The Last Look…
Posted by Colin Lambert. Last updated: July 18, 2023
Reading the ESMA review on pre-hedging a couple of thoughts came to me on two of my favourite subjects – firstly, last look as an issue has probably gone away; secondly, pre-hedging is a ticking time bomb likely to blow up and bring misery to another generation of traders.
On last look, it seems clear that the extensive and lengthy work the Global FX Committee has undertaken on the subject has paid off. The respondents to ESMA seemed happy to accept that pre-hedging in the last look window is unacceptable and seemed to think that was the only issue involved. Of course, there is much more to this, people playing around with hold times for one, but generally speaking, it seems much harder now to find someone stridently critical of the practice (I’m still there with a diminishing band of brothers and sisters of course!)
On pre-hedging, however, we don’t seem to be moving forward, and we need to, because ESMA’s extensive investigation into the practice highlights how the regulators themselves have yet to settle on a decision, but when they do, it could fall one of two ways. The fear is it falls on the side of front-running and then we have a repeat of the scandalous witch hunt of 10 years ago amongst the dealing teams of the world.
As I noted in my brief note accompanying the story on ESMA’s review, the world remains polarised on pre-hedging, with entrenched views unlikely to shift any time soon – this means we need leadership from the regulators, and pretty quickly. In case anyone is in doubt on the importance of, and potential threat from, pre-hedging, at multiple times in the paper ESMA uses the phrase “front-running” in relation to the practice.
To be clear, ESMA does not states it is actually front-running, but the very fact that it believes one of the possible outcomes of its review is to find that pre-hedging is indeed such an act should worry anyone who is pre-hedging now.
I am equally worried about the idea of treating pre-hedging on a case-by-case basis. Obviously, there are trades that are too big to quote for (or execute in a five-minute window, but don’t get me started!) as far as many institutions are concerned, and it seems people are content with the idea of getting client permission to pre-hedge ahead of time, but is that possible on every trade that will be pre-hedged? I understand it can happen, but recent history shows that it is the minority of poor conduct that taints the entire industry and causes so much collateral damage.
More notably, lest we forget, in the Mark Johnson trial it became perfectly evident that the treasurer of Cairn Energy knew, and accepted, that HSBC were going to pre-hedge the order. That seems to me to be the customer giving permission for the practice in that instance, but clearly was ignored by the US legal system.
It is hard to see any authority, especially the rather indecisive European regulators, providing strong regulatory guidance soon, however, and for all its strengths, the FX Global Code is a best practice guidance document that may not stand up in a court of law. This means the industry probably needs to create a solution, and to me, there are two possible ideas, that maybe should be combined.
Firstly, the industry should take a very close look at the recent paper on pre-hedging published by Johannes Muhle-Karbe and Roel Oomen, you can find it here. This provided a framework for judging the appropriate level of pre-hedging and for thwarting attempts at bad practice. It may be a starting point, it may be a working solution, but if there is a wider debate and understanding of the issues, then that can help the regulators – and the executing parties and their customers – better understand what is happening.
Quantitative tools have enabled the industry to better monitor other conduct risks, could it be the same for pre-hedging? At the very least, the paper’s ideas should be looked at seriously and if it is not suitable, perhaps suggestions can be made as to how the model can be enhanced to make it so?
I actually bulked a little at the word “appropriate” two paragraphs ago, because I am still unsure about the whole pre-hedging thing and whether it is, actually, “appropriate”. Pre-hedging has come about because (largely bank) management has insisted on winning more and more customer business, often at the expense of sensible market risk assessment (i.e. an appropriate spread). Allied to the associated reduction in risk appetite, this has meant that LPs are quoting more aggressively than ever for big tickets, on the understanding they can pre-hedge, and, as the Cairn Energy treasurer noted in the Johnson case, make their money that way.
It was once deemed acceptable for traders to exchange information in chat rooms – they were encouraged to do so by management. Billions of dollars in fines later a huge number of these traders were wrongly dismissed, and the definition of “acceptable” was radically changed
And that is the crux of the issue; even though I would stake a fairly hefty sum on pre-hedging being profitable for the LP on the vast majority of occasions (almost all), we actually have no way of knowing howprofitable. There is no disclosure of this as far as I can tell. Yes, the LP says they can improve the spread on the quote, but they already know the direction of the trade and if you know that you can probably quote one pip wide on a couple of yards and still be profitable!
This means that the LP has to do the right thing and be reasonable, the challenge is, one person’s “reasonable” is another’s “scandalous”. Yet again, emotion plays a role.
All of which brings me to a second option – can we insist on TCA-ing the pre-hedges? Bring real transparency into the process? That way, the client knows what has occurred, and, assuming it is in line with the recommendations in the aforementioned academic paper, can be satisfied empirically that the LP did a reasonable job.
The problem with this approach, of course, is that this means the LP will make less money and effectively will be working an order for the client, so where’s the benefit? It would also push the relationship into the “agency” field from “principal” which opens up a whole new can of worms. Equally, there are different measures of success when it comes to pre-hedging. An LP could be pre-hedging to buy, for example, Sterling and get given a hefty sum by another client that it can internalise as part of the pre-hedge. This means the final amount pre-hedged could be bigger than recommended because there was zero information leakage from the internalised trade(s). Again though, I think it is fair to ask the question, shouldn’t the LP make money out of its internalisation programme? After all, it costs a lot to build and maintain.
The bottom line in this debate is about how much money is made and where. LPs are largely in support of the practice because they make decent money from it – and, frankly, can probably ensure they never lose money on such an order. It would be interesting, as I noted in the Full FX View that accompanied the ESMA story, to hear what the buy side thinks of this, because that, ultimately, is where the issue is going to be solved.
Either the buy side accepts the practice of pre-hedging, preferably using guidelines suggested in the academic paper by Muhle-Karbe and Oomen (or an enhanced version), or it has to accept wider spreads. A third option is to go to the algos, which means holding the market risk. This is, effectively, a suggestion that the buy side accepts the status quo, or actually does something to help alleviate conduct risk amongst its service providers – for that risk exists.
If the LP can prove empirically that the end rate for the client is better thanks to the pre-hedging, all well and good, the problem at the moment is that is very hard to do
We can hide behind recommendations in a code of conduct, we can believe that the practice of pre-hedging is OK because our management says so, but neither will offer protection if regulators decide at some stage in the future, that pre-hedging is front-running because the economic rationale of supply and demand means there will be a market impact of sorts from the practice. If the LP can prove empirically that the end rate for the client is better thanks to the pre-hedging, all well and good, the problem at the moment is that is very hard to do.
To solve this, we need to embrace the quant and the analytical world to provide some leadership for the world’s regulators – and that needs industry agreement through both camps moving from their entrenched positions.
Why we need to do this is, to my mind, pretty obvious – to protect this and future generations of trading teams. Some 15-20 years ago it was deemed acceptable that traders exchanged information in chat rooms – they were encouraged to do so by management. Some, no doubt, went too far, but many others stayed within what were seen as “acceptable” boundaries. Billions of dollars of fines later a huge number of these traders were wrongly dismissed, and the definition of “acceptable” was radically changed – as I discussed in last week’s column.
It is for all these reasons that the industry needs a wake-up call to move forward and solve the issues around pre-hedging. That way it can avoid a repeat of that awful period where too many ostensibly honest people lost their jobs. We don’t want to go down that road again.