Abstract
This article explores some conceptual issues in the study of well-being using the traditional
economic approach of inferring preferences solely from choice behavior. We argue that choice
behavior alone can never reveal which situations make people better off, even with unlimited
data and under the maintained hypothesis of 100% rational choice. Ancillary assumptions or
additional forms of data such as happiness measures are always needed. With such ancillary
assumptions and additional data, however, the use of revealed preference to study well-being
can be significantly improved, so that the choices people make can jointly identify preferences,
mistakes, and well-being.
Introduction
Happiness is a very good thing. And it is a very good topic of social–scientific investigation. Indeed, beyond the study of
behavior, a central focus of Economics has always been to understand how economic behavior and institutions affect well-being.
The most traditional technique within Economics for assessing the well-being in different situations is what might (aggressively
but accurately) be called “unquestioned-rationality revealed preference”: observed behavior is assumed to reflect fully rational
maximization of utility, and this leads economists to infer that welfare is higher in one situation than another if it lets a person attain
the outcome she seems most inclined to choose. Yet other approaches to studying well-being have been used outside and (especially
recently) inside economics. In this article we clarify some assumptions underlying the use of the traditional choice-based approach
to the study of well-being, and make the case for supplementing and combining it with some of these other approaches. We do so in
light of three strands of research: 1) theoretical models of how utility may depend not just on outcomes of choice, but on the context
in which those outcomes were chosen; 2) psychological research showing ways in which people are less than 100% rational; and 3)
the growing literature that investigates the determinants of well-being by measuring happiness in different situations
Bookmarks