One of the problems with being associated with a subject for
several years is that you can feel the issue and the debate is getting a little
tired just when a lot of other people start getting involved – and you’re back
on the train! I feel this way about last look, but rather than bang on about it
yet again I thought it might be worthwhile looking at how the industry got
itself into this mess in the first place – and, possibly, how it can avert
further damage.

Monday’s column tweaked the interest of several of you –
specifically my observation that when looking at rejection rates we need a more
subjective consideration of customer behaviour. I noted that customers, of all
shapes and sizes, often try to execute trades – even hedges – when they see the
market moving in their direction of travel – hence why the cost of rejects will
always outstrip price improvement.

A friend got in contact to remind me about another impact
from customer behaviour that I did not mention – the fact they are using an
aggregator. When executing on an aggregator it is inevitable that the
aforementioned client behaviour is exacerbated as they see pricing shift in
front of their (sometimes virtual) eyes.

So I apologise for that oversight, which does also play quite
a role in the balance between reject cost and improvements.

While last look remains an emotive subject, my sense is that
within my circle of contacts and friends, it has moved on a little – specifically
those people are talking about the potential cost of last look. Clearly this is partly the result of those
against the practice seeking to raise the heat under the debate with some
numbers, but if I am right, this signals the start of the next level of debate
around last look.

But how did we get to this stage? What caused the use of
last look to trip over from being a genuine protection against technical and
wider infrastructure shortcomings into a practice where accusations are thrown
around about systematic and opportunistic abuse?

Without doubt the aforementioned customer behaviour played a
role – since I started in the industry almost 40 years ago traders have known
that certain customers would spray the market and that their business was, in
the modern parlance, toxic. The defence mechanism in pre-electronic trading
days was to quote those customers slower and wider – something that is a little
difficult to do in a streaming environment.

There are, and have been, simply too many customers abusing liquidity,
and the level of abuse stepped up when pricing not only went electronic but
also much quicker and tighter. The “machine gunner” is not a new concept in FX,
it’s just that over the past decade and more the identity of those using the
practice has been easier to hide.

To this old-as-the-hills aspect has been added speed of
course. A newer generation of “customers” emerged whose execution strategy was
very much about hitting the last of five or six prices as they moved – mostly
this was latency arbitrage, but there were other participants also acting in
this predatory fashion who were doing so on a directional basis.

I think customers – indeed most of the industry – have also,
over the past decade, become less appreciative of the value of liquidity. Last
look may exacerbate the mirage but liquidity is, according to too many people
in our industry, guaranteed and at top of book. This sense of entitlement has
also contributed to the mess.

All that said though, and accepting that certain customers
have – and continue to – abuse liquidity, the real problem lies not in their
behaviour, it lies with the liquidity providers.

The misconception that liquidity is endless and always there
has been perpetuated first by a small group of banks who were obsessed with market
share (another long term bugbear of mine!) and latterly by non-bank market
makers, some of whom claim, when it is clearly not the case, to be genuine
liquidity providers.

It is this egotistical drive to be seen as the biggest that
saw ostensibly sensible people allow their liquidity to be abused – and their
staff with it by the way – at the altar of market share.

If we weren’t so obsessed by it – and I do think the mood is
changing – then last look would not be used as it clearly is by some players today,
as a means to show how big they are and how great their (indicative) pricing
is.

So what can be done to change things? How do we extricate
ourselves from this mess?

Let’s start by accepting a few realities. Customers will
need to understand that liquidity will not be as deep as it appears (but often
is not) at the moment, and pricing will be wider (the question of whether
execution quality will be impugned is for another time).

Some LPs, especially in the banking world, will need to
accept that some of their higher turnover accounts are not customers in any
real sense. They have been abusing liquidity and as such the obvious answer is
to cut them off.

The FX industry had a problem for a short time with HFT, it
dealt with it by effectively ring-fencing the latency arbers onto a few venues
where they cannibalised each other – do the same with the liquidity abusers,
they’ll soon adjust their execution and trading style or go somewhere else.

Equally, some LPs will need to accept that liquidity
provision is a social good and not just a revenue generating exercise. The
global economy needs FX markets to function, therefore let’s get away from the
obsession with making money out of every trade.

Finally, and easily the most difficult area, the
facilitators. Too many, as I argued recently on the monthly Insight call, have
supported last look because it helps create the illusion that they are a liquid
venue. These venues know that if last look didn’t exist then they would – if
they are lucky – potentially see less pricing and volume. If they are unlucky
or just don’t fit, they are out of business.

So, as always in this world we have an issue being
perpetuated by vested interests, however I would argue that some of the
influences I have mentioned above, the market share drive in particular, can be
changed without any apparent negative impact on the business bottom line.

Make no mistake, removing last look will see some collateral
damage and some of it may be surprising. I would argue, however, that for the
genuine users (the hedgers) of the foreign exchange market the world could
actually be a better place. For others it wouldn’t, but if you are going to
shed a tear for the market share chasers, the machine gunners and the
arbitrageurs please do so – just don’t expect me to.

Colin_lambert@profit-loss.com

Twitter @lamboPnL

Twitter @Profit_and_Loss

Colin Lambert

Share This

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit

Related Posts in