Dark pool data stresses need for more clarity

Attempts to improve the transparency of trading volumes in broker-owned crossing networks in Europe appear to need greater coordination following the publication of dark pool data for June 2010.
By None

Attempts to improve the transparency of trading volumes in broker-owned crossing networks in Europe appear to need greater coordination following the publication of dark pool data for June 2010.

According to recently published figures from Credit Suisse's Crossfinder dark pool and Citi's Citi Match platform, the combined total of trading on their internal crossing engines was €9.99 billion in June. Citi reported €5 billion worth of trading on Citi Match during June, while Credit Suisse traded €4.99 billion on Crossfinder.

Credit Suisse and Citi are also among six banks that recently agreed to report dark pool trading activity to trade reporting facility Markit BOAT but the total volume for all six pools in June is barely above that for Crossfinder and Citi Match combined.

According to the Markit BOAT data, the six broker pools – which also include the crossing networks of Deutsche Bank (DBA), J.P. Morgan (JPM-X), Morgan Stanley (MS Pool) and UBS (UBS PIN) – traded €11.27 billion in June. This means Credit Suisse and Citi accounted for around 89% of the total reported by Markit BOAT, leaving the remaining four crossing networks with just 11% of flow between them. While enquiries by theTRADEnews.com confirm that Credit Suisse and Citi account for the majority of trading across the six, individual brokers have suggested that the apparent dominance of the two firms is overstated.

Markit BOAT, which validates the data from each broker prior to publication, claims its methodology is consistent with MiFID guidelines.

This means its numbers do not include crossing between in-house books and only include trades where a client order has been crossed against another client order, or where a client order has been crossed with a broker’s internal flow.

Both Citi and Credit Suisse say their reports are single counted and only include liquidity that clients could have interacted with. Therefore, they do not include large trades the banks conduct with themselves or executions resulting in routing orders to other dark pools.

The apparent data inconsistencies may be due to the differences in the types of trading that takes place in each broker dark pool. Some broker dark pools may include internal market-making flow, which can inflate reported volumes, while others may combine liquidity from the different parts of an investment bank's business.

Client facilitation trades done by sales traders may be executed in broker dark pools but are not reportable even though they are essentially done on behalf of a client.

According to Andrew Bowley, head of electronic trading product management at Nomura there is no clear distinction between the flow found in different broker crossing networks, a factor that he believes needs to be addressed in the European Commission's upcoming review of MiFID.

“It would be helpful if there was a clearer delineation between what is considered crossing, typically understood to be a two-order event that usually executes at the mid-price, and internalisation, which is generally thought of as market making or risk facilitation type business,” Bowley told theTRADEnews.com. “At the moment under MiFID, firms are making their own conclusion as to how to categorise their business, while it seems that crossing would best fit the multilateral trading facility (MTF) category, if not OTC, while market making was intended to be captured by a systematic internaliser (SI).” Nomura operates an MTF but is also registered as an SI.

Under MiFID, “firms which, on an organised, frequent and systematic basis, deals on its own account by executing client orders outside a regulated market or multilateral trading facility”, should be classified as systematic internalisers. Broker crossing networks, on the other hand, are not defined under MiFID and brokers assert that they are governed by their best execution obligations.

Different kinds of trading venue have different reporting obligations. For example, while broker crossing networks report to Markit BOAT, automated OTC trading is not split out from manual OTC trading, which makes separating traditional OTC trades from those conducted in broker crossing networks all but impossible. Quarterly stock-specific data on brokers' systematic internalisers however, is publicly available on a broker-by-broker basis on Markit BOAT's website.

This ambiguity between how brokers classify their dark pool and volumes has left buy-side traders continually frustrated and unable to make sense of OTC data to measure execution performance.

“All we want to know is what we can genuinely interact with and until the industry decides to introduce flags for different types of trades in broker dark pools, it's meaningless and spurious,” said Adrian Fitzpatrick, head of dealing at Aegon Asset Management. “Pre-MiFID we had around 27 flags that enabled us to identify what we could interact with. In its current state, we don't have a clue.”

“I simply can not trust any of the figures that are presented to me by brokers and generally tend to ignore the reports they send to me on their crossing networks,” added Sören Steinert, head of trading at Quoniam Asset Management, a quant-based investment management firm. “We desperately need standardised rules for dark pools across the whole industry. Inconsistent data will only damage the trust traders have in dark pools.”

Although cleaner OTC data is one of the most hotly-debated issues in the run up to the MiFID review, Bowley – who is also part of a MiFID working group trying to establish a more effective post-trade transparency regime – does not see a resolution coming any time soon.

“I don't think this will change in the near future, and the focus is now on how these businesses should be structured under the implementation of MiFID II, which is probably about two years off,” he said. “Once the structure of MiFID II is clear however, we expect firms to start changing their models ahead of the MiFID II go live.”

Europe-wide securities regulator the Committee of European Securities Regulator (CESR) is in the process of consolidating the feedback it received from market participants on a series of consultation papers looking at the impact of MiFID. CESR's recommendations will be sent to the European Commission before the end of this month, after which a reform of the directive will begin.