The Last Look…
Posted by Colin Lambert. Last updated: October 11, 2022
Great is the number of post-2005 entrants to the FX market that have been battered with stories and legends from the voice broking days by a certain generation. I have been known to drop the odd story into a conversation, be it those horrible days when everything went wrong (the best stories, no-one wants to hear from you when things have gone well), or, rather topically, the broker who refused to allow his banks to “sell the proud pound” as it plummeted towards what was, until last month, the all-time low.
I mention this because it seems that TP Icap has shut its G10 voice broking desk (Lee Oliver still breaking news from retirement!). The first reaction of many has been “didn’t know it was still going”; it was, but that widespread response serves to highlight how tough it must have been to maintain the business.
The last time I checked in, some time ago I have to confess, the voice desk was handling flow into the Fix, the odd trade from other desks, and, rather like the description of a very over-populated New York Timesnewsroom in the 1960s, it was sitting around waiting for something to go wrong (in this case with the machines).
Of course, this doesn’t mean TP Icap is out of the spot game, it has its own platform, and in several emerging markets there is still a thriving voice broking business – it just doesn’t exist in the top 20 currencies any more – aside from our old friends, the Scandis, who still require that specialist touch from time-to-time. If we are going to be gloomy, however, one could question how long even the Scandi and EM desks have given the creep of technology into these markets?
I am genuinely surprised that there was sufficient flow to sustain a voice broker for the past decade, and if these desks didn’t see an upturn in activity over the past year then they never will do – hence what appears to be the business’ death warrant.
In April 2022, voice brokers in the UK and US executed something like $50 billion per day in spot, I suspect a fair amount of this was ahead of the 4pm Fix and the rest in a range of EM currencies. Either way, it’s probably a $3-5 million business per month for the industry as a whole – decent, but is it enough to sustain desks that are probably well past their sell-by date?
It is probably a tribute to the professionalism of these desks that they are/were still going in 2022, for I recall in the late 1990s, spot volumes were already a fraction of what they were thanks to the rise of EBS and Reuters Matching. Since then, the rise of the so-called secondary ECNs, along with other technology advances have eaten away further at what was already a diminishing business.
I suspect the industry’s growing ability to match flow ahead of the Fix electronically has delivered the final blow, there is no small irony in the fact that the firm that Icap sold, has spawned the main platform for doing so. I am in no way suggesting the sale was a wrong decision – although it does seem strange to sell off a business seen as non-core and then a couple of years later launch a competitor – rather it’s funny how these things turn out.
There were, though, other pressures, not least the aforementioned proliferation of ECNs. The spot voice broking business also often relied upon risk-taking traders for business and as this has dried up, especially amongst the banks, another source of liquidity has diminished.
Voice broking still has a future of course, it’s just not in the major currency pairs and, possibly, it may not be in spot FX full stop
With banks now very much focused upon execution rather than risk taking (they are, ironically, called “brokers” by many on the buy side), this creates another challenge for a voice broker. There is no doubt that with a good voice broker, back in the day, a dealer could shift a serious amount without much of the market knowing. Today, however, customers (and compliance) demand a TCA report, with very accurate timestamps – how is that achievable by a voice broker unless they are also hitting a platform, in which case, why not cut out a link in the chain and hit it yourself?
Furthermore, pricing and execution engines are powered by vast amounts of data and rarely have human interaction – how does a voice broker’s market data get into these engines? It can’t in spot markets at least.
Finally, as has been the case with the primary venues that themselves disrupted the voice brokers, internalisation has also played a role. The less risk that hits the public market the better for market impact as far as today’s FX world is concerned, therefore, again, the voice desk loses out.
So, for a certain generation, it’s a fond farewell to the voice broking desk, with the understanding that few of us were actually using them anymore. Voice broking still has a future of course, it’s just not in the major currency pairs and, possibly, it may not be in spot FX full stop.
If that is the case, spot voice broking can join many other skilled trades that have disappeared from our world thanks to technology. All I can say is the people shutting the business down are lucky the certain aforementioned voice broker isn’t still around. I can just see him chained to his desk vowing “you’ll never take my phones”.