30.9 Critiques of TWFE in Staggered DiD

Authors like (L. Sun and Abraham 2021), (Callaway and Sant’Anna 2021), and (Goodman-Bacon 2021) have raised concerns that TWFE DiD regressions:

• Mix treatment effects across cohorts, leading to negative weights and biased estimates.
• Pre-treatment leads may appear non-zero due to contamination by post-treatment effects from earlier-treated groups.
• Long-term treatment effects (lags) may be biased due to heterogeneous adoption timing.

Recent evidence in finance and accounting (e.g., (Baker, Larcker, and Wang 2022)) shows that using newer estimators often leads to null or much smaller causal estimates.

---

References

Baker, Andrew C, David F Larcker, and Charles CY Wang. 2022. “How Much Should We Trust Staggered Difference-in-Differences Estimates?” Journal of Financial Economics 144 (2): 370–95.

Callaway, Brantly, and Pedro HC Sant’Anna. 2021. “Difference-in-Differences with Multiple Time Periods.” Journal of Econometrics 225 (2): 200–230.

Goodman-Bacon, Andrew. 2021. “Difference-in-Differences with Variation in Treatment Timing.” Journal of Econometrics 225 (2): 254–77.

Sun, Liyang, and Sarah Abraham. 2021. “Estimating Dynamic Treatment Effects in Event Studies with Heterogeneous Treatment Effects.” Journal of Econometrics 225 (2): 175–99.