The Last Look…
Posted by Colin Lambert. Last updated: April 22, 2024
Although strictly independent, and looking at the issue from a purely academic standpoint, the recent paper on pre-Fix hedging by Johannes Muhle-Karbe, Roel Oomen, and Benjamin Weber, strikes me as yet another argument for lengthening the current five-minute WM fixing window.
Obviously, I risk falling into the trap of acting like a politician and using facts and data and shaping them to support my argument, but the paper, which you can access here, highlights what underpins the Fix debate – and also what LSEG’s WM and its customers seem to care less about – the potential conflict of interest for dealers.
I don’t think there is any secret that the vast majority of dealers handling fixing flows want to do it the right way and they are challenged every time they execute such an order. They are looking for a healthy outcome that benefits the client, while not negatively impacting their own revenues – the problem is the current structure, especially at month-ends, rarely, if ever, allows that.
The problem is bad enough that the last few years have seen several regional and specialist banks tell me they either no longer off a fixing service, or they automatically hand off the trades to Tier 1 relationships. In an ideal world, this pooling of trades in the larger dealers would allow improved netting, but the fact remains that fixing flows remain heavily directional.
One thing I want to be clear on – and this is one finding of the research in the paper – is that large tickets cannot be hedged in the five-minute window. There is no real definition of what constitutes “large” flow, but the formulae presented in this paper allows people to work it out (if they have the volume data). Especially at month-end, however, I think we can all agree that the flows are “large” and would (do) overwhelm the available liquidity in the five-minute window.
In most “large trade” circumstances, therefore, hedging ahead of the window clearly benefits the client. The question we should be asking, however, is does the client(s) actually get that benefit? Surely, as I have argued for more than a decade now, if even 30% of the order is hedged outside the window, the vast majority of occasions sees the market move against the client, meaning the actual rate does not reflect the true cost of execution?
If firms are using the Fix, they clearly don’t care about execution quality… [They should] have a professional FX execution team within the business that could handle this for them
It would be interesting to see the approach proposed in the paper tested across the different mechanisms. As it notes, it uses the WM window, but can easily be adopted to test for the Siren FX methodology and its 20-minute calculation window, as well, I presume, as the B-Fix methodology, which uses executable quotes over its window. The other fixes used in the market, various central bank benchmarks, can’t really be considered for testing I assume, because there is effectively no window in which to measure transactions as they are ’point in time’ valuations.
Either way, I strongly suspect analysis would show that the cost to the client is actually lower using the longer window, because only the very largest trades would need to be executed over an even-longer period, and even if they did, the impact would be reduced due to the smaller percentage executed outside the calculation window.
Of course, should such an analysis be run, the results could make uncomfortable reading for certain parties, not least at WM and their buy side firms who seem to care very little about their FX execution quality. Would such a finding make for a change in approach? Sadly, I doubt it, because empirical evidence has been presented before that has been ignored – why should now be any different?
I have been thinking about those customers who, sources familiar with the matter tell me, believe there is nothing wrong with the current Fix window, and I suppose it reflects an attitude that too many in the asset management industry have towards FX. Put simply, if firms are using the Fix, they clearly don’t care about execution quality – if they did, they would either not be using the Fix, or they would be hedging in the days ahead with a small adjustment on the day. Even better, they would have a professional FX execution team within the business that could handle this for them – the outcome would more than pay for the extra bodies and/or technology, I am sure.
The fact is, due to the short window, and its inability to handle the amount customers want to put through it (and please can we stop hiding behind the fact it is a “reference rate” – yes, it is, but no-one is using it as such), everyone is making money out of this except for the end-investor. The dealers get the benefit of hedging ahead of the window (and their skill is judged, bizarrely, by how little they make); speculators are, as we have seen repeatedly, all over this and trading (legally) ahead of the window; and even some asset managers are making money by over-hedging expected flows in the days before, and then benefiting from the inevitable move on the day itself to make their adjustment trade.
I think it is fair to say that the situation is, still, a mess. One in which the people that so bother the regulators – the end investors – are being dudded. This paper highlights the value of hedging ahead of the window well – but the real question we should be asking is why we’re not using its ideas and findings to move to a more realistic – and fairer – window?