Nasdaq Pools Daily Trade Data With Electronic Auction System

A year and a half ago, Nasdaq Stock Market Inc. was facing increased competitive pressure from centralized trading platforms, which offer indexing firms a closing stock price at the end of the trading day that is representative of the Buy and Sell orders that have taken place.

Steve Randich, CIO
1pixclear.gif
Steve Randich, CIO
1pixclear.gif

Starting with a decentralized market created by hundreds of broker-dealers, Nasdaq's CIO, Steve Randich, came up with an open electronic stock-auction process, called Closing Cross. The system collects the Buy and Sell interests from 3,200 stocks in the last 10 minutes of the trading day, pools it and uses the data to establish a single fair closing price that has been vetted with Nasdaq buyers and sellers.

"You can equate this to the race ending at 4 p.m., and you have one last chance to get your orders in. It has the potential to establish a new price level after the close," says Robert Haggerty, an analyst at Needham, Mass.-based TowerGroup.

Prior to Closing Cross, the last trade that crossed the ticker tape at the end of the day set the price, which didn't reflect the true supply and demand of any particular stock, Randich says.

"This is something we struggled with—with the index firms, as well as the companies that list on Nasdaq, like Microsoft and Intel, because they, too, watch their closing price," Randich says.

Developed in-house, the Closing Cross application runs on 11 Hewlett-Packard Co. NonStop S88000 servers, each of which has 16 processors. The application uses a complex set of algorithms to continuously recalculate an indicative closing price, and then it broadcasts that closing price to the market indexes and listed companies.

At the same time, Closing Cross rationalizes the closing prices with price updates occurring in the stock, independently of the closing costs; calculates a price; executes all the trades at that price; and then disseminates the information again. The entire process takes less than four seconds.

"It permits large investors to effect an end-of-day price without having to worry about timing their orders. It also relieves some of the logjam the market sees in the last half hour leading up the the close of the exchange, because they know they can always get an order done during the Closing Cross," Haggerty says.

Closing Cross orders must be submitted prior to 3:50 p.m. Eastern time. At exactly 4 p.m., Closing Cross executes in an instant and produces a single closing price for each stock that maximizes the amount of shares executed, Randich says.

Closing Cross went live in April 2004, two months ahead of schedule, having been developed on an aggressive timetable that included actively trading test stocks with listed companies during normal trading hours, Randich says.

"There are no dollars involved. Everyone is using [the test stocks] in their production systems, but it's a dummy symbol," Randich says. "By encouraging our market participants to use these, we could rush software to production and debug it there as opposed to trying to internally make it perfect before we release it."

Closing Cross has spurred the development of two similar platforms using the same production method: one for the opening of the market, which went live this past fall, and another for the midday opening, which includes initial public offerings. That platform is expected to go live by the end of this year, Randich says.

Nasdaq Stock Market Inc.

www.nasdaq.com

red_bullet.gif
Business: This all-electronic exchange has 3,200 listed companies and had revenue of $590 million last year.

red_bullet.gif
Project champion: Steve Randich

red_bullet.gif
IT department: 500

red_bullet.gif
Project payback: Competitive edge, plus increased speed for round-trip time for transaction processing. Three years ago, a transaction took three quarters of a second to process. Today, transaction processing time is measured in milliseconds, representing a seventy-five-fold improvement.

Special Report

Best In Class IT Leaders 2005

Stories in this report:

Related:

Copyright © 2005 IDG Communications, Inc.

Bing’s AI chatbot came to work for me. I had to fire it.
Shop Tech Products at Amazon