What’s driving the systematic shift in commodities trading?

[Music]

Lee Price: Hi there, and welcome to J.P. Morgan's Making Sense. I'm Lee Price from the FICC Market Structure and Liquidity Strategy team. Today, we're looking at how commodities markets are changing as systematic trading continues to grow. Systematic trading involves using data and repeatable rules-based processes to make decisions on how to size risk, when to enter and exit, and how to execute, rather than relying purely on discretion in the moment. We'll discuss what's driving this evolution, from new technology to new products, as well as observations on what market volatility means for trading behaviors and liquidity. And finally, we'll anticipate what comes next, including the emerging influence of AI. To help unpack all that, I'm joined by my J.P. Morgan colleagues, Max Lee in the Commodities Trading Group, and Biko Agozino, who leads Commodities Quantitative Trading. Guys, thanks for being here.

Max Lee: Thanks for having me.

Biko Agozino: Great to be here.

Lee Price: Guys, commodities have been in focus this year with shifting macro expectations and elevated volatility across markets. So metals and energy markets have captured a lot of the attention. We've seen meaningful moves across other sectors as well. While geopolitics, trade policy uncertainty, and energy supply have dominated the headlines, I would argue there's a structural evolution taking shape. Thinking back to twenty twenty-two and twenty twenty-three, you had extreme volatility in many commodity markets that brought a range of new market participants. And as margins have stabilized since then, many firms have prioritized digital innovation, investing in machine learning, automation, advanced data processing technologies. So we've seen a broader macro footprint across this space with a wider variety of market participants. And in recent years, systematic participation in commodities has grown across physical markets, derivatives, multi-asset portfolios, and there are several forces behind that. Max, when you zoom out, what have you observed in terms of the growth of systematic commodities in recent years, in terms of who's participating and what are the key drivers behind that?

Max Lee: I think the first thing that's important is like to describe what is systematic trading. If you looked when I started, which would've been close to, to twenty fifteen, the line between discretionary and call it quant, I think was, was a lot larger. Now, even in discretionary pods, everyone is looking at data, everyone is modeling, and so how do you actually define what is signal-driven versus systematic has, has honestly become a key point. Generally, when we think about the world, systematic has pretty much a full automation component to it. Everything from data consumption to signal origination, to risk management, to the market participation, all of that should be end-to-end and assigned some degree of risk budgeting from a top-level, kind of organism. As we look at like who started to participate in, that format, I think that people have gravitated towards the ability to show, you know, not just a back test, but kind of how an idea can evolve and perform through different regimes. And especially if you look at the, the commodity asset class, that's really important, right? I mean, we've-- you spoke about some of the volatility that has been in the asset class since really COVID. I mean, I think if you looked prior to COVID, it was a generally low vol regime. We had idiosyncratic shocks that would occur, but generally, it was hard to get meaningful return. As soon as you have, you know, the headline that everyone knows where oil went to negative through some of the inflationary recovery, then into Russia-Ukraine, extending all the way then to obviously the Iran, uh, war this year, more and more people have become interested in commodities, but a lot of those people are not necessarily commodity experts. And one of the ways that then people get comfortable with managing that exposure is, can I do something that is fully process-driven, that then I can go to my CIO or my PM or whomever it may be and show exactly like what we expect the distribution of that return to be, which is really, really important, in this asset class. I think that gives a generally like a good overview of like how we see participation evolving. I mean, when I first started, a lot of what we saw from the client side was real asset owners that were looking for small alpha additivity to their inflation protection. Now you have whole hosts of, uh, hedge funds dedicated to the commodity space, whether it's just in the financial, but as well as getting to physical, as you mentioned. You have a lot of cross-asset class PMs that are looking for protection. So it's really evolved. And again, like a lot of that, that comfort and the ability to deploy comes from this idea of being able to trade or, or sort of risk manage in a way that is fully systematic.

Lee Price: Right. So it's not just more participation, but it's this shift in how risk is taken and managed. So you'd be moving from, concentrated specialization towards more quantitative portfolio style approaches. And Biko, from your seat in quantitative trading at J.P. Morgan, as the systematic participation broadens in certain commodities markets, precious metals comes to mind, what is the impact you see most commonly? Are there, positives and negatives of, of what we're seeing?

Biko Agozino: Yeah, it creates both opportunities as well as challenges. Um, there's a lot of opportunities, as Max was saying, that if you apply a systematic sort of overlay to your strategies, you're able to identify statistical arbitrage or relative value across multiple markets that typically are cointegrated or stable correlation. And so when there are moments of shock, that correlation can of course break down. And noticing that, identifying that can drive statistical decisions in terms of your risk-taking. And when you're trading across many markets, many exchanges and many domiciles, all with their own potentially different regulatory frameworks, and in commodities, we have additional concerns that we have to worry about the physical element of the commodity as well, and what you're carrying in terms of a, a future into the physical delivery. And so that challenge of integrating with all of those different exchanges or markets, requires a systematic framework to apply on how you, take risk, how you execute,

Lee Price: Makes sense. If you acknowledge we have this shift taking place that's enabled by the advancement of trading technology, you kind of have to counterbalance, the complexity here. There are some challenges. Max, we've established that rules-based approaches have started to expand, and quantitative investment strategies or quant trading has been in demand, and as firms seek to bring, more sophistication, more efficiency to commodities trading, there are some patterns that emerge, in terms of how participants interact with the market. As these changes take shape, are there things that are still distinctive about commodities when you compare it to other macro asset classes as an example?

Max Lee: Yeah, I think you've seen that, generally when you think about systematic trading or even quant trading, a lot of what the first focus is, is can you extract alpha from the signal? And where that was exceptionally powerful in the commodity space was when you have generally low interest in, in the asset class, as well as hard operational compliance and regulatory constraints to participating, that did lead to a lot of inefficiency. And in general, you know, alpha comes from inefficiency and volatility. However, you start to see more interest in the space coming out of 2020, what that does is it brings that efficient market hypothesis kind of back to the forefront, and the alpha you can get from the signal itself, it's going to diminish, right? Where I think people have then turned is that, well, the additive yield that you can get from execution is exceptionally powerful. And I like to think about it in terms of just like the general profile. When I first started, the only thing I could trade was live cattle for-- which was, you know, a very arbitrary thing. But now if I look at like the liquidity that was in, let's say, the prompt live cattle calendar structure over a decade ago to now, it's, you know, more than I would say three to fourfold, generally, you see a hundred lots on the bid-ask, one tick wide, uh, and that kind of evolution. And then again, it makes it harder for every subsequent strategy that comes out to really achieve any yield. Because of that, then, I think you're seeing sophistication move away, and investment into, groups like even like Biko’s, where the importance of how you access the market is almost equal now to like why you're accessing the market, whether that's in algorithmic form, you know, API connectivity, et cetera. You're starting to see the asset class and commodities in general start to feel a little bit more macro. You can see like how people connect to WTI, potentially seeing about how they function with like the S&P futures or how they may interact with, let's say, like the Euro USD spot market.

Lee Price: You give some of the historical background, both from, your own seat on the trading desk, but also how the market's evolved over time. And you touch on the idea that the strategies themselves are becoming smarter. And Biko, I want to elaborate on that point. When you look across systematic commodity activity today, are there behaviors you feel are most important to understand or where participants are choosing between, listed or OTC markets, for example?

Biko Agozino: Yeah. So I mean, just on the sort of evolution of the asset class commodities and, comparing it to some of the other asset classes, FX equities, commodities has actually has been just as electronic potentially as those other asset classes. As a lot of that activity is on electronic futures exchanges, and futures brokers offer systematic execution via TWAP, VWAP style algos in order to execute your interest. However, we're also seeing that same evolution start to evolve in the OTC liquidity space for commodities. By trading an OTC market, the participants are getting something that they may not be able to access on screen on a liquid futures exchange. But by then trading that OTC products, you're then trading directly with a counterparty who is able to minimize potentially that your market impact on that market and your information dissemination. And so as systematic trading becomes more prevalent in commodities, your market footprint is also just as important. And to Max's point, it's not just your alpha, it's also how you execute that and how you minimize your market impact. And so executing with a provider that is attempting to internalize the liquidity that allows you to minimize how much other participants can see that you're executing at that particular time, which in the long run can translate to lower cost, which gives you a, a significant edge over competition. However, OTC is not directly fundable with futures, and so you may have some basis between OTC and futures if you, trade both of those products. And as a result, having visibility and, , ability to manage your risk around that is very important.

Lee Price: That's helpful, On the OTC side, notable that you mentioned the benefits of internalization and minimizing, info leakage, which is, always an important consideration that we, focus on in terms of growth of e-trading really across asset classes. And, we touched on it earlier, but given the market conditions this year, the volatility, it's no surprise that, we're seeing trading volumes up across commodity exchange-traded derivatives as well. Max, how do market participants navigate these structural changes in these markets? Does it add complexity to what you're trying to build when it comes to systematic strategies?

Max Lee: Honestly, yes. I mean, the answer is you go to a bank. I mean, it, it is hard, right? I mean, the asset class has... Especially because when you think about the fact that the underlier is a physical good, and that physical good has implications across all sorts of consumers, right? Where I do think banks can be helpful is that, you know, we are built around this architecture of how do you provide liquidity? And where in the commodity space, again, people try to gravitate towards is like where is that, that opportunity exists. Going back to that earlier comment was, you know, opportunity exists in inefficiency, so how do you execute or find edge in things that maybe are less picked over? Thinking about something very actually common today, there's a lot of ask of like how do people execute in some of the metal space in, in Asia? It's not a trivial thing to get set up. So I think what you'll see is that either people outsource some of that and are willing to pay a fee for it, that connectivity, that operational footprint, that, that regulatory satisfaction. Um, you outsource that to large institutions that are able to sort of facilitate that across everything. Or you see that there is a concentration in some of the liquid space, um, and people just accept that that's where they're going to be comfortable. But I think overall what it does is that it is-- makes it exceptionally difficult from someone, I think from scratch, to say, "We are going to start a commodity systematic trading operation that can do everything." You really kind of have to pick and choose what is the right balance of investment in infrastructure versus what is going to be your return again, whether it's from your signaling, whether it's from your execution style, like Biko was mentioning. But that fragmentation is something that comes as just a requirement, I'd say, of the asset class

Lee Price: We also have expansion in the trading toolkit itself, so new protocols, new mechanisms to bring enhanced automation to the trading process. And I want to look as liquidity providers develop electronic capabilities, what do you see, Max, as the key near-term milestones versus longer-term goals? And maybe the biggest constraints that are specific to commodities around how you build it out, into the future.

Max Lee: Sure. I mean, I think sticking with that idea of sort of like fragmentation between what is the liquid macro space and then what is the call it micro space. I, I think in the former, you know, the both the short and long-term objectives pretty much align to what you would see in other asset classes, especially like maybe like equities and, you know, kind of FX probably being the most liquid version of FICC. You probably want to see heightened participation from algo footprints. You want to see, you know, I think banks working around internalization, which is a, is always a key theme that you're seeing, not only just in the sell side, but even in the larger buy-side firms. And in general, I'd say you probably want to see some sort of like health of the market emerge. I mean, going then a little bit further to like what is the challenge, what you see is that a lot of algo and E-frameworks are built again on a framework that it would probably references again, FX or equities. The volatility in those asset classes just isn't the same, right? And I think that that comes from not only just realized moves, but also bid-ask explosions, gapping. In the more micro space or areas that, trade differently than like the WTI, so thinking about either your cleared swaps or your more, uh, your power markets. I think that the objectives there is just to see higher participation. One of the things after the ERCOT freeze in '21 is that, again, people were interested in like the, the potential of that volatility, uh, and how that could potentially benefit the portfolio, how that could provide opportunity. And you start to see just incrementally more liquidity come in from some participant, whether it's funds, potentially, you know, banks get more involved, through client activity. And there's this concept of like volume begetting volume, and that I think is probably like the right objective is to have some of those markets which currently are on the fringe, kind of similar to how cattle was, you know, in 2015, move more into this like macro container of general accessibility.

Lee Price: Yeah, Biko touched on it earlier, but just the idea that commodities have physical foundations. So, you know, supply chain, storage, logistics, it can create real limits from a statistical model standpoint. And Biko, I want to kind of go a little bit deeper into, from an e-trading perspective, you spent time, building trading capabilities across FICC asset classes. But the thing that I think that is notable that we've seen client interest in really across asset classes when it comes to e-capabilities is direct connectivity, via API. And is there anything kind of specific to the commodities evolution that you think about when you're talking about API integration?

Biko Agozino: Yeah, absolutely. If we take precious metals, a lot of the precious metals API proliferation happen naturally because we can onboard precious metals just like any other FX pair. And so when clients are integrating with us for FX, they can just ask what other pairs are able to trade over this exact same specification, this same API, and we can give a list of non-deliverable forwards, we can give a list of precious metals pairs, et cetera. Now with commodities unfortunately it's a slightly different fixed specification, and so that requires slightly different integration. However, what we're seeing a lot of interest in is also that API integration. Whereas previously a lot of the trading may have been over voice and the futures market with futures brokers. If you want to access OTC liquidity, you don't necessarily need to just go to voice or just go on a single dealer platform. You can also access, via an OTC provider giving you an API. What we're, um, seeing more and more interest in on is be it cash settled, physical or even cleared, as a give up on exchange, can we provide this API, capability so that our clients are able to manage their systematic trading on their side integrated into J.P. Morgan's technology to be able to trade on an automated basis, without necessarily having to go through our single dealer platform execute or by trading with a voice, salesperson.

Lee Price: Thanks, Biko. Max, jump in on anything product development.

Max Lee: All I think it does is it really reduces the, the time horizon that people can think about, right? I mean, before when everything had to be point and click or even again, further back to, to a phone call, you were really limited in terms of what liquidity the market could give you sort of throughout the day, as well as you were limited to like what you could just feasibly do. So there's a lot of sort of unidirectional trading, a lot of buy and hold, or sell and, and I think the asset class functioned much more in a high volatile, kind of like terminal value focus, from an ideation perspective. Now, if you have the ability to transact at a much tighter time horizon and, and efficiently, you expose like a whole, I think, area of research and a whole time horizon that just previously was, was not as accessible. And then if you play that forward, well, then everyone can do that. That brings more liquidity to the market in periods that are not just the, you know, the settle or the open. So I, I think it's an exciting time, and it does unlock a whole body of research, and sort of innovation which, which has been long possible, again, in like the FX and equity space and, and bring it to, to commodities.

Lee Price: So before we wrap up, let's look forward. As technology continues to advance, participation in systematic approaches may broaden further. So key question is, where the durable edge remains and how AI changes the toolkit. Now, AI obviously comes up in a lot in conversations about commodities because it has the potential to reshape how teams research, monitor risk, and execute. And it's difficult to talk about the next generation of trading workflows without acknowledging that influence. We're seeing increased experimentation across the industry, from early pilots to more operational use cases, and the practical question is, what actually changes day-to-day?

Biko Agozino: It'd be crazy for us not to use it as a productivity tool in order to, um, help develop our algorithms and expand to new markets quicker. However, it still does, provide the same, bottlenecks, with regards to delivery in general as like we have to ensure that the, the system is safe and secure before we put it out to and that is like a sort of manual oversight that is required in order to ensure that, we're operating on safe footing. It doesn't necessarily speed up the, the rollout to production phase, but we're able to prototype and iterate much quicker in a non-production environment, thanks to some of these new AI tools. But look, um, quantitative trading has always been very data-driven, and one of the aspects around that is that like you need data in order to make decisions and to update your view of where the market is going. But if there is some idio shock and it moves, expansively beyond the points of regular mean reversion, how do you detect that? How do you detect that quickly? You need enough data to be able to say with certain that it has actually left the usual, distribution of the, the price. And so, in order to detect that, I mean, like using the modern AI techniques, we're able to leverage machine learning in intelligent ways to detect those dislocations and make those decisions quicker and, and operate with high conviction even if we have relatively limited data.

Lee Price: It's cutting-edge stuff. We've covered a lot of ground today. From breaking down recent commodities market conditions, examining the growth drivers behind systematic trading, and anticipating further technological advancement within financial markets. Max and Biko, thanks so much for joining today.

Biko Agozino: Thanks for having us.

[Music]

Voiceover: Thanks for listening to J.P. Morgan's Making Sense. If you've enjoyed this conversation, share your feedback by leaving a comment or review wherever you listen to podcasts. And be sure to follow our channel so you don't miss an episode!

The views expressed in this podcast may not necessarily reflect the views of JPMorgan Chase & Co, and its affiliates, together J.P. Morgan, and do not constitute research or recommendation advice or an offer or a solicitation to buy or sell any security or financial instrument. They are not issued by Research but are a solicitation under CFTC Rule 1.71. Referenced products and services in this podcast may not be suitable for you, and may not be available in all jurisdictions. J.P. Morgan may make markets and trade as principal in securities and other asset classes and financial products that may have been discussed. The FICC market structure publications, or to one, newsletters, mentioned in this podcast are available for J.P. Morgan clients. Please contact your J.P. Morgan sales representative should you wish to receive them. For additional disclaimers and regulatory disclosures, please visit www.jpmorgan.com/disclosures

Copyright 2026 JPMorgan Chase & Co. All rights reserved.

[End of episode]