The analysis that accompanies the final tables of the BIS Triennial survey makes for some interesting reading, however I do think it possible to come to the wrong conclusions when looking at the likely impact on the market structure.
At face value, the excellent analysis highlights the reduced role of the Dealer (in survey speak this is Reporting Banks), the increase in non-bank market makers, the increased role of prime brokerage, and the increase in “hot potato” trading. This screams “exchange” when assessing the likely market structure and indeed it could be that we are heading that way, however I think this overlooks other aspects of the authors’ findings and, indeed, the data itself.
Without a doubt, the rise of “hot potato’ trading signals a reduction in risk warehousing, one of the traditional services of the banks (and as a friend recently pointed out to me, the rules over proprietary trading are so ambiguous this could actually be seen as prop trading and therefore banned soon).
One of the major changes in the market structure of recent years has been reduced holding time, this is inevitable given the rise in influence of non-bank market makers and high frequency traders (that can be one and the same of course), but that does not necessarily mean the FX industry is favouring an exchange-type solution.
I was particularly interested in the rise in Direct Electronic methods. Activity on single dealer platforms rose pretty much in line with the overall increase in activity, but the “Other” (non-SDP) segment was huge – responsible for nearly a quarter of all spot business and one-fifth of all activity.
This is clearly the impact of aggregation and to me confirms the rise in importance of relationship trading, which is, of course, the antithesis of the exchange. Yes, the big story maybe the breakdown of the inter-dealer market, but the reality is this has been happening for almost 10 years since EBS Prime was mooted and Hotspot started gaining some traction in spot markets.
What the BIS report seems to confirm is a market structure undergoing change, we are seeing the final breaking down of barriers between the B2B and B2C markets, resulting in the end of the inter-dealer market, and alongside that customers are building what are effectively their own multi-dealer platforms.
It strikes me, however, that what we are really seeing is the same market structure, it’s just the names have changed.
The FX market is still made up of a group of public markets from which most people derive a price (think the voice brokers and then an enlarged EBS/Reuters Matching group in the 1990s) – it’s just the dominant players in that space have changed and perhaps (I have to stress perhaps) people are deciding they want to trade on different public venues.
It is ridiculous to say that, for example, a GSA Capital or Virtu Financial has replaced, again for example, a Bank Austria or Chemical Bank of days gone past, because the business models of both are very different. That said, however, both those banks were known as market makers in their day and as such to end users, a price is a price – it doesn’t matter in the age of PB who delivers it.
Equally, the profile of the end user has changed. What were once major players in the FX market are now content to have a lower risk profile, service their own customers, and leave the rest to the big boys to fight over. Again, it is only the names that have changed. Whereas before, those “using” the market rather than “making” it were predominantly asset managers and corporates, this group has now expanded to include regional and local banks.
In effect then, although the price is more visible, the underlying structure of “haves” and Have-nots” remains in place. The real concern for the market is likely to be concentration risk, but even then, this is little new for the industry. I am often told of the “problem” of the FX market being held in the hands of three or four banks, but is it really a problem?
I would suggest that three or four players controlling the FX market is a problem, but would also argue this is not the case and that there are probably three non-bank players who all hold decent shares in market activity. Yes, they are prime broked by the same banks who dominate the banking segment, but there are strict Chinese Walls between dealing and prime brokerage, so although the bank “sees” the flow, in reality, the people that can actually influence markets – the dealing desk as well as the price and risk engines – often cannot.
There are justifiable fears over non-bank market makers grabbing too much market share for this does raise the risk of sharp movements. As I have pointed out before, these firms have no “customers” in the traditional sense of the word (although some do now through liquidity provision arrangements with some banks) so have every right to pull out of the market when things get too hot. Banks don’t have that privilege.
I suppose in an ideal market scenario, there would be eight or nine banks in the very top group, but the reality is (and with apologies to some who think they are very top tier) there aren’t. In this situation, if non-bank market makers are taking up the slack and providing an alternative source of liquidity over the preferred mechanism of those customers (including lower tier banks) then that has to be good for the market doesn’t it?
The only thing is, and taking this piece full circle, customers seem to prefer trading on a disclosed basis with people that can offer them “full amount” liquidity. That would seem to rule out several of the non-bank providers.
So overall, yes there are changes in the market microstructure – this column has been discussing them for literally years now – but it can also be argued its just the names that are changing. On one hand we have fragmentation in the multi-dealer venue industry (but the new model is no different to the old) and on the other we have concentration of activity in fewer market participants (whose modus operandi has also not changed over the years).
The only real difference is the technology we are using, but that said, these remain interesting times…
Colin_lambert@profit-loss.com Twitter @lamboPnL