Still greater light must be shed on dark volumes, say buy-side

European buy-side traders have welcomed recent steps by brokers to provide more transparency on dark pool trading activity, but more accurate real-time post-trade data remains a distant aspiration.
By None

European buy-side traders have welcomed recent steps by brokers to provide more transparency on dark pool trading activity, but more accurate real-time post-trade data remains a distant aspiration.

Last week, Credit Suisse said it would start publishing historical activity executed in Crossfinder, the investment bank’s systematic internaliser (SI) or internal crossing engine. According to Credit Suisse, €10 billion was executed in Crossfinder in Q4 2009, making it larger than Chi-X Europe’s Chi-Delta, the biggest non-displayed multilateral trading facility (MTF), which traded €8.1 billion in the same quarter, according to Thomson Reuters’ dark trading market share report.

“We wanted to provide greater clarity to the market on how much is executed in Crossfinder on a monthly basis,” George Andreadis, head of liquidity strategy for Credit Suisse’s Advanced Execution Services division in Europe, told “We hope other brokers will follow suit.”

Credit Suisse’s Crossfinder reports are single counted and only include liquidity that clients could have interacted with. Therefore, it will not include large trades the bank conducts with itself using Crossfinder or executions resulting in routing orders to other dark pools.

Investment banks have shown more of a willingness to report activity in their internal crossing engines in recent months. In Europe, Nomura’s NX dark pool started reporting its daily activity this month following reclassification as an MTF.

In the US, Goldman Sachs has reported all volumes traded in its SIGMA X dark pool on a single-counted basis since June 2009 in response to concerns expressed by the Securities and Exchanges Commission over the transparency of dark pool trade reporting in the US equities market.

In Q4 2009, the Committee of European Securities Regulators (CESR) asked Europe’s national regulators to collate data on broker dark pools within their jurisdictions. A source that participated in research conducted by UK regulator the Financial Services Authority estimated that the final total for dark trading across Europe is unlikely to exceed 4%.

CESR’s investigation into dark pool volumes and reporting was launched following a longstanding row between European exchanges and brokers over the level and significance of non-displayed OTC trading in Europe.

The Federation of European Securities Exchanges (FESE), a trade body representing Europe’s exchanges, estimated in December that 38% of European equity flow was executed over the counter, but acknowledged the difficulties of separating out broker dark pool volumes. FESE argues that brokers’ internal crossing engines perform the same function as regulated exchanges and MTFs and should be governed in a similar way. Although MiFID requires trading venues to be registered as exchanges, MTFs or SIs, some brokers’ internal matching engines are registered as neither. Furthermore, brokers say their dark pools are already covered by their fiduciary duty to deliver best execution for clients using the tools available to them.

Some brokers argue that monthly or quarterly data is of little use to clients.

“The real issue that needs to be solved is the wider debate about the granularity of the post-trade regime in Europe,” said Brad Hunt, managing director, equities, Goldman Sachs. “Until those issues are resolved it is difficult for market participants to ascribe meaningful context to any activity numbers.”

While noting that historical broker dark pool information is ‘nice to have’, Sören Steinert, head of trading at Quoniam Asset Management, a Frankfurt-based buy-side firm with €12 billion assets under management, says that what the buy-side really craves is more accurate real-time post-trade data.

“We would use monthly or daily figures to track the long-term growth of a broker dark pool, but this doesn’t help me for the trades I am doing today,” said Steinert.

Steinert also notes that restricting the reporting of trades to a broker’s own dark pool could underestimate the true amount of flow available to investors if dark pool linkages are discounted. Credit Suisse’s Crossfinder Plus service is connected to Instinet’s Blockmatch dark MTF, while investment banks UBS, Morgan Stanley and Goldman Sachs signed a deal in May 2009 to allow each other’s algorithms access to their respective crossing engines.

Brokers already post all their non-exchange equities trades to reporting venues such as the OTC-specific BOAT. But trades executed on an automated basis in dark pools or systematic internalisers are not identified separately from trades handled manually by the brokers’ ‘upstairs’ trading desks. This means venues such as BOAT cannot help buy-side firms to compare brokers’ dark pool volumes. Both the buy-side and the sell-side are reluctant to provide too much transparency as real-time information on volumes passing through a particular bank’s dark pool could potentially adversely impact quality of execution for clients.

One way to solve this would be to split out automated OTC business from manual OTC business in real-time data but only reveal specific venue information at the end of the day.

Various sell-side firms have said they are working with each other and regulators to come up with a solution, but some buy-side traders remain unconvinced.

“Historical data will undeniably have an influence on buy-side dealers that may want to tap previously untried liquidity sources, but that’s still a fair stretch from brokers adopting a more granular post-trade reporting regime for their dark pools,” said one senior London-based buy-side trader. “I don’t see brokers voluntary stepping forward to provide real-time data so it will be down to pressure from their customers and regulators to do so.”