The Last Look…
Posted by Colin Lambert. Last updated: February 27, 2023
This is the 100th column since we formed The Full FX, I would like to sincerely thank everyone who provides feedback to these missives – both negative and positive. Without it, the world and certainly this publication, would be a poorer place.
One of the most enjoyable aspects of my professional life is the time I get to spend educating, and learning from, the next generation of people in FX, not only is it rewarding in itself, but I also get a reality check on some of the issues that I, and many others, think are still unsolved in FX markets.
I am talking about my help running the ACI Australia Dealing Simulation Course, the latest of which was last week in the UK. Part of what we do to prepare the next generation for the challenges and opportunities of working in our industry is through what we call “Ethical Debates” but are in reality a discussion about one or more areas that we think they should be aware of.
Last week, we covered repapering trades, last look, and the benchmark fix. I asked an old and trusted friend – yet another person guaranteed to give me the unvarnished truth! – whether my approach was balanced; after all, we all know on two of those subjects I can go off the deep end! I am assured I gave a balanced view of both sides’ arguments and, given I have pretty much had the same approach to the last six or seven courses where we have conducted these sessions, I would like to share with the readership the collated, and anecdotal, thoughts of the next generation.
I should stress that their hands on experience is very limited in FX markets, but they are uniformly very intelligent and thoughtful – and highly thought-of by their employers, so the views are considered and, frankly, I think it is important we listen to them.
On repapering trades, we talk them through SNB-Day, certain parties can look away now, and the furore over the repapering of trades and the setting of the low at what was considered an artificial point. We also talk them through the arguments of both sides in the UK court case between Bank of America and CFH.
The general consensus is that the establishment of the low at 0.85 was probably a little high, but fair. There are questions asked as to what happens to those traders who took risk on lower, understanding as they did, that EUR/CHF at 0.5 and lower levels was frankly ridiculous.
On the court case, each team of two or three people act as members of a jury and they have to cast a vote. On one of the seven courses, the group found unanimously that the trades executed by CFH below 0.25 should not be re-papered, and on the others it has been a large majority (most often 8-1, 7-2, or 8-2) the same way.
The last look debate is an eye opener for me. I am not exactly a fan of the practice as everyone knows, but the next gen accept the argument presented that there sometimes needs to be protection when quoting to so many channels; there may be latency to certain geographies; and some liquidity consumers are predatory and need subduing.
What was a surprise to me on this argument was how it seemed a 50-50 split over asymmetric last look, with some no finding an issue with it – even when they are told of the guidance alongside the FX Global Code, that cites best practice as matching trades as quickly as possible with no additional hold time.
Clearly, asymmetric responses (and I am talking more than a one or two millisecond difference over a month) indicate that some hold times (often on rejects) are being artificially extended.
This suggests to me that, and, not for the first time, I could be wrong on this, that while I believe the FX Global Code has done a great job in cleaning up last look, that job is not yet over – as highlighted by this column on January 12. Perhaps the GFXC should be looking at asymmetric responses? For a start it would help if the views in the “guidance” were actually inserted into the Code itself, rather than in something “to be read alongside the Code”. The next generation could probably do with some help and guidance here from an independent and authoritative body.
If you execute $1 billion from 2pm to 2:30pm, the TCA counts every trade; why aren’t they counting every trade in the $1 billion at the Fix?
Finally, on the subject of benchmarks – and I am very careful about how I present this argument on the course – we talk through how it used to work, why it went wrong; what the changes were; and how it works today.
We have been presenting on the Fix for some years now and the response is always the same – mystification. It has been pointed out by some delegates that there is likely to be a huge cost to change (but it is far from a majority view) and most acknowledge that the industry is between a rock and a hard place where executing parties have to trade ahead of the Fix to avoid excessive market impact.
Last week, however, someone (a member of the next generation I should stress) got themselves very much on my Christmas card list, by making the simple observation, “if you execute $1 billion from 2pm to 2:30pm, the TCA counts every trade; why aren’t they counting every trade in the $1 billion at the Fix?”
We have had plenty of delegates question the use of the Fix full stop, but this is a more subtle observation that really does get to the crux of the matter. We are executing trades where the TCA takes place over perhaps one-third or one-quarter of the time horizon used – how can that make sense?
I’ll leave it there – we all know I could go on for another hour on this – with the observation that this is sad news for those of you who think I will one day shut up about the Fix. Thanks to the views of the next generation to pass through our markets, until there is real change, I am not going to!