Thursday’s column elicited quite a lot of feedback, so to keep the ball rolling I thought I would discuss it here. As always not everyone agreed with me and I didn’t agree with every argument made, but the very fact that people are talking about it can only be a good thing – it was noted by several correspondents that as things stand, customers are among the collateral damage.

Where to start? Well more than one person got in contact to argue that tighter spreads are exacerbating the problem, because not only are banks seeing reduced revenues from lower customers volumes, but they are not making as much money on those trades even when they happen. I am not sure much can be done about that, it is a competitive market and simple aggregation means that customers can receive very tight spreads. Not only that, but as more customers use their data (or start actually using the data provided by the various platforms) they can easily see which of their liquidity providers are trying to widen out and simply tap into the extended list of LPs waiting in a queue behind the current crop.

Of course for some customers at the sharper end of the spectrum this is not an option, but then the banks’ revenues probably benefit from these customers doing less business! In the main, however, your average customer has more LPs than they need, which will mean spreads remain tight.

This is not necessarily a bad thing, for one of my points on Thursday was that by putting too many eggs in the basket marked “client” and ignoring opportunities presented by the markets (as opposed to the customer base) banks are effectively saying they want to make a lot of small incremental gains from each trade. In these circumstances the spread is still relevant, but not as much as the notional volume of business. I wonder if we are approaching a point at which spreads are at a level that even “good” customer business starts being marginalised? We have seen it amongst the more active, professional trading segment of the banks’ client bases, who is to say it won’t drip over to the asset manager space where best execution is becoming more important, or even the corporates (which is a limited market anyway in volume terms)?

We may not be there yet, but I have some bad news for the banking industry – spreads aren’t likely to widen, they are probably going to be squeezed even further. How does the customer-centric, volume-based model look then?

In many ways this topic is a digression from one of the key premises of my argument last week – that this lack of risk taking is negatively impacting liquidity in the market. Here there was whole-hearted agreement. As several people noted, order books at the banks, for whatever reason, are not that deep and even those orders that do exist on the take profit/market entry side of the ledger are often not put into a primary venue ahead of time. 

I have to say that I don’t actually believe that customer orders should be part of the trading business anyway, indeed a few banks are moving their client order books to a separate business unit. This greater confidence engendered amongst the client base – the knowledge that the orders are not being used to inform the trading business – is not having the positive knock on effect of deeper liquidity at the moment it seems, because some desks handling the orders are not placing them into the market until the last moment, and of course in a flash event the market has already moved through the level.

One correspondent suggested that the orders could be placed on the bank’s proprietary platform – most have order management capabilities – but it kind of breaks down the information barrier between segregated order desk and the trading business when a bid or offer appears on the latter’s screen or is available to its pricing engine!

It should be pointed out that sources at platforms, including primary venues, tell me that often there are resting orders on the venue and that these are filled very easily in flash moves – almost always to the benefit of the customer. During event at the start of the month, sources at banks and platforms say their customers picked up some real bargains in certain currency pairs, to the extent that by the time the customer had actually picked up the information their order was filled the market had already bounced back 50-100 points above/below the level.

Imagine then, how much deeper the market would be if trading desks at the banks were free to reinforce this resting liquidity by executing trade ideas of their own? As one friend of mine noted, machines are designed to be safe and will not guess where the market is if the data doesn’t fit their parameters, whereas a human trader will take a calculated guess.

By taking out customer orders and providing a prop desk with the same information the banks do their customers in terms of flow data the environment can be created wherein a new generation of risk takers helps bolster liquidity in FX markets. They should not be afraid of the high speed traders because after all, they should be looking to make more than just a tick here and there. 

And I think this is the important factor in this whole discussion. If the human prop trader is to make a comeback in banks then the role has to offer value and a differentiator, and to me that is in their ability to take risk over very different time horizons and with broader price targets. A human cannot price the FX market as well as a machine, nor can it manage risk over very short time horizons better. It can however, provide something different when it comes to actually trading (as opposed to pricing) the markets by looking longer term with wider parameters.

So there is a role there in a new structure – one that would help build more robust liquidity and provide a few more safeguards against silly moves, as well as help bolster FICC revenues at times when customers simply do less business. The problem is convincing people it can happen and will not bring a ton of problems with it.

Having spoken to a few people in regulators that deal with this, as well as bank internal markets compliance, I sense an understanding that this could be a good development, but it is unlikely to go anywhere. For while the aforementioned often have some trading experience and can appreciate the bigger picture, the minute the issue is escalated those managers with a more legal background see red flags and hear klaxons.

That is a problem of the banks’ own making of course, but as a vibrant and innovative industry it would be nice think that the FX market could reinvent itself again. Not so much back to the future but maybe somewhere halfway there.

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