This study evaluates the impact of the COVID-19 pandemic and the associated “epidemiological traffic-light” system on homicide rates in Ecuador. We exploit monthly panel data at the canton level for 2020–2021 and estimate an event-study difference-in-differences design using Poisson pseudo-maximum likelihood with two-way fixed effects. The treatment is defined as the first transition from the red (strict) alert level to more flexible yellow or green stages, and we control for COVID-19 incidence and mortality, local economic activity, mobility indices and demographic characteristics. Our preferred event-study specification suggests a temporary increase in homicides in the month of the first relaxation of restrictions, but the estimated effects in subsequent months are imprecise and statistically indistinguishable from zero. The average post-treatment effect across the first five months after the transition is small and not statistically significant. Linear specifications in levels and in log(1 + homicides), as well as wild-cluster bootstrap inference, corroborate the absence of robust effects of the traffic-light policy on homicide rates. These findings indicate that, conditional on epidemiological and economic conditions, the traffic-light mobility regime did not generate systematic or sustained changes in lethal violence at the canton level. We discuss possible mechanisms and implications for the joint design of public-health and public-safety interventions.
Chucuri et al. (Fri,) studied this question.