Predictive Power in Behavioral Welfare Economics
41 Pages Posted: 2 Mar 2017 Last revised: 5 Apr 2017
Date Written: April 3, 2017
When choices appear inconsistent due to behavioral biases, there is a theoretical debate about whether or not it is necessary to impose the structure of a model in order to provide meaningful welfare guidance based on such choices. To address this question empirically, we evaluate the predictive power of the "model-free" approach to welfare analysis proposed by Bernheim and Rangel (2009) using two standard data sets. For most of the demand functions that could possibly be observed in this data, their approach does not offer clear welfare guidance. However, we find that for most of the demand functions that are actually observed, their approach can be used to make tight predictions, even when these demands exhibit inconsistencies. For the experimental choices of Manzini and Mariotti (2010), we show that the welfare guidance provided by such predictions are largely consistent with delay aversion, while the guidance provided by revealed preferences is ambiguous.
Keywords: Welfare economics, behavioral economics, revealed preferences, limited attention, experimental economics, scanner data
JEL Classification: I30, C91, D12
Suggested Citation: Suggested Citation