Collaborative Research: Caution and Reference Effects

Understanding how we make decisions is essential to predict economic outcomes and to design optimal policy. Decades of research documented how people often deviate from rationality and tend to do so following specific behavioral patterns. Three of these patterns have been widely documented and are of particular importance in economics: the endowment effect (where the act of ownership increases the value that people associate to a good), loss aversion (the tendency to avoid losses over achieving equivalent gains), and the certainty effect (a bias towards risk-free options, in violation of Expected Utility theory).

All three phenomena have been shown to play prominent roles in many contexts, including investment, medical, and educational decisions. In this project, the research teams introduce a new approach to study all three phenomena jointly as stemming from one single intuition: it suggests that all three can originate from the fact that individuals are sometimes unsure of what they prefer and act with caution.

This approach makes novel predictions about the relationship of these phenomena and offers a novel tool to organize existing empirical evidence. The team will also conduct also conduct an experiment to test the new predictions of this theory. The results of this project may give us new insights into how to encourage people to save for retirement and protect their health.

The research team wants to demonstrate that all three phenomena above (the endowment effect, loss aversion, and the certainty effect) are captured jointly by the Cautious Utility model. In this model, the individual has a set of utility functions—as if unsure of which one to use—and adopts the most pessimistic one to evaluate each option. The project shows that the Cautious Utility model is characterized axiomatically from imposing a (known) weakening of the independence axiom of Expected Utility that rules out the opposite of the certainty effect.

It then shows that this model implies loss aversion and the endowment effect—ruling out the opposite. This establishes a strong theoretical link between these phenomena. They will also show how this model can be used to organize existing empirical evidence, including observations at odds with leading alternative models.

Finally, they plan an experiment that tests the critical novel predictions of the Cautious Utility model on the empirical relationship between these phenomena.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.