The news this week that Citi served a 90-day notice to the largest non-bank market making firms in FX – including Jump Trading, XTX Markets, Virtu Financial, HC Technologies and, at an earlier point in time, GTS – that it is no longer willing to offer them FX prime brokerage (PB) services, certainly got tongues wagging.
But before analysing the broader impact of this decision, it’s important to consider why it was made in the first place.
“I think this is Citi finding a convenient exit from a business they were over-stretched in,” is the assessment from a senior manager in the FXPB division at another bank. “Seems to me they want to look like most of the other PBs regarding their client base.”
Other sources claim that one of the key concerns for Citi is the perceived risks associated with offering PB services to these types of firms. A New York-based senior figure at one firm which was forced to move away from having Citi as a PB, says that Citi wants greater pre-trade credit controls, but that such controls are incompatible with the strategies employed by some trading firms because they add too much latency.
Moreover, they speculate that Chris Perkins, who took over Citi’s FXPB business when it was merged with prime finance and securities services (PFSS) in December 2018 following an estimated $180 million loss, is getting pressure from within the bank to reduce risk in the business.
“He’s obviously been given latitude to lose a fair amount of revenue to eliminate risk, so he’s simply fulfilling this mandate. He’s getting pushed from higher up to dump people,” said one source.
Another source, who has a background in FXPB but now works elsewhere in the market, claims that Citi’s market share in the FXPB landscape, which this source estimates at more than 40% at one point, meant that it came under heavy scrutiny by internal risk staff at the bank.
“As a result, they had to make certain changes, and if you look at the HFTs, they look very risky unless you have good risk controls and management processes in place,” says this source.
In a similar vein, a source at one of the firms impacted by Citi’s decision says that the move was clearly a response to the losses that the business sustained at the end of last year, which in turn prompted the change in management.
“Citi has lost a lot of domain knowledge over the past year or two in FXPB, I am not sure the people running the business even want to be in it”
Indeed, multiple sources cited that incident as a driving force behind the recent changes in Citi’s FXPB business, including the decision to change its pricing structure, but also observe that the HFTs that it is cutting mainly trade spot FX, whereas those losses are understood to have been from a leveraged fund trading structured options. “It’s ironic that it was a hedge fund that behind that loss, but it’s the non–banks paying the price,” notes one HFT.
“I don’t get this one,” adds a New York-based source. “This is the fear of what? The runaway algo? No one has ever lost a boatload of money trading spot FX intraday. Ever.”
The source adds: “As a PB you get paid to carry some level of credit risk, and there’s always something that can go wrong. But I think that Tower is unlikely to be one of the firms that blows up, they’re not carrying a lot of risk at any moment in time.”
This is a reference to a move Citi reportedly made earlier, cutting Tower Research Capital as a PB client.
A spokesperson for Citi declined to comment on any questions regarding its FXPB business, and issued the following statement: “We are fully committed to our foreign exchange prime brokerage business and frequently review and evaluate our platform.”
A risky business?
But aside from the risk concerns, sources suggest there were also question marks around how profitable this high volume, low latency business was for Citi.
“Generally, these high volume accounts don’t fit the ideal profile of today’s PB. It works being a PB to high volume accounts when you’re running a billion dollar business with good margins, but when that revenue is $500 million and the margin is squeezed, you inevitably start to look at where your costs are,” says a senior figure on the clearing side of the industry.
A counter argument to this put forward by the source in New York is that actually these firms are trading predominantly spot FX and generally not holding huge positions for long periods of time meaning that, in a post-Basel III environment, they will consume less balance sheet and capital than another client that trades less frequently but puts on, say, a large FX options position.
However, the source adds that Citi’s decision may well have been driven by the fact that these high volume trading firms don’t provide much value for the bank beyond its PB business.
“Over the past week, I’ve realised there’s a lot more banks doing PB in FX than I thought. Non-banks aren’t dangerous or bad actors and I think a lot of banks are starting to understand that.”
There were also grumblings from a variety of market sources that the real issue here is that Citi’s FXPB business is now being run by individuals with less experience in the OTC markets.
“Citi has lost a lot of domain knowledge over the past year or two in FXPB, I am not sure the people running the business even want to be in it – they’re clearing advocates and don’t really understand FX,” says a senior manager at one FX platform.
This comment was echoed in various ways by others, with the general theme being that the current management of the FXPB business “doesn’t understand” how this business has traditionally worked within the FX market.
A slight variation on this was that one market source conceded that many of the concerns being raised by Citi – such as the lack of systemic risk controls and problems around credit allocations – are valid, but they also say “that’s not the way the industry works”.
“The broker has the risk tools that the client uses….Citi’s saying the clients are not willing to do that, but they don’t have the risk tools,” this market source says, adding: “All the issues Perkins raises are valid issues, but that’s the business of PB – that’s what the business is.”
Having ascertained the reasons for Citi’s decision to cut some high-profile clients, it’s worth also considering the impact that this could have, both on these trading firms and the industry as a whole.
The larger trading firms seem largely unperturbed by the situation. They claim to have seen the writing on the wall when Citi’s PB business suffered the large loss last year and so made sure they had backup PBs in place or added relationships so they would have options if and when Citi pulled back.
The biggest challenge for them seems to be getting all the right legal paperwork in place before the 90–day notice period from Citi expires. A senior figure at one trading firm describes the “cascade of documentation” that FXPB requires as a “paperwork nightmare”, while another at a different firm complains that they will now have to spend their summer negotiating ISDAs.
“FXPB paperwork is exhausting,” this source adds.
But the sources suggest that the outlook might not be as positive for small– and mid-sized firms that Citi has cut from its FXPB client base in recent months.
“I think that the big guys are going to be ok, they’ll find homes with the other FXPBs, and in fact we’ve already seen that happen. It’s the guys who have under $100 million in capitalisation that will struggle,” says the source with a background in FXPB.
“I am genuinely surprised that the non-bank firms don’t have a plan in place for something like this. They look at the four or five PBs in place and think they can just switch over – it won’t be that easy”
Perhaps a bigger question is whether other bank FXPBs will follow Citi’s lead or try to scoop up the firms that are now looking for an additional PB. Overwhelmingly, the latter seems to be the case right now.
“There isn’t a bank that hasn’t called me to ask me about us moving our PB to them,” says the source in New York.
While from the sounds of things a large number of banks are trying to pick up some of these firms as FXPB clients, it seems that NatWest, HSBC and Deutsche Bank in particular have been moving aggressively to expand their FXPB client base this year. Additionally, at least one large non-bank firm says they’ve also heard from Bank of America Merrill Lynch, Societe Generale, BNP Paribas and State Street.
“Over the past week, I’ve realised there’s a lot more banks doing PB in FX than I thought,” says a large non-bank source. “Non-banks aren’t dangerous or bad actors and I think a lot of banks are starting to understand that.”
Impact on liquidity
Despite this though, a number of market sources still see the potential for market disruption following Citi’s decision.
“There is a price at which these firms will get PB–ed, the question is: will it work for them? Competition might mean one or two firms step up and try to take over the business, but Citi had the broadest product set out there (the risk management of which I think in part drove their decision), but eventually they are likely to come to the same conclusion as to its value. The price increases necessary to fully cover the costs for a PB are not 31 times, as Citi suggested, but they are multiples of what is currently being charged,” says a senior FXPB source.
This last comment apparently references a report published by Citi earlier this year, which argued that the uncleared margin rules (UMR) coming into effect in September this year and next, will significantly increase the cost base for operating an FXPB business.
The source continues: “The regulators might want to take a close look at some of the PBs that step up to make sure they have the right risk framework in place – the firms being talked about here have very fluid and volatile risk profiles, managing these is a complex business and takes a huge investment – look at what Citi have invested in PB over the years to just stay at this level of the business.”
Indeed, a senior figure at one trading firm that was dropped by Citi recently says that while there have been some unexpected banks stepping up to offer FXPB services to them, and bigger banks at this, they are “getting their eyes opened to what technology they need”.
They add: “The problem is, it’s not a 90–day solution.”
“Citi had a great PB franchise, so this is definitely shaking things up. I’m hopeful the sector could become healthier because now there’s more competition.”
Another senior source in the FXPB industry suggests that the non-bank firms cut by Citi are being short-sighted by thinking that they can simply switch their business to another PB and carry on as normal in the FX market.
“I am genuinely surprised that the non-bank firms don’t have a plan in place for something like this. They look at the four or five PBs in place and think they can just switch over – it won’t be that easy. And what happens if, probably when, someone wakes up at one or more of those four or five and says ‘we’re not going to do this anymore’? What’s the non-banks’ plan then? They’re a huge part of the market but they’re not thinking this through all the way at the moment. The technology investment necessary is huge, I don’t think most of the other PBs will want to commit to it,” they say.
Given the position that some of these trading firms occupy in FX today this could have potentially significant implications for market liquidity. After all, the rise of firms like Jump, XTX, Virtu and HC Tech as market makers in FX has been well documented and they make up a sizable percentage of the ECN trading volumes.
This is why the platform senior manager reckons that we could be witnessing “a landmark event” that could have significant repercussions for FX market liquidity.
“A couple of the firms involved have told me they don’t think fees will increase, but that’s rubbish – Citi was already the cheapest PB on the street so just by pushing more volume to the others they will be paying more. Citi was also the only bank whose trading business actually thought the PB business was beneficial to them, that’s something else the other PBs are going to have to overcome at some stage,” they say.
The source continues: “Liquidity could be hit by this. If PB fees do go up, as I think they will, then some, if not all, of these dealers will look again at their businesses. Their margins and revenues are already tight thanks to the low volatility in the market, if it becomes 50% more expensive – and that could be a conservative estimate – I wonder how committed they will be.”
Reversing PB concentration
Not everyone shared this outlook though. For example, the head of e-FX trading at one bank in London claims that even if the firms cut by Citi reduce their trading volumes by 50% there would still be “plenty” of liquidity for “those who need it”. The banker notes that the optics probably won’t look good for the multibank trading venues because their volumes will drop slightly and the top of book pricing on their platforms will widen out, but also suggest that this could lead to firmer liquidity and less “pipping”, which they claim would be a good thing.
The banker adds: “The interesting piece might be how these trading firms behave on aggregators if they have to pay more for PB – given how people generally like that business they will probably stay competitive on those channels by focusing more resources there.”
The source with the background in FXPB says that there could be an impact on liquidity, but suggests that it is likely to be a short-term one.
“I think that initially, while the PBs are ramping up with these clients, you’re going to see a little less liquidity in the market. Citi provided these guys with huge limits and who knows what size limits they’re going to have at these other PBs,” they comment.
“I think this will create ripples throughout the industry, because we hear people questioning the PB model. It was a good idea back in the day, but we have better solutions now that are more efficient and lower firms’ risk profiles. What is the value of PB to the market?”
In fact, some of the trading firms themselves claim that while, in the short term, Citi’s decision to stop offering them FXPB services creates some disruption, in the longer term it will prove beneficial to the overall market because it will mean that less flow is concentrated within Citi’s business.
A source at one trading firm says that concentration risk amongst PBs has been their biggest concern for the past four or five years. The source expresses hope that Citi’s decision reverses some of this concentration risk that had built up.
“People don’t change PBs very often, so when one of the major PBs exits, there are opportunities for those that had trouble growing the business,” the source notes. “Citi had a great PB franchise, so this is definitely shaking things up. I’m hopeful the sector could become healthier because now there’s more competition.”
Similarly, a source at another trading firm says: “Morgan Stanley used to be the be-all-and-end-all of FXPB and then when they cut clients, everyone went to Citi. Now, I hope that everyone spreads out, so we don’t have the concentration risk that we had before.”
Changing the status quo
Elsewhere, there seems to be a consensus that prime-of-primes could find life harder following Citi’s decision. The platform senior manager thinks that many of them will see their pricing widen out and the senior manager at the FXPB reckons that they could be next in the firing line if Citi decides that it wants to further reduce its FXPB client base.
The source working in clearing, meanwhile, questions what this could mean for the FXPB model more fundamentally.
“I think this will create ripples throughout the industry, because we hear people questioning the PB model. It was a good idea back in the day, but we have better solutions now that are more efficient and lower firms’ risk profiles. What is the value of PB to the market? It creates a lot of different types of risks with minimal profit margins – there are people out there who think it’s a bad market model,” the clearing source says.
This source continues: “There have been concerns raised in the PB world about the consequences of a real event in the current environment – FXPB has seen a couple of small incidents, like the problems Citi faced last year in Hong Kong, but it hasn’t had a high impact fat finger event, for example, since SNB – and that was a mess.”
As a consequence, some sources are suggesting that Citi’s decision could represent an opportunity for CME. The logic being that if market access becomes more restricted for a number of firms in the FX market they will look towards CME, as the operator of FX trading venues and a clearing house, as an alternative solution.
Other sources were less convinced by this idea though. Many seem sceptical that clearing houses are ready to take on the cash markets, and argue that there are already drivers pushing more NDF and FX options activity into these venues. Further, the e-FX head in London points out that most of the business on CME’s futures platform is from professional traders, which they claim these non-bank market makers do not really want to face up against as counterparties. They add that these market makers use CME for exiting risk, not assuming it.
A few sources also name-checked Capitolis and Traiana as firms that could potentially benefit from the shifting dynamic signalled by Citi’s decision.
“I think the FXPB model – the clearing model – has been broken for some time and is ripe innovation. Capitolis, Traiana and anyone else that could potentially have a unique solution has an opportunity. Dynamic credit tools that view credit changes based on the amount of liquidity in the market, the time of day and how much you’re trading would be a real value. Auto reducing credit during inactive periods – that solution doesn’t exist and innovation around that would be great,” says a source at one trading firm.
Looking at the bigger picture, they suggest that more innovation could be one of the most important outcomes from Citi cutting its FXPB client base back so dramatically.
“Sometimes you have to have events like this to get innovation. Things in FXPB had been status quo for a long time,” a source notes.