Two-Way Fixed Effects Estimators with Heterogeneous Treatment Effects
63 Pages Posted: 24 Mar 2018 Last revised: 6 Aug 2018
Date Written: August 2, 2018
Linear regressions with time and group fixed effects are widely used to estimate treatment effects. We show that they identify weighted sums of the average treatment effects in each group and period, with weights that can be estimated, and that may be negative. In two articles that have used those regressions, half of the weights are negative, thus suggesting that those regressions are often not robust to heterogeneous treatment effects across groups and over time. We propose another estimator that is robust to heterogeneous treatment effects. In an application, it is of a different sign than the linear regression estimator.
Keywords: Fixed Effects, First Difference, Differences-in-Differences, Heterogeneous Treatment Effects
JEL Classification: C21, C26
Suggested Citation: Suggested Citation