Ahead of my normal duties at Profit & Loss Mexico on Thursday evening and Singapore next Wednesday afternoon (both local time), I had the all-too-rare pleasure of being on the other side of the “divide” this week, as a speaker on an excellent ACI UK panel on conduct and the FX Global Code. I was pondering that discussion this morning when I came to the conclusion that we, as an industry, have actually not moved on that far over the past two years on some critical issues.
This and next Monday’s column are therefore going to come across as something like a “best of” (some of you might substitute “worst of”) that deal with regular issues in these pages, because I feel it is important to summarise where we are, and where we need to go, if the FX industry is to ensure it is not to be dragged into another scandal.
Today we will deal with hold times in last look, Monday will be the biggie in pre-hedging. Neither of these issues has, to my mind, been dealt with adequately, especially against a backdrop – as highlighted by my esteemed fellow panellist Jack Drohan on the ACI UK webinar – of increased US regulatory activity.
Something Jack said tweaked my antenna and while I may, as has been known, be reading too much into something, my first thought when he mentioned a principles-based regulation from the US National Futures Association (which has apparently been picked up by the CFTC) that included the word “honour”, was about last look.
Of course this could simply be the use of a somewhat old-fashioned phrase meaning people are expected to conduct themselves in an honest fashion, but my immediate thought was, “are there regulators out there who will apply that to a trade?’ As in, market participants should honour their trade? If there were people thinking that way, then the FX industry has a whole heap of headlines coming its way with reject rates in some places as high as 60%.
So what can be done about it? The first port of call has to be the Global Foreign Exchange Committee, but I honestly can’t see the deeply entrenched interests on either side of the GFXC’s P17 debate shifting. At the moment, and this was highlighted on the ACI UK panel, the most important tool in the box is the disclosure. I am sure they were much looser, and they certainly weren’t public, but I would be amazed if the actions brought against the banks weren’t actually covered by some sort of disclosure document drafted early in the century.
My point is, a disclosure didn’t stop the authorities going after the banks and, occasionally, individuals, over last look. It didn’t help that the banks rolled over and paid up while the individuals often fought the case via the proxy of an unfair dismissal claim, and won, but the facts are the facts – an action that few were thinking about and that probably was very loosely covered in a disclosure document of sorts, came back to bite everyone on the backside.
Last look as an issue should be easier than this to solve. It is a data driven problem, so surely there should be data driven answers? I understand that some consumers of liquidity act in a less than trustworthy manner and the LPs need protection against that at times. The data, however, indicates the average cost of this client (and it can, no doubt, be projected across the entire institution to take into account profits on, for example, equities), so why not just widen that client out to take that into account? Some will throw their hands up in horror at the loss of business, but why do you want business that systematically costs you money? That is a question that should be asked anytime there is a problem with a particular consumer.
The data also tells us where the problems lay in terms of anonymous venues, so either turn the tag off if it is possible, or stop quoting to that venue. Do you really need to be quoting across dozens and dozens of venues (and a tip of the hat to Citi here, naturally) when you don’t actually make enough money from them?
Switching the argument around, any LP worth its name in the G7 currencies is taking a 5 millisecond feed from the primaries and a 25ms feed in lesser currency pairs. Why then, are they holding trades for upwards of 100ms? If it’s a “dodgy” counterparty, then I refer you to my earlier point about cutting them, or the platform, off. If it is a technology issue then we need firms to actually share their hold times for the counterparty and how that sits with a peer average (and no, I do not think active hedge fund streams should be the benchmark for corporates or asset managers).
If we have transparency around this process, then consumers can make their own choice and certain connections come with a very clear “buyer beware” rider. If something goes wrong there is nowhere to lash out and get recompense – the buck stays right where it should be in those circumstances. Equally, if the LP is being fair, they have the right to go to the customer and show the latter exactly why they are suddenly on their “D” stream.
I don’t like last look, I think that much has been obvious since I first wrote about it in the mid-2000s, but I understand how there is a need in some quarters for a defence mechanism against speed or execution style. We are now in a position, however, to account for the impact of certain trading styles and indeed last look, in a much more accurate fashion than ever before.
This is something for the GFXC to discuss and deal with and I implore those members of the committee who refuse to consider a change to reassess their position. It was argued on the ACI UK panel that the FX Global Code benefits from the teeth that comes with a regulator backing it and using it as its baseline for expected behaviour. That baseline can be quantified where it perhaps could not have been three or four years ago.
Best practice around last look should evolve to a recommended ceiling on hold times (it can be dependent upon the speed of a player’s primary data feed and be, for example, 2X the average data pulse). It should also promote fair asymmetric behaviours, such as hold times on accepts and rejects. Most importantly, it should then recommend that LPs and platforms share in an audited fashion their average hold times with the counterparty and the peer average.
If we can achieve that as an industry, then last look will not come back to haunt us. My fear is that with volatility returning and, inevitably, reject rates going up and hold times being extended, the entrenched positions on the GFXC will stop the Code doing what it was designed to do – evolve with the markets.
My understanding is that a recording of the ACI UK event will be available on the ACI UK website in due course