Managed Futures Returns: Skill-Based or Inherent?
Copyright 2000 Barclay Trading Group, LTD.
Terms, Conditions and Trademarks Apply.
A hypothetical investor is examining his portfolio results for 1995. Among his investments, he holds shares in the ABC Fund, an equity mutual fund that invests in large capitalization stocks. At the end of the year, his shares have appreciated 20%. This investor also has units in a managed futures fund. At year-end, those units have also appreciated 20%.
For the new year, our investor must decide whether or not to make changes in these sectors of his portfolio. Although a 20% return for the year is quite high for the mutual fund on an historical basis, he is reminded that the S&P 500 index measured a gain of 37.5% for the same period. As a result of this apparent under performance, he decides to switch from ABC Fund to another mutual fund and feels that he has made the right decision.
However, when he tries to extend this line of reasoning to evaluate the performance of his shares in the managed futures fund, he finds that he is somewhat at a loss. Although this fund also under performed the S&P 500, he is not sure whether the comparison to the S&P index is relevant. Upon discovering that the Barclay CTA Index measured a 13.5% gain for the year, he feels somewhat more comfortable that the futures fund added value, but still isn't quite sure.
Institutional investors find themselves in a similar quandary. While it is fairly easy for them to evaluate the source of returns of equity and fixed income managers, they are still somewhat at a loss when it comes to managed futures. It is unclear to them whether the returns from managed futures investments are the result of capturing an inherent return, manager skill, or some combination of the two. To better understand the underlying issues, we've invited two experts in the field to present their views.
Charles A. (Tony) Baker, Trilogy Capital Management, LLC. Mr. Baker is Executive Vice-President and Director of Institutional Services at Trilogy, a registered investment advisor and CTA firm that specializes in the use of derivative-based investment strategies. He is responsible for all services provided to institutional investors including marketing and product development. Prior to involvement with the establishment of Trilogy, Mr. Baker held positions at SEI, Leland O'Brien Rubinstein Associates, Western Asset Management Company, and Union Bank.
Richard A. Pike, RP Consulting Group, Inc. Mr. Pike is President and founder of RP Consulting. Established in 1990, RP Consulting is dedicated to the very specialized application of managed futures and derivatives to institutional investors' portfolios. The firm works with domestic and international banks and brokerage firms to develop client products, as well as with pension fund sponsors in there search, development and administration of their managed futures investment programs. Mr. Pike has been employed in the futures industry for 17 years and has been a member of senior management of four major Wall Street firms.
Q: What do you believe to be the source of returns for managed futures? Is there an inherent return that is being captured by CTAs or are the returns skill-based?
Baker: The primary purpose of a futures contract is to serve as a risk transfer vehicle for hedgers. That is, futures serve as a principal tool for hedgers to shed price volatility stemming from commodities used in their businesses. This acquisition of greater stability in their business is a valuable function, and one which would be expected to have a cost. This is true independent of the skill of those assuming the price volatility. Thus, fundamental inherent returns are generated by investors purchasing price volatility and providing a risk transfer service to commercial hedgers.
Just like inherent equity returns, inherent futures returns can be captured through passive, non-forecasting strategies. Through their active management techniques, CTAs not only earn this inherent return, but also add a skill-based return, or alpha. In this way, CTA returns are quite akin to traditional active equity and fixed income managers. In fact, recent studies utilizing style analysis techniques (an approach which has gained wide acceptance largely through work by Nobel-laureate Bill Sharpe), have shown that very large proportions of CTA returns can be explained by an inherent futures, or market, return.
Pike: The source of the returns for managed futures is manager skill. CTAs (the managers providing the skill) trade most all markets worldwide, including stock and bond indexes from many diverse economies. There is ample research supporting the notion that inherent return is present in the markets they trade. Even the commodity markets have exhibited inherent return attracting investor capital. The presence of inherent return, however, does not explain the CTA motivation for being in the market, nor do the inherent returns have any relationship to the historical returns provided by CTAs.
CTAs are in the market in order to exploit what they perceive, through their proprietary research, to be short- term price inefficiencies that exist as the market forces seek equilibrium. Collectively, they research (and trade) each individual market as it relates to time and as it relates to other markets. They form opinions of those relationships. If prevailing market prices are not in total sync with those opinions, trading opportunities exist. Free short selling makes it possible to engage in countless trades that are independent of the price level of the market alone. Free leverage allows CTAs to capture very small price changes and magnify them into a substantial effect on their portfolio. Add the exchange-listed and OTC options strategies available, and the list of market independent trading opportunities would appear boundless. It is those opportunities that motivate CTAs to trade and are the source of their returns.
To capture inherent market return an investor need only hold a long unleveraged position in a market. Not much research and very little trading skill is necessary to capture the inherent return.
Q: Whether the returns are judged to be inherent or skill-based would affect the choice of a suitable benchmark. What would you consider to be an acceptable index or benchmark of managed futures returns?
Baker: Given my previous argument, it's obvious that an acceptable index or benchmark would first and foremost attempt to passively capture the risk transfer premium available to investors. That is, the index must reflect the returns available in the futures market from passively buying volatility from hedgers. Just as equity market indices are diversified to show the overall returns available from providing equity capital, a futures index should also be broadly diversified to demonstrate the returns available across a broad representation of commercial markets.
The BARRA/MLM index (which is now reverting back to its original MLM index name) has been a reasonable first try at providing this kind of benchmark. It is simple, unmanaged, diversified, and roughly approximates the returns available to investors from passively owning volatility. However, it has several defects, which can and should be improved upon. We are currently in the process of constructing a new index, which is intended to correct these flaws and serve as the new futures industry benchmark. This project is being carried out with the help of an advisory committee made up of an impressive list of representatives from the institutional investment industry and the futures industry. With their help, we are confident that a sound, acceptable index will soon be available.
This is not to say, however, that peer group universe benchmarks are not useful. Just as with equity managers, these kinds of benchmarks provide invaluable insights into the performance of individual managers relative to their peers. Both inherent return indices and peer group benchmarks, such as the Barclay CTA Index, should continue to coexist as each serves its own function in performance measurement.
Pike: Benchmarks are useful for the institutional investor for a number of analytical tasks. One is to be able to decompose an active manager's return into 1) returns provided by the asset class' inherent (systematic) return, and 2) returns provided by the manager's skill (selection of holdings and timing of holding periods). Another analytical task is to predict returns, volatility and correlation among asset classes for use in asset class allocation optimizers. Optimizers use only the systematic returns of asset classes and their sectors as the independent variables.
Most benchmarks are therefore composed of systematic return data. In fact, very few benchmarks have ever been developed using isolated and documented manager skills. Managed futures are tracked by several peer group performance indexes which are satisfactory for analysis. Since the returns are all manager skill- based and not systematic return- based, they should not be used in the regular optimizers. These days tools exist for optimizing systematic and skill-based returns. Institutions are just now beginning to focus on that sort of analysis.
The peer group indexes that are currently published within the managed futures arena are suitable for use in constructing skill return benchmarks which are very useful for determining the value managed futures can provide for alpha transfers, alpha overlays and alpha optimization strategies.
Q: Steven Strong at Goldman, Sachs & Co. recently argued that the S&P 500 index or an international bond index would be appropriate indexes in regression analysis of CTA performance for purposes of calculating alpha and beta. Do you agree with this approach?
Baker: Absolutely not. I am presuming that the general purpose of this kind of analysis is to evaluate a CTA's relationship to a market (beta) and the amount of skill he demonstrates (alpha). Given that, I cannot understand the rationale behind such an approach. This is the same as saying that it would be appropriate to use a bond index to calculate equity managers' alpha and beta. Unless the underlying index is somehow representative of the same asset class as the manager, results will be spurious.
If we were to regress an equity manager's returns against a bond index, there would be little relationship found and the alpha and beta calculations would be misleading since we are dealing with two separate asset classes. In the same way, regressing CTA returns against bond or equity indices will show little or no relationship (it all looks like alpha), but the measurements are really meaningless. The only appropriate index for this kind of analysis is one which reflects the underlying inherent returns in the futures markets, just as the only appropriate index for an equity manager is one which reflects the underlying inherent returns available in the equity market he or she operates in.
The only case in which it would make sense to use a traditional asset index is when the manager was using positions in financial futures as a substitute for the underlying securities, such as stocks and bonds. However, in that case, I would argue that the strategy is not managed futures in the way we understand the term.
Pike: Equity and bond indexes are only useful for the portfolio managers at the highest echelon to determine the correct allocation optimization between assets with systematic returns. These indexes (or any other beta indexes for that matter) provide no clues for the performance attribution of CTAs. Systematic returns provided by various asset class benchmarks are also useful when combined with skill benchmarks to attempt to optimize the mix of beta and alpha exposures sought by investors.
Managed futures are a set of skill strategies and are correctly represented by peer group index-based benchmarks. Very few asset classes have skill or alpha benchmarks readily available for analysis. Managed futures are unique in this respect.
Q: What does the ability to implement futures strategies as an overlay imply about the source of returns for futures?
Baker: Nothing. The ability to use futures as an overlay is simply the result of their operational efficiency. Since futures require only a small deposit as collateral margin, the bulk of an investment portfolio can be invested in other securities. Traditionally, this investment has taken the form of cash. However, it can be invested in other securities such as bonds or equities. This ability has been used by some to construct portfolios in which the underlying investments are in equities (managed by someone other than the CTA), and the CTA manages an additional futures allocation utilizing only a small portion of the total investment as margin. All that has been done is to change how cash is invested and that implies nothing about the source of returns for futures.
Pike: Overlays are strategies that can be employed by investors without the need for funding or disturbing an existing manager mix. The fact that futures trading requires no funding is one of the clues that observers have to identify the nature of the return source. The level of return and risk sought by managers of futures trading is not constrained (or motivated) by available assets. The desired level of exposure to any one trade is determined by its fit into a total portfolio matrix of risk and return probabilities provided by all the other existing trading exposures. A disconnection to any funding constraint allows traders to create exposures with beta relationships to the underlying market that range from multiple positive to multiple negative. The betas also vary greatly over short time frames. This disconnection to the underlying markets gives intuitive evidence of the skill of the trader providing the dominant influence of the returns.
Q: Can managed futures be widely accepted as a legitimate investment and play a significant role in investor portfolios if the returns are perceived to be the result of manager skill?
Baker: No. This is an issue that has been baffling us for some time. Much of the futures industry has continued to maintain that the returns generated are solely due to skill. Yet this argument has clearly not been successful in getting significant allocations from investors. Virtually no serious inroads have been made in the institutional community, and the traditional retail industry is static, at best. The main reason for this is obvious, and it's not fees.
When you examine the portfolios of any of the major investor markets, it does not take a rocket scientist to notice that every investment which plays a significant long-term role in those portfolios has its own economic rationale for generating returns, independent of any manager skill. Equities and bonds--both domestic and international, cash, and even real estate, all have an assumed inherent return which gives investors comfort that they will earn long-term returns even if their selection of managers is not the best. Investors simply will not make major, long-term commitments to anything which they perceive does not have an inherent return. The perceived risk with a pure skill investment is too high. This is particularly true as one deals with larger and larger pools of capital, where an intermediary (such as a pension officer) is likely to play a significant decision-making role. In effect, you are asking them to make a bet solely on your skill. If things turn out poorly, they will be in the line of fire and have nothing to fall back on. On the other hand, if things go well, they don't get a commensurate participation in the rewards. It's a bad risk/return trade-off any way you look at it.
We also hear the lament from the futures industry that much of the money coming from retail sources is "hot money" that is, money which has a high turnover rate. But when faced with only manager skill as an argument, consider how an investor might feel when performance is poor. With no economic rationale to give them comfort that returns will be there in the long-run, is it any wonder that these investors will pull money quickly? It is indeed ironic that the plague of hot money is doomed to continue as long as so many continue to cling to the manager skill argument.
Now we do have a sound, compelling economic rationale for inherent returns in the futures market, and one which coincides perfectly with the primary function of the futures market. In addition, when this argument is tested using actual data, the support is incredibly strong both across time and across markets. Finally, this inherent return can be captured in an implementable, indexed investment. In short, with futures we now have for the first time all of the ingredients that are required by large investors to make significant, long-term commitments.
Aside from the compelling investment logic and appeal, one would think that the industry would embrace this concept warmly, if only for business reasons. CTAs are active managers in the futures markets earning a market return and an alpha based on their skill.
Huge, permanent investments are made in other asset classes with managers who provide precisely this kind of investment and are paid quite handsomely. It is obvious this is what the investor market wants. Yet, incredibly, there are many in the futures industry who still take issue with this argument. There are many reasons why, I suppose, some of which are not particularly flattering. However, without this inherent return argument as a foundation, there is no hope of a futures investment playing a significant role in investors' portfolios.
Pike: Capturing manager skill is the objective of the majority of institutional investors. Even after a strong trend toward indexing the majority of institutional assets are actively traded. In our opinion, the recognition of the real source of managed futures returns as being manager skill clears up some confusion surrounding the strategy. The confusion has hampered the understanding and acceptance of managed futures in the past. Remove it and we open the door to broad use of the strategy in institutional portfolios.
Copyright 2000 Barclay Trading Group, LTD.