The Last Look…
Posted by Colin Lambert. Last updated: January 22, 2025
We had a quiet 10th anniversary in FX last week, and I’m not surprised it was quiet, for it was hardly the market’s finest hour, but it has been a big driver of market structure change over the last decade.
Obviously I am talking about the de-pegging of the Swiss franc from the euro on January 15, 2015 – still the most stunning event I have seen in my near-47 years in the industry. It was such an amazing event that I still use it for a case study on the ACI Australia Dealing Simulation Course – and I always start with a price formation to guess the low traded…and it always takes a couple of dozen prices until we get anywhere near the right answer!
The collapse to near-zero was, naturally, a flash crash, although I suspect EUR/CHF still ended the day significantly lower than the Swiss National Bank intended when it pulled the plug. The carnage that was unleashed by that move is still felt today in the market – try getting a bank prime brokerage account if you are an up-and-coming fund – but it also provided many valuable lessons.
Firstly, I think it made people smarter about how they sourced their pricing data – we have all heard about the bank that forgot the CME circuit breaker! It also, talking to people about the aftermath, made LPs more skittish and quicker to widen out. I do not think this is actually a bad thing, because what that reflects is LPs putting a better value on their liquidity. True, prices are still very tight (although as I reflected last week, they may not be executable), but when the smelly stuff hits the fan, LPs are better equipped to deal with it.
A couple of people I have spoken to counter this argument, observing we have had flash events since, most notably in Cable and AUD/JPY, but I would observe those events also reflected the lack of liquidity that has always been a factor in the early Asia-Pacific market. As my good friend Steve Flanagan likes to point out, the markets traded at regular and frequent price points in these events, and he is right of course, but the bottom line is just there was never going to be enough liquidity to meet demand at that time of the day. Of course, the market didn’t trade at regular and frequent price points in the EUR/CHF debacle.
I think we have also learnt a lot about execution algos, and I suspect that the lingering memory of January 2015, is behind what still seems to be a disappointing take up of these products in the FX market. There has been progress, without a doubt, and some providers are doing nicely with their agency algos, but overall, I suspect the majority have not seen return on investment – especially those who spent heavily on building more sophisticated strategies.
Algos are being used, of course, but when it comes to execution, my feedback is it is still heavily-focused on the core strategies that have existed for more than a decade. Opportunistic algos have done better, largely thanks to March 2020, but overall, I wonder if there is still a fear factor about their use? As always, I think the fear level should be aligned with the quality of the provider, some of the offerings I have seen have superb risk controls and probably should be used more (they probably are in fact), whereas there are still those whose value proposition seems to be “get it done and don’t worry about the market impact because we’ll hit top of book all the way up or down”.
SNB-Day can also be seen as the real trigger for a host of LPs deciding they didn’t want to manage client orders on a principal basis – because the risk-reward balance had shifted dramatically to compile the existing conflict between client and provider. It took some time, but the number of LPs who tell me they simply pass the orders on, or stick a whole lot of rules around it to make it less attractive to the client, has grown exponentially over the last decade.
This is a shame, because some clients need access to third-party order management, although I would say that in some cases, the client demands around execution quality haven’t helped the wider cause. When the market drops 20 points in milliseconds – as can still happen – it’s unfair to expect the LP to execute the order at, or one pip from the rate…unless of course, they have “(pre)hedged” it, but we don’t want to go down that rabbit hole today!
The number one impact of SNB-Day, however, has to be what I have already referred to – the destruction of the easy credit model. From retail brokers who went under, to those who pursued their clients for margin (including taking their house), we saw the best and worst of human nature when it comes to money, but the result has been everyone getting nervy about extending credit.
To a degree, the industry has helped get over this with the prime-of-prime model, but all that has really done is pushed the problem further away from regulator eyes. There are some very good, well-regulated, PoP providers out there, backed by solid banks in some cases, but there are also those regulated in places that I have never heard of that list their LPs as being other retail brokerage platforms. If things go wrong, there will be further carnage, just at the lower end of the spectrum.
The reality is that the retail world has always been a little murky in foreign exchange, but SNB-Day changed how many hedge funds, run by traders with years of experience, were viewed by the banks. Most are well-serviced by reputable PoPs, but you do wonder at what point the environment changes from “safe” to “uncertain” and then “risky”? Is it at $100 million AUM? Or $50 or 10 million?
The hope is we never find out, of course, but markets have a nasty habit of surprising us just when we think we have seen it all. I genuinely don’t think we will see another event on the scale of January 2015, but we will see them – especially given what can politely be termed the “uncertain and potentially volatile geo-political environment”.
January 15, 2015 was a spectacular day, and one we will talk about for decades to come I suspect, but was it good for the FX industry? The optics weren’t – and aren’t – great, and there are the lingering effects on credit. The market structure has also changed, so we don’t know, for example, how the clearing mechanism will hold up in the face of a catastrophic event such as the de-pegging (or US Treasuries flash crash).
Overall though, in terms of market functioning – and by that I mean the availability of a price when required – it has not only recovered well, but, I would argue, become even more robust. There is an infamous quote from a US army officer in the Vietnam conflict – “it became necessary to destroy the town to save it” – that attracts ridicule to this day, but for the FX market, the absolute destruction of EUR/CHF for a few minutes 10 years ago has led to a better, more robust market.
There is nothing wrong in things going horribly pear-shaped as long as you learn from the mistake, and I think everyone in FX, from central bankers, traders and the oversight functions, has demonstrated they have learnt. It hasn’t been a straight line, and I personally would prefer banks to have more capital to offer to hedge funds, but it could be argued the latest, more robust, iteration of the FX market started just over a decade ago in Switzerland.