Behavioral Implications of Uncertainty in Macroeconomics Capstone Conference

There has been a recent surge in research that explores how expectations and uncertainty impact financial markets and the macroeconomy. This capstone conference, part of the Behavioral Implications of Uncertainty in Macroeconomics Project (BUMP) generously sponsored by the Alfred P. Sloan Foundation under the Macro Finance Research Program of the Becker Friedman Institute, will explore some of the latest research findings on modeling and measuring expectations and their resulting implications for economic dynamics. Researchers explored the role of memory, information acquisition, and uncertainty as contributing factors for forecasting and facilitating decision making in dynamic settings.

Event Summary

On September 22-23, 2022, the Macro Finance Research Program (MFR) hosted the Behavioral Implications of Uncertainty in Macroeconomics Capstone Conference generously supported by the Alfred P. Sloan Foundation. This two-day conference, held at the David Rubenstein Forum at the University of Chicago, marked the culmination of a series of events and conferences exploring some of the latest research findings on modeling and measuring expectations and their resulting implications for economic dynamics.

As part of the larger Behavioral Implications of Uncertainty in Macroeconomics Project (BUMP), the conference collected an intentionally broad set of promising new research, convening both early-career economists and those with years of research related to the various sources and conceptualizations of uncertainty. Specifically, the research explores behavioral implications for the field of macroeconomics.

How people conceive of and respond to uncertainty is a critical behavioral ingredient of dynamic economic models. In many macroeconomic models today, uncertainty has only modest impacts. This is because these models embrace the assumption of rational expectations that says that people know the probabilities implied by the model. The rational expectations assumption is a valuable tool for evaluating many problems but is dubious for analyzing many of the important situations we face today. Concerns about long-term growth including the potential for secular stagnation, climate change, limited technological progress, and altered demographic composition are on many peoples’ minds.

Therefore, the BUMP project was launched in 2018 to push beyond the usual rational expectations approach in macroeconomics by attributing uncertainty to the probabilities that economic agents are facing. The project was launched by Professor Thomas J. Sargent, W.R. Berkley Professor of Economics and Business at New York University and Lars Peter Hansen, David Rockefeller Distinguished Service Professor in Economics, Statistics, and the Booth School of Business and director of the Macro Finance Research Program. Together, the two long-time collaborators launched the BUMP project with their joint research on the topic of uncertainty, broadly conceived, providing the foundation for the intellectual ambitions of this four-year project.

Since its launch, the BUMP project has pushed beyond the conventional risk-based analyses by probing more general paradigms coming from decision theory and mathematical control theory. Furthermore, it complements and extends other behavioral research that emphasizes psychological mechanisms. It has accomplished this by using statistical theory to formalize how environmental complexities of the model framework can influence individual behaviors. The intellectual advances not only contributed to macroeconomics, but they also have positive spillovers to other areas of research.

“There’s been a real push toward specialization within economics, which is productive, but it creates at least some intellectual isolation and separation,” Hansen said. “That’s why we like to do things like this to get people talking. It’s fascinating, and we think it’ll have positive effects down the road. Events like this encourage conversations between people that will have influence on their research for years to come,” he said.

The presentations themselves spanned from game theory to macroeconomics, microeconomics, fiscal studies, economic history, climate economics, finance and entrepreneurship, and more. Tim Christensen, Professor at both New York University and University College London and an active participant in BUMP workshops throughout the course of the project, presented new work proposing methods for learning externally valid policy rules using experimental or observational data.

“Borrowing insights from the robustness agenda of Lars and Tom [Sargent] has been quite productive in terms of framing how to handle aspects of the model that econometricians are not confident about—and thinking about how to develop methods to quantify the effects of policies that are robust to specifications of those pieces,” Christensen said.

Christensen was among several early-career scholars to participate in the capstone conference and in other BUMP events, such as the Blue-Collar Working Group and the 2019 Conference on Robustness in Economics and Econometrics. He appreciated the in-person capstone event with its social breaks that allowed for additional discussions and connections.

“I was initially a consumer of Lars and Tom’s work,” he said. “I’m an econometrician, but there are commonalities between how they set up problems and how you’d like to frame optimality criteria in other settings as well, so I found their work inspirational and thought-provoking. That led me to work on a bunch of other things – somewhat unrelated to their work but still inspired by it.”

Christensen’s presentation, in many ways, is representative of the larger theme that emerged from the conference of connecting seemingly dissimilar research areas to create more rigorous models that are both honest and ambitious in their approach to uncertainty. Several presentations combined economic theory and game theory with empirical evidence in exciting new ways. Others provided a deeper understanding of the frontiers of such theory.

Another active participant in the BUMP project, Jaroslav Borovička, Associate Professor at New York University, noted how productive BUMP and the capstone conference have been for his own work.

“This is all great long-term accumulation of insights,” said Borovička. “The presentations give you ideas, and now you need to go back to the papers, study them carefully, talk to people about their research to see how to productively push things forward and build on those.”

His co-author Anmol Bhandari, Associate Professor at the University of Minnesota, presented their recent work on robust bounds on optimal tax progressivity, addressing key sources of uncertainty in tax policy, such as the correlation of productivity and labor supply elasticity. Without incorporating uncertainty into models used to make such policy decisions—including as Hansen noted after the presentation, the sources of uncertainty around individual productivity—certain assumptions may work against policy goals.

“In this profession, when you’re early in your career, there’s not much reward for thinking broadly, so as an additional goal of the conference, we want to help provide that encouragement,” Hansen said.

Hansen and Sargent’s own joint work, as presented by Hansen to close the first day of the capstone conference, extends notions of uncertainty beyond risk in ways that make contact with applied econometric challenges by distinguishing concerns about potential misspecifications of likelihoods from concerns about robustness of alternative priors.

“What Lars has done very well his whole career is build statistical models that are about the behavior of people making decisions in the face of uncertainty. Everyone at this conference —and there was a wide range of people with different perspectives— they’re doing things that Lars and I think are pushing the frontier of economics,” Sargent said.

In all, the capstone conference—and the BUMP project more broadly—provided the rare opportunity for scholars across research interests and career positions to exchange ideas and network, which will no doubt enrich thinking in this field of research for years to come. The incorporation of various types of uncertainty into dynamic models remains an exciting challenge for the field, and as the capstone conference made clear, will require such exchange across traditional research lines and vantage points.

“This conference reminds me of where I was 40 or 50 years ago at the University of Minnesota,” Sargent noted in his closing remarks. “It was a small department with a mixture of people in different fields, and we didn’t have macro seminars or micro seminars. We had one seminar for everyone, and we’d be learning from people from different fields. That’s the spirit of this conference. I enjoyed every single paper, and it makes me optimistic about the future.”

Hansen and Sargent both see these types of connections as having vital implications for formulating sensible economic policies in the future.