A couple of months ago I wrote a piece highlighting what I suggested was the first signal of a real shift away from anonymous trading toward disclosed. The column received, as is usual I should point out, a mixed response, with several of you pointing out that without the anonymous pools of liquidity price discovery would be severely hampered and liquidity would dry up.
In extremis this is a very good point, but I think it ignores the important fact that all we really need for price discovery is one or two CLOBs. To argue, as some did, that increased disclosed trading is a threat to liquidity levels just seems plain wrong to me. The recent paper from the BIS Markets Group highlighted how, while volume was shrinking on the primary venues, they were still the main source of price discovery. This is apparent to anyone who watches the markets reasonably closely – it doesn’t take long to watch a dealer’s aggregator with the primary and secondary venues in it, to see the pattern emerge – pricing on a primary venue is replicated down the ladder. It’s always been that way and I don’t think it is changing at all.
What is changing, and this will have an affect on the primary market venues’ activity levels, is that LPs are just so much better at pricing these days. They simply don’t need a huge amount of volume on the primary venue to price off thanks to their correlation matrices, increased internalisation flows and their order books, which are also congregating into a few hands.
So not only – in my opinion – do we not need so many anonymous venues, their value in terms of liquidity levels has also been diminished. This is not to say the model will continue to shrink and disappear, more to say that I honestly cannot see a recovery in activity levels unless something very dramatic happens – and it is hard to see what that would be given how in stressed markets volume seems to go to the primary venues.
My opinion piece earlier this year cited data from the FX committees’ semi-annual surveys – more specifically those from the UK and US, the other committees mystifying reluctance to provide more data in their survey meaning those two centres’ numbers are the only available – but data from the platform world seems to be reinforcing it.
I should point out that comparing the ADV of those platforms to report indicate that most have added about 10 yards of volume compared to the 2017, so overall the picture remains healthy, but it is where the volume is being added that I find interesting. Sources familiar with the matter say a great deal of CboeFX’ growth has come from disclosed trading, similarly, NEX Markets is benefiting from having added disclosed channels some years ago.
Where it gets really interesting is when you look at the performance of FastMatch – often associated with the ECN model – and FXSpotStream, which is very much in the disclosed space. Comparing year-to-date volumes with 2017, FastMatch has added just under two yards a day, which at 10% is the sort of growth most firms would be happy with.
Over at FXSpotStream, however, the growth has been quite startling in percentage terms at least. The platform handled something almost 10 yards a day more in the first 10 months of this year, an almost 50% increase.
This is not to analyse the performance of these two platforms I ought to stress, more I am using what I consider to be the best two proxies for trading methods amongst those platforms who publish data. FastmatchFX does support disclosed trading and its business is about more than just one platform – as the growth of the Tape would suggest. There may also, of course, be other factors at work that are not public, that make it hard to accurately compare performance – and one should not ignore the economics involved.
Either way, however, it would seem that the disclosed model is finding favour among LPs, who are clearly happier pricing there. What is interesting is how more buy side firms are embracing the model, having shifted away in the wake of the chat room scandals. My sources tell me that the pricing is simply better via disclosed channels and larger amounts are available. That has to be attractive to any liquidity consumer beyond those looking to turn a quick buck – and for those firms, the ECN venues are still there. The question has to be, how much growth is there in the speculative segment, especially when the trading methods within that community can sometimes prove very unpopular with the LPs?
That aside, I find myself looking at the data and finding it hard to escape the conclusion that the numbers we all seem to hold so much store by, are reinforcing a sense that has grown and grown over the past couple of years. People on both sides of the liquidity divide have crunched the numbers and looked into the broader issues and they are backing the disclosed channel with that most powerful of voices – their business.