January 9, 2017
Since lack of correlation to global markets is often the driving force when adding alternative investments to a portfolio, many investors start their search for alternative investments by screening for funds with low to negative correlations. I listed off many of the possible alternatives strategies in my previous Student of Finance post, titled What Your Portfolio Needs: There’s Simply No Alternatives, but one of those strategies in particular stands out from the pack as a top alternative strategy that I believe should be implemented as a part of everyone’s portfolio: trend-following managed futures.
The return profile for trend-following managed futures over the past twenty years can be seen as both boring, and incredibly impressive, depending on what stats you’re focused on. This strategy characteristically aims to cut off losing trades and let the winners run, resulting in few periods of sharp drawdowns like were experienced in domestic equities during the early 2000s and in 2008. During this time period, the correlation between the Credit Suisse Managed Futures index and the S&P 500 total return index was -0.08, practically non-existent. Furthermore, the trend-following managed futures index had a -2.0% down capture ratio during this period, signaling that it actually profited during the months that the S&P 500 lost money. Considering the current market environment of high valuations and the resulting low future expected returns for most equities and fixed income markets over the next 7-10 years, there’s an especially strong case for increasing investments in trend-following managed futures.
(Source: Morningstar Direct)
Although many investors shorten the name and simply refer to them by what underlying securities the fund trades, “managed futures,” most investors more properly refer to this investment strategy group by what it does, “trend-following” or “trend-followers.” They are even sometimes referenced by those who operate these trend-following funds, the commodity trading advisors, or “CTAs” for short. However, there is an important distinction to not just lump all “managed futures” funds together, since there are multiple different types of non-trend following alternative funds that either trade futures contracts or have some form of either “managed futures” or just “futures” in their fund name. More on these other distinct futures trading strategies in coming blog posts. For now, we are focusing on the trend followers, which has its roots in momentum – a concept that most people can easily understand and get on board with.
It intuitively makes sense that once things get moving in a particular direction, they will keep heading that direction in some capacity for a period of time. Without going into great depth of the definition as it applies to physics, Newton’s first law of motion basically says that objects in motion tend to keep doing what they’re doing (i.e. inertia). It’s easy to make examples of a ball rolling down a hill or a truck driving down the freeway, but our focus as it relates to investing is applying this concept to stock, bond, and commodity asset prices. The momentum factor in investing is the generally accepted tendency for asset prices that have performed well (poorly) in the past to continue to perform well (poorly) for some period of time in the future. The performance of the S&P 500 over the previous two decades, seen in the chart below, does a great job of illustrating this tendency for investable asset classes to exhibit long-term trends. The up-trends have persisted for many years in a row, before experiencing shifts into prolonged down-trends that lasted a year or more.
(Source: monthly return data from Morningstar Direct, my graphic)
As with many of the opportunities or factor premiums available to investors, the momentum factor persists due to the various behavioral biases that are innate in all of us. Humans have the tendency to be very short-sighted, often over-reliant on the most recent information when making decisions. Even though investors intend to act rationally, most tend to be swayed by fear and greed, which support the momentum of asset prices. Herd mentality and confirmation bias amplify the amount and duration of these trends, further explaining why momentum exists and supporting the persistence of this factor into the future.
Narasimhan Jegadeesh and Sheridan Titman, who authored Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, were a couple of the first to assemble research that showed the benefits of buying winners and selling losers. Since their paper was written in 1993, countless white papers, books, and other research have been written on the subject. Although the term momentum can often be used to reference relative strength when investing, by investing in the asset with the highest relative performance, the more relevant definition of it as it relates to this discussion is in the context of time-series momentum or trend, which deals with absolute performance. One of the most widely accepted ways to measure trend is to use a simple moving average (MA) of the stock’s previous monthly prices. For illustration purposes within this paper, I’ll use the most common moving averages to show you the benefits of trend following: the twelve month moving average, which drops the most recent month because it tends to experience a trading reversal (i.e. only using previous months two through twelve). The moving average is what’s called a lagging indicator, because a stock’s price will have to already fall a significant amount before shifting from a positive trend to a negative trend. You’ll notice the slight lag in the following graph of the moving average applied to the S&P 500.
(Source: monthly return data from Morningstar Direct, my graphic)
Overall, that 12 month moving average does a pretty good job of showing the long-term trend of the S&P 500 at any given point in time. However, the fact that momentum is a lagging indicator can be seen when looking closely at the price movement of the S&P 500 during points of inflection in the moving average. These inflection points, times when trend shifts from either positive to negative or vice versa, show that the trend signals are always a little late to enter a bull market or exit a bear market, but at least the majority of the trend is captured and often the maximum drawdown is muted. If you were to use the 12 month moving average from the previous chart as a trading signal by only owning equities during periods when the S&P 500 price is above its moving average, you would’ve experienced higher return and less risk.
(Source: monthly return data from Morningstar Direct, my calculations)
(Source: monthly return data from Morningstar Direct, my graphic)
Although the previous example does a decent job of helping understand why and how to implement quantitative trend signals, most trend-following investment strategies actually improve upon this simple model that only held the S&P when it was in an up-trend. Trend-followers generally have the ability to go “short” an investment if the signal is suggesting a down-trend, as opposed to only being able to go “long” during up-trends. Another major difference is that trend-followers are often highly diversified by applying these trend signals and the ability to go either long or short in dozens of distinct markets with highly liquid futures contracts. Consider AQR’s Managed Futures Strategy Fund, which has the ability to invest in over 100 futures and forwards contracts, all of which can be classified within the four broad categories of global equities, fixed income, currencies, and commodities. For instance, this fund could be profiting from being short the S&P 500 while equities are experiencing a bear market, while at the same time profiting from a rise in bond prices and a rise in the price of cocoa if it held long positions and the prices continued to increase. More often than not, trend-followers have the ability to profit from directional movements in practically any market in the world that has enough liquid futures contracts.
It’s worth noting that trend-followers that invest in futures contracts will inherently have the imbedded leverage of trading futures. This can often be beneficial for trading, especially if the trend-following fund is very large and needs liquidity of underlying holdings. One other thing to note is that the collateral needed to be held within each fund can be earning interest – another return generator within the fund. In past decades with fixed income yields in high single- and low double-digits, this was a major tailwind for managed futures funds. However, in recent years, with rates yielding almost nothing, these funds aren’t receiving the benefit from holding collateral that they once did. Managed futures funds stand to benefit when rates normalize in the future to higher yields.
As with any investment strategy, there are always some drawbacks to trend-following. The key issues that trend-followers periodically experience are periods of either no clear trend or shifting trends. Since trend-following is a lagging indicator by definition, the price of an investment would need to reverse significantly to shift investment signals. For example, even if the S&P 500 was going down in 2008, signaling a down-trend that resulted in taking a ‘short’ position in the S&P 500, it bottomed in early 2009 and still took another 6 months before the signal shifted back to an up-trend and suggested a ‘long’ position, resulting in missing the 50% gain that occurred over that timeframe since the S&P 500 bottomed in February 2009. In this example of how shifting trends often hurts returns, the trend-follower would have actually lost roughly -2% in calendar year 2009, whereas a buy-and-hold investor in the S&P 500 would have gained over +26%. Aside from seeing trend-followers struggle when trends shift often in short periods of time, many trend followers recently experienced some tough years in 2010 through 2013, when many global markets were range-bound during many rounds of quantitative easing and central bank actions.
Considering only short periods of time, any strategy can be shown as an incredible ‘hero’ or an equally incredible ‘zero’. However, trend-following strategies have been successfully implemented by many investors for nearly half a century, with some of the biggest success stories coming from a handful of investors who started in the 70’s and 80’s. John Henry, one of the most well-known trend followers who’s now worth roughly $2.3 billion, used profits from trend-following to purchase the Boston Red Sox for $700 million in the early 2000s. Many other trend-followers over the past decades have used the strategy to grow very little money into massive wealth, or used it to experience great profits during periods such as 2008 when practically all other investors lost their shirts. For extensive proof of the prevalence of trend, take a look at either of the two white papers: J. P. Bouchaud Capital Fund Management’s Two Centuries of Trend Following and AQR’s A Century of Evidence on Trend-Following Investing.
Even though there have been some big-shot trend-followers who stand out from the crowd over the past decades, implementing trend-following on your own is possible if you prefer a more hands on approach or direct control over your assets. Similar to the example of applying a 12 month moving average to the S&P 500, you could apply any variation of an intermediate- or long-term moving average (i.e. anything longer than a 50 day MA) to many different asset classes, monitor the signals regularly (e.g. daily, weekly, etc.), and place any buy or sell trades accordingly. Although implementing trend-following is possible for almost any investor, a more realistic take on implementation recognizes that individuals will often not be able to implement trend-following in a full capacity on their own. For a lot of investors, it’s either very difficult to get the short exposure of some commodities and less-mainstream asset classes using ETFs, or more likely the effort and costs associated with placing the trades is too large of a hurdle to overcome. On a related note, individual investors likely don’t have the time to dedicate to creating and monitoring the strategy, they might not have the knowledge about trading/security selection/Excel modeling/backtesting/etc., and they also probably don’t have the same access that scale provides for accessing liquid futures contracts in dozens of markets all at the same time.
The good news is that there are plenty of investment options for retail investors to access trend-following managed futures. There are lots of options in mutual fund format, many of which were introduced in the years following the huge relative success of managed futures in 2008, when the S&P 500 fell more than -37% and the Credit Suisse Managed Futures index returned about +18%. The following is a chart of many managed futures funds that are available to invest in today. Keep in mind that not all funds classified within Morningstar’s Category “US Fund Managed Futures” are actually trend-following managed futures, and that some invest in other strategies such as counter-trend, option/volatility harvesting, systematic global macro, or multi-strat alternative strategies. I tried to remove many of the funds that weren’t primarily trend-following to help keep the comparison as useful as possible.
(Source: Morningstar Direct)
With so many different options, investors may find it hard to narrow down the search to pick a fund that best suits their needs. The AQR Managed Futures Strategy fund AQMIX stands out from the pack because of the sheer size of the fund ($12.1B AUM) compared to its peer group. AQR’s success at raising assets within this fund is probably due in combination to an early inception within the liquid alternatives space in 2010, in addition to its relatively low expense ratio (1.22%), as opposed to the majority of investors actually believing that AQR’s process is better, or even notably different, from most of its peers. The data suggests that pretty much all trend-followers move in sync, although each of these funds probably has a slightly different way of applying trend signals, selecting asset classes, sizing trades, and managing portfolio risks. For instance, see the monthly correlation data for the list of trend-following funds over the past two years:
(Source: Morningstar Direct)
Few, if any, of the wholesalers or PMs of these funds will admit to advisors that it’s impossible to prove whether one trend-following fund is better than another, since each of them has worked hard to constantly improve their strategy or process over the past decades. However, the truth is that all of these trend-following strategies are black-boxes to an extent. Even if one particular fund has done better than the other ones over the past few years, there’s no way of knowing whether it will continue to be better than the other strategies that use slightly different signals, asset classes, or size of exposures. The outperformance during any period could have been due to either manager skill (unlikely since they each employ many incredibly smart PHDs in math, stats, etc.), or more likely just pure luck.
When selecting the best trend-following managed futures fund for my investments, I use the following framework:
CHARACTERISTIC: MY PERSONAL CRITERIA:
Each investor probably has different required characteristics that they’ll first use to help initially screen out the managed futures funds, so let me be as clear as possible by saying that these are not the ‘perfect criteria’ that every investor should use. There is no such thing as having a magical set of ‘perfect criteria’ anyways. I only provided my criteria as a reference to help you understand my process for narrowing down the list to select what I consider to be a topnotch trend-following fund. It doesn’t really matter whether you have a stronger preference for more manager experience, assets invested in the fund, or anything else for that matter! The key things are that you take the time to understand generally what each one of these unique managers does, so that you can rule out the strategies you don’t want exposure to (i.e. short-term trend-following, counter-trend, option trading, etc.), and that you compare the remaining list of similar black box strategies based on risk-adjusted fees.
Considering investments based on risk-adjusted fees isn’t very common for most advisors or investors; however, it proves itself very useful when comparing a handful of funds that we assume have the same opportunity set and are all equally expected to benefit from trending environments. Seen in the chart below, the two AQR trend-following funds actually provide a perfect illustration of this, since AQR’s most popular fund AQMIX targets a 10% “moderate volatility” and charges 1.22%, while the sister fund QMHIX targets a 15% “high volatility” and only charges 1.68%. That means investors who hold QMHIX get 0.7% more standard deviation (a.k.a. risk), and the related returns that potentially come along with that additional volatility, for each percent of fees paid to AQR in fund expenses. Stated another way, since AQMIX and QMHIX hold roughly the exact same portfolio with different risk levels due to the amount of gross exposure, an investor could pay less to achieve the same AQR 10% “moderate volatility” trend-following exposure by holding 67% QMHIX & 33% cash, which results in only paying an effective 1.12% expense ratio.
(Source: Morningstar Direct, my calculations)
In a perfect world, all the funds would publish their target volatility numbers for us to easily make the comparison. For sake of time efficiency, I’m going to use the 3yr standard deviations for each fund and compare them to their prospectus net expense ratios to calculate the “bang for your buck,” measured in Volatility Per 1% Paid in fund expenses. Ignoring funds that are too small and adding rough standard deviation estimates for a few funds that don’t have a full 3yr history, we see that AQR “high volatility” fund QMHIX is actually cheaper than many other funds when you take into consideration the volatility you receive. QMHIX ranks among the most expensive in absolute terms, but it’s the 3rd cheapest fund in the list if you consider Volatility Per 1% Paid. Along with QMHIX, EBSIX & MFTNX are impressively cheap.
(Source: data from Morningstar Direct, my calculations)
Based on this analysis, I’d set up conference calls with the portfolio managers from each of the top five fund strategies, ranked by either of my two ratios that measure the cost of volatility within each fund. Talking directly with the PMs would allow me to ask deep questions about each strategy’s risk exposures, asset class weights, and implementation of trend-following period variations to verify whether it’s still a good fit within my portfolio. Bottom line, investors should really think twice before paying more in fund expenses just because the fund’s marketing team claims that they’re better than any other similar trend-following manager – a claim that’s impossible back up with evidence, since no one knows what will happen in the future! All else equal (which is more of a theoretical concept applied here), investors should select the one with the lowest fees.
In conclusion, almost all investors should have an allocation to trend-following within their portfolio to improve risk adjusted returns. The trend-followers are especially valuable to have during periods of market stress when equities experience sustained drawdowns, since they have the ability to profit from either up-trends or down-trends in dozens of futures markets. Although it’s possible for investors to create & implement a limited form of trend-following in their portfolios, it’s probably best for the vast majority of investors to implement via a low cost mutual fund that covers a diverse group of asset classes across various intermediate and long-term trend signals.