Model Behavior: How a Financial Theorist’s Insights Help Explain Activity in Markets
Professor shares research insights in interactive talk with students during memorial lecture series.
Brian Waters explains his research interests and approach during the most recent Steven Lindstrom Pardo Memorial Finance Lecture Series talk on April 4. The series was created by the family of Pardo, an engaging and intellectually curious student who died of cancer shortly after graduating from Leeds in 2020. Below, Waters leads two students in a game to demonstrate strategy and speaks to a crowded lecture hall.
Trial by fire? Brian Waters knows all about it.
Upon finishing his economics degree from Vanderbilt University, Waters didn’t get the graduate school offers he wanted, so sought out additional experience to round out his résumé. A tip from a professor led him to the White House, where at 22 he found himself alongside some of the brightest minds in finance and economics as they confronted the worst recession since World War II.
“I went from being a student to getting thrown right into the fire—in a way that I probably should not have,” Waters, an assistant professor of finance at the ’s Leeds School of Business, said with a laugh.
Waters’ experience in the Council of Economic Advisors—an in-house think tank to help the president understand pressing economic issues—involved performing extensive financial analysis to help guide the nation’s response to the financial crisis. It also gave him focus for his academic career, which shifted to financial theory and examines contracting and the role of private information in markets.
“I try to look for those things in the market that don’t have obvious explanations, and then I write a mathematical model that explains how market participants behave in the real world,” he said.
Playing game theory
His research involves some complex mathematics, but in a recent research talk with Leeds undergrads, Waters invited the audience to participate in a couple of games to better understand strategic interactions and decision-making, as well as how insider knowledge can unravel markets.
The lecture was the latest in the annual Steven Lindstrom Pardo Memorial Finance Lecture Series, which challenges faculty to make their research presentable and engaging to undergraduate business students. It’s named for Pardo (Bus’20), an intellectually curious student and academic standout who died of cancer shortly after graduating. His family created the research series in the hopes of honoring Pardo’s curiosity and inspiring his love of learning in future generations of students.
Watching a room full of students compete in strategy games validated how approachable Waters’ work is, though he did get laughs and a few gasps when he showed students the actual mathematical model developed to validate his work on how private information can influence things like the sale of a used car—especially the value of learning within that framework.
“When I’m selling my car, I have better information about it than the buyer does,” Waters said. “But the value of that information really comes from how long I owned the car,” which both helps the owner determine a price while acting as a signal for a buyer.
Consider home sales. If two identical homes hit the market at the same time, they should be priced roughly the same. But say one’s been lived in two decades, as opposed to two years. The owners who’ve lived there 20 years, Waters said, “have really good information about the house. If it were a bad house, they wouldn’t have stayed so many years, and so the fact that they stayed indicates the value of the house must be good.”
“The best part about being a theorist is writing something down and saying, ‘Yes, that obviously applies to my behavior.’”
Professor Brian Waters
His research demonstrated that home prices—when controlled for price appreciation—follow a U-shape that’s highest when a home sells immediately or after the owners have lived there a long time, and is lowest when it turns over quickly.
Pricing points
“If you see a house for sale after two years, you’re concerned that there’s something they learned in that time that’s driving them to sell,” like flooding or bad neighbors, Waters said. That will drive down the price, “especially if you see the home comes on the market every couple of years, in which case you can probably make a lowball offer for it.”
Being able to model those agent behaviors has allowed Waters to make some interesting finds, which have been published in respected journals such as the Journal of Finance and Management Science. He has also done notable work in what he calls optimal contracting—for instance, determining the conditions under which social investors can catalyze companies to invest in greener practices, even at a loss of profit.
Just because that work is theoretical doesn’t mean it isn’t applicable. Waters drew laughs when he shared his own experience of selling his first home in Boulder after living there for two years.
“The best part about being a theorist is writing something down and saying, ‘Yes, that obviously applies to my behavior,’” he said.
Steal Away
To better explain backward induction and unraveling, Brian Waters took a page out of Major League Baseball’s opening day. A rule change now limits pitchers to two attempts to pick off a batter attempting to steal a base; a third unsuccessful attempt by the pitcher results in a balk, which means all baserunners advance to the next base.
So runners now know pitchers are unlikely to attempt a third pickoff, meaning they’re likely to be more aggressive about steal attempts after two failed pickoffs. Knowing this, though, a pitcher is unlikely to attempt a second pickoff, because it will trigger more aggressive stealing from a runner. But: That signals to runners to be aggressive after a first pickoff attempt, since the pitcher is unlikely to throw a second pickoff.
Back and forth the strategy goes until you realize runners will be more aggressive from the start, knowing the pitcher’s attempts to pick them off are limited. Waters said you could expect a much higher probability of attempted and successful stolen bases. On opening day this year, runners succeeded 21 of 23 times they attempted to steal a base. Last year, five bases were stole on on nine attempts.