2023 Reports

Estimating Separable Matching Models

Galichon, Alfred; Salanie, Bernard

Most recent empirical applications of matching with transferable utility have imposed a natural restriction: that the joint surplus be separable in the sources of unobserved heterogeneity. We propose here two simple methods to estimate models in this class. The first method is a minimum distance estimator that relies on the generalized entropy of matching introduced in Galichon and Salanie (2022). The second applies to the more special but popular Choo and Siow (2006) model, which it reformulates as a generalized linear model with two-way fixed effects. Both methods are easy to apply and perform very well.