Regular readers will know that I have long held reservations about exactly how much various platform operators supported the FX Global Code – specifically, would they bite the hand that feeds them by shutting a client off if they were disruptive or behaving at the margins of respectability?
My concerns are aimed at both sides of the LP/LC relationship, for the former I want to know, or rather the LCs and wider world should know, who, if anyone, has excessive rejection rates and/or hold times. I fully accept that the LP in question may be contravening the best practice set out in the Code on another venue and therefore it is difficult for the original venue to police that, but I would argue they should at least flag the activity so that other market participants are aware.
On the LC side we need to know who is creating market impact deliberately by, to use one example, machine gunning LPs, or in a second instance, taking what might be delicately called a flexible attitude to full amount trading.
As a digression, the latter is actually something I think we will be hearing more of in the coming year and will be an area of conflict between LPs and LCs – at least for a period of time before a happy medium is arrived at.
The challenge with full amount trading is simple – several LCs look at it differently to the way it used to be viewed. In the past it meant, if I was a client looking to trade 100 million I would ask one or more LPs for a price in 100. Now, this trade is broken down into, for example, 10 lots of 10 million and each tranche is traded as a full amount. In other words, they don’t ask multiple LPs for a price in 10, they ask one, therefore it is full amount to them.
The potential conflict arises from how long both parties believe the LP should have to manage that risk. For some LCs they are happy to let the LP internalise it over 20 seconds or more (obviously it depends upon the currency pair and other factors like time of day). Others, however, like to give the LP two or three seconds to exit the risk before hitting them (or someone else remember) “in full amount” again.
This creates market impact if the window allowed the LP is too short, because the latter will not want to hold the risk for too long if they know another (to use my example) 80 million is going to hit the market. Luckily we have the data available to make this a short term issue, but I have little doubt that there will be a period of tension between certain participants.
Anyway, as noted, I digress – because again the issue of what constitutes “full amount” or the length of time that is appropriate to allow an LP to clear the risk has nothing to do with the platform.
What does have plenty to do with the platform operators is how people execute and trade on their venues, and too many still hide behind the argument that they don’t know what is going on elsewhere, therefore they cannot effectively monitor behaviour for signs of shady practices. As noted before, though, I disagree because there are certain signals, such as excessive reject rates, hold times, or market impact, that indicate something may – and I stress the word ‘may’ – need investigating.
I cannot get away from the feeling that some – and again I have to stress the word ‘some’ – operators are hiding behind the bland statement that they have a rulebook. Often these are not public, so you have to be a client to find out what is and is not acceptable, but I sometimes wonder exactly how often the rules are applied. It is all very well having a rulebook, but if you don’t enforce them, or do so in an asymmetric fashion, then what value do rules have?
Some of you may be thinking I am reading too much into the issue (it wouldn’t be the first time!) but private conversations I have been having were reinforced during a session at the Profit & Loss Singapore conference when certain speakers appeared to support the notion of treating LPs and LCs differently. The basis of this approach seemed to be that the buy side had not created the problems that required the creation of the Code, therefore they should be treated differently. LPs, to put it bluntly, were the ‘bad guys’ and as such they had to be treated harsher.
This is patently unfair, and once again I find myself taking issue with the viewpoint that the customers had nothing to do with the misconduct issues we have seen. Firstly, without asset managers’ obsession with tracking error, leading to their use of the benchmark fix, they would not have pressed the banks into an overly-competitive position where the only route to salvation for dealers was to invent something called “pre-hedging”.
Secondly, without certain clients machine gunning the LPs (often with the sanction of the latters’ senior management thirsty for market share) and generally abusing liquidity, there would be much less need for last look. It would have been created as a hedge against latency from certain client connections and not as a protective measure that led to its use being overly-proscriptive.
To return to the original point, however, what is fair in a platform operator treating its LPs and LCs differently? If a rulebook is in place, is public, and is enforced, then there should be no need for this approach – an approach that does, frankly, indicate a casual, condescending and presumptive attitude to liquidity. Anyone with this view is taking it for granted, and one can only hope that one day they understand how wilfully wrong this is.
So perhaps we need to get to a situation wherein platform operators are required to demonstrate enforcement of their rulebooks? Again, I accept that certain behaviours cannot be effectively monitored but they can be hinted at, perhaps what we need is an industry ombudsman that can gain access to combined data, take a pan-platform view, to raise issues of concern? In such a scenario, there would then be a responsibility of the platform operator to report any concerns they may have and, more importantly, they would not be able to hide behind an obscure and difficult to obtain document.
There are platform operators out there with a very transparent approach to enforcing the rules on their venues, but clearly there is a broad range of attitudes and approaches. What we need is a set of guidelines (sorry GFXC – more work!) that set out exactly what level of transparency is required?
The GFXC is looking at disclosures as I write this and will report this week I believe, it strikes me that here we have that working group’s next task.