Risknet Wintons David Harding On Turning Away From Trend Following Jean Wimmerlin 528149 Unsplash
19 July 2018

By Rob Mannix, Risk.net

When asked the ideal dollar figure for assets under management in Winton’s newly launched trend-following fund, David Harding’s answer prompts a double take. “Nought would be perfect”, the firm’s founder says. He is only half joking.

After 20 years, Harding is steering the $30 billion firm he built increasingly away from trend following. Winton is cutting the weight of the strategy in its main fund – the Winton Fund – by half, from 50% to around 25%.

The new launch is for clients that want to keep a higher exposure, but frankly Winton is hoping few of them do.

“It’s a significant commercial move, a significant reorientation of Winton’s resources internally and market positioning,” says Harding. It is also a big step for one of trend-following’s pioneers. Harding is the ‘H’ in AHL, the quant firm he co-founded in the late 1980s and sold to Man Group before launching Winton in the late 1990s.

He talks about “steeling” himself before making the decision. Yet this is no dramatic rift with the past, he asserts.

Trend following might have worked “fantastically well” for more than three decades, but the wane of the strategy underscores Winton’s view that “there are no immutable laws in finance”, he says.

Speaking to Risk.net in his office at Winton’s headquarters in London, Harding damns the view rooted in the work of academics Eugene Fama and Kenneth French that markets can be seen as efficient price-discovery machines where enduring rules apply – the so-called efficient market hypothesis.

“Everything we’re doing,” he says, “is centred on the ways in which efficient market theory isn’t true”.

The last straw

Harding’s long-held beliefs on market inefficiency met a more pressing reason for a change of tack during February’s market turmoil.

February was “the straw that broke the camel’s back”, Harding says, convincing him trend following had become “further overcrowded”.

“My estimate of the level of slippage caused by trend followers’ position adjustment after February edged up. And when I say ‘edged’ I mean a 10–20% increase.”

Trend-following hedge funds typically scale positions based on volatility, leading to fears about what would happen if funds were forced to cut exposures due to a sudden increase in volatility. During the first week of February, the Vix index of implied volatility in equities spiked from around 13 to 37.

At the end of the month, the S&P 500 was down close to 4%. The Winton Fund lost 5% in February and Societe Generale’s index of trend-following hedge funds posted a 6% loss, its second worst since inception in 2000.

The episode highlighted the growing difficulty of satisfying the twin objectives typical of investors in trend-following funds, Harding says – maximising risk-adjusted returns, but also minimising the correlation of the portfolio with equity markets, to which many of these investors have high exposures.

We have become famous as a trend follower. That’s a great thing. It gives you this reputation and franchise. But it’s also a bit of a prison. It’s your enemy and your friend at the same time.

“The tension arising as a result of trying to satisfy these two masters” had become “steadily more stretched”, Harding says. “After February we decided we had to make the distinction clearer to our clients.”

Winton will reduce the weight of trend following in the Winton Fund, a multi-strategy fund that manages more than $9 billion in assets, and in other funds managing a further $3.5 billion by “about” January next year.

The new stand-alone trend-following fund has a 1% management fee and no performance charges – Harding says the returns no longer justify hedge-fund type fees. The strategy delivered roughly a 0.5 Sharpe ratio over the past 10 years, Harding estimates, versus 0.8 for the Winton Fund overall.

“We have become famous as a trend follower. That’s a great thing. It gives you this reputation and franchise. But it’s also a bit of a prison. It’s your enemy and your friend at the same time.”

Harding says he worries that “all the clients might say: ‘Oh we’ll have the thing that’s cheaper’ ” – but so far the signs are encouraging. He thinks 5–10% of assets might have switched to the new fund by year end.


A cynical reading of this could be that Winton is responding to the commoditisation of trend following by banks and asset managers offering alternative or smart-beta products.

Broadly, these lower-cost products are based on the idea that funds that systematically tilt towards securities with price momentum can harvest a lasting premium from investors’ tendency to favour winners and shun losers.

Several of Winton’s peers have responded to the competition and to sluggish sector returns by targeting new markets or offering simpler, cheaper versions of their strategies.

Man AHL has led moves into more esoteric assets, where competition is less intense and trends are arguably stronger. Aspect Capital launched a low-cost version of its trend-following strategy in 2016.

But Harding rejects absolutely the idea Winton is being undercut or that trend following can be thought of as beta at all.

Momentum is not a fixed law of markets, he says; it was a good trade and it got crowded. “If any trade gets very crowded then it can backfire,” he says. “It’s a standard market thing. If everyone tries to do the same thing at the same time it goes wrong.”

The implications for providers now looking to feed on what hitherto was Winton’s bread and butter are evident. “I expect trend following to gradually deteriorate and I’ve been on record saying that for years,” Harding says.

Equally he rejects the suggestion that because trend-following returns have waned, the opportunity for firms like Winton has shrunk. Such thinking comes from the flawed assumption that markets fit the efficient market model, he says.

Here Harding sets out his view of market returns as “in general the product of a complex, multidimensional chaotic process” impossible to characterise by any simple mathematical model. Once he hits his stride on the topic, there is little stopping him.

“The track record of firms such as Winton and AHL over the last 30 years is a living refutation of the efficient market hypothesis put forward by Fama and French,” he says. “Seeing your hypothesis falsified and then trying to defend it by retrospectively introducing extra variables and fitting the data to them is hardly the height of good scientific practice.”

The idea of markets as efficient can be useful in the same way as Newtonian physics is useful, he says. But losing sight of the theory’s flaws risks allowing dogma to stand in the way of progress. “It’s a set of intellectual ideas that, if I were being really critical, I’d say forestalls all thought.

“There are no laws of finance. Many people in finance seem to be trying to treat it as a physical science. This is a mistaken approach. Their philosophy is wrong.”

The Winton way

These views inform Winton’s investment approach in some intriguing ways.

For one, it means championing the value of wisdom in investing. Harding references Warren Buffett several times as an example of someone with an admirably broad base of knowledge. Charlie Munger’s book On Success shows he is “incredibly widely read”, Harding says.

The quant culture in finance, conversely, he sees as often dogmatic. “Many mathematicians and programmers don’t read a lot of history, philosophy, religion,” Harding says. “That’s what you need to be a successful investor. You need to have an eclectic outlook on the world.”

Secondly, Harding is not hung up on the interpretability of signals.

“If I could find strong evidence in some data that there was a phenomenon we could make money from, I wouldn’t care whether I understood it or not.”

People always say to me: ‘This is what works.’ I say: ‘Worked’

As a hypothetical example he suggests the Bank of Japan might rebalance its foreign currency reserves at the same time each week. “You might find that anomaly, find that it is as strong as hell,” Harding says. “You don’t know why it works. But you can make money out of it.”

A third point of influence is Winton’s insistence on the stringent testing of ideas.

Strong evidence of a tradeable signal requires more than “running a five dimensional optimisation back over the data” and getting a strong result, Harding says.

“People always say to me: ‘This is what works.’ I say: ‘Worked’.

“‘Works’ implies a property of a population. ‘Worked’ implies it is the property of a sample. ‘Works’ implies this is a fundamental underlying property of nature. ‘Worked’ implies it might have been significant or might simply have been a random feature of the data.”

As for the future, Harding sees no comparative advantage in seeking out strong signals which mostly will be quickly identified and traded on by others. Instead, Winton’s focus is on continuing to seek out weak signals but more of them.

“For weak signals there’s relatively little time and effort devoted to looking for them by others,” Harding says. The approach, which leads Winton to draw its researchers often from academic disciplines where the problems of using data with low signal-to-noise ratios are well understood, has served the firm well.

Winton’s full-year return for 2017 was 7.9%, compared with less than 2.5% for SocGen’s index that tracks funds with a purer focus on trend following. The Winton Fund is up 2.5% net year-to-date, whereas the SocGen index has lost more than 4%.

In other commercial moves, Winton has successfully registered with the Asset Management Association of China as a private securities investment fund manager, enabling the firm to develop onshore investment products for qualified investors in China.

The firm also recently span off its Hivemind data science unit to run as a stand-alone business. Harding describes Winton’s core skill as “using powerful computers to look for empirical patterns and relationships in data” and the company describes itself as an investment management and data science company, possibly pointing to more ventures in areas beyond finance.

Winton already invests in businesses in fields such as cybersecurity and healthcare where data science is set to play a pivotal role in future.

As for Winton’s core business, though, the impression Harding leaves is clear: he is not regretful about decoupling from the strategy that made his reputation.

“When Mick Jagger gets up on stage he’s always got to play Paint it Black and Brown Sugar,” he jokes. “We’ve had to play Paint it Black and Brown Sugar over and over again.”

Reprinted with permission from Risk.net.