If there were any doubt where the focus of ASIC has been to date in relation to enforcing it’s Market Integrity Rules, a further look at the published infringement notices tells us a clear tale of two key areas.

One is Automated Order Processing (AOP) and related filters, where putting aside the exact rule breach alleged, at the time of writing more than 85% – with penalties totalling $872,000 – have involved an AOP filter element which either caused or could have prevented the breach.

The second area is the creation of disorderly markets (ASX MIR 5.9.1), mainly via order entry errors, where more than 60% – with penalties totalling $532,000 – have involved price or volume errors caused either by direct entry algorithm or a Designated Trade Representative (DTR).

So the most obvious two questions are:

1. why has ASIC been focusing on this type of misconduct?; and
2. why does it continue to occur, given most are preventable with appropriate AOP filters and related processes?

Part of the answer to the first question may stem from ASIC needing to ‘get some runs on the board’ after taking over supervision from the ASX in August 2010, and breaches of ASX MIR 5.9.1 are generally seen as quicker to prove and progress. Another is that AOP related misconduct is a prominent issue for regulators globally, and ASIC realises that having appropriate AOP filters at both the participant and operator levels is key to preventing breaches and therefore maintaining market integrity in a world where upwards of 80% of all order creation is automated.

Another answer for disorderly market events is that they are simply easier to prove, particularly when market operators have worked hard to define more clearly when trades are considered to have occurred in an extreme trading range. If a trade occurs in this range then ASIC can clearly show it would have had a ‘detrimental impact to the integrity of the market’ which is an element that must be satisfied for a breach of most trading related MIRs. It can be argued that the erroneous traded price, when transmitted to the market at large, sends a misleading signal as to the market for, or the price of, the security in question. When that price also represents the official open, high, low or closing price for that security, in the absence of cancelling the trade the market is mislead into thinking that the price is genuine.

The second question is tougher to answer. Market participants are still used to a world where limited resources are allocated to market integrity compliance, both from a human expertise and importantly a technology standpoint. Even those who have implemented surveillance software are finding out that some of the focus has shifted to pre-trade and intra-day AOP filters, which prevent damaging trades occurring, as opposed to post-trade alerts and reports that are often under calibrated or misread on an end of day or T+1 basis. There is also a tendency for participants to blame the ASX for not having hard filters in place to enforce its extreme trading ranges. Although this would be nice and should be a focus for ASIC given the Chi-X platform seems to have these in place, it has also been made clear by ASIC that the first line of defence lies with the market participant, as the gatekeeper of all orders passing through it’s AOP gateway.

Also interesting is the change in approach by both ASIC and market participants to this type of breach since the first infringement notice, against Austock (published back in December 2011), where ‘DTR error’ was accepted it seems without further scrutiny as to whether an appropriate AOP ‘hard filter’ should have been in place to prevent the order reaching the market, versus the latest infringement notice for creating a disorderly market, against UBS (published on 22 October this year), where even though DTR oversight was again involved, this time the remedies put in place by UBS were highlighted in the notice, and included:

  • a review of all filter limits applicable to the relevant trading desk;
  • implementing a number of new filters applicable to that desk; and
  • implementing a quarterly review process whereby the filter reject levels and the application of the filters to clients and internal users in relation to that desk are affirmed or amended.

What is clear is that ASIC now expects participants to undertake at the very least a full review of their AOP filters, along with other organisational and technical measures designed to prevent this type of breach wherever possible.

To say the industry is playing catch up is an understatement, but with multi-breach infringement notices and escalating penalties attached, the real risk to a participants bottom line is starting to bite, as is the increased damage to reputation that comes with every published outcome.

In an ideal world the 2 tier defence system made up of appropriate filters at both the participant and operator level should make this type of breach an endangered species, thereby forcing ASIC’s focus to shift to other forms of market integrity misconduct.