The Last Look…
Posted by Colin Lambert. Last updated: November 1, 2021
There are many reasons I hope the FX market does not follow the path of equities – and the start of a court case in the US last week reminded me just why.
A US court started hearing arguments relating to Citadel Securities suing of the SEC over its decision to allow exchange operator IEX to offer an order type that predicts the next movement on the US NBBO (national best bid and offer) tape to protect market participants from latency arbitrage. The so-called D-Limit order monitors price changes on other exchanges and in expectation of a change in the NBBO, adjusts the price it displays.
I am not sure where to start, because frankly I have problems with most aspects of this – but let’s get the tricky one out of the way first. Should an exchange (or platform) operator be able to change the prices it displays without checking first with the placer of that order? Presumably with this limit type the placer agrees to have the price adjusted, but what happens if the anticipated change in market is so fleeting the buyer or seller doesn’t actually get filled? You could argue they are taking that risk by using the limit type, but I am uncomfortable having the venue actually change the price unilaterally.
I believe that MQLs and speed bumps are healthy for a market, so in principle I have no problem with a strategy that seeks to limit the advantage of speed, but the purist in me thinks, “if you don’t want to buy or sell there why are you placing the order?”
It’s different in an RFS environment, a market maker is being asked for a price so may be ambivalent as to what happens, but in an exchange/CLOB environment it’s different because, in its purest form, it’s an expression of interest venue. In my mind, some form of randomisation or speed bump seems cleaner than the venue itself changing the price – admittedly based upon data.
Of course, I am not sure, given the opening arguments, that the plaintiff is a supporter of MQLs or speed bumps either, because and this is the bigger issue for me,– and probably the main reason I don’t want FX to follow equities – it reflects an obsession with speed that certain market players have. The order type was created because some firms – Citadel Securities among them presumably – were able to trade quicker than the data feeds from other exchanges changed.
You could argue this is the reward for investing in microwave towers, satellite networks and any other strategy that trims a microsecond or two off round trip times, but is it healthy for the overall market? I must confess my intimacy with US equity traders is anything but close, however I can’t see why a retail (or institutional for that matter) investor feels they are being disadvantaged because the trading process is being slowed down by a few milliseconds. The only people who seem bothered by this are those firms who are predicated upon speed, which brings me to another reason to hope FX ploughs its own path – risk absorption.
The high-speed players want to leverage their technological advantage to exit risk almost the microsecond they are given it. I actually shouldn’t use the word “risk” there, because frankly there is so little of it actually taken. Complaining about not being able to hit a resting bid or offer just ahead of a market move seems, at face value, reasonable, but when you consider that most of the time this happens it is because the market maker has just been given or taken you are left with the inevitable conclusion that it’s really only part of the merry-go-round that is latency arbitrage.
These “market makers” would not be making the prices they do elsewhere if there wasn’t a better bid/offer elsewhere on the system that they can take advantage of, so are they really actually adding anything beyond an illusion that the market is deep and liquid? Equally, their risk holding times is microscopic unless things go wrong, which is clearly not that often looking at the money they make, so are they really providing a valuable service?
I absolutely accept that speed is vitally important in delivering trade data to risk systems, but that’s really a post-trade process, albeit in micro- or milliseconds, when it comes to actually being able to hit a price, however? If people are made to stand in behind their prices on these public venues and are kept on the hop by randomisation and other techniques, then surely a better market environment for everyone emerges?
In the old days in FX, traders that hit on the change were called snipers and actually got themselves a bad name. In US equity markets today, they’re called HFTs and seem feted in some circles.
What strikes me about this debate, and others that have gone before it, is that I cannot recall the end investors actually being asked what they think? Some exchanges aren’t going to do it because they want to continue milking the HFT cow that provides them with so much brokerage; the HFTs themselves certainly aren’t because they recognise a good thing when they see it; and, it seems, the authorities have either got bigger fish to fry or can’t be bothered.
If we believe – as I do – that markets exist for the end-user, then surely we should ask them? Have the SEC or whichever of the seemingly 10,000 US regulators is in charge of this stuff run a consultation amongst actual end investors. If nothing else, at least then we should have a view as to whether the people that matter care. Currently, it feels like two billionaires, standing in the middle of a poverty-stricken slum, arguing over a ten dollar bill they found on the ground.
There is little skill in taking advantage of price changes, in the old days in FX, traders that hit on the change were called snipers and actually got themselves a bad name. In US equity markets, they’re called HFTs and seem feted in some circles, and that is the biggest problem of all and something for the FX industry to consider. US equity markets seem like a playground for HFTs to show off the latest technological advance that gains them a microsecond and enables their latency arbitrage to thrive. They don’t seem a place that is beneficial for anyone else – and even the HFTs must, at some stage, start to question the value of investment in shaving a microsecond off a round-trip time?
It won’t come as a surprise to many of you to hear that I think the FX industry has, largely, got it right – fast enough is good enough. Technology that delivers speed is largely used for risk monitoring and measuring, while when it comes to trading there is, generally, a fair and balanced playing field for everyone to play on.
It also has genuine risk absorbers, and that should never be forgotten, or be overlooked by clients seeking to hedge (or enter) risk. The alternative is a high-market impact environment, where protecting yourself suddenly costs a lot more – that should be food for thought for those who believe the FX market structure needs to change and, more pertinently, more reflect that of the equities world.