Markets under the microscope

Market bubbles. Stock prices. Transaction costs. Day-traders.

Jeffrey H. Harris delves into some of today's hottest stock market topics.

Harris, who joined the Lerner College of Business and Economics last year as an assistant professor of finance, is a nationally known expert in the new field of market microstructure. Market microstructure is the examination of how traders interact, what costs they face, how they behave when rules change and how the market is affected by the presence of informed traders, big institutions and other players.

The research takes a magnifying glass to market forces, examining the interactions of buyers and sellers--rather than the overall economy or an individual corporation's balance sheets--to explain market prices and transaction costs. The work involves computer analysis of billions of price quotes and transactions every day.

If Harris' research seems esoteric and extraordinarily detailed, its results are of great interest to investors. "Most of my work is in transaction costs and how investors are impacted by those," he explains.

Harris started out with a research project that had explosive results. He was a member of a team in 1994 that uncovered irregularities in the prices dealers were quoting for major NASDAQ stocks, including Intel, Novell, MCI, Cisco, Microsoft, Apple and Oracle.

"We basically got some data that suggested they weren't being entirely competitive," Harris says.

His role had been to download and write code to analyze millions upon millions of daily stock transactions. The data revealed that NASDAQ dealers, called market makers, were consistently maintaining spreads of 25 cents, rather than the minimal 1/8 (12.5-cent) increments, on NASDAQ's highest-profile and most actively traded stocks.

The study jolted market prices as soon as it became public.

The U.S. Justice Department launched an investigation into alleged dealer collusion. Investors filed multiple class-action suits against big-name investment firms. The suits ultimately were settled for more than $1 billion, and the Justice Department filed an antitrust action that also eventually settled. NASDAQ was forced to restructure.

This was heady stuff for Harris, who was still finishing his doctorate at Ohio State University.

Next, Harris and one of his collaborators began looking into a new group of market players known as SOES (Small Order Execution System) bandits, who eventually became commonly known as day-traders. NASDAQ had established SOES, permitting automatic execution of orders involving fewer than 1,000 shares.

"It basically opened the door for day-trading," Harris says. "For the first time, instead of having to call a market maker to execute your order, you could put your order into the NASDAQ Small Order Execution System, and you could become a day-trader."

Harris, who by then was on the faculty of the University of Notre Dame, and his co-author concluded what many day-traders learned from bitter experience. "One of the things we found there was that it was very difficult to make money," he says.

Harris' research covers new territory resulting from the rise in NASDAQ's importance in the last two decades and the ability of newly powerful computers--the very industry that propelled NASDAQ--to analyze moment-by-moment market transactions. Previously, Harris says, researchers only had end-of-the-day data to study.

"There were little things in the market that nobody did understand," he says. Then, markets started producing CD-ROMs recording every quote. The amount of information soared as markets began permitting quotes in 1/16ths and eventually in pennies. Today, the data has moved from CDs to the more capacious DVD-ROMs.

"NASDAQ had grown into a huge market in the 1990s," Harris says. With the phenomenal rise of such computer titans as Intel, Apple and Microsoft, which hadn't qualified for the New York Stock Exchange, NASDAQ was transformed from a venue for small-capital companies into a major player.

Harris honed his knowledge of NASDAQ as a Visiting Academic Fellow there from 2000-01, when he joined the UD faculty. Among his recent papers is "NASDAQ Trading Halts: The Impact of Market Mechanisms on Prices, Trading Activity and Execution Costs." NASDAQ halts trading on a stock when news is pending about the company, which occurs four to five times on a given day, Harris and his collaborators found.

"Jeff Harris' work has helped us and investors understand the markets better, particularly the NASDAQ stock market," says Mike Edleson, senior vice president and chief economist at NASDAQ. "He's a recognized expert on the workings of markets and how investors and companies benefit from them."

Harris often shuttles between his office in Purnell Hall and the Washington, D.C., area, where the SEC and NASDAQ are located and where his wife, Kirsten Anderson, BE '89, is a professor of financial accounting at Georgetown University.

One thing that Harris doesn't spend time on is analyzing his own stock holdings. "I trust the professionals to manage my portfolio," he says. "The truth of the matter is that the more you know, the less you know."

He notes that individual investors face some of the same hurdles as day-traders, including having to pay broker fees and trying to get ahead of specialists with supercomputers who scour world markets on a minute-by-minute basis.

There's something to be said for the long run, Harris says, because today's top 10 stocks won't be in the top 10 a few years from now. "Price today is always expectation about the future," he says. "If I knew exactly what was going to happen, I wouldn't be teaching."

--Sandy Dennison James