The Global FX Committee is to soon embark on its first three year review of the FX Global Code and this represents a great opportunity to clarify – and hopefully eradicate – what I still believe to be the biggest legal risk facing the FX industry. Whereas I think prudent tweaks to the Code and the stressing of the need for total transparency around the practice means that one of our controversial issues – pre-hedging – will be taken care of; the other remains a problem. The issue is, of course, last look, but I am not so much concerned about its existence (I still don’t like it as regular readers know) but more about how it is being used.
I should stress I am not talking about the use of information or trading in the last look window, which were the initial problems with the practice – those, I believe, have largely been dealt with. The problem that exists now is one of asymmetricity of responses, both in terms of the counterparty and the outcome of the trade. Not only do I not understand how it can take radically different times to reject or accept a trade, but how can the same LP have different hold times for different clients?
To the first point, I started writing about asymmetric response times in 2017, you can read the first column here, and it still bothers me that the same LP, down the same pipe, can take multiple times longer to reject a trade than accept it. Back then I was told by some LPs that their longer response time was to give the client’s trade a greater opportunity to come back into court so they could fill it – and recently I have been getting the same response. My response is also the same as it was in 2017 – nonsense! If the client flow is non-toxic why are you even price checking it – surely you make enough money from that client over a portfolio of trades not to worry? And if the flow is toxic, what are the chances of it coming back into court in another 100ms and how is it helping the client by leaving them in the dark over their fill for even longer – meaning they may miss an opportunity elsewhere?
The technology and pipes are the same, the response times should be broadly similar – and I accept there is the odd nuance that means there may be a few milliseconds difference. People keep on telling me they can credit check in microseconds and the primary venues update, at the slowest, every 25ms, so why should it take longer than that?
This feeds into the second issue I raised – how can an LP have different hold times for different clients?
Again, I will stress that not all LPs do this – I spoke with several who have a firm policy in terms of for how long, if at all, trades are held before the checks are made and this is symmetric across clients – but enough do. Of course, officially firms have nothing to say on this, but privately senior figures in the LP world have told me whether they support such an asymmetric policy and it bothers me not only that several do and that anecdotally there are many more out there, but that few of them think there is anything wrong in it.
In no way am I suggesting that these people are deliberately trying to get an advantage, most often they tell me it is a defensive mechanism against the trickier flow they see. The archetypical example is the systematic corporate hedger who is subject to a zero hold time versus the rather aggressive hedge fund who is fond of spraying the market, who gets a 100ms hold time.
The Code states,
“If utilised, last look should be a risk control mechanism used in order to verify validity and/or price. The validity check should be intended to confirm that the transaction de- tails contained in the request to trade are appropriate from an operational perspective and there is sufficient available credit to enter into the transaction contemplated by the trade request. The price check should be intended to confirm whether the price at which the trade request was made remains consistent with the current price that would be available to the Client.”
Holding a trade artificially in the circumstances highlighted above is for neither a credit or market risk check – it is commercially driven. The apologists will reiterate their argument that they are protecting themselves against toxic flow – and I have some sympathy for their plight – but the appropriate response to that is to widen the spread quoted to that counterparty, not hold their trades for longer than is necessary. The counter argument continues that they are actually conducting a market risk check because they need longer to check whether the counterparty is trading away than one standard primary venue update, because sometimes the flow takes a little longer to feed through. This is still a commercial decision – it is saying “we know this party is likely trading away and we don’t want to be hurt by them”. What it is not saying is “when I conducted my standard check of [fill in time here] where was the market?”And don’t get me started on asking, why make a price that you’re not confident in? If you think the market is moving, skew. If you think the client is away, widen…or, and this is anathema in today’s world I know, don’t quote them at all.
I would also point out a fundamental flaw in this tactic anyway. Generally speaking the maximum hold time is established by whoever governs the trading mechanism and if the flow is toxic then most LPs using this tactic will hold that tricky flow for the maximum possible time. If that happens then the market checks at 25, 50, 75 and 100ms will show nothing, they will all accept the trade and the market will move after that. In other words, they will be in exactly the same position as they were when the trade was first executed! This practice only serves to potentially urt those firms that are doing the right thing and that can’t be good.
One alternative could be that the LP has traded during this hold window elsewhere and would no longer be hurt by the flow because it is matched off, but that is stepping into really tricky territory and again, it doesn’t change anything, it is a commercial decision, not a risk check.
I don’t care (I do of course) whether the hold time is 100ms or 1ms, but surely under the terms of the Code it should be the same for everyone, no matter how they trade?
There is also the question of deciding who is subject to what hold time?
To quote the Code again.
“A Market Participant should be transparent regarding its last look practices in order for the Client to understand and to be able to make an informed decision as to the manner in which last look is applied to their trading. The Market Participant should disclose, at a minimum, explanations regarding whether, and if so how, changes to price in either direction may impact the decision to accept or reject the trade, the expected or typical period of time for making that decision, and more broadly the purpose for using last look.”
I would stress there the use of the word “client” and not “clients”, for that tells me that each client should be told what their average or intended hold time is and why – it is not about putting out a generic disclosure to cover all clients, it needs to be more granular than that. I am assuming that all LPs who have signed up to the Code adhere to this, if they do not – again, we are running the risk of the legal industry feeding of us again.
If an LP is not totally transparent with the counterparty with regards to how, if at all, their hold time differs from the norm, then we could see some clients (not for the first time, I should point out) being unfairly impacted by a too-liberal application of last look rules.
The starting point for rectifying this issue (or condoning it) is the FX Global Code – either it has to explicitly say this practice is OK, or it should include a “negative example” to reinforce a tweaking of the language in Principle 17 which would make it clear that using last look for, admittedly oblique, commercial reasons is taboo. By doing this, clients (if they bother to look at the Code – and this would be another reason for them to do so) would then be fully informed or could at least ask the right questions of their service providers.
If the change is to explicitly frown upon using last look for commercial purposes then this would also represent a strong signal to certain trading platform owners and facilitators. I have long complained about these parties’ arguments that they can’t enforce behaviour because they can’t see what goes on away from their domain, well this is easy to spot. If a platform is not asking an LP why they are taking longer to reject than accept, or why their hold times are inconsistent, they are not being good market citizens. It is a signal that something is wrong and needs investigating, and that should supersede the commercial decision that may – and I stress the word ‘may’ – upset a big bro payer.
As an industry, we are in a period of reform that will not last forever. At some stage the regulators and legal eagles will have expected us to get our house in order which is why there can be no more prevarication over strengthening Principle 17 and making sure that last look guidelines are as watertight as we can get them. The message should be simple – don’t adjust the hold time, adjust the spread, and if you keep on getting stuffed then adjust it again!
I fully accept that the review of the Code is not about a major re-write and I don’t believe this is such a thing, it is an important adjustment to one of two Principles that remain controversial and difficult to build consensus upon.
I have a problem with last look mainly because it is a practice that effectively says ‘we only deal is the customer is wrong’, but beyond that it remains the one issue that could come back to bite the industry where it hurts. There was genuine shock in the FX world when a practice that had been largely condoned, sharing information in chat rooms, was suddenly seen by the outside world as wrong and legal and regulatory actions were brought that effectively sought to retro-fit behaviour.
Last look, and in particular treating clients in different fashions within the same framework, is such an issue that could easily be seen in the same fashion, and we need to do all we can to ensure that it doesn’t become the issue of 2021 in the law courts of the world.