Risk & Return Quantitative Investment Forum
Delegates attended forum 27 - 28 May in Toronto, Canada.
Toronto - It's been a while since I attended an event where so many talented quant researchers were gathered in the same room for two intense days of geeky presentations and panels. Managing directors and heads of Quantitative Research from BlackRock, UBS, BMO, RBC, TD, CIBC, CPP IB, OMERS and many other prestigious three-letter financial institutions lectured on the industry's hot topics.
The first day focused on the usefulness of quantitative approaches, strategic asset allocation and smart beta. Jay Vyas, managing director and head of Quantitative Investing at CPP Investment Board kicked off the conversation with a reference to an April 2012 CFA Institute blog post titled Does Quantitative Investing Have a Future?
"Performance is not enough," he said, urging quantitative portfolio managers and researchers to better understand the intricacies of marketing their strategies to institutional investors. "You have to put yourself in the investor's shoes now."
Contrary to most institutional investors, CPP IB actually increased its allocation to quantitative strategies after 2007. Vyas has seen a recent uptick in re-investment, leverage usage and assessment of strategy crowding risks. And while investors used to chase performance before the crisis, he said that they are now chasing continuous innovation - a never-ending quest.
Smart beta approaches and low-volatility in particular have been perceived as one such area of innovation since the turn of the century when the first smart beta ETFs were listed.
When asked about his methodology to research and develop such strategies, Julien Palardy, VP Quantitative Equity Group at TD Asset Management replied: "Define your risk and returns goals first, as a utility function. Only then define assumptions and sensitivity".
"Goals generate strategies, not the reverse", Serguei Zernov, director, Quantitative Investments OMERS Capital Markets added.
Smart beta semantics
Factor-based allocation was also cited as one of the recent buzzwords that animated the community. For Palardy and Zernov, it's simply a new way of expressing the same ideas with a different language: more of a change in terminology and marketing than a fundamental discovery.
Expanding on the theme of smart beta and after a panel on the potential crowding risk for such strategies, Bill Tilford, head of Quantitative Investments at RBC Global Asset Management further gave his assessment on the suitability of low volatility approaches in a rising interest rate environment.
"High beta is a bad investment proposal since 1957, even neutralised for other measurable factors", he said, building on his firm's research on more than 50 decades of curated Canadian equity data.
"Our research shows that the low-volatility factor has no sensitivity to rates in a rising rates environment," he added, citing a study comparing quarterly returns of that strategy with Canadian 10-year government bond yields.
MF's performance drivers
The second day started with a panel discussion on mixing quantitative and fundamental analysis and how each type could benefit the other. The highlight was a presentation on managed futures strategies and their recent performance.
Jason Russell, president & CIO, Acorn Global Investments addressed the fable of the death of managed futures with a thorough presentation of the strategy performance drivers.
"The risk-on / risk-off environment of 2009 - 2012 has produced an unusual regime of high average absolute correlations among the asset classes traded by managed futures portfolio managers," he said. "After 2013, we've seen these correlations come down and the industry's performance improve."
Indeed, numbers don't lie for quants. The Newedge CTA Index published daily by Societe General's brokerage arm, one of the key managed futures performance benchmarks, was one of the best hedge fund strategy performers in 2014.
After losing 2.25% between 2009 and 2013, the gauge gained 15.66% last year alone. "2008 was a structural crisis, so it led to an elongated recovery. In our experience, such recoveries are accompanied by transitions in the market that make it difficult for trends to emerge," explained Roland Austrup, CEO and CIO of Integrated Managed Futures.
In early October last year, a director at an asset allocator to alternative strategies in Canada told me that he had received no convincing explanation for CTAs' underperformance during the previous five years.
Russell provides one: "Oil is a concrete example of how most markets became after the heights of the crisis. It went mostly sideways between 2010 and mid-2014. Investors generally look at the fixed income and equities markets, but CTAs actually spread their risk between several dozen markets in parallel. Currencies, grains and energies didn't display the same obvious trends as the S&P 500 and the US 10-year did. These are just two of the forty markets we trade."
Yet, this diversification and flexibility is one of the most important assets of managed futures compared to other alternative strategies.
Take what is now the traditional long/short equity approach as an example. The overwhelming majority of long/short equity portfolios are net long, most of the time. In crisis periods like 2008, these strategies didn't provide the hedge investors had expected from them.
That same year, Newedge's benchmark not only delivered 13.07%, but it was one of the only alternative categories that yielded positive returns, providing sophisticated investors a way to capitalise on a multitude of factors when the equity market mostly tracks GDP over long periods of time.
But it's no risk-free bet either.
"CTAs are about managing risks versus traditional long-only investors which are only accepting risks," commented Austrup. "Risk management is required because of the leverage inherent in managed futures. This leverage is required for low-volatility asset classes to deliver a meaningful contribution to the portfolio, for example in the bond market."
Global macro is another such hedge fund category that trades a wide variety of global markets, including currencies, fixed income, equities and commodities. The presenters noted that investors have witnessed increasing correlation between global macro and CTAs. "Global macro approaches are becoming more systematic, while managed futures have been incorporating more and more data as part of their decision-making process", commented Austrup. "So we've seen increased correlation across these two broad strategy categories."
The session was informative for a general audience but unfortunately, time ran out and didn't allow for deeper review of the principles, modern technics and innovations used post-crisis and post-recovery by managed futures portfolio managers. Some will say that everything is in the numbers, but I was rather looking for an inside view of how Canadian managed futures practitioners adapt to perpetually changing market environments.
Seeking answers - Q&A
Following the conference in Toronto, I decided to sit down to interview one of the attendees who could provide me with his opinions: Pierre Thauvette, head of Quantitative Research and global multi-asset portfolio manager at Claret.
Thomas Pasturel: What is your role at Claret? What is the Global Multi-Asset fund you manage?
Pierre Thauvette: My main role consists in heading the quantitative research department. As part of that role, I develop quantitative investment models both in the traditional and alternative space. I manage a team of analysts and act as portfolio manager for the Claret Global Multi-Asset fund (GMA). The GMA is an alternative investment product that falls in the following classification categories: CTA, managed futures, systematic diversified and systematic global macro.
The GMA was incepted in February 2012 with the mandate of offering a diversified alternative that would complement the bottom-up value approach that Claret is known for and good at. In other words, the GMA seeks to provide returns for investors irrespective of market conditions. We've seen great interest for these types of strategies lately as the GMA has returned 53.27% since inception which translates to a 13.67% annualised return. The fund is up 16.43% year to date, while the S&P 500 is essentially unchanged.
Q: How would you describe managed futures and your investment approach? What makes it unique?
A: I see managed futures as a catch-all expression that encompasses any investment approach that requires the management of the various risks arising from trading futures contracts (and their OTC cousins) to gain exposure to multiple asset classes, both long and short, and benefit from evolving market conditions.
In traditional investments, synonyms for this expression would be "Active Equity" or "Active Fixed Income". It means something but you're generally left thinking: "what kind?" When describing my investment approach, managed futures would be the conversation starter. From there, I gradually dive into the following topics to paint a clear picture of my investment approach: what to trade, when to enter, how much to trade, risk control, when to exit and finally where to trade.
It's all about providing diversification and managing risks using a systematic and clearly defined framework. Having a clearly defined framework designed to handle various market environments is quite appealing and comforting when the outlook for traditional asset classes is uncertain. In simple terms, I trade 140 markets on a trend-following basis where the risk on each position represents a small fraction of the fund's capital, the sector and portfolio level risk are constrained, the losers are stopped out, the winners are scaled out and the positions are held at more than one futures commission merchant to further diversify counterparty risks.
Q: My perception is that the quant research world is split between those who approach the markets as yet another time series to be dissected with sophisticated formulas and those who have hands-on market experience and decide to quantify their approach. What do you think and in which category would you find yourself?
A: I would have to say I'm mostly in the former camp, except for the sophisticated part. I do find myself treating markets as simple time series to be dissected more often than not. While doing so, I do my best to keep Einstein's advice in mind where "Everything should be made as simple as possible, but not simpler."
Our team is continually assessing new strategies and we're looking to add additional markets to the investment universe of the fund. The goal is to further reduce the size and impact of each position in the portfolio. Additional markets provide both diversification and scalability. There are plenty of markets worth looking into: FX forwards, interest rate swaps, forward freight agreements, credit default swaps, electricity, emissions, factor-tilt ETFs, REITs, just to name a few. Although each market does require an attention to detail and a reasonable comprehension of the underlying asset classes and return drivers to which an exposure is sought, I would tend to treat them as simple time series in the end.
Q: The panelists today discussed the risk of crowding for smart beta strategies, especially given the AUM inflows low-volatility approaches have experienced over the past few years. A huge run-up in AUM also occurred in managed futures strategies since 2009. Are you concerned about similar risks for managed futures? I remember that some of the futures traders featured in Jack Schwager's Market Wizards book shared that concern for trend following strategies. That was 26 years ago and CTAs are now one of the leading hedge fund segments with more than $312 billion in AUM.
A: I am not concerned at all by the capacity issue. Newedge put out a great research piece on July 31st, 2013 called "Capacity of the managed futures industry" where it eloquently demonstrated that there was no cause for concern on the capacity front. The paper also mentions the fact that the futures markets volume has grown significantly over the past several years and that there are no major issues in sight that could derail a continuation of that growth.
Some good food for thought I came across on this topic comes from Winton Capital in a research piece dating back to February 2014 called "The growth of the CTA industry". The paper mentions the simple fact that all of Bridgewater's AUM is generally included in the $300 billion you are referring to. Given that it cannot be ascertained that all of Bridgewater's AUM falls under the managed futures category, the paper guesstimates that $50 billion can be easily be taken off that number.
Another excerpt from an interview with David Harding has him pointing out several misconceptions in regards to the capacity of the industry. Misconceptions about the size and capacity of the underlying, one's ability to observe that capacity through market inputs and treating all volume equal all contribute to obfuscate the true capacity of managed futures strategies.
A similar analogy can be made on exchange-traded funds. When judging the capacity of ETFs, one should consider the volume of the underlying securities and not only the actual volume of the ETF itself. An ETF showing little historical volume does not prevent significant liquidity from being available on both side of the quote on any given day.
Q: The state of heightened correlations across asset classes during and following the financial crisis was also discussed today. Have you taken any steps towards better preparing for such hostile environments going forward? How does the "quest for continuous innovation" mentioned by Jay Vyas of CPP IB translate at your firm?
A: Although I cannot deny the fact that heightened correlations can probably explain a large portion of the latest period of underperformance that managed futures has just gone through, I believe many more factors must be considered when attempting to explain what caused this underperformance.
There are several items on our research agenda that could yield additional insights: what was the average volatility of markets during the period, was that volatility increasing or decreasing, was the high correlation mostly observed in some market sectors or across the board, what was performance like when correlations were increasing, was it any different when they came back down, does history offer a comparable market environment?.
All these topics probably have at least some explanatory power. As it pertains to the "quest for continuous innovation", two quotes come to mind: "It's not the strongest of the species that survive, nor the most intelligent, but the one most responsive to change. Darwin " and "Those who cannot remember the past are condemned to repeat it. Santayana."
Both quotes point toward a need for a constant research effort aimed at increasing our understanding of past and current events in order to be better prepared to act as the uncertain future unfolds before us.
Q: A topic that was not touched upon much during the conference was the combination of multiple systematic investment approaches to achieve better risk-reward profiles. Is this an area of interest for the Global Multi-Asset fund or are you sticking to the more traditional trend-following strategies CTAs are famous for?
A: It is definitely in the research pipeline. Our research projects coming out with the most convincing results are actually in that area. I have researched several strategies with different underlying return drivers using different instruments throughout the years. I have found that the most stable results can be observed when combining simple robust investment strategies versus adding complexity in a single strategy.
Some of the themes that yield interesting results when combined include carry, trend, momentum and value. These themes can be implemented on various asset classes. Several good research papers touch on these investing themes both individually and in combination.
About the writer: Thomas Pasturel is an independent consultant in finance and technology with expertise in capital markets and quantitative research. His focus is on alternative investment strategies, risk management and global macro analysis.