managed futures' long-term net returns were only slightly better than cash, as much of the gross excess return was paid to the investment manager in asset management and performance fees 3. The reality is that is hard to develop generalized expectations for any active strategy. Having already established that managed futures is an active strategy, we believe that return expectations should be a function of managers skill, risk profile, and market environment. Undoubtedly, the returns of managed futures will cover a wide range. The risk profile of managed futures has been more consistent with volatility in the low teens*. Importantly, the expectation is that low long-term correlations with other asset classes and hedge fund strategies will persist, including potentially negative correlations during major market dislocations. 2 Brian Hurst, Yao Hua Ooi and Lasse Pedersen, Understanding Managed Futures (AQR Capital Management, February 2010) 3 Geetesh Bhardwaj, Gary Gorton & K. Geert Rouwenhorst, Fooling Some of the People All of the Time: The Inefficient Performance and Persistence of Commodity Trading Advisors (Yale Working Paper No ) * Managed futures programs can be run at various volatility levels 2011 Rocaton Investment Advisors, LLC 7

9 Risks of Managed Futures Managed futures investments are exposed to a variety of risks. These should be considered along with the anticipated benefits of including a managed futures allocation in an investment portfolio. Leverage. Managed futures programs use implicit leverage, which serves to magnify returns as well as risks. The amount of leverage depends on the strategy employed and the volatility of the market being accessed. The leverage used in a managed futures strategy may lead to volatility that typically ranges between 10-20%. Counterparty risk. Futures contracts trade on organized exchanges which have successfully managed counterparty risk issues through account position limits and the posting of margin. A typical managed futures portfolio will have notional exposures several times the portfolio asset value and all of the positions will move money to and from the exchanges on a daily basis. Currencies trade over-the-counter through a network of global banks and these banks require margins to be posted. The managed futures investor is exposed to counterparty risk through profitable positions when funds are due from a bank and also from the investment of cash balances. These cash balances are mostly in the investor s own portfolio, but also in excess margin accounts held at a bank or at the exchange. Complexity. As mentioned earlier, managed futures strategies can be transparent in terms of detailing actual positions. This characteristic can be diminished, however, due to the complexity of the systems and modeling that make up the investment process and govern portfolio construction. Therefore, these strategies can seem less transparent to investors and harder to research and evaluate. Poor performance in choppy markets. Trends are easy to spot after the fact, but they are very difficult to identify before they occur. The predictive power of even the most sophisticated trend models is usually poor and few managers achieve batting averages of over 50% on their trades. Over the long term, this is mitigated by the aforementioned return profile (winning positions tend to make more than the losers lose), as false starts are usually stopped at a pre-determined level and winning trades tend to offset the magnitude of the cost of losing trades in terms of performance. However, if markets do not exhibit strong trends for a period, the cumulative effect of small losses can add up to significant peak to trough declines. This was the case in mid- 2004, when broad managed futures indices were down 11-13%, cumulatively, between March and August and posted negative returns for six consecutive months. Trend reversals. Some major market moves come not as part of a trend, but due to a trend reversal. Rapid upside or downside re-pricing of the market that goes against the trend may catch trend followers in the wrong position and lead to outsized losses. For example, one of the worst months for trend followers in the last 20 years occurred in November 2001 when, following a series of positive events, risk assets temporarily snapped back sharply after months of trending downward. Managed futures indices registered declines of 7-9% in that month alone. Results were also underwhelming in October of While most CTAs were flat during that month, they did not protect investors from the substantial decline in the equity markets. Performance dispersion. Another risk that managed futures investors face is performance dispersion across individual managers. Despite the fact that much of the manager universe invests in the same markets and pursues similar strategies (i.e. trend following), returns among managed futures managers have varied significantly. This dispersion is particularly significant during years when a variety of very strong trends are present in a number of markets. For example, in 2008 a year 2011 Rocaton Investment Advisors, LLC 8

10 that witnessed strong trends in energy-related commodities, equity markets, currency markets, and bond markets there was a large dispersion of returns among managed futures managers. It is important to note that much of this dispersion appears to be random in nature and driven by secondary investment decisions. These may include sector weights, liquidity constraints and portfolio optimization, rather than the trend following models themselves. For example, consider two hypothetical managers who used identical models to identify and capitalize on a trend. For liquidity reasons, one manager is more biased towards financial (stock and bond) futures than the other. In such a case, the manager with a heavier financials exposure will outperform in an environment where stocks and bond markets exhibit more significant trends than the commodities market. Fig. 7: The Dispersion of Returns within the Managed Futures Universe Max 60 % st Quartile 2nd Quartile 3rd Quartile 4th Quartile Min 5th/95th percentile spread: 59.8% Sources: HedgeFundNet, Rocaton Performance dispersion among managed futures strategies is significant, especially in periods when a variety of strong trends are present in the markets, such as Implementation Considerations Due to the potential for dispersion in results, Rocaton believes any managed futures allocation should incorporate a number of managers through a fund-of-funds or utilize highly diversified strategies. There are a small number of funds-of-funds and a more significant number of large, wellestablished diversified managers among which an investor can choose. While fund-of-funds solutions may entail higher expenses, they may compensate for this by removing the direct due diligence expense, as well as providing manager and strategy diversification. Fund-of-fund managers may also be able to negotiate lower fees with underlying funds to reduce overall expenses. The appropriate target weight in a diversified portfolio to managed futures will vary depending on the overall asset allocation, objectives and risk tolerance of the investor. Normally, individual active investment strategies should be limited to 1-2% of a portfolio. Given the diversification properties of managed futures, however, allocations between 2-5% may be considered to help mitigate other portfolio risk factors. This allocation could be part of an opportunistic allocation, a diversifying strategies allocation, or part of a hedge fund allocation Rocaton Investment Advisors, LLC 9

11 Conclusion Most institutional investors are understandably skeptical about trend following as an investment strategy. However, the successful long-term track record generated by the now-$250 billion managed futures industry over several decades suggest that institutional investors take a fresh look at these strategies. Managed futures can provide valuable diversification benefits to institutional portfolios that cannot be easily obtained through the use of other strategies. In addition, the markets in which managed futures managers operate are highly liquid and the managed futures community has experienced little of the misalignment of interests that has plagued the hedge fund industry in the recent past. The ability of managed futures to produce attractive positive returns in crisis environments is not guaranteed, but it is a rational expectation when one understands how these strategies work and how they have performed in various market environments. In Rocaton s opinion, the probability of diversification benefits in crisis periods, coupled with the long-term return potential, justify consideration of managed futures strategies as a strategic allocation in institutional portfolios. The information included in this publication has been taken from sources considered reliable. No representations or warranties are made as to the accuracy or completeness of this information and no responsibility or liability (including liability for consequential or incidental damages) is assumed for any error, omission or inaccuracy in this information. This information is subject to change over time. This publication is not intended as investment advice. Before acting on any information contained in this material you should consider whether it is suitable for your particular circumstance. Any opinions expressed in this publication reflect our judgment at this date and are subject to change. No part of this publication may be reproduced or redistributed in any manner without the prior written permission of Rocaton Investment Advisors, LLC Rocaton Investment Advisors, LLC 10

12 Appendix Index Proxies (Fig. 1 6) Asset Class / Strategy Index Name Source Managed Futures (BTOP 50) The BTOP 50 Index BarclayHedge Managed Futures (DJCS*) DJCS Managed Futures Hedge Fund Index Credit Suisse (hedgeindex.com) U.S. Equity S&P 500 Index BNY/Mellon Asset Servicing Int'l Equity MSCI EAFE Index BNY/Mellon Asset Servicing U.S. Fixed Income Barclays Capital Aggregate Bond Market Index BNY/Mellon Asset Servicing High Yield Barclays Capital Corporate High Yield Index BNY/Mellon Asset Servicing Emerging Debt JPMorgan EMBI Global BNY/Mellon Asset Servicing Commodities Dow Jones-UBS Commodity Index Bloomberg Equity long/short DJCS Long/Short Equity Hedge Fund Index Credit Suisse (hedgeindex.com) Global macro DJCS Global Macro Hedge Fund Index Credit Suisse (hedgeindex.com) Distressed DJCS Event Driven Distressed Hedge Fund Index Credit Suisse (hedgeindex.com) Fixed Income Arbitrage DJCS Fixed Income Arbitrage Hedge Fund Index Credit Suisse (hedgeindex.com) Convertible Arbitrage DJCS Convertible Arbitrage Hedge Fund Index Credit Suisse (hedgeindex.com) Merger Arbitrage DJCS Event Driven Risk Arbitrage Hedge Fund Index Credit Suisse (hedgeindex.com) Multi-Strategy DJCS Suisse Multi-Strategy Hedge Fund Index Credit Suisse (hedgeindex.com) Short Bias DJCS Dedicated Short Bias Hedge Fund Index Credit Suisse (hedgeindex.com) *Dow Jones Credit Suisse The BTOP 50 Index was used as a proxy for managed futures returns in Fig Portfolio Composition (Fig. 5) All portfolios are rebalanced monthly. Portfolio weights are as follows: "60/40" Portfolio Index Name 1/87-9/10 S&P 500 Index 60% Barclays Capital Aggregate Index 40% "60/40" Portfolio + 10% CTA Index Name 1/87-9/10 "60/40" Portfolio 90% BTOP 50 Index 10% Diversified Portfolio Index Name 1/87-12/87 1/88-1/91 2/91-12/93 1/94-9/10 S&P 500 Index 30% 30% 28.5% 28.5% MSCI EAFE Index 30% 25% 23.75% 23.75% MSCI EM Index 5% 4.75% 4.75% Barclays Capital Aggregate Index 30% 30% 28.5% 28.5% Barclays Capital Corp. High Yield Index 10% 10% 9.5% 4.75% JPMorgan EMBI Global 4.75% Dow Jones-UBS Commodity Index 5.0% 5.0% Diversified Portfolio + 10% CTA Index Name 1/87-9/10 Diversified Portfolio 90% BTOP 50 Index 10% 2011 Rocaton Investment Advisors, LLC 11

13 Appendix (Cont d) Managed Futures Universe (Fig. 7) The Rocaton Managed Futures Universe consists of funds that report into the HFN (HedgeFund.Net) database and identify their primary strategy as CTA/Managed Futures. This broader subset of funds is filtered to exclude funds with assets less than $25 million, duplicate entries (such as multiple share classes of the same fund) and misclassified strategies (as deemed by Rocaton); it may also be supplemented with funds that do not report into HFN but that Rocaton has access to. A total of 93 funds were included in the analysis. The Rocaton Managed Futures Universe is dynamic in nature and is subject to change in terms of constituents and in terms of inclusion/exclusion criteria used to create it Rocaton Investment Advisors, LLC 12