Trend following – raincoat or coat rack?

Before we can declare anything dead, we first have to define what exactly it was that was living. Which heart supposedly stopped beating? Historically, there have been different ‘life forms’ known as trend following.

Trend following

The famous term trend following was established in the eighties and nineties by a colorful group of traders who made money by trading trends in futures markets in a systematic way. Some of these traders shared a common background. A group of U.S. traders originated from the Turtle project set up by Richard Dennis and William Eckhardt in the early eighties. In the UK, many of the early CTAs emanated from Sabre Asset Management in the eighties and nineties. Meanwhile, across the U.S. and Europe, various independent traders started to employ similar trading approaches. This included Transtrend, which started in the late eighties as a research project within a physical commodity trading firm. All these CTAs had one thing in common: they focused on trading futures markets. That’s what the ‘T’ in CTA stands for. All they did was generate trading performance.

Seeing trend following as an investment style started to become popular after the Credit Crisis following a series of papers published in academic journals that showed that the returns of these CTAs could be largely explained by the simulated returns of time series momentum strategies. These publications paid minimal attention to the diversification aspects within the universe of traded markets and made rather naïve assumptions about the market impact of actual trading. The degrees of freedom in this particular definition of trend following were largely constrained to the choice of parameters – i.e., speed – and the composition of the investment universe. The main embodiment of this life form continued to be the publication of academic papers presenting simulated returns as ‘empirical evidence’.

Explaining returns, however, is not the same as describing the trading activity that actually delivered those returns. We cannot exclude that some did apply the suggested time series momentum strategies, but at least we never did within our Diversified Trend Program. Nor would we recommend implementing most of the other operational assumptions made in these publications.

With respect to the source of returns that trend following CTAs aim to capture, there also seem to be different flavors. To mention just two of them:

• ‘Stop-and-reverse principle’: This is based on the idea that individual markets tend to drift in a certain direction for a sustained period of time, among others due to crowding behavior of market participants. The disintegration of such crowds is breeding ground for the formation of new crowds. This trend following strategy moves with these crowds.
• ‘Diversified portfolio principle’: This is based on the idea that market-impacting developments typically start as local developments with only local market impact before they broaden. The expansion of such developments typically causes market reversals in some markets, while directional moves in other markets accelerate. We at Transtrend are supporters of this principle.

The easiest way to distinguish between these two schools of trend followers is by listening to their over-simplified sales pitches. The stop-and-reverse pitch is that you cannot lose with trend following, because when the market reverses, the program will reverse its position as well. The diversified portfolio adherents preach that you cannot lose with trend following, because when one market reverses, losses on positions in that particular market will be compensated for by profits in other markets.

Again, both of these stories are over-simplified. But the different principles do lead to different choices made in an investment program. Does this then define two distinctly different investment styles? We wouldn’t say so. Most importantly because we have reached the conclusion that the importance of style has become grossly overrated. And not only with respect to trend following.

Investment style

If we forget about all the elaborate reasons behind every buy and sell for a moment and just focus on where it all leads to, we believe there are essentially only four different basic investment styles:

• Long only – the investor or investment manager controls a portfolio of long positions in assets and trades around it. Substyles can be defined, determined by subsets of assets to which this style is applied.
• Short only – the investor controls a portfolio of short positions in assets (most often only equities) and trades around it.
• Trend – the investor controls a portfolio of positions in the direction of anticipated trends in global markets and trades around it.
• Market neutral – the investor controls a portfolio of positions in only idiosyncratic market moves and trades around it, while staying away from the major global directional risks.

These different styles are certainly relevant. They explain in what kind of market environment an investment program can be expected to do well and in what kind of market environment it will likely struggle. They therefore explain a large part of the correlation between different investment programs. But they do not explain their performance. Investment performance is – apart from certain regular streams of income (such as interest on cash balances, dividend and lending fees on stock holdings, rents received) – completely determined by the act of trading: by the differences between buy and sell prices. The real art of investing is simply selling high and buying low.

However, over the last two decades, the investment industry has gradually become captivated by the idea of style determining everything. This development has been fueled by different factors, including scale advantages of large players in the industry, academic fashions and a narrow focus on costs by among others regulatory authorities. One after another, (also) alternative investment strategies have been mimicked by (hypothetically) applying yet another naïve rule-based approach to yet another subset of tradable assets. Within this school of thought, trading activity is mostly regarded as a source of costs rather than as the foundation of investment returns, let alone as the force that makes markets move and function (or not function).

Considerations regarding market functioning, how markets are imbedded in society, and the important role that investors have in the marketplace don’t seem to be part of this style-focused approach to investing. And even more alarming, nor do they play a role in many recent academic publications. In an earlier article, we discussed one of the fruits of this school of thought: factor investing. A world described as being driven by some highly abstract factors, completely disconnected from the real factors that drive real-world people like you and me, our fellow citizens and the billions of people that nowadays are just as closely digitally connected to us. Their supply and demand, as well as their hopes and fears ultimately drive markets.

In the early seventies, many composers occupied themselves with serial, atonal music. Among them the Dutch composer Simeon ten Holt. But then he realized, as he wrote later:

“Composing had become table work, merely arranging mathematical series. There was a pseudo-creative process taking place in a kind of present without a past.”

We feel that this description more and more applies to the contemporary investment industry. After suffering from what he called “artistic and creative anemia”, Ten Holt returned to tonality. With our Diversified Trend Program, we’ve in the past few years made a comparable move away from this narrow arrangement of mathematical series, back to its origin: trading markets in a non-conventional way.

Around us, our original peers have entered into different directions as well. Some have added alternative markets, others have not. Some have added alternative data, others have not. Some have lengthened their time frames, others have not, or have even chosen to shorten them. Some are now applying machine learning techniques. Some have outsourced their trade executions, while others are strengthening their own control with respect to market access. And so on. Is trend following dead? In nature, such a growing diversity is seen as a sign of vitality, not a sign of looming extinction!

Coat rack

Non-conventional trading starts by acknowledging that the chosen style, whatever style that may be, does not determine what should be done and what cannot be done. Moreover, the degrees of freedom in the investment program should not be dictated by the chosen techniques. A non-conventional trader has the freedom to develop whatever technique is necessary to achieve the ultimate goal: trading results. In a well-designed trading operation the technique serves the trader, not the other way around.

In essence, the chosen style should just be the coat rack that serves as a convenient tool to organize and store coats – i.e., trades. Nothing more. If the coat you are wearing doesn’t protect you enough against the current weather conditions, you don’t change your coat rack – you change your coat. Nor would you argue: “I cannot use an umbrella because I have a coat rack.” We believe that every investment program requires continuous adaptation to stay in sync with changing markets, but all these upgrades are pretty meaningless if they essentially constitute nothing more than a makeover of the coat rack.

Instead, we prefer to focus on fresh trading ideas. Nothing fancy, just good old selling high and buying low. If we are not restricted by too narrow a definition of our investment style, there aren’t really that many trades that wouldn’t fit on our coat rack. Certainly not within trend following. Whereas in long only the buying still has to be done before the selling, in trend following we aren’t even restricted by that.

We often hear that the growing regulatory burden is making it harder to invest. There is some truth in that, but let’s be fair: in the current investment management industry, most restrictions are self-imposed. By the ultimate investors as well as by the investment managers themselves. The main outcome is often just a false sense of control. To free ourselves from these restrictions is the ultimate demonstration of life.

Which brings us back one more time to our initial question: Is trend following dead? It could be. But if it is, it didn’t die. The coat rack is clearly dead. It’s made out of dead material. But the people who are using that coat rack to store their coats, their vests, their hats, their umbrellas, or whatever suits them, the people who are replacing these instruments whenever they feel they should, these traders are very much alive and kicking.