The Last Look…

A couple of months ago I observed that the multi-dealer platform world was in a state of flux – the “primary” venues would find it impossible to recreate the glory days and the other platforms would likely face – and struggle – in a price war due to low yields. That price war appears to be happening now.

We should not be surprised that firms are being pressured to cut fees, any market involving brokerage inevitably sees price wars – perhaps the surprise is it took so long? This is not a new story of course, three years ago Citi undertook an analysis of platforms which led to it cutting off many of what were admittedly way too many connections. I wrote in subsequent months that other banks were conducting the same analysis, this has now spread to more and more predominantly bank LPs.

What these players are finding out is what Citi did three years ago, yields on the multi-dealer platforms, that have been diminishing for some time, have reached the point where the costs of connecting, trading and maintaining, offer a negative return. Inevitably, and for the first time armed with hard data, these firms’ management have decided to squeeze the platforms.

This represents a significant shift in mood, especially from banks, who have traditionally been happy to be on the platforms to ensure they could continue to engage with clients – many of whom had initially led them there. At first it really was a case of studying the clients on the various venues when assessing their worth, but now it is very much about making it clear to clients that they are no longer willing to engage on these venues due to the lack of return.

The attitude of the clients is interesting. More of them are telling me they are concerned about the denigration of the liquidity (and spreads) they will receive and are willing to discuss alternate mechanisms and, in some cases, pressure the platforms to reduce fees. A substantial minority, however, observe that such a small percentage of their flow actually goes to LPs outside the top three or four, that they won’t notice any change.

The later point is an interesting one, because another recurring theme in these pages – and Profit & Loss beforehand – has been smarter aggregation, which normally means smaller panels of LPs, as per the series of academic papers by Roel Oomen in 2016-17. If smaller aggregation pools work, as the evidence appears to suggest they do, do the clients actually need to be on the platform in the first place? Especially if the platform offers little in the way of workflow?

So there has been a consolidation of flow into fewer LP hands for quite a time now – it was interrupted when the non-bank players made their move, but with one, maybe two exceptions, the real economy clients are still not engaging with this sector, they remain with the banks. Overlay this concentration, however, with the growing realisation on the part of lower tier players that they are not making any money from these venues and you have the perfect storm for the platforms – one group can muscle their prices lower, the other can threaten to walk away altogether.

The platforms that will survive what is likely to be a highly competitive time are going to be those that can demonstrate value to the LPs as well as the buy side.

This represents an interesting challenge for the platforms because to me it is not just about cutting brokerage – what we are seeing is a sea change, one that will only intensify. If brokerage is cut from $5 (or more) to $3 per million on these venues for example, will it make enough of a difference? Another underlying theme of the FX market in recent years has been the compression in spreads on these venues to the stage where “winner’s curse” is a regular feature. Yes, cutting brokerage will help, but it cannot compensate for the increased level of competition which is undermining LP yields to such a degree. The logical response from LPs is to price these venues wider and save their best streams for internal and “preferred” channels, such as tight aggregation pools.

The platforms that will survive what is likely to be a highly competitive time are going to be those that can demonstrate value to the LPs as well as the buy side. This puts more pressure on the liquidity managers at these venues, but the reality is even the best curated liquidity will not make enough money for some LPs.

I mentioned how the challenge will intensify and it will – in what could be a very important area, FX swaps. SA-CCR is going to hit the banks’ FX businesses hard, someone told me recently it will be “the new G-SIB”, which is widely seen as the major contributor to widening spreads and diminished liquidity at quarter ends. SA-CCR will increase costs to the banks, which means wider spreads or, as will happen with some, them walking away.

Another area of competition? The major banks themselves. White labelling is seen as largely a mature business in FX, and it is, however those banks that have built a serious technology stack that can meet the needs of, for example, a regional bank, are actively looking at how they can make this an outsource service. I am aware of two banks that are out and about discussing such a partnership with regional and smaller players, more may emerge.

The latter development actually brings me back to my arguments with the CLOBs more than 10 years ago, when I asserted that the bank single dealer platforms and algos would become competition for those venues. That obviously occurred, just look at the volumes on the “primary” venues, so can it happen in the multi-dealer world? Absolutely, and it probably will.

Volumes in this space are not really going anywhere, FXSpotStream is growing still it seems, but then that venue more reflects an aggregation/single dealer model with its one connection model to a limited number of LPs, with capped fees. Elsewhere? Well, they’re not really going down, but they aren’t exactly going higher either.

The biggest LPs emerging as the platform’s biggest competition is not a new phenomenon of course, but what is different this time perhaps is the quality of the tech stack available, how much easier it is to deploy, and, crucially, the increased willingness of the LPs to demonstrate tighter, deeper pricing through their preferred channels.

Paradoxically, and this is again something I have mentioned previously, this issue generally by-passes the firm CLOBs, especially Matching and EBS Market. As I noted in September, the LPs understand they get real value from these venues because it’s where they execute their exhaust flows – they want to trade there. The fact that the volume numbers are lower does not diminish in any way the value of those venues – sometimes, and this is hard for a platform I know, it’s about more than average daily volume!

So how this plays out will be interesting to see. It is easy to say that aggregation providers will come to dominate, but I don’t see that. Not only do some of these venues charge brokerage at the same level (and should be regulated the same I would argue), but there are also concerns over their ability to maintain investment and innovation in their tech stack. This means we are likely to see winners and losers emerge more clearly over the next year or two amongst the platforms, and former will be, I suspect, those who have embedded processes in the buy side systems.

Ultimately, the decision may lie with the clients, though, will they support their LPs’ drive to reduce costs? They should, because what the multi-dealer platforms are starting to understand – and the customers themselves will find this out – is that liquidity is much more valuable than it has been given credit for.

colin@thefullfx.com

@ColinlambertFX