Nasdaq Duels NYSE for Best Execution

In a war of numbers, the Nasdaq Stock Market and the New York Stock Exchange are at it again, battling over which type of market gives the investor best execution.

Nasdaq claims that large capitalization stocks listed in the Standard & Poor's 500 index are, on average, executed in 2.5 seconds, or 9.7 times faster than the 24.3 seconds it takes to execute comparable stocks on the NYSE.

The analysis was provided in October 2002 by Market Systems Inc. (MSI), an independent, third-party data-analysis firm that calculated a year's worth of public data (for orders up to 10,000 shares) that the market centers are required to report under the Securities and Exchange Commission Rule 11Ac1-5.

Nasdaq's Senior Vice President and Chief Economist Mike Edleson attributes the results to Nasdaq's distributed, open-architecture platform. Nasdaq is faster, contends Edleson, because, instead of routing orders to a central market, market makers and electronic-communication networks execute the orders without human intervention. "We've streamlined the auto-execution piece of it, to take the human-decision element out of it."

"It's pretty easy to send an order over wires," says Edleson, who contends that the NYSE has good delivery technology. "The question is what do you do with it when you get it, and so there's still a lot of human intervention on the New York side," says Edleson.

In response, Paul Bennett, the NYSE's chief economist, doesn't dispute the accuracy of the numbers, per se. "MSI is a very credible organization," he says. However, he adds that Nasdaq has taken "averages across a large number of very different (execution venues)."

"Almost anybody involved in the business understands that trading floors have a lot of advantages, although they are slower." On the other hand, the NYSE has automated systems that will execute in one to two seconds, he says, citing Direct Plus, a system that clocks in at 1.5 seconds. "If what you want is speed, and you don't care much about the price, then we have a system that is as fast as the Nasdaq," says Bennett.

But the MSI study contends that trading costs are 55.9 percent lower on Nasdaq for S&P 500-listed stock, where the average-effective spread is 1.6 cents on Nasdaq versus 3.7 cents for NYSE-listed S&P 500 companies. (An effective spread measures the execution price and then compares it with the midpoint of the bid/asked spread at the time the order arrives.)

However, Bennett contends that because Nasdaq has chosen to volume weight the spreads, it gives emphasis to a small number of stocks (such as Microsoft) that have very large inter-dealer markets and are actively traded on ECNs.

Nasdaq also has fewer stocks - 75 versus the NYSE's 421 - listed in the S&P 500. These same stocks, if moved to the NYSE, would have much lower turnover, he asserts. Another reason why effective spreads are lower is that, "The Nasdaq market has fallen in price - from a high of 5,000 to 1388 in February. "It's fallen below 40 percent of its value before," he says. A lower price stock, say a $10 stock as compared to $50, has a smaller bid/ask spread, he adds.

"If you just equal weight them, then you completely reverse their results, it tells me they're just spinning something," says Bennett.

On the other hand, the whole subject of best execution is open to debate. "There isn't any agreement on the definition of best execution," says Sang Lee, manager of the Securities and Investments Group at Boston-based Celent Communications, who asks: "What is best execution? Is it the speed of execution or price of execution? There really isn't a single industry-wide definition of it. That gives people the loophole to back up all kinds of statistics." Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio