LX framework targets toxic trading

Barclays has introduced a 'toxicity framework' to its LX dark pool that it claims makes predatory trading easier to identify.

By None

Barclays has introduced a 'toxicity framework' to its LX dark pool that it claims makes predatory trading easier to identify.

Instead of segmenting profiles according to trader type - institutional investors and broker-dealers, for example - LX tracks activity according to a trader's aggressive, neutral or passive liquidity profile.

"We don't have traders coming in with a million orders to trade five. If we did, we'd tell them to change their behavior to enhance their interactions," said Barclays head of electronic trading Bill White. "For high impact flows you can work with the broker to balance their participation levels."

White said the framework marked a shift away from screening out specific firms - an approach he said made little sense because their trades were in any case likely to be transmitted via a broker.  

A recent report by Barclays execution consultants Jeffrey Alexander and Linda Giordano suggested buy-side traders had underestimated execution costs generated by venue "nuances" that could make it easier for predators to find and step ahead of institutional orders.

The report urged the adoption of a framework, developed with 20 buy-side clients, that compares venues using aggregate broker data on short-term alpha, sub-routes and post-fill market volumes. It described current buy-side attempts to measure best execution according to monthly broker reports as "akin to measuring the speed of light with a sundial".

Recent data from Rosenblatt Securities puts Barclays LX volumes at 90.8 million shares behind only those of Credit Suisse's Crossfinder with 120.9 million. Matthew Samelson, principal at firm Woodbine Associates, a research firm that recently published a comparison of 25 dark pool venues, attributed LX's performance to the fact that it offers access to hedge funds and high frequency traders excluded from restricted venues.

He contrasted it with networks such as Liquidnet, which limit activity to specific types of investor.

"There's no good and no bad model," said Samelson. "But those that wrest greater liquidity will have greater volumes." 

«