Impact of national lockdown on COVID-19 deaths in select European countries and the US using a Changes-in-Changes model

Abstract: In this paper, we estimate the impact of national lockdown on COVID-19 related total and daily deaths, per million people, in select European countries. In particular, we compare countries that imposed a nationwide lockdown (Treatment group); Belgium, Denmark, France, Germany, Italy, Norway, Spain, United Kingdom (UK), and the US, to Sweden (Control group) that did not impose national lockdown using a changes-in-changes (CIC) estimation model. The key advantage of the CIC model as compared to the standard difference-in-difference model is that CIC allows for mean and variance of the outcomes to change over time in the absence of any policy intervention, and CIC accounts for endogeneity in the choice of policy intervention. Our results indicate that in contrast to Sweden, which did not impose a national lockdown, Germany, and to some extent, the US were the two countries where nationwide lockdown had a significant impact on the reduction in COVID-19 related total and daily deaths per million people. In Norway and Denmark, there was no significant impact on total and daily deaths per million people relative to Sweden. Whereas in other countries; Belgium, France, Italy, Spain, and the UK, the effect of the lockdown was in the opposite direction, that is, they experienced significantly higher COVID-19 related total and daily deaths per million people, post the lockdown as compared to Sweden. Our results suggest that the impact of nationwide lockdown on COVID-19 related total and daily deaths per million people varied from one country to another.