Simple Difference-in-Differences Estimation in Fixed-T Panels

Nicholas Brown¹, Kyle Butts², and Joakim Westerlund^3,4
1Florida State University, ²University of Arkansas, ³Lund University, ⁴Deakin University

Paper

Abstract

We propose a simple approach to treatment effect estimation in panel data 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. The procedure allows the covariates to be affected by treatment and enables separation of the part of the estimated treatment effect that is due to the covariates from the part that is not. The asymptotic properties of the new approach are established, and their accuracy in small samples is investigated using Monte Carlo simulations. The procedure is illustrated using as an example the effect of increased trade competition 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.

Replication

Monte Carlo Simulations

1. code/simulations/sims.R

• Simulations presented in Table 1, 2, and 3 as well as additional simulations presented in the appendix

Application

1. code/Trade-Liberalization-and-Markup-Dispersion/analysis.R

• Takes the data from Lu and Yu (2015) and recreates Figure 2 in their paper.
• Estimates the TECCDE model to estimate the effect of World Trade Organization ascension on markup dispersion in (previously) high-tariff industries.
• Decomposes the effect into the mediated effect via decrease in marginal cost dispersion.

Citation

@article{brown2023difference,
  title={Difference-in-Differences via Common Correlated Effects},
  author={Brown, Nicholas and Butts, Kyle and Westerlund, Joakim},
  journal={arXiv preprint arXiv:2301.11358},
  year={2023}
}