Brokerage firms are racing to create new trading algorithms to hawk to the buy side, but the buyside isn't buying the hype. Funds are overwhelmed by the number of new products on the market, but still don't get enough service to be informed about correct use of the programmed trading strategies, traders said at last week'sWorld Research Groupconference. There are 28 algorithm providers peddling five to seven algorithms each, and buy-side traders say they're hard to compare.
"There is not enough education," said Frank Loughlin, director of global equity trading for Bernstein Investment Management and Research. "As much as 85% of electronic order flow is still executed through the VWAP, the least value-added algorithm," added Harrell Smith, analyst with Celent.
Moreover, buysiders often blame brokerages for an algorithms' poor performance when in fact the trader either misunderstood how the algorithm achieves results or made mistakes in setting benchmarks or selecting strategies. "There are times when we end up canceling out [the algorithmic order] and doing it ourselves," saidBen Sylvester, a trader at $112 billion fund Babson Capital Management. Brokerages need to devote more time to buy-side traders, either via tech support or personal seminars, other buy-side traders added.
Mary McDermott-Holland, senior v.p. with Franklin Portfolio, said the sellside should become more pro-active in offering pre-trade analytics, as part of the algorithmic trading package. Pre-trade analytics make predictions about how a particular algorithm will perform based on past execution data.
"There is a lot of buzz about these tools, but not a lot of talk on how to choose the right metric, incorporate it into your trading strategy, and follow it up," she said. Currently, about 48% of buy-side traders use pre-trade analytics to determine which algorithm to use, but that number is expected to grow to 67% by 2007, according to a recent Tabb Group report.
Despite the drawbacks and confusion, the buy side wants more algorithmic offerings. Use of algorithms is expected to jump to 22% from 15% of all order flow over the next two years, market experts predicted.