The Dynamic and Heterogeneous Effects of COVID-19 Vaccination Mandates in the USA

Mandatory vaccination for COVID-19 has received intense political and ethical debates, while the literature on the causal effects of vaccination mandates on vaccination outcomes is very limited. In this study, we examine the effects of the announcement of vaccine mandates (VMs) for workers working in three sectors, including health, education, and state governments, on the uptake of first-dose and second-dose vaccination across 50 states in the United States of America. We show that VM announcements have heterogeneous effects; hence, standard two-way fixed effects and difference-in-differences estimators are biased. We present evidence for the heterogeneous treatment effects using recently developed estimators of de Chaisemartin and D’Haultfœuille (2020b) in single and two-treatment settings. In the setting of a single treatment, when treating all VM announcements equally, our results show that VM announcement was associated with an increase of 20.6% first-dose uptake from 1 July to 31 August 2021. In two-treatment settings, our results suggest that VM announcements for workers in health or state government sectors have significant causal effects on first-dose vaccination. Additionally, VM announcements do not have significant causal effects on second-dose uptake. Our results are robust to the choice of differing outcome variables and periods after controlling for state-level covariates, including COVID-19 death, unemployment, and cumulative two-dose vaccination.