“From the start we were repeatedly told that the world does not need another CTA, and that trend following is dead and does not work anymore, as markets have become more and more efficient. Yet we have established an impressive track record over more than 16 years now that is based on what we call a ‘trend following market inefficiency’. Trend following is alive and well,” says Quantica Capital founder, Dr Bruno Gmuer.
The strategy is not predicated upon the multi-decade bond bull market continuing. In the first quarter of 2021, trend following CTAs including Quantica are delivering strong returns amid a sharp selloff in government bonds, thanks not only to bond downtrends but also commodity and equity uptrends and currency moves, which partly reflect investors discounting inflation. Quantica’s return profile is recognizably that of a trend following CTA, with a long-term correlation of around 0.70 to the SG Trend Index. Correlation measures only the direction and not the quantum of returns however, and Quantica has outpaced peers for years, including in phases that were challenging for some of them, such as the post-GFC period 2009-2013. Its outperformance of larger trend following peers present in the SG Trend Index has averaged almost 5% p.a. since inception 16 years ago and has been almost 10% p.a over the past five years – a period that was lacklustre for some trend following approaches that have been approximately flat.
Quantica Managed Futures (QMF) has received The Hedge Fund Journal’s UCITS Hedge performance awards in the trend following category on several occasions. The QMF Program has also delivered strong absolute returns: an annualized net return of 8.0% per year on a volatility of 10.5% p.a in USD. It has generated cumulatively more than $450 million in gross P&L for its investors since inception in 2005, yet assets remain below $1 billion, illustrating how performance is prioritized over asset gathering.
While many seasoned CTAs and quantitative strategies are clustered around London, New York and Chicago, Quantica is today amongst only a handful of long-standing systematic investment managers in Switzerland. This helps to explain its differentiated investment philosophy, process and culture – which eschews optimization. Quantica is also something of a maverick compared with some other funds in Switzerland started by traders who worked for, or spun out of, some of the world’s largest hedge fund managers. Quantica’s investment process was developed independently and autonomously: “We had absolutely no trading background and no idea what other CTAs were doing and built the investment process in a fully independent and autodidactic way, without any mentor or external adviser. Of course, I was inspired by the work and achievements of Ed Thorp, which I have always truly admired, and like any ‘wannabe trader’ at that time by the Market Wizards series of books and by the Turtle Traders story. I was naïvely thinking if they can do it, we should have a chance too,” says Gmuer, a mathematician and actuary with a PhD in financial economics.
Gmuer has been CIO and portfolio manager throughout. His career began in academia, teaching financial markets and game theory at the University of Zurich, but he rejected theoretical faith in the orthodox efficient markets hypothesis, in favour of a distinctive and conceptually different trend following approach based on quantitative signals related to risk-adjusted, relative outperformance. The genesis of Quantica’s models lay in quantitative tactical asset allocation models developed by Gmuer while he was leading the quantitative team for the CIO office of Julius Baer.
Gmuer resolved to form his own firm in order apply his models as a fully-fledged standalone strategy and to assert the independence that he cherishes. He did not want to join a larger organization that might have interfered with the process, and nor did he want to give away seed economics in the management company. Gmuer was fortunate to secure seed capital of $100 million in 2005 from one of the largest global macro managers at that time, which did not have systematic capabilities in house. This provided permanent capital for six years, which permitted Quantica the luxury of focusing on the investment process without worrying about marketing or asset raising, though that suddenly changed.
He recalls how the loss of the seed in 2010, resulting in a 90% decline in assets, marked the most difficult moment in Quantica’s history: “Despite having built a great track record, we had to change the business strategy to attract outside investors for the first time. Luckily, we very quickly won sophisticated and experienced investors mainly from the CTA fund of funds side, and added to our assets, regaining our former level by late 2011. Then by 2012 we reached $300 million, which increased every year, as we launched the UCITS in 2015, became fully regulated in Switzerland and raised more institutional capital”. Today the investor base is much more diverse, including institutional investors such as insurance companies and foundations, wealth managers and banks.
Investors have been drawn to Quantica by its realistic objectives, consistent performance profile before and after 2008, ongoing style consistency, transparency, and efficient operational structure.
Quantica believes that a gross Sharpe ratio target of close to one is realistic for a medium-term strategy trading in liquid markets, and this has been achieved. Gross volatility, calculated using daily data, has also averaged very close to its 12% target (the performance fee reduces net volatility).
Quantica has not changed risk or return targets for the post-GFC period, and views central bank intervention as a double-edged sword: “Central bank activity can be a negative in breaking or slowing down some trends but can also be beneficial in creating stronger trends for systematic diversified trend followers. For example, the Japanese approach termed “Abenomics” with its structural reform, fiscal expansion and monetary easing had sizable medium to long-term effects on Japanese equity, bond and currency markets that translated into solid profits for Quantica and other systematic trend followers. In a similar way, we believe that interventions by the ECB and monetary policy by the Fed since the great financial crisis have significantly amplified asset inflation across equity and bond markets and certainly added to trend following returns,” says Gmuer.
The strategy is intended to be an attractive standalone investment, offering uncorrelated long term absolute returns, and provide “smart diversification” versus conventional asset classes – Quantica’s biggest drawdown, in 2015, was not in a difficult year for conventional asset classes. It aspires to be an all-weather allocation thriving through different longer-term cycles. “The concept of “smart diversification” entails dynamic correlation and risk during different market regimes. We aim to be positively correlated with rising markets and negatively correlated with falling markets – on average over long periods, while also sometimes capturing a positive risk premium in rangebound markets. The beta and correlation in bear markets are on average negative – but not always,” says Gmuer. “The diversification benefit does not always work – particularly during short lived market corrections – so “crisis alpha” would not be the right label – and trend following should not be mixed up with tail risk strategies, which provide more predictable payoffs at the expense of more regular insurance premium costs, which often make them expensive for investors to hold for long periods”.
In common with tail risk strategies, traditional trend followers can be characterized as being long volatility. A simple trend following strategy can be replicated through owning an at the money straddle, whereas Quantica is not aiming to be long of volatility. “We are neutral volatility, and the return pattern is pretty symmetrical with near zero skew,” says Gmuer.
If trend following CTAs – including Quantica – are not a guaranteed source of bear market diversification, they have thus far been more frequently reliable diversifiers than have many alternative risk premia (ARP) strategies. Quantica observes how many ARP approaches have markedly underperformed during equity market corrections, and Quantica Director of Research, Artur Sepp, contributed an article entitled Trend-Following CTAs vs Alternative Risk-Premia to The Hedge Fund Journal on this topic, arguing that ARP approaches are too static, not adaptive enough, and are exposed to significant tail risks, such as selling credit and volatility. Many ARP strategies – and some trend following strategies – are offered at flat fees with no performance fee, whereas Quantica is not prepared to offer the limited capacity of its distinctive strategy without a performance fee.
Returns rather than assets remain the priority; assets are the consequence of a job well done.
Dr Bruno Gmuer, Founder and CIO, Quantica Capital
Quantica’s models are differentiated from other trend following CTAs in three main and somewhat interrelated areas: signal generation, portfolio construction and implementation.
A relative risk-adjusted trend follower
The most striking differentiator is how Quantica defines trends and therefore generates signals. “Quantica measures trends on a risk adjusted and relative basis within its investment universe, which entails ranking trends for its 60 markets on a matrix of about 3,600 coefficients. Each instrument is considered in the multi-dimensional context of its performance relative to the other 59 markets, rather than in isolation. This more efficiently identifies decorrelation, divergence, dispersion and outperformance or underperformance, over a medium to long term time frame. It can also swiftly adapt to changing patterns of correlation and market regimes, such as when risk on markets move to risk off,” says Gmuer.
For instance, Quantica reversed to a short stance in US Treasuries already in September 2020, since in relative terms they were underperforming European government bonds and other asset classes such as equities, FX and commodities as the correlation structures were changing. In absolute terms, bond yields were only a month away from their all-time lows, and Quantica believes that some other types of trend models took longer to switch from long to short.
What is more surprising is that Quantica had a small short in some US and European equity indices in March 2021. Since some of these markets were at or near all-time highs (or at least multi-year highs), this might sound like a countertrend or mean reversion trade, and it would be classified as such in a traditional framework of absolute trend following. But in Quantica’s worldview of relative trends, the fact that relatively volatile equity markets had started to show weaker risk-adjusted trends than commodities, currencies or bonds, was enough to warrant a small short position. This should also be considered in light of the overall portfolio risk budget: the short in equities allowed for more risk to be deployed into commodities and short fixed income, where the models produced stronger signals.
Opportunistic portfolio construction
Following relative risk adjusted trends also means that cross-instrument and cross-asset correlations feed into signal generation at the start of the process, rather than the more typical approach of considering them in portfolio construction later; Quantica is not carrying out mean variance optimization, though the volatility budget applies an overriding volatility control filter at portfolio level, which could adjust position sizes proportionally to stay within defined volatility bands.
Portfolio construction is bottom up and opportunistic, which means that asset class and instrument weights fluctuate considerably according to the strength of the special trend signals. Quantica has found that this dynamic approach adds to performance by underweighting exposures to sideways markets and overweighting those with the strongest trends. For instance, in 2016-2020, an overweight stance in US equities versus European and Asian equities added to outperformance, though this has shifted towards Asian and emerging markets equities since November 2020, when the value style of equities also started to outperform. Gmuer believes that a fixed diversification constraint – while appealing to some investors – would hamper long term returns.
Research prioritizes three areas: expanding the investment universe, signal generation and portfolio construction, and improved execution.
Nicolas Mirjolet, CEO, Quantica Capital
Turnover and transactions costs
The third source of outperformance is an efficient implementation process, minimizing portfolio turnover and transaction costs, partly through trading only liquid markets, but also through lower turnover. As well as tying in with portfolio construction, this also circles back to the signal generation. Quantica exponentially weighted methodology detects trends over periods of 6-8 weeks, though holding periods can be years in some cases.
“The lower turnover becomes more important in sideways markets. When correlations are more stable, we trade less and pick up medium term and long-term premia rather than getting whipsawed, and this is when our biggest outperformance occurs. This is important because markets are in this normal regime about 65% of the time,” explains Gmuer.
Quantica estimates that its total implementation costs are below 0.50% per year (including bid/offer spreads, brokerage and clearing fees and market impact). “We estimate that some other CTAs could be spending three times as much – 1 percentage point more – partly due to their higher portfolio turnover, and because their larger size restricts implementation. On turnover, our round turns per million of 400-450 including rolls are probably 50% below the average for the industry,” says Gmuer.
Even in stressed market conditions, around the coronavirus crisis in 2020, Quantica’s June 2020 quarterly insight, Crisis Liquidity, found that various metrics of trading costs, including bid/offer spreads had not increased relative to market volatility – and aggregate market impact was only gauged as 6 basis points.
The investment approach has remained remarkably consistent over the years, with only one core flagship investment strategy offered (which can be customized to different investment universes). Quantica has stayed style consistent in capturing relative trend inefficiencies in liquid markets over medium term time frames. Its cautious philosophy carefully balances the potential benefits of any innovation against costs, in terms of one or more of: implementation, complexity, return profile or transparency. However, the research agenda is broad, and some nuanced refinements could be contemplated.
“We have not reduced average timeframes nor added shorter term timeframes, since these would not only increase implementation costs but could also dilute the smart diversification and consistent style qualities,” says Gmuer. In some market corrections, including the coronavirus crisis in March 2020, shorter term models would have cut and reversed to short equities whereas Quantica took about three weeks to reduce long exposure from 120% to 20% – and did not actually go short. However, Quantica’s quarterly insight paper, Why Speed Matters – April 2020? finds that shorter term models would not have performed best in various other historical crises.
The research team is now co-headed by Gmuer and CEO Nicolas Mirjolet, who manages the research agenda and priorities, in an open plan office format and with an open mind. “We were very fortunate to attract research team members with a diverse background, experience and an impressive track record and reputation in the industry. They do a fantastic job in constantly challenging the validity of our investment hypothesis, and to gradually improve the quality of our investment process in all aspects from signal generation, risk management, portfolio construction and last but not least implementation,” says Gmuer. A proprietary in-house research tool is used by all researchers, which allows for faster analysis of new data or universes. The research team propose ideas to improve the risk/return profile, which then need to be reviewed and approved by the investment committee. “The strategy and models will remain consistent while research prioritizes three areas: expanding the investment universe, signal generation and portfolio construction, and improved execution,” says Mirjolet.
Since 2008, some over the counter (OTC) and alternative markets, which can sometimes be less liquid, may have generated stronger trends, according to traditional trend definitions. “However, Quantica has always had an investment universe of the most liquid, exchange traded futures contracts globally, which has expanded from around 40 to 60 contracts since 2005. We have not expanded the universe to include illiquid or any OTC markets, which would also increase implementation costs, not to mention operational complexity,” says Gmuer.
The QMF Program has delivered strong absolute returns: an annualized net return of 8.1% per year on a volatility of 10.4% p.a in USD
The investment universe will stay liquid: “We are evaluating 15-20 futures instruments as candidates for potential inclusion in the flagship universe. We are also researching Chinese futures markets and market access, including both onshore futures and internationalized contracts, as potential sources of diversification. We could also contemplate applying our models to a purely Chinese investment universe,” says Mirjolet.
On signal generation, Quantica continues to hone and refine how its key differentiator – the use of cross-instrument and cross-asset correlation and volatility measures – feeds into generating signals and constructing the portfolio. “We are looking at relative trend dynamics and evaluating the potential to broaden signal generation by adding dimensionality to instruments and broadening the definition of trends followed by combining instruments or synthetic risk factors,” says Mirjolet.
Though holding periods are at least several days, Quantica executes trades throughout the day, and uses high frequency tick data for execution analytics. “We are constantly striving to improve slippage metrics, using its proprietary reporting and analytics. Machine learning might enhance some parts of the process. For instance, trade implementation could incorporate term structure, liquidity and carry,” says Mirjolet.
These projects have not yet led to any changes however: “There is a big difference between the research agenda and actual implementation. There is a very high hurdle to improve on the 16-year track record, given the constraints. We want to thoroughly analyse complexity but stay as clean and efficient as possible and avoid it as much as we can. But research does not need to be applied to be useful – we are always challenging our process,” says Mirjolet.
If the investment due diligence process routines for Quantica have been steady, the firm ticks more boxes on the operational due diligence side now than it did 10 years ago. “The strategy is still – to a large degree – the same as when it was launched back in 2005. What has changed most significantly, though, is the environment. We had to adjust as a company, acquire the necessary regulations, become an institutional set-up, grow the team and define clear roles and responsibilities,” says Gmuer.
The firm started out as two staff in Schaffhausen and has steadily grown the team. Beautiful offices near the river in Zurich help to recruit and retain talented technology and quantitative professionals.
A key hire in 2020 was Mirjolet, who is driving forward the next stage of institutionalization, with investments into research, technology and operations. The culture is open, transparent and collaborative between and within four business units: research; trading; technology and business development.
Yet Quantica is still a relatively lean team of 13. “We consider the size of the team – given our assets under management – a pledge of efficiency. We try to remain lean and nimble and continue to hire opportunistically,” says Gmuer. Ownership of the firm has been extended to include all members of the executive and investment committees. Team members are encouraged to think entrepreneurially like a business owner.
The firm, principals and employees also have substantial capital in its investment products. The flagship QMF Program is offered in a variety of structures: a suite of separately managed accounts, an offshore Cayman fund, a Luxembourg UCITS umbrella structure and on a few dedicated fund platforms. “We want to make the best possible solutions for investors,” says Karin Jans, Partner and Head of Business Development, who featured in The Hedge Fund Journal’s 50 Leading Women in Hedge Funds 2020 report.
The strategy can be customized to a different investment universe, subject to liquidity and diversification criteria. For instance, there are UCITS and managed accounts both with and without commodity exposure – and the strategy could trade only commodities. Meanwhile the QMF Cayman fund now offers USD gold denominated share classes for both the 1X and 2X leveraged versions. “This could be viewed as an inflation hedge, is a highly capital efficient way of investing in two strategies at the same time, and it helps to overcome the disadvantage of gold not delivering any dividend or interest income,” says Jans.
The flagship strategy can also be adapted to a stronger focus on tail risk protection.
If asked for ESG criteria, five people could give ten answers and there are no clear guidelines for systematic managers.
Karin Jans, Partner and Head of Business Development, Quantica Capital
The complexity of catering for institutional investors across multiple jurisdictions globally (North America, Europe, Asia) and managing different managed accounts, funds and legal entities with different service providers, executing brokers, prime brokers, custodians, clearing and administration counterparties, regulatory needs and constraints could lead to tracking error. Quantica’s objective is to deliver a minimal tracking error between its different investment solutions. The firm’s QMF UCITS strategy, for instance, trades pari passu and with negligible tracking error to the QMF Program implemented through the QMF Cayman fund.
Operational excellence and technology are one benefit of the efficient operational structure, which has continuously automated and improved investment processes to make them robust and scalable. “Our dedicated COO and Chief Technology Officer seek to keep the firm at the forefront of technological developments. In everything we do, we focus on building highly cost-effective, resilient, scalable and automated solutions and processes. The technology unit has full ownership and responsibility over data management, integration, integrity/consistency and accessibility,” explains Mirjolet.
Operations and technology also provide the foundation for workflows, scalability, robustness and expediting the research agenda. “The time taken to evaluate a new strategy or instrument, the integration/addition of new data sources or new security types into the research framework, the evaluation of their impact on the existing strategy, or the application of an existing strategy to an entirely new set of instruments has come down considerably, from initially a few months to just a few days,” says Mirjolet.
Operations and technology also feed into investor communication. “Since its inception, Quantica has strived to offer maximum transparency for its investors when it comes to understanding the investment process and related portfolio allocations. Our investment process and portfolio allocations over time are intuitive to understand – from data sourcing and normalization, signal generation, portfolio construction to trade implementation, cost optimization and operational procedures. Investors may be given access – if requested – to more granular information, such as signal information for each market, and how positions evolved over time,” says Gmuer.
Since 2020, Quantica has also been publishing the “Quantica Quarterly Insights”, a publication series which strengthens transparency and communication by providing more detailed explanations of thinking behind its strategies and models including the broader macro and risk environment. The topics are often sparked off by investor meetings. A few of them are highlighted above, and Quantica’s September 2020 quarterly insight, Trend-following in a low-yield environment, points out that over one third of the return from owning bond futures between 2005 and 2020 has come from rolling down an upward sloping yield curve. Quantica’s relative risk-adjusted return forecasts consider carry or term structure, which can be particularly important in fixed income.
“We are UNPRI signatories and want to be at the forefront of thinking on ESG and are regularly discussing this in our ESG committee meetings. We are looking at possible solutions and want to be transparent and honest about what can and cannot be done,” says Gmuer. “We find that it is still challenging to implement an ESG conscious investment approach for a systematic manager. The liquidity of ESG equity index futures remains well below Quantica’s minimum criteria. The use of ESG equity swaps in response to investor demand would need to be carefully weighed up against Quantica’s usual criteria such as costs and complexity,” says Mirjolet.
ESG is clearly subjective: “If asked for ESG criteria, five people could give ten answers and there are no clear guidelines for systematic managers. It is not clear what ESG compliant means. But we recognize that ESG driven by investor demands has a lot of traction,” says Jans.
“Returns rather than assets remain the priority; assets are the consequence of a job well done. We expect our AUM momentum will be positive, based on our exceptional 3, 5 and 10 year track records,” says Gmuer. Quantica is increasingly a contender in competing with much larger firms for institutional mandates. “Our main ambition is to deliver the best possible risk-adjusted net returns for our investors,” says Gmuer.