FEATURED MEDIA
The Behavioral Shift
This year’s student-organized panel discussion, titled “The Behavioral Shift,” invited University of Chicago students and faculty to hear how leading behavioral economists Colin Camerer, Andrew Caplin, and David Laibson place the growing field of behavioral economics within the broader context of more traditional economic analyses. Hosted by the Becker Friedman Institute for Research in Economics and moderated by undergraduate students Alex Foster and Nicole Gorton, the panel painted a view of behavioral economics as a field that works within the formal framework of traditional economics while extending its scope and challenges some of its fundamental assumptions. The panelists suggested that with novel tools and new data sources, researchers can empirically validate models that complicate traditional rational agent assumptions. Such extensions, they stressed, should lead to a nuanced understanding of human behavior more in line with economics’ role as a descriptive social science.
Before presenting on their own research, each panelist presented his perspective on what precisely defines behavioral economics, since economics as a field already focuses on human behavior. Camerer emphasized behavioral economics’ focus on psychological motives to behavior while retaining traditional economics’ strength in formal modeling. Caplin and Laibson both agreed, and clarified that behavioral economics tweaks the assumptions but not the utility-optimizing framework of traditional economic models.
Delving into how this difference is reflected in their own research, each mentioned papers he had worked on recently and explained the research process of a behavioral economist. All three panelists mentioned the radical shift enabled by new data sources like brain scans that give insight into what was previously a “black box” of human decision making. They also all provided examples of how intuitive insights from everyday life that challenge the assumptions of traditional economic theory can provide the basis for new models with greater predictive power, particularly where traditional economics has not been successful.
Then the moderators began the main portion of the discussion, probing the panelists’ perspectives on the key economic concept of rational choice theory. The panelists agreed that it was a very difficult concept to define precisely. Laibson suggested that predictive power should displace adherence to assumptions of rationality as the benchmark by which a model should be measured. However, all of the panelists agreed that even in the context of behavioral work, models that assume rationality are useful.
Later, the panelists were asked what they believed the most robust result in behavioral economics was. Camerer chose to highlight a couple of common psychological phenomenon familiar to economists, “hindsight bias” and “sunk cost fallacy,” that are very robust but “haven’t been studied in modern behavioral economics as much,” suggesting room for further research. Caplin’s answer described what he called “thoughtlessness”—that “every single day we are affected by things that are so casual relative to the things that should be affecting us in the theory.”
Asked whether they would characterize behavioral economics as a toolkit or a subfield, the panelists agreed that it is a toolkit meant to augment classical economics. Caplin touched on themes from the previous year’s talk, “How Big Data is Changing Economies” and highlighted how behavioral economics is utilizing increasingly powerful biological data. For example, Laibson explained how some of his research has involved looking at genomic data to understand educational outcomes and “unravel the biological mechanisms” behind decision-making.
Lastly, the moderators asked whether the panelists felt that behavioral economic methods—often lab experiments—have an external validity problem such that these experiments are not generalizable on a large scale. Camerer, said that there is no such external validity problem, citing studies that have successfully linked lab experiments to personal traits. For example, results from experiments testing risk aversion correlate with individuals’ stock purchasing and gambling habits. Similarly, experiments matching the same subjects in lab and field settings have yielded consistent results.
Caplin and Laibson discussed the external validity problem in economics more generally and the difficulty of replicability for even well-known and acclaimed results. In particular, Laibson cited “p-hacking”, where researchers run many regressions until one significant result is uncovered and subsequently published. He expressed hope that the field will move more towards registering studies with journals and research discussion portals ahead of time so that these troublesome methods become less common.
Following these questions from the moderators, the panelists fielded questions from audience members on the role of behavioral economics for central bankers and results in neurology that are most crucial for behavioral economics.
Overall, the panel helped the audience to achieve a nuanced understanding of behavioral economics as a field, as well as how behavioral economics both challenges and complements traditional economic analyses.
—Daniel Roberts and Nicole Gorton