This article was first published by Pensions & Investments on 15th June 2026: Link Here
This article was first published by Pensions & Investments on 15th June 2026: Link Here
As deglobalization, policy divergence and geopolitical uncertainty reshape markets, institutional allocators are rethinking the impact of macro regimes on their portfolios. P&I spoke with Christian Dery, head of macro strategy at CFM, a Paris-based quantitative manager specializing in systematic global macro strategies. He explains the firm’s process and how it translates macro narratives into data-driven signals, how artificial intelligence is expanding the opportunity set and why institutions are increasingly on board with systematic macro as a strategy for diversification and portfolio resiliency across all market regimes.
P&I: Could you define systematic global macro? What are the key characteristics of CFM’s strategy?
Christian Dery: Systematic global macro uses quantitative models to express views across liquid global markets: rates, foreign exchange, equities, commodities, credit and volatility. It is highly diversified and scalable by design. Where discretionary macro trading relies primarily on human judgment — often very effectively — systematic strategies codify the research contributions of many views across multiple domains, including alphas — at CFM we call them predictors — portfolio construction and execution.
CFM was founded over 30 years ago by a group of scientists and is designed to scale the impact of research. There are no pods, silos or star traders; there is only one team. All the research — from predictors to portfolio construction to execution — flows through our flagship multi-strategy fund. We then leverage that full engine to design compelling strategies for investors, including our systematic global macro strategy. For allocators, the result is a genuinely diversifying return stream across market environments.
P&I: Why now? How is the current macroeconomic environment making it more important to consider systematic global macro?
Dery: By construction, a well-run systematic global macro strategy should exhibit diversification properties and low correlation to broad asset classes. In that sense, it is not a strategy that you time; investors need to identify managers that exhibit skill — admittedly a difficult task to do ex ante.
Many institutional portfolios are typically overweight U.S. assets, including illiquid assets. The relative outperformance of U.S. equities over the last decade has delivered a major tailwind to investors, led by the technology sector and its hyper-scaling business models. But if we look at longer historical periods, you find that the equity risk premium has failed to deliver over long stretches of time. Now, it would be fair to say the world is entering a heightened period of uncertainty, whether from rapid technology disruptions or shifting geopolitics. In this environment, strategies that offer compelling performance, genuine diversification, adaptability and attractive liquidity terms warrant consideration. Well-designed systematic global macro strategies exhibit these properties.
P&I: Do you feel institutional allocators have underestimated how exposed diversified portfolios can be to macro regimes? Is there greater understanding of macro strategies benefits today?
Dery: I have a running line that no portfolio escapes macro. I describe macro as a black hole that eventually pulls in all strategies. It might not matter today, but eventually it will. Portfolios contain latent risk factors that emerge in certain environments. A good example was the 2022 Fed hiking cycle. Many long-short equity managers were caught wrong-footed, but experienced macro investors understood well how to position for a credible central bank faced with an inflation threat. The driver of performance for those poorly hedged equity managers was a macro risk factor, not idiosyncratic risk tied to their stock picks.
I would say most investors understand the portfolio benefits of macro strategies. As a former allocator, the difficulty is finding managers that can deliver a sustainable track record. Part of the appeal of systematic approaches is the ability to achieve higher levels of diversification and what I’d call the ruthlessness of algorithmic strategies. They don’t get tired or emotional, and they continue executing on carefully designed logic. Well-designed systematic firms also have more longevity than models built around a small number of individuals with significant key-person risk. A scalable process with contributions from a larger number of talented researchers wins in the long run.
P&I: How should allocators think about alpha and risk when comparing systematic and discretionary macro strategies? Where does the human element come in?
Dery: Historically, the major edge of discretionary trading was the ability to change your mind in response to sparse data. One data point can be sufficient to reposition a portfolio aggressively. This does suffer from small sample properties and there is survivorship bias — the investors who made bad binary calls are no longer in business. The question for allocators is whether those insights can be captured with greater consistency and better risk properties. That’s where systematic frameworks add real value.
On the risk side, the difference is equally important. Discretionary strategies tend to carry more concentrated positions and overlapping risk exposures. At CFM, portfolio construction is an important area of research and a source of alpha for our strategies. For example, we have a portfolio construction approach we call agnostic risk parity. The idea is that a portfolio can be decomposed into orthogonal statistical modes, which are more accurate representations of the risk a portfolio is taking. By allocating risk more effectively across these modes, we are able to achieve a higher level of diversification.
Beyond that, the structural advantages of world-class systematic approaches continue to compound with the explosion of data, advanced technology stacks and the emergence of powerful AIs. But this isn’t a zero-sum argument. Discretionary human insights become significantly more powerful when combined with the systematic research stack. The machine doesn’t replace the insight — it scales it.
P&I: Could you describe CFM’s approach to systematic global macro: What are the main return sources, and how do they interact in the portfolio? What are the key benefits of the strategy?
Dery: CFM’s research is generated by one unified team and has three main pillars: predictors, portfolio construction and execution. Think of these as building blocks. At CFM, a given strategy can have hundreds or even thousands of de-correlated ideas embedded within it. The basis of a good predictor is an idea that has a persistent statistical signature and adds new information to a portfolio. New information means an idea that is de-correlated with the balance of ideas in a given portfolio.
To give investors a sense of the types of ideas embedded in a portfolio, we classify predictors into clusters. These clusters include trend and mean reversion, fundamental macro, sentiment, networks, patterns and carry. In our systematic macro strategy specifically, we also include a defensive layer — a strategy designed to perform in stress environments without punitive carry — that is expected to improve the overall convexity and risk-adjusted returns of the strategy.
P&I: How is CFM harnessing AI and machine learning to evolve this strategy?
Dery: I am a believer in AI. One could argue it is a generational technology — perhaps the most significant since the internet. At CFM, there is a lot of tinkering and exploration already underway, and we are finding multiple use cases as it rapidly evolves. We recently established a machine learning lab. Part of its function is to fill in knowledge gaps for researchers who are thinking about practical applications of the technology. The pace of innovation is extraordinary. So much human and financial capital is finding its way into AI, and the rate of progress is accelerating.
Our use cases include predictors, classification and network discovery, and automated research. We can extract rich contextual information from various forms of media: text, video, audio. That information can be codified into new predictors, for example, extracting sentiment and nuance from earnings reports or central bank statements.
A large language model, or LLM, can propose novel network and classification structures that humans might not consider. And agentic approaches can test and code ideas at scale.
P&I: What types of institutional allocators are showing strong interest in systematic global macro, and how are they using it in portfolios?
Dery: We are seeing interest in systematic global macro from a wide range of allocators — pensions, sovereign wealth funds, endowments, family offices and, increasingly, insurance companies and wealth platforms. The use case varies by portfolio. For large institutional investors with significant illiquid allocations like private equity, real estate and infrastructure, systematic global macro serves as a liquid diversifier that can also act as a source of rebalancing capital during dislocations. For allocators with heavy equity beta, it offers genuine de-correlation. And for those building out an alternatives allocation, it sits naturally alongside other hedge fund strategies as a complement that behaves differently from long-short equity or credit.
Any description or information involving modes, investment purposes or allocations is provided for illustrative purposes only and does not constitute investment advice nor an offer or solicitation to subscribe for any security or interest. Any statements regarding correlations or modes or other similar behaviors constitute only subjective views, are based upon reasonable expectations or beliefs, and should not be relied on. All statements herein are subject to change due to a variety of factors including fluctuating market conditions and involve inherent risks and uncertainties both generic and specific, many of which cannot be predicted or quantified and are beyond CFM’s control. Future evidence and actual results or performance could differ materially from the information set forth in, contemplated by or underlying the statements herein. CFM accepts no liability for any inaccurate, incomplete or omitted information of any kind or any losses caused by using this information. CFM does not give any representation or warranty as to the reliability or accuracy of the information contained in this document.