The Last Look…
Posted by Colin Lambert. Last updated: December 2, 2025
The CME outage last week is the latest example of how the FX industry has, in the main, managed to wean itself off the primary venues – 10 or 15 years ago things would have been a little messy in the markets, now it seems, any disruption is minimal. The reality is, in OTC markets at least, with a few exceptions, the issue was one for CME Group, rather than many of its customers.
To a degree, this is a risk that the Merc is always going to have by having everything running off the Globex infrastructure – everyone faces third-party risk, and this was, clearly, an issue with a third-party, but a couple of people have raised the redundancy question with me, observing that, for example, if EBS was still hosted in London, at least a good chunk of the CME FX business could have continued operating.
This is an interesting question, especially given how CME’s FX business has evolved. If EBS was untouched by the cooling issue at a data centre, could CME have, for example, pointed all its Spot+ and a good number of its FX futures clients at that venue? It is not the silliest idea, but the reality is such a model is probably inefficient in economic terms, so won’t happen (and some have noted the irony of it happening just as CME was cutting multiple FX product people in London, see Entries & Exits in this week’s newsletter)!
In terms of market participants, this outage merely served to highlight genuine LPs from recyclers. Dealers I spoke to reported initial disruption, but noted they have solid fallbacks built from experience and quickly shifted their models to ensure minimal disruption to pricing. Where the problem was felt, apparently, was at the more retail end of the business where firms have for years been very loose in their interpretation of “liquidity provider” (and “institutional” for that matter!), resulting in them halting trading in some instances.
A couple of people got in touch to suggest that some players at the sophisticated end of the scale also had issues, because they use CME futures a lot, however private enquiries found no evidence of this, beyond a couple of non-bank market makers relatively new to the business who apparently priced wider.
I have always thought that a firm’s ability to price through an outage at one of the primary venues has demonstrated its status in the market. Put simply, if any disruption to their pricing persists, they are a recycler, no matter what they might say!
Disruptions akin to this are going to happen with greater regularity, because liquidity is not the only thing being fragmented
As was the case with the last time there was a major outage on a primary (EBS in 2023), this is a data, rather than a liquidity, issue. Other platforms undoubtedly stepped into the breach, it’s not as though there aren’t enough other CLOBs to trade upon in the market, although given it was a month-end, when most get a boost in activity, it is hard to know exactly how much of a bump they received. Equally, the fact it was the day after Thanksgiving, when many in the US are taking a long weekend and there were no data or speeches, would have muted any impact.
Sources I have spoken to suggest that no one platform hoovered up the EBS (and futures) flow, and that it was likely spread amongst three or four. This is good news for protagonists of fragmentation, less so for those who think FX could do with some consolidation – for the latter to happen, a clear number one alternative to both EBS and Matching needs to emerge.
I think I am right in saying that it was good that EBS was up and running before the London 4pm Fix, but given my antipathy for that mechanism, I am not sure! Either way, the disruption was dealt with before that time of day.
Hopefully multiple OTC venues don’t all converge at one point the way they often do in exchange-driven markets
Ultimately, beyond CME, which lost a good deal of brokerage thanks to the trading halt (and obviously it was in more than FX) that lasted up to 11 hours in some markets, it is unlikely buy-side customers noticed anything, and the LPs mostly managed to cope. This is a good thing, of course, because it shows how well inured to disruption the market structure can be.
This is turn should provide some comfort for the FX industry, because I sense that disruptions akin to this are going to happen with greater regularity, because liquidity is not the only thing being fragmented. Different support functions and technology services are being farmed out to a host of third-parties, meaning every platform doesn’t have control of at least part of its infrastructure. The next working day after the CME issue, for example, Australia’s stock exchange had an issue (ASX has had its own problems to be clear), and more are likely. Exchange groups and other major platform owners are unlikely to bring everything in-house, which means this technology risk will persist – in spite of efforts by regulators to control it.
The bottom line is, things go wrong with technology and the entire system depends upon its weakest link. This won’t change, but for OTC FX at least, the good news is that it has seen another major outage at a primary venue, and continued operating with barely a murmur. All we must hope is that multiple OTC venues don’t all converge at one point the way they often do in exchange-driven markets.
