Direct And Indirect Treatment Effects With Time-Varying Covariates

• In Event: Innovations in Causal Inference: Quantile Mixtures, Empirical Bayes Aggregation, and Synthetic Controls

Abstract

We propose a simple approach to treatment effect estimation that is valid when the

number of time periods is small and the parallel trends condition is violated due to the

presence of interactive fixed effects. We show that if there are time-varying covariates that

are linear in the same factors as the outcome variable, we can estimate not only the usual

dynamic average treatment effects on the treated, but we can also separate the effect of

treatment into different causal channels. The asymptotic properties of the estimator are

established and their accuracy in small samples is investigated using Monte Carlo simula-

tions. The procedure is illustrated using as an example the effect of increased trade com-

petition on firm markups in China. We estimate that about half of the impact of China’s

entrance into the WTO on markup dispersion came from the changes in industry-level

productivity.

Author

• Kyle Butts, University of Arkansas, Fayetteville
  Presenting Author