If you’re advising people to make better decisions, you’d better give them appealing alternatives to choose from. That’s one takeaway from PhD candidate Alex Ballyk’s behavioural economics research. The Department of Economics’ student explores decision making through experiments at the Toronto Experimental Economics Laboratory (TEEL). In the experiment at the heart of her working paper Paternalistic Persuasion, Ballyk designated participants as experts (called “Advisors”) to try to help a second group of participants (called “Choosers”) to make better decisions by giving them advice. However, persuading Choosers was challenging: Choosers weren’t sure whether all recommendations came from those well-meaning Advisors. Therefore, Advisors needed to send recommendations that Choosers could trust to be in their best interest.
“I study whether, in an environment where Choosers are reluctant to accept certain recommendations, Advisors can make Choosers better off by persuading them to change their behaviour,” Ballyk explained. “Think, for instance, of a financial advisor trying to help their client save more for a large purchase. The client could be reluctant to save more for a number of reasons: perhaps they would need to break certain spending habits, or are wary that the financial advisor’s recommendation is influenced by commission incentives. In order to persuade the client to save more, the advisor should account for the root of the client’s reluctance. For example, a client who isn’t used to saving a lot of their money may be more willing to make small yet frequent contributions to their savings account, rather than large and infrequent contributions.”
Ballyk approached her question in two steps. First, she wrote an economic model of a “recommendation game” describing a situation like the financial advice example. In the model, Choosers’ reluctance stemmed from wariness about Advisors’ incentives: they weren’t sure whether some advice was from the well-meaning Advisor, or someone whose interests weren’t always aligned with the Chooser’s.
“The main prediction of the model is that the Advisor should take the Chooser’s wariness into account. If the Chooser rejects certain pieces of advice due to this wariness, the Advisor should give advice that’s similar, but that the Chooser is actually willing to take,” Ballyk said. Armed with these predictions about the recommendations Advisors should send, she then ran lab experiments to test whether individuals acting as Advisors actually send those recommendations.
“The main result of my first experiment is that many Advisors send sub-optimal recommendations. Instead of recommending actions that account for Choosers’ wariness, most Advisors recommend actions that would only be appealing to Choosers who don’t fear any conflict of interest. Even after playing the game multiple times, and getting feedback about the recommendations Choosers are willing to accept, many Advisors still don’t send persuasive recommendations.”
Ballyk conjectured that Advisors may not have realized that they should try to make their recommendations appealing to Choosers. To test this hypothesis, she ran a second experiment where she prompted Advisors to think about how Choosers might react to each possible recommendation. “I found that in this version of the experiment, Advisors tend to send more persuasive recommendations. This suggests that, when prompted, they realize that they should account for how Choosers may respond to their recommendations,” Ballyk said.
Beyond financial advice, there many other contexts in which the insights from Ballyk’s research can be applied. Consider, for instance, health-related behaviours such as exercise, eating and alcohol consumption habits. “For people who haven’t developed healthy habits, recommendations that capture gradual behavioural changes – initially exercising a bit more or drinking a little less, for instance – may be most appealing. Health policymakers may wish to take this into account.”
Ballyk credits the teaching practice she has developed as a course instructor with helping her to refine her approach to this research.
“Teaching game theory definitely influenced my approach to my job market paper. It gave me the insight that the situation I was interested in could be modeled using a game-theoretic model,” she said. Ballyk also believes that her teaching experience has helped her refine her research communication skills. “Teaching involves sharing information you’re already familiar with, with others who are less familiar with it. A fact that might seem obvious to you as an instructor might not be very obvious to students. I have strived to present concepts in a very intuitive way while teaching, and I’ve taken that approach into my presentations as well.”
Learning to speak to generalist audiences has also been important for Ballyk to meet the needs of various bodies who have helped to fund her research, including the Institute for Gender and the Economy (GATE). GATE helped fund a second project she worked on with her colleague Annabel Thornton. Together, they studied how gender differences in job search behaviour may make women less likely to apply to higher-paying yet more competitive jobs.
“I’ve really enjoyed having access to so many different communities at U of T,” Ballyk said. “GATE brings together researchers from many different fields. Annabel and I received feedback from sociologists, psychologists, and people who study organizational behavior. It’s really rewarding to interact with so many people who are passionate about what they do and to get their perspectives.”
The grants Ballyk has received have allowed her to pursue her own research, and also to support two of her fellow PhD students interested in behavioural economics. These students provided invaluable research assistance while Ballyk ran the experiments that formed the basis of her job market paper.
“They were at an earlier stage of their PhD, so I thought it might be helpful for them to gain experience as experimental researchers before running their own experiments,” she explained.
In addition to her research and teaching responsibilities, Ballyk has also taken a leadership role in co-organizing roundtable discussions for graduate students for the Women in Economics, UofT group.
Taken together, Ballyk’s research, teaching and mentoring have left an impression on faculty and the Department of Economics as a whole.
“Alex is an impressive scholar whose research is pushing the frontier of behavioral economics as she documents using experimental methods how people’s choices and actions in a variety of environments are affected by what they believe are others’ preferences and beliefs,” said Professor Yoram Halevy, Ballyk’s thesis supervisor. “Alex is a very independent researcher, an excellent experienced teacher who is self-motivated to conduct innovative research.”
Return to the Department of Economics website.
Scroll more news.