Rob Daly: Where Mifid II Goes Wrong on HFT

A few weeks ago, the European Commission released its second draft of its Financial Instruments Directive, Mifid II, and, according to Rob, the proposed regulations look designed to fail.

Have you ever known from the outset that a business plan wouldn’t work? Either those who developed it did not think through the process entirely or they made assumptions that are totally detached from reality. The Markets in Financial Instruments Directive review (Mifid II) makes me wonder what regulators were thinking when it comes to high-frequency trading. In the entire 196-page document, high-frequency trading is mentioned a total of seven times in only four sections of the text. 

First, the European Commission wants all organizations providing market accesses to high-frequency traders, as well as the trading venues themselves, to implement the appropriate safeguards to “mitigate disorderly trading and ensure the resiliency of their platforms.”

Member states must have their regulated markets put systems in place to “reject orders that exceed pre-determined volume and price thresholds” and clearly erroneous trades if there is a significant price move in the instrument on that market or related markets during a short period of time.

The Commission also will require markets to have in place systems and procedures to make sure those algorithmic trading systems “cannot create or contribute to disorderly trading conditions on the market” and that they are able to slow down these trade flows if there is a risk of the trading system reaching capacity. The regulated markets will also need to set appropriate standards regarding risk controls for those trading via direct electronic access to the market, and be able to stop the order flow if necessary.

Indecent Proposal?
Another proposal is that member states must ensure that their regulated markets provide a transparent, fair and nondiscriminatory fee structure for their respective co-location services.

Finally, the Commission will observe the effects the new regulations have on high-frequency trading by examining the number of new authorized firms operating under the new regulations as well as tracking the amount of “disorderly trading” that can be sourced to high-frequency trading firms.

It seems to me these proposals can best be described as noble but impractical to implement on a state-by-state basis. The first issue—and this is an area you would assume the Commission would have spent a significant amount of time on—is that there isn’t a definition of high-frequency trading in the entire Mifid II document.

Coming up with a threshold that is a ratio of executed and canceled trades is a good way to detect the typical signs of high-frequency trading. But will the regulatory onus be on the individual broker-dealers who are executing high-frequency trades on behalf of their clients, or will it be the trading venues’ responsibility to see which participants are passing the threshold? I assume it will be the former, since only the venues know broker-dealers’ identities and if they slowed down the executions of the broker-dealers, that could affect non-high-frequency traders as well.

Harmonization
The most important issue I see is the simple harmonization of these standards across the member states. Given how uneven the original rollout of Mifid was in November 2007, I see nothing but late-adopting states gaming their “early” counterparts in order to attract firms and liquidity to their markets. Of course, I could be cynical and think that the EC officials hope that the national regulators make a total hash of the implementation of Mifid II so they can use their power under Article 290 of the Treaty of Lisbon to dictate all the details to the member states and their respective market regulators.

Not that this would ever happen—never. A supra-national regulator would never inflict unnecessary pain and grief onto its member states in order to consolidate its power. I’m not sure what I was thinking!