Quant investor of the year: GAM Systematic Cantab

Risk Awards 2020: Cantab benefitted as diversification started working once more

Ewan Kirk
Ewan Kirk

After faltering in 2018, diversification came good again this year for quants. GAM Systematic Cantab was a big beneficiary.

Cantab’s Quantitative Programme was up nearly 31% for the year to 25 October 2019. Its Core Macro Programme was up nearly 19%. That compares with gains of 7% for the HFRX Equity Hedge Quantitative Directional Index. Hedge funds overall have returned about 6% on average year to date.

Head of research Frédéric Desobry says the improved performance is down to the firm’s broad spread of models – trading across hundreds of markets, using different trading styles and a spread of forecast horizons. “Different types of diversification are at play in the models and they all contributed in their own way to enhancing the performance,” he says. With its broad spread of bets, Cantab was able to “lever up in a good market”.

The multi-strategy Quantitative Programme, which combines Cantab’s best ideas, trades 300 fixed income currency and commodities markets, for example, and more than 3,000 equities.

Like many quant firms, Cantab went through a period of soul searching at the beginning of the year – researching how correlation structures and volatility might have changed and whether its assessment of the market environment was off base.

In 2018 Cantab’s flagship fund had taken a 23% loss amid cross-asset reversals in February and October, which dragged down performance for quant funds across the board.

The firm found little that it thought needed fixing, though. “We didn’t find reasons to either change the signal definitions or the portfolio construction methodologies,” says Desobry.

Ewan Kirk, Cantab’s chief investment officer, told Risk.net in July that blaming quants’ tough times on changes in correlation or volatility was to fall into “retrofitting stories” to the facts.

Instead Cantab stuck to its guns, benefitting as much from what it chose not do as what it did.

Take trend following, the investing approach Cantab historically was best known for but which in its basic forms has generated negative returns in seven years of the past 10. Trend following constitutes around 25% of holdings in the multi-strategy programme. That contrasts with some other quant funds that remain “dominated” by the strategy, Desobry says.

Discovery has done well. That has proved helpful in a year where more middle-of-the-road futures markets have proved less helpful
Frédéric Desobry, Cantab

Cantab was able to generate returns far ahead of pure trend managers in a year when trend following has experienced positive but patchy performance. Societe Generale’s trend-follower index is up more than 8% but has seen losses in almost as many months as gains.

“The fact trend has been quite mixed and had some strong months and some bad months has helped us on a relative basis,” Desobry says.

Likewise, equity value positions have dragged down performance in some big-name funds in the quant space in 2019, but here also Cantab’s positions are “relatively small”.

“Equity value is one of the many factors in the single-name equities cluster, itself one of several clusters in the multi-strategy Quantitative Programme,” Desobry says. “Multi-asset value has actually done OK this year. But value in equities has done terribly for a couple of years, so for anyone who has outsized exposure to value equities, it’s been a challenging time – even including the positive performance that came in value in September,” he adds.

Cantab derived further diversification from less well-trodden markets including power, Chinese offshore futures and interest rate swaps outside the US and western Europe.

These are traded in Cantab’s Discovery Fund, a trend-following strategy in so-called frontier markets. Discovery, in turn, is also included in the unit’s multi-strategy Quantitative Fund. Cantab added more than 10 new Chinese markets to Discovery this year.

In a world where investors want lower fees, and everyone wants minimum transaction costs, for us to have something that saves us many tens of basis points a year in trading costs is very helpful
Frédéric Desobry, Cantab

“Discovery has done well. That has proved helpful in a year where more middle-of-the-road futures markets have proved less helpful,” Desobry adds. Discovery delivered 14.4% year to date, as at 31 October 2019.

Meanwhile, Cantab’s carry model was a strong performer in both the Quantitative Programme and Core Macro Programme. Correlations with trend following stayed low and the model benefitted from raising fixed income exposures early in the year. Desobry describes the carry model as “very opportunistic” and “able this year to rotate rapidly” between asset classes.

The change in fortunes has been seen also in asset inflows. GAM, which acquired Cantab in 2016 for $217 million, was forced to write down the value of the acquisition by $60 million in 2018 due to lower than hoped-for asset inflows as quant funds fell from favour with investors. Year to date, GAM Systematic, the division of GAM that houses the Cantab funds, has seen assets under management grow by $700 million.

As for other causes of Cantab’s improved returns, machine learning contributed to the firm’s success, Desobry says, but was not a primary driver.

“The highest value we get out of machine learning is something that looks very dull, which is using machine learning within our execution algorithms and making very short-term predictions on price moves that impact whether we trade into positions fast or slow,” he says.

“In a world where investors want lower fees, and everyone wants minimum transaction costs, for us to have something that saves us many tens of basis points a year in trading costs is very helpful,” he adds.

Desobry picks out the firm’s code infrastructure – all of it programmed in Python – as a more critical source of advantage, though.

“That might sound like a trivial thing, but it’s not. Very many competitors have lots of legacy systems and legacy databases, and the fact that at GAM Systematic everything is in one programming language means that our infrastructure and our end-to-end processing of signals to positions to reconciliations is all very low touch,” he says.

“That means valuable research time can be spent on data sources and new algorithms, and not on multiple legacy systems.”

Looking forward, Cantab will be hoping those new ideas bring still more diversification.

  • LinkedIn  
  • Save this article
  • Print this page  

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact [email protected] or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact [email protected] to find out more.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: