Back to Forum

Avoiding the Next Train Wreck

Financial Advisor

By Scott A. MacKillop | February 6, 2013

Articles By Us

The problem with trend-following investment systems.

Related Resources

Link to Online Article

Investors have not forgotten 2008. They are still fearful and worried about a repeat of that devastating experience. Recent headlines have done nothing to allay their concerns.

The result is that they have focused more on avoiding risk. Certainly, they are still interested in making money, but they are far more concerned with not losing it.

In this environment, they are understandably drawn to investment products and strategies that avoided disaster by being “out of the market” in 2008. This has led to the rise of a number of asset managers or “third-party strategists” who found safety on the sidelines for much or all of that year. Most were quite small before then, but their assets under management swelled afterward.

My firm competes with many of these, so from time to time we are asked a question, which, when stripped of its niceties, boils down to: “Aren’t these guys smarter than you? They saw the train wreck coming in 2008 and got out of the way, but you didn’t. Why should I trust my clients’ assets with you and not them?”

Because we have been asked this question a few times since then, I have done quite a bit of research into the offerings of these firms, and I have learned a lot about them that should be of interest to advisors.

Simply Following The Trend

In fact, most of these firms did not see the train wreck coming at all. They had no idea what was about to happen in 2008 or why. They simply used trend-following systems to determine whether they wanted to be in or out of the stock market. Since the market was trending down by the end of 2007, they were mostly sitting on the sidelines by the time the real trouble hit in 2008. They were not analyzing events. They were simply following the trend.

Trend-following systems use technical indicators to determine the direction of the assets they invest in. If those assets are trending up, they invest. If the assets are trending down, they get out. Every system is different, but each tries to identify directional trends using a predefined set of parameters and react to perceived trends using a set of mechanical rules.

The basic concept seems like a good one on the surface. The problem, of course, is putting this simple concept into practice.

Easier Said Than Done

A trend-following system is only as good as its ability to identify and respond appropriately to a trend. Unfortunately, sometimes there is no trend to follow. On Black Monday, October 19, 1987, the U.S. stock market fell by more than 20% in one day. There are circuit breakers in place today designed to prevent declines of this magnitude. Though future drops may be smaller, it is still difficult for any system to detect and act upon short-term swings. If the market plunges before a trend is established, the system may provide minimal protection.

Trend-following systems can suffer from the same problem in determining when to get back into the market. Because by their very nature they do not get back in until they detect an upward trend, time must pass before they reinvest. If the market turns quickly, trend-following systems can miss out on the upward swing while waiting for a trend to be established.

Trend-following systems also have trouble in volatile but sideways markets like those of 2011. They tend to react to these markets by jumping in and out as the choppy markets send false trend messages. In essence, they react to trends that never materialize. In these environments, they can perform poorly and may generate higher-than-usual transaction costs.

Trend-following systems use historical data patterns. They do not make a reasoned assessment of asset valuations. They do not consider the impact that crises might have on those valuations. If the historical patterns hold up, all is well. But if the future unfolds in a way that varies from past patterns, the system may not react appropriately.

Also, because some mutual funds impose restrictions discouraging short-term traders from purchasing their shares, trend followers are limited in the universe of funds they can use. That means they lose access to many of the skilled managers that other fund investors have.

How Was It Done?

Nor do trend followers build fully diversified portfolios using modern portfolio theory. A trend-following firm may observe that many asset classes become highly correlated in down markets, but it won’t recognize that falling values in one asset class can be offset by rising values in another.

Instead, these firms seek to invest only in asset classes that are trending upward. Some will invest in multiple asset classes simultaneously, but not because of the traditional principles of diversification. They only invest in asset classes that they believe will produce positive returns in the near term.

Because they do not believe in the wisdom of broad diversification, some trend-following firms have become highly specialized and may use as few as two asset classes. One approach that has become popular in recent years as interest rates have fallen switches between cash and high-yield bonds or other fixed-income securities.

If you place client assets in a product like this, make sure you understand which asset classes were used to generate the historical returns. If the program uses only two or three asset classes, it should not be used as a lifetime, core portfolio for your clients. Also, think about how the asset classes in the program are likely to perform in the future. An asset class that generated strong returns in the past may not do so going forward.

Examine The Record

Be sure to carefully examine the entire track record for firms that were “out of the market” in 2008. Don’t be transfixed by the performance that year. It was a traumatic time and everyone would like to avoid a repeat. But one year does not a track record make. Look for evidence of a manager’s consistent skill in varying markets.

What you may find is that the benefits gained by being out of the market in 2008 were given back by lagging performance in other years. Many of the programs I have reviewed did well in 2008, but didn’t fare well in other years, and that lesser performance essentially neutralized the benefits of missing the 2008 downdraft. A client would have done as well in a simple buy-and-hold strategy.

This raises an important point. In general, trend-following programs seem more successful at avoiding big losses in years when there is a clear downward trend than in generating superior long-term returns through varying market conditions. They are intended to be risk-management tools. When they get it right, they can make life more tolerable for nervous clients during turbulent times. Just don’t expect them to keep up in good times or over complete market cycles.

When examining the track records of trend-following programs, be sure to focus on the length and quality of the track record. Many of these firms have relatively short track records and some have no track record at all.

I was surprised when I looked through the materials of one firm at a recent conference. The firm had given a presentation, and what a reasonable person would have assumed was its real performance turned out, on closer examination, to be hypothetical.

Make sure you read the fine print. Back-tested performance is not performance at all. It only shows what would have happened if a set of rules had been applied to historical data. In fact, the rules were probably developed by looking back at that very data and tailored to fit the patterns that emerged from that data. The future may look very different.

Another more subtle thing to focus on when you’re looking at a firm is the amount of assets it managed during the period covered in its track record. If you look closely, you will find that some firms “out of the market” in 2008 actually managed very little money at that time. They might have managed a handful of client accounts using their strategy, but once they “called the market” in 2008 they packaged up those accounts as a “product” for wider distribution.

The problem is that you don’t know how many other strategies these firms were managing that didn’t make the grade and were simply buried by the side of the road. When you put your clients’ assets at risk, you want to entrust them to a firm that has a well-thought-out, disciplined process that will work in different investment environments. A firm that simply incubates strategies and then pushes its “winners” out the door does not fit this description.

Behavioral Considerations

There is a behavioral issue associated with trend-following systems. Clients take comfort from being “in sync” with what is happening in the markets. They may not be happy when their portfolio declines in value in a down year, but they understand that they are subject to market forces. They are comforted by the fact that their performance is consistent with that of others around them. There is safety in the herd.

Obviously, clients are more than willing to be out of sync during down years. That is, they would gladly sit on the sidelines when the market is tanking, even if others around them are suffering losses. But they are not so happy if this comes at the cost of giving up returns in a rising market. They don’t like being on the sidelines when others around them are making money. This often happens with trend-following systems.

Check Out The Team

The usefulness of a trend-following system is entirely dependent on its ability to identify and appropriately react to market trends (and not react to false trends). Obviously, this is not an easy task. There is good reason why Jane Bryant Quinn famously quipped: “The Hall of Fame of market timers is an empty room.” Markets are complex, ever-changing and full of surprises.

It is hard to know how many trend-following firms have produced solid, long-term track records for their clients over the years. But it is clear that devising systems that consistently identify price trends in various asset classes and profit from them is an incredibly difficult task.

If you are considering investing your clients’ assets in a trend-following system, you should take a close look at the team that developed and maintains the system. Do they have the type of qualifications and experience to keep timing the market successfully when so many others have failed?

Find out where the system came from and who maintains it. I have encountered “investment firms” that are nothing more than marketing organizations that buy their “signals” from an independent provider. There is nothing inherently wrong with this approach as long as there is full disclosure and you are satisfied with the qualifications and experience of the independent provider. Just make sure the firm you are dealing with has guaranteed access to the signals so you don’t wake up one day to find that your clients are drifting in a rudderless boat.

Don’t be afraid to ask how the system works. In the investment world, luck and skill can look very much alike for long periods. Do your best to determine if that great 2008 performance can be repeated. You can’t invest in yesterday’s track record. But you can tell whether the system is likely to produce good results in the future. A Final Word In a video posted on the Web site of a well-known trend-following firm, the firm’s principal says that before 2008, “We listened very, very closely to the markets on a daily basis and the markets were screaming at us and telling us something really, really bad is coming. And if you listen closely to the markets, you can hear what they’re saying and you can make those important moves.”

Balderdash!

Investing is not some Disney movie where Uncle Remus talks to Br’er Rabbit and Br’er Bear about what’s happening up on Chickapin Hill. Markets don’t talk to us no matter how closely we listen and they certainly don’t scream at us about really, really bad things that are coming. This kind of nonsense trivializes the difficulty of protecting clients in volatile markets and wrongly suggests that trend following offers some sort of safe haven.

It is understandable that we would be drawn to a firm that missed the carnage in 2008. Clients still remember the pain and want us to protect them from future trouble. But in the rush to find shelter for our clients, it is still important to critically assess every approach to investing no matter how well it did in 2008.

When I was in Little League, in the South Palo Alto Lions, we played against a pitcher who could throw an amazing curveball. It came toward the plate in a giant arc, like a big, fat, slow-motion rainbow. It was the juiciest pitch you have ever seen. No one could resist swinging at it. The problem was, just as you took your swing, that thing would break and your bat would slice harmlessly through thin air while the ball plopped softly into the catcher’s mitt.

In investing, what looks like a “fat pitch” is usually an illusion. If the trend unfolds in the right way, a trend-following program can provide downside protection. But you should understand that these programs have their risks and limitations. None can totally avoid declines in portfolio value while capturing all the upside. No investment process can produce that result.

If you decide to invest some of your clients’ assets with a trend-following firm, here is a thought: Divide the money between that firm and a firm that remains more fully invested at all times. Fully invested portfolios have historically done well when the trend followers have done poorly. They can soften the blow if the trend followers don’t catch the trend just right.

Scott A. MacKillop is president of Frontier Asset Management, a firm that constructs and manages portfolios for financial advisors and their clients. He is a 36-year veteran of the financial services industry.