Algos From Merrill, Others Target Smaller-Cap Stocks
February 26, 2007
Last year saw the launch of several algorithms designed to stealthily hunt liquidity via multiple crossing networks and alternative trading systems (ATSs) that traders can utilize for automated small- and mid-cap trades. These included offerings from Citigroup, Instinet and JP Morgan Chase & Co. Analysts, however, are not surprised to see more along the same lines.
"There continues to be a huge demand for a truly effective small-cap algo," says David Mortimer, a principal with consulting firm Vodia Group of Concord, Mass. "No one has the small-cap algorithm solution completely down and it remains an open challenge."
The effort has
been driven, in part, by the fragmentation of liquidity among a
growing number of ATSs and the tendency of human brokers to focus
on large-cap trades that generate more lucrative fees, says Nitin
Gambhir, CEO of New York-based research firm and technology
provider Tethys. Brian Schwieger, London-based head of Merrill's
algorithmic execution unit for Europe, the Middle East and Africa,
agrees. Since the fourth quarter last year, his division has been
working on a new algorithm specifically for small- and mid-cap
trades.
"Our current algorithms, like some of our competitors', react to trade frequency, liquidity and spread," Schwieger says, adding that such a strategy's effectiveness diminishes down the market-cap curve. "Once you start getting into the really tricky, low-cap stocks, it's not just a question of adapting the trading style of the algorithm. A totally different trading style is required--just as it is with a human trader."
International Lessons
To achieve this goal, Schwieger says his development team will be incorporating lessons learned from Asian markets, including some of the infrastructure choices and order-book handling logic employed where traders have encountered much longer queuing times. "The U.S. is probably the most liquid of markets, but when it comes to equity trades, Asia is at the other extreme," notes Schwieger. At the same time, he says that the new, yet-to-be-named algorithm will leverage some of the probing characteristics of the firm's OPLX, or Optimal Price Liquidation Extreme, implementation shortfall strategy.
Merrill's small-cap project involves a development team of three in London, with input from developers in Asia and New York. Schwieger says he is also working directly with traders to ensure that their concerns are addressed. He also notes that while the focus is on small- and mid-caps in his overseas regions, his counterpart in the U.S. is working on a similar strategy there. He says that the new small-cap algo should be on internal trading desks in the latter half of March and available to clients by midyear.
Jefferies, a New York-based investment bank known for its small- and mid-cap expertise, has taken a different approach. Ross Stevens, co-head of equity products, says the firm has spent about a year working with a development team of 20 to upgrade its core suite of about eight algorithms to fully incorporate Jefferies' latest internal research and strategies appropriate for small-cap trades.
