Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects

de Chaisemartin, Clément; D'Haultfœuille, Xavier

doi:10.2139/ssrn.3148607

Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects

35 Pages Posted: 24 Mar 2018 Last revised: 1 Nov 2019

See all articles by Clément de Chaisemartin

Clément de Chaisemartin

University of California, Santa Barbara (UCSB) - Department of Economics

Xavier D'Haultfœuille

Center for Research in Economics and Statistics (CREST)

There are 2 versions of this paper

Date Written: October 30, 2019

Abstract

Linear regressions with period and group fixed effects are widely used to estimate treatment effects. We show that they estimate weighted sums of the average treatment effects (ATE) in each group and period, with weights that may be negative. Due to the negative weights, the linear regression coefficient may for instance be negative while all the ATEs are positive. We propose another estimator that solves this issue. In the two applications we revisit, it is significantly different from the linear regression estimator.

Keywords: Linear Regressions, Fixed Effects, First Difference, Differences-in-Differences, Heterogeneous Treatment Effects

JEL Classification: C21, C26

Suggested Citation: Suggested Citation

de Chaisemartin, Clément and d'Haultfoeuille, Xavier, Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects (October 30, 2019). Available at SSRN: https://ssrn.com/abstract=3148607 or http://dx.doi.org/10.2139/ssrn.3148607