Speakers at the Forex Network New York event highlighted that credit provision and intermediation for non-spot FX products could be set to undergo some significant changes in the near future, although there was some debate about how these changes would play out.
Christopher Perkins, global head, OTC clearing and FX prime brokerage at Cit, set the tone for the discussion by declaring that FX prime brokerage (FXPB) businesses have not been pricing their return on capital effectively and that, as a consequence of this and incoming regulations, these businesses need to fundamentally rethink how they charge for their services.
In particular, he argued that FXPBs should start charging execution brokers (EBs) rather than just clients for their services, a view that has since been advocated by Citi in a recently published whitepaper.
“When you step back and look at the value proposition of FXPB for clients, we’re giving them operational efficiency, capital efficiency and maybe some legal efficiency. Now, if you turn around and look at the value proposition we offer EBs, we’re giving them operational efficiencies and capital efficiencies because we’re posting margins to them and consolidating their exposures with the prime broker, in many cases reducing their margin call, assuming the client credit risk and taking on the client-facing RWA. So although we offer value to both the clients and the EBs, traditionally the client pays for everything. So I would argue that this business model needs to change – people who are given value need to pay and therefore the EBs need to start paying,” said Perkins.
Further, Perkins claimed that the current pricing model of FXPBs is about to become even more unsustainable in the face of the Uncleared Margin Rules (UMR) that are due to start hitting some buy side firms in September 2019, and then even more in September 2020. Under these rules, for non-spot FX products FXPBs will have to begin posting out collateral to EBs and, as Perkins pointed out, every time margin is posted out of the door it costs money.
He added: “PB is going to get more expensive if you’re trading in derivatives and swaps, whether you like it or not. Should we saddle the buy side with all of these costs? I would argue: no.”
The real credit challenge
By contrast, Leah Mallas, global head of FX prime brokerage and FX clearing at J.P. Morgan, does not think that the overall FX PB market is significantly mispriced right now.
“I think clients appreciate that there is value to the FXPB business which involves us taking on risk and providing credit”, she said.
Mallas added: “For a while it felt like there was a race to the bottom with FXPBs competing only on price but we tried to right-size that a few years ago.”
Pressed on whether the changes brought about by UMR could impact JP Morgan’s FXPB pricing model, Mallas said that they already consider these rules when considering client portfolios and the bank is ready to offer more central clearing to clients, should the market move in that direction.
Offering a buy side view on this issue Michael O’Brien, head of global trading at Eaton Vance, said that while he understands why FXPBs might want to split costs across both the clients and the EBs, ultimately no matter how the up-front costs are structured it is the clients that will end up footing the bill.
“From my perspective, I like to see explicitly what I’m paying. So long as that’s transparent, I can manage that charge in whichever direction it goes,” he said.
Importantly, O’Brien added: “I think it’s the access to credit that’s the challenge, not necessarily the cost of it. The minimum volumes needed to be an attractive PB client can be challenging even for large asset managers. So I think that one of the potential solutions here is central clearing in addition to the FXPB model, a combination of both could work for asset managers, especially if your FCM and FXPB are the same institution so that you get more operational efficiencies. We’re at a point now where we’re clearing virtually all of our deliverable and NDF trades that are eligible for clearing, partially because it’s a different way and an easier way to access credit than we’ve had in the past.”
FXPB Vs. Central Clearing
This in turn led to a debate amongst the panellists about whether central clearing could eventually become the prevalent credit model in FX for non-spot products.
“I think the two models will co-exist. Hedge funds will probably be slower to move away from the FXPB model because it’s already providing them most of what clearing would give them and there’s no regulatory mandate for clearing FX – there’s nothing forcing them towards this model. Asset managers will probably start clearing FX products first, especially the ones that are already clearing other products they trade. Over time, we might see dealers pricing it differently, with one price for clearing and another for FXPB.” said Mallas.
To which Perkins responded: “I slightly disagree with that view. I think that when the final UMR threshold is reached in 2020, it will be a book-end and the NDF market is going to essentially clear by that point in time because you can’t ignore the costs that will arrive. I don’t know what will happen with FX options, there’s still a big question over that, but if a trade can clear, then it’s going to clear because the economics are so profoundly different.”
However, Perkins was quick to add that he thinks the FXPB model will continue to exist, and that indeed under the UMR rules this service will actually become more valuable to market participants because of all the capital efficiencies it can provide to help clients cope with the new requirements.
For his part, O’Brien agreed that NDFs will largely move towards central clearing but that in most cases the decision about whether to use FXPB or central clearing will be product specific and situation specific.
“I can imagine a scenario where buy side traders are deciding whether they are going to use FXPB or central clearing when executing a trade on almost a trade-by-trade basis. So at the extremes I think that spot will always be traded via FXPB and I think that NDFs will go towards clearing, but for everything else in the middle I think there will be a divergence with firms using each service for different types of trades,” he said.
Infrastructure still lacking
However, the panellists did explain that there are some sizable infrastructure challenges that are preventing many FX products from becoming centrally cleared right now. For example, Perkins noted that while LCH has launched a service for clearing FX options, it is currently only for dealers and it will take time to build out the ecosystem necessary to manage the risk associated with clearing physically -settled options for buy side firms. Or, that perhaps liquidity will alternatively develop in a cleared, cash-settled options product.
He added: “Essentially, in order to clear an FX option it has to be cash settled, but that introduces basis in and of itself. Think of how complex the world could get when you have all these different products with different economics but that are really the same product. Let’s take forwards, for example. You could have a physically settled forward, a cash-settled bilateral forward, an LCH cleared cash-settled forward and a CME cleared cash-settled forward. That’s four forwards with slightly different economics and basis, it could get very complicated.”
“There’s really no solution yet for FX options,” agreed Mallas. “The dealers are testing out the LCH service but it’s going to be a while until that is sorted out, let alone getting clients on there. Some clients want to be clearing FX options sooner but there’s no viable solution today for them to do so.”
O’Brien also agreed that the infrastructure is lacking for asset managers to clear many FX products, stating that while the FCMs themselves are ready to clear the executions desks at the banks are not.
“The problem isn’t that the executions desks don’t want to do this, the problem is: how do you get a trade from an execution desk into clearing? How do you go from one to the other? In rates and credit there were good solutions to this issue, but there isn’t such an obvious one in FX,” he said.
The good news, added O’Brien, is that most buy side firms won’t be impacted by UMR until 2020 and so the FX industry has some time to work on building a scalable solution to this problem.