Doctoral Dissertation Research in Economics: Reference-Dependent Preferences, Expectations, and Dynamic Choice

Harvard University, Cambridge MA

Investigators

Abstract

This award funds laboratory experiments testing the effect of expectations on reference points and intertemporal choice. Reference dependent utility models (for example, prospect theory) have been widely applied as alternatives to the expected utility models widely used by economists. However, there is little evidence on how the reference point is actually determined. Previous researchers have often assumed that the reference point is the status quo, but recently other economists and decision scientists have argued that reference points are determined by expectations of future outcomes. The coPI and a collaborator have recently developed empirical evidence that expectations do indeed determine the reference point in a static decision making context. This new project considers how expectations affect choice in a dynamic intertemporal context. The research provides new evidence on how individuals make intertemporal choices while also distinguishing between competing theories of reference-dependent preferences. The study of intertemporal choice has been a fertile field of research, as evidence has mounted that suggests that individuals have time inconsistent preferences: instead of discounting utility at a constant rate, many individuals appear to have a discount rate that declines with time. Such preferences can lead to internal conflict, over-consumption in the present, and a desire to use commitment devices to bind future behavior. At the same time, some theories of reference dependent preferences also predict that individuals will over-consume in the present because they desire to give themselves favorable surprises. The new research funded by this award manipulates expectations at various points in time to determine whether intertemporal choices are affected as predicted by reference-dependent preference theories. By providing evidence on how reference points are determined over time, this project enables researchers to make better use of reference dependent models in analyzing economically important behaviors such as labor supply, savings decisions, and whether or not to purchase insurance. The broader impacts of this research include the design of more effective interventions (by policy makers and by individuals) aimed at encouraging individual behavior change and self control.