Abstract

(co-authored with Martin Browning, Laurens Cherchye, Thomas Demuynck and Frederic Vermeulen)

We present a methodology for the structural empirical analysis of marital matching patterns that allows us to evaluate individuals' trade-offs between material benefits (i.e., real consumption) and immaterial benefits (i.e., unobserved match quality) from marriage. We focus on a setting in which agents can be divided into observable types, with common preferences within a type. Our general methodology includes both general utilities and quasi-linear (i.e., transferable) utilities of the individual household members, and assumes stability of marriage as the equilibrium condition that characterizes the observed partner choices. The basic ingredients of our methodology are testable revealed preference conditions for marital stability that are intrinsically nonparametric, which enables an empirical investigation that is robust to functional specification error. We also demonstrate the practical usefulness of our methodology through an application to the Belgian MEqIn data.