It was 10 years ago yesterday that one of the more higher profile ventures of recent years in the FX space shut its doors in that October 17 2008 was the last day of trading for FXMarketSpace.
There have been times in the past five years when people in the industry – and I have to stress they did not work for FXMS at any time – have told me they thought the centrally-cleared, anonymous model would have been ideal for the current environment. I am not sure whether it would have succeeded had it come to market in, say, 2014, but there is no doubt that a model that alleviates the pressure on prime brokers in terms of the capital rules, does get more of a hearing today than it ever did a decade ago.
To me the problems with FXMS were simple: it attracted the wrong sort of trading firm at the time; it failed to get a netting solution off the ground; and it was too transparent – and by that I mean it was on a hiding to nothing by publishing volumes at such an early stage. I cannot think of a single platform that has published volume data so early apart from perhaps FastMatch, which did so three or four months into existence.
So would the FXMS model work now? Possibly it would, but I am unconvinced because, to repeat the doubts I expressed at the time, a successful FXMS would be in direct competition with its two owners, CME and Reuters. It could be argued that the two years the two firms gave FXMS was too short, after all the most successful platform launch this century (and yes, there is plenty of time for another one to surpass it) was probably FXall, and even that took over two years to achieve break even.
In the current world, the fungibility between cash and futures continues to increase, most recently via CME’s FX Link product, so would there really be a need for a centrally-cleared spot solution? I understand the argument that prime brokers would gain some relief and there is no doubt that PBs facilitate a lot of volume in spot FX, but would it be a game changer? I am not so sure. Of course, it could be argued that CME’s recent surge in volumes indicates that there would have been room, however I have never been one to see positives in cannibalisation!
Another problem is that the nature of the market and its participants has changed and without a different approach, FXMS would run into the same problems today it did a decade and more ago. There is a rapidly-dwindling minority of liquidity providers willing to price aggressively into an anonymous venue that is populated by small, tech-savvy firms – especially one on which their speed is bolstered by the absence of credit constraints.
In many ways the failure of FXMS actually signalled the start of what would be a very slow change in attitudes in the FX industry. This was really the first time the major LPs took a stand and said they didn’t like pricing into a certain liquidity pools (and the lack of attractive buy side customers certainly didn’t help). It also represented physical evidence of the major LPs refusing to interact with the HFTs, something that finally nailed the door to the market shut for many of these firms.
HFTs also found on FXMS that their worst nightmare, in FX terms at least, was being in a liquidity pool with their own type. In many ways it highlights the predatory aspect of these firms strategy that without the “soft” (probably that should read “slow”) flow from the banks and other less-sophisticated participants, they simply cannot succeed – they end up eating each other. Of course, at a philosophical level, this can also be argued about banks and how their top end market making businesses are effectively funded by the P&L generated from less sophisticated customers.
It is hard overall though, for me to see that FXMS would have really had a significantly better experience had it been launched in the past five years, but that is not all the story.
Obviously given the amount of resources being thrown at the credit problem in FX there is still an issue to be solved. The thing is, that issue is not really in spot, it’s in the forwards or swaps – and that is where I think FXMarketSpace would have succeeded. I remember sitting down with someone senior at CME a few months before the closure and it was obvious that the Merc was worried about its lack of traction and considering pulling the plug. I couldn’t think of anything to say in defence of a platform that was struggling to put up numbers normally associated with the larger retail platforms, let alone institutional, but I did wonder aloud whether a move into FX swaps could work?
This had, my conversant told me, apparently already been considered but here timing and the Lehman debacle hurt FXMS, because it was decided that the short term outlook was too uncertain to feel confident in executing such a drastic change of direction (not to mention the cost involved). Thus the idea never got off the ground.
Roll forward to the present day and we still don’t have either a really successful CLOB for FX swaps, I would estimate that about 25% of FX swaps are executed in this fashion and the vast majority of that is in the short dates, or a cleared solution. The problem is who becomes the central counterparty? No prime broker is really going to want that job – or more specifically will want to do it at the price necessary for it to be economically viable – and clearing houses are still battling away on other products to worry about something that is not mandated.
Of course, it could be argued (and yes, this is tongue in cheek) that Deutsche Borse have spent more than EUR 800 million and CME potentially $5.4 billion in acquiring OTC platforms that may help to feed an FX clearing solution and therefore it would have been cheaper to switch FXMarketSpace to an FX swaps platform and keep spending $50 million per year on it.
At some stage I do see an FXMarketSpace-type solution in FX swaps, however I just cannot see where it comes from in the short term, unless, perhaps, it is one result of any Deutsche Borse and Refinitiv deal. There is a need for such a solution and that need is only likely to grow more urgent – especially if banks pull back from the FX swaps market in the face of competition the way they did in spot.
It was often said, by me and others of a similar background (in age terms), that FXMarketSpace was the Martin Peters of FX platforms (for the uninitiated, Peters, an English football player was once lauded as being 10 years ahead of his time by then England manager Sir Alf Ramsey). The fact is though, we were probably wrong because, 10 years on, no such solution exists. It probably will do in five years’ time, but sadly, to continue the analogy, I can’t think of any sportspeople who were that far ahead…