Read time: 5 min

And Finally…

What on earth would make me argue for another platform? A different mechanism that’s what.

It was suggested to me in an email last week that I really don’t like regulators. Absolutely wrong, I just don’t like some of the regulations that are being pushed through as they are only likely to degrade, rather than improve, the market infrastructure.

While I think some CFTC rules are likely to precipitate a race to the bottom in terms of regulatory arbitrage, a conversation last week reinforced my belief that the US has got it right over retail FX. There are countless other regulations that are sensible and will make the markets a safer place.

As if to reinforce my balance when it comes to regulators, I come this week to praise another – the Australian Securities and Investments Commission – regarding Dark Pools and HFT.

I think it was back in 2004 in Profit & Loss when I first actively promoted the idea of a Dark Pool in FX and in spite of a couple of reasonable efforts, the concept has yet to really take hold, and I recognise (but don’t agree with) the argument of one or two regular correspondents that the entire FX market is a dark pool.

My perception of a Dark Pool then was a mechanism on which participants can clear large risk, indeed it remains so to this day. It is, unbelievably, coming up to a decade throughout which I have argued that matching large risk in a dark environment means less market impact, lower processing costs and the chance to attract larger client flow through DMA mechanisms.

Although it was not accepted at the time, one thing that has become an overriding concern is the cost of processing thousands of tickets. I accept the efficient frontier argument when it comes to FX processing, but there has to be more costs, even if it is in building excess capacity in repeatedly having to break $50 million trades into 50 or more tickets.

More to the point, the “sniffing” technology in the market has become better so that it is easier than ever for a market participant to identify an order being worked, and skew the market accordingly. Algorithmic execution strategies are a good advance, but speaking to someone on this issue last week, they were confident that it was easier to identify a large order being worked than it was to mask it. 

This all argues for a Dark Pool with the appropriate rules on minimum size and time in market, as well as the flexibility to allow discretion around orders.

I repeat this (and one day I will be proved right with a successful mechanism!) because the ASIC paper last week had some of these facets as part of its recommendations. The paper looks at Australian equity markets, but it’s not irrelevant to FX markets because a dark mechanism, is a dark mechanism – and it’s all about the rules.

The consultation paper – anyone wishing to comment can do until May 10 – lays out a series of proposed rules for Dark Pool and Crossing System operators. For Dark Pools it suggests a trigger point for minimum size orders in the dark mechanism, “where there is evidence dark liquidity has caused degradation in the price formation of a security or group of securities”.

Interestingly, another proposal is for a minimum resting time “for small and fleeting orders”, something that seems directly designed to combat order “flashing” and excess messaging that is associated with some HFT players.

The proposal is that if dark liquidity reaches 10% of the total liquidity in a instrument, and there is an increase in the “lit” spread of 4% (ASIC offers an alternative of 20%, and a 15% (or 20%) decrease in depth of market at the top five price points, then minimum order sizes would come into operation.

These make sense it would seem to me in the equities market, but as is well known I am not that much of an expert of that ecosystem, so what should the appropriate regime be if the FX market were subject to a similar review?

Well actually I don’t think the FX markets need triggers, I would argue that minimum size thresholds would work just as well (perhaps with partial fills allowed subject to certain fill ratios over a period of time) as a basic rule of thumb. In spite of all the segment’s success in improving the standard of technology in FX markets, it seems apparent that as many players as possible are determined to avoid HFT players, or at least force them into trading larger lots.

One of the problems with operating a Dark Pool in FX is that the provider needs to be paid and brokerage levels, due to the smaller amount of trades, are not sufficiently high. This makes it difficult to convince someone of the economic benefits in building a dedicated venue, but the ASIC proposals for equity markets could offer a potential solution.

A mechanism could impose minimum resting times on orders under a certain amount, but again I would suggest this amount be at the level where the vast majority of HFT firms that still engage in excess messaging and arbitrage cannot or will not participate. There are some honourable exceptions to the rule in the wider HFT community that are co-existing with the banks on some of the platforms, and I am sure they can/will stream in larger amounts that those operating at the margins of rectitude.

This all tells me that a venue operating with a minimum amount of, for example, $2 million (and equivalents) could see some decent traffic volume and be open to all in the market that satisfy that criteria. If that proves unworkable because bad behaviour still exists, then simply up the minimum size, as ASIC is suggesting.

The ultimate solution remains, to my mind, one where participants can match risk in much larger amounts, block trading to all intents and purposes, or an aggregation of aggregation. The various multi-dealer platforms are all circling the issue as things stand, but too many are going out of their way to be HFT unfriendly, or perhaps too bank friendly.

This is not the way things should be – the right mechanism should be open to all who qualify and who can operate within the rules. So I would argue that while we have more platforms coming to market with various solutions, we still have a need for a credible, well-managed and correctly regulated Dark Pool. I have said before and I will say it again, it’s all in the rules.

This column is taking a short break over the Easter holiday, but will return early next month. I clearly need a holiday for if I read this right I have just argued for greater fragmentation!

Profit & Loss

Share This

Share on facebook
Share on google
Share on twitter
Share on linkedin
Share on reddit

Related Posts in