Cobalt Pilot a Big Moment for FX Industry?

As a Cobalt pilot programme delivers a more than 50% reduction in credit utilisation, Colin Lambert asks, has the FX industry finally solved its longest standing bottleneck?

Credit as a bottleneck in FX has been discussed for so long it has almost become a cliché. Everyone has known about the problem, but due to a blend of negligence and other priorities, little has been really done to solve the issue. Platforms continue to push providers to carve out credit and those providers are still over-allocating. Furthermore, real exposures are over-stated thanks to a lack of netting, meaning a central plank of an efficient market, credit optimisation, is foregone.

Against this background, financial market infrastructure provider Cobalt, has released the results of a pilot programme run with seven major FX banks across 10 trading venues, which indicate that credit utilisation could be reduced by more than 50%. For the pilot, Cobalt received live data from all 10 venues for spot interbank transactions among the banks in the group. It says that from one weeks’ worth of data there was a 47% reduction in the utilisation of credit with one highly active bank achieving a 60% reduction.

These are serious numbers by any standard, especially if, as would inevitably happen, the efficiency gains are enhanced by a wider group of participants. In what could be a significant moment for the modern foreign exchange industry, the technology to ease the credit bottleneck has not only been rolled out, but it has been proven to work.

As previously reported by The Full FX, GFMA’s FX division (GFXD) has formed a working group to investigate the systemic risks created by the inefficient distribution and management of credit, the PBs are the spearhead of this, but in reality the challenge is one faced by just about every market participant. Therefore, in what is perhaps an important, but subtle shift, and while the PBs do indeed benefit from significant efficiency gains, the pilot offers benefits to any trading house.

Changing the Game

At the heart of the credit issue is, it could be argued, a lack of intent on the part of too many industry participants, and a lack of understanding around the real issues. Andy Coyne, co-founder and chief product officer at Cobalt, argues that only one credit line is important – the global limit. “It’s a simple statement, but it doesn’t matter how or where I trade with my counterparty, all I really care about is the global limit with that firm,” he explains. “Everything else is process – and it is being managed badly, because currently it is fragmenting credit.

“There is also no standardisation,” he continues. “Different firms measure NOP (net open position) and DSL (daily settlement limits) differently and there has been a lack of investment at the platforms.

One of the factors why platforms have been slow to invest is that they don’t own the credit and associated risk. They previously have focused on persuading credit providers to carve out a portion of their global limit and allocate it to their platform. This is an inefficient way of handling the issue because it does lead to over-allocation – and natural latencies in the process mean credit providers rarely have a real-time view of their credit utilisation with individual counterparties. Changing the allocation to a venue is also a laborious process.

With efforts to build electronic trading levels in swaps continuing, credit efficiency becomes an important part of the process. Effectively having dynamic credit and netting processes available can boost the automation push, not least because it is, in effect, a first stage of optimisation.

In short, as has been noted for several years in the FX industry, there is a need manage and allocate credit better, something Cobalt is seeking to address with its solution. The pilot used a combination of dynamic credit distribution and automated switching. The dynamic credit engine distributes a single limit simultaneously to multiple venues and thanks to real-time trade matching, something that Coyne says “simplifies everyone’s world, including the venues”.

He adds, “The credit officers only have to worry about the global limit, the platforms are more efficient and get access to more credit as and when it is needed because the system handles the underlying allocations dynamically and simultaneously. This solves the problems of credit fragmentation, carve outs and administration – and it creates a common standard.”

It is not only about credit officers and the platforms, however, for dealers themselves can benefit. As Darren Coote, CEO of Cobalt observes, when you get away from CLS members and currencies, limits tend to be smaller, therefore more precious. “Limits are valuable in these markets and dealers don’t want to miss trades,” he explains. “As things stand, a bank could be constrained from trading because their limit is full with a counterparty, but in reality, there are offsetting trades. We can net those trades and free up credit for more business to be done.

“In the major markets the e-traders are less concerned by this, because there is often another bid or offer a fraction of a pip away and it all comes out in the wash,” he adds. “Get away from G3, however, and the price gaps can be meaningful.”

Because Cobalt runs a shared ledger, it can also run an automated switching process, which nets down exposures across counterparties adding, Coyne suggests, at least another 10% efficiency to credit utilisation. These switching sessions can be run on a regular basis, for example at market time zone changeovers, but could also be adjusted for market regimes, for example in highly volatile market conditions where positions are changing quickly, the gap between runs would probably be reduced.

Although the pilot did focus on spot, there is an obvious opportunity in non-spot FX products, especially FX swaps. With efforts to build electronic trading levels in swaps continuing, credit efficiency becomes an important part of the process. Effectively having dynamic credit and netting processes available can boost the automation push, not least because it is, in effect, a first stage of optimisation. Solutions for reducing capital exposures continue to be rolled out, but generally speaking, Coyne observes, there is a need for reconciliation because compression runs, for example, can be initiated. “Credit should be the first line of optimisation, it can be run as often as needed,” he observes. “From there, the more sophisticated optimisation processes, such as multi-lateral compression, can be run.”

Why Now?

There is a sense that the timing of Cobalt’s push and the successful pilot, is very good. The firm has been live with 24 Exchange for almost a year and a second platform is due to join soon, The Full FX understands. That said, take up has not been as quick as perhaps it should have been considering the scale of the problem being solved, or at worst eased.

For the banks, a big driver of adoption has to be regulation – already under pressure from the implementation of SA-CCR and UMR, capital exposures have become a big topic in these institutions, and maximising credit can play an important role in reducing them. Perhaps the challenge for Cobalt is that banks, largely speaking, are not worried about their credit exposures in spot markets, where much of the data and focus exists in the industry. As noted, however, more forwards and FX swaps are being traded electronically, and this provides opportunities in a product set that the banks care about much more.

The drive for change has to come from the banks…with FX markets getting busier, the need for an efficient credit process is heightened – it is in busy markets that the good trading businesses differentiate themselves.

If the concept, which has been proven in spot, is extended into forwards and swaps, then the argument for adoption across the FX product suite becomes compelling. To a degree, the adoption, or otherwise, of the solution in spot depends very much upon developments in FX swaps.

The question of why more platforms are yet to engage is an interesting one, because with the exception of the two primary CLOBs and those venues with a central counterparty, platforms continue battling for credit crave outs, which is an out-dated and defensive tactic.

There is the argument that some venues prefer not to ease the credit bottleneck because they have a reasonable level of business that is protected by credit carve outs, but such a head-in-the-sand stance is unlikely to stand the test of time. Inevitably, easing the credit bottleneck opens the field for more competition and, certain traders look away now, more fragmentation of liquidity in FX markets. It also, however, offers opportunity for greater competition – one in which platforms with good technology, operating procedures and processes, will thrive.

In all likelihood, some platforms will have to be forced to adopt a streaming credit solution with dynamic distribution, but if they really want to get in front of their customers, sell and buy side, they are able to do so now.

More importantly perhaps, for all participants, is the growing demand from regulators and market bodies for something to be done about a problem that has existed ever since electronic trading hit the mainstream. Rarely is a regulator-imposed measure a better solution than something developed by the industry.

Ultimately the most important calculation in this instance – those who would benefit versus those who would not – falls very firmly on the side of the beneficiaries. Equally, a more efficient FX industry would emerge – one that does not perhaps, as some believe will come to pass, inevitably lead to a listed, cleared, market structure. It’s one thing an OTC platform trying to retain market share through credit, but it is ultimately futile if the result of continued inefficiency, is a regulatory push away from that model.

In all probability, the drive for change has to come from the banks, starting with those who have the data and results from the Cobalt pilot programme – and at the heart of that push should be the e-trading teams. After all, with FX markets getting busier, the need for an efficient credit process is heightened – it is in busy markets that the good trading businesses differentiate themselves and being able to deal more, with more counterparties, is at the centre of that.

It may be a small step for now, but turning theory into reality with a streaming credit API and shared ledger that have proven their worth in this pilot programme, could, in a few years’ time, be seen as a significant event in the FX industry.