Is the trend your friend?

It’s not the risk events that we see as possible that hurt us, it’s the unforeseeable ones that cause the real damage as they aren’t factored into our psychology, market prices or anyone’s precious models that our industry likes to rely on. These unforeseeable, improbable, but high impact events are known as black swans. Examples include: the 2011 Japanese earthquake, 9/11, COVID and arguably LTCM and the GFC, depending on if you see these as foreseeable or not. The point is that hugely impactful events that are completely unpredictable and unprecedented do occur². It’s in these sorts of events that you’ll be glad to own ‘alternative’ assets that have different drivers of return and that can benefit from market volatility and the prices of stocks and bonds decreasing. In the good times, these assets are often a drag on performance but over the long term, you’ll be glad you had them. Like buying home insurance, you hope you never need it and for most it is unutilised, but it’s certainly not a waste of money. One such alternative investment strategy is trend following; so, is the trend your friend?

Trend following is ostensibly a very simple concept: identify the assets that are trending. For example, the price has been steadily going up or down for a period (around 3 months for a medium-term trend follower), build a position that benefits from the trend continuing (i.e go long or short), hold them whilst the trend continues and exit the position as the trend weakens and before it reverses. It sounds too simple to work, but over a 200-hundred-year period from 1814 to 2014, trend following was proven to work in financial markets across multiple geographies and asset classes³. And it’s not just a return for taking extra risk, it’s a consistent and persistent market inefficiency⁴; there’s very few investment strategies that have 200 years of supporting data. Importantly, this period includes many market and economic cycles, drawdowns and bull markets, so it has proved itself across good and bad periods. Ultimately, it is a more reliable data sample than samples that go back only 10 years for example.

Over a 200-hundred-year period from 1814 to 2014, trend following was proven to work in financial markets across multiple geographies and asset classes. And it’s not just a return for taking extra risk, it’s a consistent and persistent market inefficiency; there’s very few investment strategies that have 200 years of supporting data.

Most market inefficiencies are competed away by hedge funds and investors quickly, but what makes trend following different? Some academics suggest that it’s people’s initial under-reaction to fundamental news that causes trends to persist. This may be true for markets driven by fundamental news where most participants have access to this news, but this wouldn’t explain trend following working in markets that are speculative, not supported by fundamental information, or markets where there is little fundamental information or access to it. So, in addition to under-reactions, human behavioural errors may also cause trends to persist, errors such as herding bias, confirmation bias, action bias, loss aversion, over confidence and many more⁵. Unsurprisingly, I’ve started to see trends everywhere, which is at least partly due to confirmation bias. One example is friends investing in cryptocurrencies; when they were going up, they loved them and bought more and used the price going up as confirmation that they were right and this was the death of fiat currency, but when they were going down, they hated them and sold - all of this despite no real fundamental changes. Investment fund closures and launches follow a similar pattern. Asset managers build what sells, if growth funds are selling, they launch growth funds, if they aren’t, they close them and launch value funds, thus reinforcing trends in equity factors (coincidentally, equities were the best performing asset class for trend followers over the long term²).

There are other reasons why trends may persist such as the homogeneity of risk management systems, risk targeting and the use of stop losses but let’s not get too ‘into the weeds’⁶. Warren Buffet once said, “What we learn from history is that people don’t learn from history”, and until we’re convinced that they do and that risk models can be perfected, which they can’t be due to their understandable inability to predict the future⁷, it seems sensible to believe that trend following can continue to deliver returns.

Providing a good absolute return is great, but by providing a return that is uncorrelated to traditional assets is why it adds values to portfolios and why it is classed as an ‘alternative investment’. Trend followers can go long (benefiting from a price increase) or short (benefitting from a price decrease). In fact, trend followers tend to deliver their best returns in the most extreme return periods (when equity market returns are very good or very bad) as that’s when strong trends in various asset classes exist and trend followers profit the most. For instance, the global equity market’s worst drawdown (using the MSCI world index hedged to USD) was -50% from October 2007 to September 2009⁸. Over this period, a trend follower index (the Barclays BTOP50) was up 17% and in the popping of the tech bubble from 2000 to 2003, world stocks were down -37% whilst trend was up 46%⁹. Why is this? As trend funds could short assets that were showing strong negative trends (equities) and go long assets positively trending (government bonds, for example). These trends persisted for some time, hence the ability of trend funds to latch onto them and deliver returns. Therefore, an allocation to trend following helped reduce the maximum drawdown of a 60/40 portfolio notably. Since 1986, trend followers have delivered an equity like return (7.0% annualised vs 7.7% for equities), with two thirds of the volatility, a max drawdown of -16% vs -50% for equities with almost no correlation to bonds, stocks, commodities or the dollar¹⁰.

What’s the catch? In a swift reversal of a trend (e.g when equity markets or any asset class drops suddenly in value) trend followers don’t have time to react and returns can suffer. At these points, trend following funds can have some correlation to equity markets. However, they can be long or short at these points, so they could be lucky and avoid this, they are highly diversified across asset classes and the most serious economic events and market impacts usually last long enough for the models to latch onto the new downward trend (but its worth remembering that no strategy can deliver in all environments). The other main drawback is, like equities, it can have periods of uninspiring returns with the last ten years being a good example. However, with 200 years of supporting data, do we really think the market and people have changed so much that trend following won’t work again? I think not.

Overtime, trend follower funds have had to find new ways to differentiate themselves and improve returns relative to peers. One way of doing this has been to add new alternative markets to their funds. Many funds trade over 400 markets, offering great diversification and the potential to profit from trends in many markets. The best markets to add to a portfolio have low correlations to the existing trend following portfolios and traditional asset classes. These assets often have idiosyncratic price drivers that are uncorrelated to traditional drivers of financial markets, examples include cocoa, cheese, robusta coffee, hogs and orange juice. From the end of September 2023 to April 2024, cocoa rocketed from just around $2k a tonne to just under $10k driven by lack of supply due to a drought in West Africa and chronic under investments in cocoa farming¹¹. So next time you’re annoyed at the price of your chocolate bar doubling from one pound to two try to focus on the positives, you might have made a lot more than a pound from that price increase because the trend is your friend (well most of the time at least).

¹ The Black Swan (Taleb, 2008)
² Two centuries of trend following (Lempérière et al, 2014)
³ Two centuries of trend following (Lempérière et al, 2014)
⁴ Risk management Garleanu and Pedersen (2007) as cited by AQR 2024
⁵ Risk management Garleanu and Pedersen (2007) as cited by AQR 2024
⁶ The Black Swan (Taleb, 2008)
⁷ https://www.man.com/maninstitute/trend-following-what-not-to-like
⁸ https://www.man.com/maninstitute/trend-following-what-not-to-like
⁹ https://www.man.com/maninstitute/trend-following-what-not-to-like
¹⁰ https://www.jpmorgan.com/insights/global-research/commodities/cocoa-prices

The information in this document is believed to be correct but cannot be guaranteed. Opinions, interpretations and conclusions represent our judgment as of this date and are subject to change. Past performance is not necessarily a guide to future performance. The value of assets such as property and shares, and the income derived from them, may fall as well as rise. When investing your capital is at risk. Copyright Investec Wealth & Investment Limited. Reproduction is prohibited without permission.