That we are still debating the positive or
negative impact of non-bank market makers on the FX market doesn’t surprise me
– what does is the simplistic level of debate over what I consider to be a
fairly complex issue.
Last week in Copenhagen it struck me how
many people I spoke with who thought the big prime brokers could shut the
non-banks down at the drop of a hat. I suppose technically they could of
course, a hefty rise in prime brokerage fees or a withdrawal of credit totally
would send most of these firms back to where they first emerged – the cleared
world with its very limited spot market opportunities.
I don’t understand why they would want to,
however, because the banks are not only (I assume) making money from the
non-bank PB clients, but there are smaller prime of prime operators (who exist
to a degree at the behest of the major banks of course) who will step in.
The whole bank/non-bank debate is much more
complex than that, however, for while I don’t think it is necessarily healthy
for XTX’s future expansion to have such a prominent position in the Euromoney rankings because it just
raises attention above FX business level (imagine a major prime broker like
Citi making it easier for its big rival JP Morgan to grab some of its business
and vice versa for example), it is a
fact of the modern market that XTX and firms like it are having a bigger impact.
Those who are still reluctant to embrace
the non-bank sector are, however, conveniently ignoring that a gap has emerged
in the industry for these firms to fill. There are a bunch of lower tier one,
tier two (and beyond) banks that have just stepped away or decided to go
agency, and this inevitably created opportunities.
To me the big issue is, and has been for
the past 18 months since I first really started focusing on it in the wake of
the WMR reforms, market impact. It has been interesting to see this issue grow
in the minds of more people in the industry as the months have passed, to the stage
that now, as the algos grab more market share, it is close to becoming an
So we’re all worried about the market
impact of what were once considered run of the mill orders and too many seem
content to merely blame the non-bank segment. That is a nonsensical argument in
my book because there are several banks who handle risk in a churn and burn
Non-bank firms will tell you they are
holding larger risk for longer periods than ever before but speaking to sources
familiar with the matter and with a bunch of counterparties, I don’t think this
larger and longer strategy qualifies for risk warehousing. Yes their behaviour
has changed, but not to the degree where if I was a corporate treasurer or head
of an execution desk at a large asset manager, I would consider executing a
large ticket with them (unless I was after market impact of course, which is
another matter entirely!)
To reiterate though, there are plenty of
banks that behave in just the same fashion – the number of major risk warehousing
operations has shrunk dramatically over the past couple of years.
Which brings me to a rarely stated aspect
of this whole debate.
We blandly accept that a customer has the
right to expect little or no market impact on their larger orders, but do they?
After all, it’s not as though they are “full only you” on these orders.
They have accepted the empirical argument
that breaking a larger ticket into smaller amounts is the way to go – which
means they are diluting their relationship with the liquidity providers by
giving them all a piece of the action. Which really means, if the customer is
particularly large or smart over market timing, they’re kindly inviting all
market makers to lose money when they trade!
Another facet is the skill or directional
nature of the client’s business. If that client knows they are smart and
generally are going to see the market move their way – why on earth should they
expect a liquidity provider to hold the risk for any length of time? The LPs
have businesses to run as well and if they don’t make money (or lose about 80%
of their capital due to a fine – are you listening DoJ?) they will simply stop
operating and concentration risk and signalling risk go up, while liquidity
levels go down.
So when I hear, as I did last week, genuine
end users of the FX market complain about market impact I have less sympathy
than many. By breaking an order down across multiple LPs the “customer” is
assuming market risk, just as they are if they use an algo. Elements of this broader
risk are signalling risk and market impact – and that is an issue across bank
and non-bank LPs.
Some of these buy siders are unhappy at the
growth of the non-bank sector, but they are missing the point that their
changing behaviour is partly responsible. The only way to avoid market impact
on an order as an end user is to RFQ the trade – and if you ask more than one
LP and the “winner” gets caned on the deal, your spreads will be wider and
liquidity thinner next time you return to the table.
There is a lot of debate over the changing
market structure, but I don’t believe it should be along the bank/non-bank
lines. Customers are more than ever asking how an LP handles their flow – the
LP has just as much a right as to ask the customer how much of that flow they
More pertinently the line between how a
bank and non-bank firm manages the risk continues to blur. There are many
flavours of non-bank firm of course, just as there are banks, so this really is
a matter of individual taste. You want to execute cleanly and not have to worry
about impact? Trade with one counterparty.
You want “best” execution (it rarely is of
course but don’t get me started on that one)? Then ask around – but don’t for
one minute think you have a right to be protected from commercial interests. My
sense is that particular aspect of the age of entitlement in FX terms is coming
to an end – and I for one think that is a good thing.