Papers › On the Use of Two-Way Fixed Effects Models for Policy Evaluation During Pandemics
On the Use of Two-Way Fixed Effects Models for Policy Evaluation During Pandemics
Germain Gauthier
In the context of the Covid-19 pandemic, multiple studies rely on two-way fixed effects (FE) models to assess the impact of mitigation policies on health outcomes. Building on the SIRD model of disease transmission, I show that FE models tend to be misspecified for three reasons. First, despite misleading common trends in the pre-treatment period, the parallel trends assumption generally does not hold. Second, heterogeneity in infection rates and infected populations across regions cannot be accounted for by region-specific fixed effects, nor by conditioning on observable time-varying confounders. Third, epidemiological theory predicts heterogeneous treatment effects across regions and over time. Via simulations, I find that the bias resulting from model misspecification can be substantial, in magnitude and sometimes in sign. Overall, my results caution against the use of FE models for mitigation policy evaluation.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Results from the paper archive 2025-07-28
No leaderboard rows for this paper in the archive.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections