Millions of Dollars on Thousandths of Seconds

In the world of high-frequency trading, submitting your trade a thousandth of a second faster than your competitor can be the difference between gaining or losing millions of dollars in the stock market. Trading firms therefore spend fortunes in an arms race to get everfaster communications infrastructure and technology.

“What is it about market rules that makes it a good idea to spend hundreds of millions of dollars on thousandths of seconds?” wondered Eric Budish ’00. And are these market rules best for investors and society?

Eric Burdish
Eric Budish '00; Majors: Economics and Philosophy.

A professor of economics at the University of Chicago Booth School of Business, Budish researches answers to these types of questions. In November the Quarterly Journal of Economics will publish his paper “The High-Frequency Trading Arms Race: Frequent Batch Auctions as a Market Design Response.” In it, he argues that the current rules are economically inefficient and proposes a new system.

To explain the inefficiency, Budish gives an analogy: Imagine you’re selling your home. At the open house, 10 buyers show up at the same time, all willing to pay the asking price. What would you do? Most likely, you’d see if any of the buyers is willing to pay more. But what if, instead, the 10 buyers have to run laps around your house, and whoever wins the race gets the original asking price? “It’s a competition, and buyers might invest in fancy running shoes, but it’s a form of competition that isn’t constructive,” Budish says.

A more efficient system, he says, would have the 10 potential buyers competing in an auction. In the paper, Budish suggests exactly that: he’d replace the current design of the stock market with auctions held as often as once per tenth of a second.

“I’m not a communist,” Budish says. “I like free markets; I like competition. But I’m trying to encourage competition in a more constructive dimension—price instead of speed—which ultimately leads to better outcomes for investors and for society.”

There are three main benefits of frequent batch auctions, he says: They enhance liquidity for investors, they stop a socially wasteful arms race and they simplify the market computationally.

At Amherst, Budish wrote his senior thesis on Internet auctions. Professor Walter Nicholson, the Ward H. Patton Professor of Economics, Emeritus, remembers it as “the best development of an original model I ever encountered in teaching 44 years at Amherst.” Budish’s paper has already won substantial attention, with many in the business world regarding his proposal as a serious and creative alternative to the current market design. For example, SEC Chair Mary Jo White described such auctions as a “competitive solution … to minimize speed advantages.”

Today, Budish continues to present his research and argue for the merits of his idea. “A lot of the discussion about high-frequency trading,” he says, “has been moralistic: are high-frequency traders good or evil? I’m studying the system at a more structural level. Sometimes technocratic details can make a big difference in how well our markets work.”