"I think there has been an increase in program trading over the last couple of years," said Eric Noll, director of research at Bala Cynwyd, Pa.-based Susquehanna International Group LLP, an institutional sales, research and market-making firm. "Most of the users of it are buy-side institutions. They have become the biggest source or users of program trading."
Thomas Van Riper, a managing director at Bank of America Securities, New York, agreed that more institutional investors are likely to use program-trading techniques and said he expects the market share of program trading and algorithmic trading of single stocks to reach 60% in the next few years.
"I would say the trend is definitely increasing for program trading," said Mr. Van Riper, who is former president of Vector Partners, a broker-dealer specializing in the creation and execution of portfolio trading strategies that was acquired by Bank of America last January. "In addition to program trading, algorithmic trading is catching on like wildfire across all brokerage firms."
Algorithmic trading uses computer programs to make trades on single stocks within certain limits such as risk levels the client can set.
"We continue to see more business flow that way," Carey S. Pack, president of BNY Brokerage, New York, a unit of Bank of New York, added, referring to program trading.
He said that while increased program trading was a function of liquidity — the more liquidity, the more effective program trading can be — the "big push" toward more program trading was driven mainly by an increase in electronic trading and advances in electronic trading platforms.
"A lot of it (growth) is from the coming of age of technology," Mr. Van Riper added. "A lot of people had been doing it a certain way manually. Slicing and dicing orders was more manual and it was not scalable. Today, the fact that you can leverage technology makes it more scalable."
He said that with existing technology, a buyside trader could immediately identify which trades he or she can best handle through a program, which trades should be handled through an algorithm and which trades he needs to work himself.
"Traders are trying to be more efficient. They're looking at the total day — what orders or groupings can be traded electronically or algorithmically and what groups are best served by more traditional relationship trading," he explained. "He'll say ‘I can package these up as a program, I can send these other orders to a lower commission algorithmic trading (system) and for these 300,000 shares of a stock that trades 100,000 shares a day, I need to get a sales trader involved to get best price."
Mr. Noll at Susquehanna added that volume weighted average price, or VWAP, which is the average price paid per share during a certain time period, has become increasingly important to institutional investors as they scrutinize more closely their stock purchases and sales. Increased use of VWAP has also helped boost the use of program trading.
"There's been big growth in VWAP programs and that wasn't possible a couple years ago because the technology wasn't sophisticated enough," he said. "We've developed technology that allows us to offer VWAP execution on large baskets."
Mr. Van Riper added: "The buyside trader trying to come up with the most efficient, intelligent way to attack his trading day in 2003 means using technology, which frequently results in program trading or algorithmic trading."
But despite improved connectivity through high tech order routing tools, increased transparency and control, program trading is not expected to take the place of the traditional phone call method of trading.
"As much as program trading has grown, the need for the typical buy side sales trader/sell side sales trader relationship isn't going away," Mr. Noll said. "The business is evolving in two very distinct ways — one is the macro program trading and the other is relying on the intelligence of smart people."