Managed Futures: Understanding a Misunderstood Diversification Tool.

Managed Futures: Understanding a Misunderstood Diversification Tool.

August 24, 2016 $fut, Tactical Asset Allocation Research, Managed Futures Research
Print Friendly
(Last Updated On: December 21, 2016)

In my two previous blog posts (here and here), I analyze the performance of bonds during really bad months for US stocks (“Crisis Alpha” months), and I analyze the performance of US stocks during really bad months for US bonds.  A quick summary of the results from those prior studies:

  • Bonds have historically provided some diversification benefit during bad months for stocks. Bonds have historically had a positive return in excess of cash of about 0.12%, on average.  The excess return has historically been positive about 55% of the time.  However, most of that performance has been driven by the income return component of bonds, and not their price appreciation.  This might make future diversification benefits harder to come by for bonds.
  • Stocks have historically provided “meh” diversification benefit for bonds when bond returns have been bad. Stocks, historically and on average, have provided negative returns in excess of cash during the worst 50 months for bonds.  In addition, stocks have provided positive returns in excess of cash only about 40% of the time.

One key takeaway point from the previous posts is that investors may need to look beyond traditional stock/bond portfolios in an attempt to increase the diversification of their portfolios. Among the more compelling “outside the box” diversification options are managed futures, which Wes briefly mentions in an older post here on crisis alpha.

What in the heck are managed futures?

Managed futures is a catch-all term for strategies that trade futures contracts. In many respects the label “managed futures” is similar to “equity investor”–the label doesn’t really narrow things down. That being said, futures-based trading strategies have typically been pursued by hedge funds and commodity trading advisors (CTAs) and many of these strategies are based on some sort of trend-following rule(s).

Trend following (or absolute momentum investing) can be best described as going long assets that are going up in price and selling (shorting) assets that are declining in price.  The managers of managed future strategies apply this trend following investing on many (thirty or more) liquid and exchange traded future contracts across commodities, equity indices, currencies and government bonds.

To learn more about Managed Futures, please check out the following links:

How do Managed Futures Work in a Portfolio?

Data

For the time period 1/1926 through 12/2012, we use Ibbotson Associates US Large Cap TR index for US Stocks, Ibbotson Associates Long-Term US Government Bond TR index for US Bonds, Ibbotson Associates 30 Day T-Bill TR for Cash and the data from the AQR paper “A Century of Evidence on Trend-Following Investing” net of the paper’s estimate of fees and transaction costs (which are estimated net excess returns) for the Managed Futures return stream.  We do an additional robustness check with additional data, as described below.

To extend the analysis on Managed Futures, we also use the Barclays Top 50 Index (BTop50) which is a net of fees index created by Barclays Hedge, Ltd. to represent the Managed Futures industry.  It is not possible to invest in the index, but in order for a manager to be included in the index they must be open to new investments (among several other requirements).  The Barclays Top 50 index is used as a proxy for net of fee, live and out of sample performance of the Managed Futures strategy.  The BTop50 has data going back to 1987.

Unless otherwise noted, all returns will be excess returns (total returns less the return on cash) and all averages are arithmetic.

Correlations

First, to determine if a Managed Futures allocation might diversify a portfolio of traditional stocks and bonds we use the traditional tool, correlation analysis. The results are as follows:

1 - Correlation Table
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

The correlation table, above, shows that US Stocks and US bonds have had very little correlation with one another over time (as expected).  In addition, Managed Futures (ManFut) shows very little correlation to US Stocks and US Bonds.  This indicates that they have the potential to be a great diversifier to a portfolio of US Stocks and US Bonds.

However, as the two previous blog posts have shown, one must also look past summary correlations and also look at how assets have performed during “crisis” periods in order to determine if the diversification is helpful (i.e., the returns aren’t all bad at the same time) or if it is pseudo-diversification (i.e., the cross correlations are low but the returns are bad at the same time).

Performance When Stock Returns Are Bad (“Crisis Alpha” months)

As in my previous blog post, we will divide history into two categories 1) US stock total returns are -5% or worse (“Crisis Alpha” months) and 2) US Stock total returns are greater than -5% (“Normal” months).  We will then study the historical returns of US Stocks, US Bonds, Cash and Managed Futures during those time periods.  I have summarized the return data in the table below:

2 - ManFut Crisis Table
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

And a visual summary of the data:

3 - ManFut Crisis Graph
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

As you can see from the table and graphs, most of the time (90%) US stocks have had total returns that were greater than -5% in a month.  During those 90% of months, US stocks averaged total returns of 2.08%.  However, during “Crisis Alpha” months, stocks have averaged total returns of -9.05%.

In order to provide attractive diversification for stocks, an investment doesn’t only need to have low correlation, but it must deliver attractive returns (on average) during those “Crisis Alpha” months…and do so with some consistency.

As we look at the performance of US Bonds, we see that they did in fact provide positive excess returns, on average, of 0.12% and did so with some consistency (positive excess returns in 55% of “Crisis Alpha” months).

However, given the difference in magnitude between US stock returns during “Crisis Alpha” months (-9.05% total return) and US bond returns (0.12% excess return), it is a drop in (the diversification) bucket.

As we study the Managed Future return series during “Crisis Alpha” months we see that it has provided an excess return of 1.30%, which is more than 10X the performance of bonds during “Crisis Alpha” months!  In addition, the Managed Futures return series has provided positive excess returns in “Crisis Alpha” months 67% of the time.

So to summarize so far – Bonds have been an OK diversifier for stocks (low correlation and small positive excess returns during “Crisis Alpha” months), but Managed Futures have the potential for offering much better diversification (low correlation with larger and more consistent positive excess returns during “Crisis Alpha” months).

This finding isn’t new, but confirms the findings of researchers like Dr. Kathyrn Kamisky and Dr. Alex Greyserman that managed futures tend to provide positive excess returns during periods when US stocks perform unusually poorly.

Performance when Bond Returns Are Bad

To analyze the performance of how US stocks and Managed Futures perform during bad months for bonds, we look at the 50 worst months of excess returns for bonds over the period 1/1926 through 12/1926.  A summary table of the data is below:

4 - Bad Bond Table
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

And visually:

5 - Bad Bond Graph
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

As you can see from the data, above, US Bonds have historically averaged a -5.35% excess return during their 50 worst months.  During these tough months for US bonds, US stocks have also struggled with an average excess return of -1.47%.  In addition, during the 50 worst months for US bonds, US stocks have only generated a positive excess return in 38% of those months.

Said another way – when US bonds struggle, US stocks haven’t historically provided much help for portfolios.

As we look at the performance of Managed Futures (ManFut) during the 50 worst months for US bonds, however, we see that they have averaged a 0.96% excess return.  In addition, Managed Futures have generated a positive excess return in 30 of the 50 worst months for bonds.

This finding seems to be a less discussed attribute of Managed Futures – they are an investment (or investment strategy) that has historically provided a positive long-term return and also provides diversification for both US stocks AND US bonds.

Robustness Check

There is always a chance that the findings discussed above are due to chance or data mining or other issues that might make the Managed Futures strategy appear more attractive.  To help address that concern, we do a robustness check on the findings by using the Barclays Top 50 Index (BTop50) and repeating the analysis using data from 1/1987 through 12/2015.

For US stocks the data table is below:

6 - BTop Crisis Alpha
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

As you can see in the graph, US stocks since 1987 have about the same ratio of “Normal” months at about 90% and “Crisis Alpha” months at about 10% as they do in the longer data sample (1/1926 through 12/2012).  In addition, the returns during “Normal” months is very similar at a positive 1.53% excess return and the return during “Crisis Alpha” months is also very similar to the longer sample at -8.55%.

As we look at the Btop50 excess return performance since 1/1987 we can see some items that are different from the AQR data and some consistencies:

I notice two main differences between the BTop50 performance and the AQR performance:

  1. Frequency of positive excess returns – The AQR data shows that excess returns are positive about 66% of the time. However, the BTop50 shows that excess returns are positive only about 52% of the time.  These are obviously over different time periods, but it shows that perhaps some of the returns from the Managed Futures strategy have been eroded away.
  1. Average Returns are less during All Months and “Normal” Months – The notion that returns are less for the Managed Futures strategy is further supported by the data in that the average excess return during all months is about 0.4% less in the BTop50 data than it is in the AQR data. In addition, the returns during “Normal” times is also about 0.4% less in the BTop50 data than in the AQR data.

Despite the differences, one striking and very important similarity remains:

  1. Returns during “Crisis Alpha” months – The average excess return during “Crisis Alpha” months is 1.64% which is about the same as the excess return in the AQR data (although over different time periods). In addition, the frequency of positive excess returns is about the same in the BTop50 data (about 65%) as in the AQR data (about 67%).

This robustness check on Managed Futures helps strengthen the case that Managed Futures are capable of delivering attractive diversification benefits for stocks.

For the 50 worst months for US Bonds since 1/1987, the data table is below:

7 - Btop Bad Bond
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

As we can see from the table, above, US bonds have averaged an excess return of about -4% during their worst 50 months between 1/1987 and 12/2015.  This is largely consistent with the performance during the worst 50 months for bonds between 1/1926 and 12/2012.

One notable difference between the AQR data and the BTop50 data is that the BTop50 hasn’t provided a positive excess return during bad months for bonds.  There could be any number of reasons for this and it might be the subject of a future blog post.

A Poor Man’s Method to Attain Managed Futures Exposure?

Ok, enough with the theory…how do you add Managed Futures exposure to your portfolio?  There are a couple of choices in varying degrees of complexity, fees, etc.  I’d revisit Wes’s great article on FACTS to determine where you feel most comfortable in the spectrum of the FACTS framework.

  • Option 1 – Direct exposure to CTA hedge funds: This might be the route preferred by most institutional investors, but this isn’t an option for most individual investors.
  • Option 2 – Invest in a mutual fund that uses a Managed Futures strategy: There has been a large increase in the offering of mutual funds that pursue a managed futures strategy.  However, there can be added complexity due to choices about how the fund is managed (does it invest directly or is it pursuing a fund of funds type strategy, etc.)
  • Option 3 – Replicate a Managed Futures type strategy in your own portfolio: One could identify a group of ETFs that approximate exposures available via futures (e.g., equity, fixed income, commodities, and FX) and run a simple trend-following model to determine long/short exposures. If the short side is too complicated, a long/flat (go long or go to cash, but never short) approach could potentially work (see the RAA system for an example). The DIY approach is potentially cheaper, but one also risks diluting the performance of a more pure managed futures exposure.

Summary

Studying bond returns during tough times for stocks (and vice versa) shows that correlation might overstate the diversification benefits of stocks and bonds.  This post introduces a new investment strategy, Managed Futures, and shows that Managed Futures have promising diversification benefits for stocks AND bonds using one data set.

Using another data set for the Managed Futures strategy shows that Managed Futures have been a great diversifier for stocks but a mixed bag diversifier for bonds.

Hopefully, this post serves as a good starting point in understanding some of the diversification benefits Managed Futures may provide.


Note: This site provides NO information on our value investing ETFs or our momentum investing ETFs. Please refer to this site.


Join thousands of other readers and subscribe to our blog.


Please remember that past performance is not an indicator of future results. Please read our full disclosures. The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. This material has been provided to you solely for information and educational purposes and does not constitute an offer or solicitation of an offer or any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The factual information set forth herein has been obtained or derived from sources believed by the author and Alpha Architect to be reliable but it is not necessarily all-inclusive and is not guaranteed as to its accuracy and is not to be regarded as a representation or warranty, express or implied, as to the information’s accuracy or completeness, nor should the attached information serve as the basis of any investment decision. No part of this material may be reproduced in any form, or referred to in any other publication, without express written permission from Alpha Architect.


Definitions of common statistics used in our analysis are available here (towards the bottom)




About the Author

Andrew Miller CFA, CFP

Andrew Miller, CFA, CFP: Andrew is Chief Investment Officer at Miller Financial Management, LLC where he is primarily responsible for investment and financial planning research, asset allocation, and integrating client’s financial plans with their investment portfolio. Andrew is a Chartered Financial Analysis and is a Certified Financial Planner® practitioner. Andrew graduated from the Indiana University Kelley School of Business in 2004 with a Bachelors degree in Business Administration with a concentration in Finance. Andrew’s research interest is in using academic investment research to create investment portfolios that improve the withdrawal rates and financial outcomes for clients.


  • Zach

    Andrew, thanks for sharing! Are there any mutual funds in particular that are worth looking into further?

  • Andrew Miller

    I don’t want to mention any names but if you go to Morningstar and search through the managed futures category there are quite a few. I think Morningstar has awarded several of them Analyst ratings of Bronze or better…but that is Morningstar’s opinion.

  • Also not inclined to mention specifics, but some rules of thumb: focus on process-driven (ad-hoc is dangerous in mgd fut space), affordability (fees can get wild), trading-frequency (higher is actually better), and current AUM (lower is better). Good luck

  • Thomas Burnstead

    Are there any glaring reasons to avoid using an ETF structure to access managed futures strategies? I’m thinking specifically of the the big one in the space, WDTI WisdomTree Managed Futures Strategy Fund, but am curious about the ETF structure generally w/r/t access to managed futures.

  • They are generally sub-optimal to be honest. Lots of constraints in the ETF structure–not ideal for l/s, high leverage, high frequency rebalance, etc.

  • Thomas Burnstead

    Got it, much appreciated! Thanks for another great article.

  • happy to chat mgd fut offline.

  • Aaron

    Most of the mutual funds out there have silly high fees. They don’t necessarily tell you up front that they outsource the strategy to someone else who puts on another layer of fees. The mutual fund is really a fund of funds. Of the 50 mutual funds out there, there might be 5 that directly run the strategy and don’t add another layer of fees. Buyer beware.

  • Also, if you plan on replicating a managed futures strategy by building a portfolio of ETFs and overlaying some trend following rules, you may run into some problems – namely, the ETFs not performing like the actual underlying futures contracts they try to mimic.

  • Remember that Morningstar penalizes volatility in their ratings. Managed futures tend to have higher volatility (standard deviation) than some other more traditional investment options. As a result, they have lower Sharpe ratios (CAGR / Std Dev) in general. Morningstar does not reward survivability in their ratings, so a fund with a two year track record of great returns may receive a higher rating than a 20-year fund with solid returns.

    You may consider using the MAR ratio when evaluating managed futures funds. MAR = CAGR / Max Loss. For context: In general, trend following strategies produce MARs of 0.30 – 0.70 over time while traditional buy and hold stock market strategies produce MARs of 0.10 – 0.30.

  • Jack Vogel, PhD

    This comment (not mine) is being passed on from Scott Berglund (President of Elk Hill Advisors in Roanoke VA) —

    I think the monthly loss > 5% parameter may understate the true value of managed futures (MF) as a stock diversifier. Using returns in down calendar years for the S&P 500 gives much better results. Using the returns for the S&P and managed futures (BTOP 50 + AQR before 1987) in Meb Faber’s backtesting module from 1929 – 2015 shows the following. There were 24 down years for the S&P. MF returns were positive in 23, or 95.8%. The 24 year CAGR was a robust 15.3%. Narrowing the parameters further to include only years down more than 5% gives even better results. 20 out of 20 such years were positive for MF with a CAGR of 17.6%. I’m not certain, but maybe the better results are at least partially attributable to the longer measuring period. Nice work, Andrew!

  • Andrew Miller

    Please pass along my thanks! A couple of thoughts: 1) It is important to always use excess return so that different time periods are more comparable. 2) I choose 5% losing months to increase the frequency of the “Crisis Alpha” month observations. There are many ways to define a “Crisis Alpha” month. You can use drawdowns (of any length) greater than say 15%, months where VIX spikes more than 20% (or some other threshold), etc. The problem with choosing other definitions is one of limited data (VIX only goes back to 1990) or few observations (losing years or drawdowns are somewhat rare for drawing statistically significant inferences).

    That said, my own research is consistent with those findings, in that, as the length of the measurement period increases the power of managed futures to diversify (provide positive returns) increases with it. This just shows how robust their diversification power has been historically (and hopefully will be in the future)!

  • Andrew Miller

    I think that managed future funds should come with their own set of evaluation criteria as the traditional performance metrics don’t really apply to managed futures due to their non-linear return profile. Just as an example, a managed future manager could improve a statistic like Sharpe Ratio or MAR by including some amount of long biased equity (or carry trade). However, this is not why you are paying or using a managed future manager! If someone is going to use a managed future manager they should be paying for the pure trend following, risk managed, convex payoff not for a high Sharpe or MAR.

  • When you follow trends, diversify, cut losses short, ride winners (and follow your system like a Marine), you tend to produce a higher Sharpe, MAR and Sortino over time.

    In trend following, you manage risk through diversification, using stop losses and applying a responsible position sizing methodology.

    You receive a convex payoff when you cut losses quickly and let winners run. You limit your downside and set no limit on your upside. In traditional fundamental “buy low, sell high” investing, you do not receive as much convexity since you allow losses to run against you (because you believe it will come back higher later on).

    See here for a performance experiment of a pure trend following strategy that manages risk versus the S&P 500. http://www.michaelmelissinos.com/blog/one-simple-rule-can-make-all-the-difference/