Understanding the Heterogeneity of Adverse COVID-19 Outcomes: The Role of Poor Quality of Air and Lockdown Decisions

Abstract

The uneven geographical distribution of the novel coronavirus epidemic (COVID-19) in Italy is a puzzle given the intense flow of movements among the different geographical areas before lockdown decisions. To shed light on it, we test the effect of five potential correlates of daily adverse COVID-19 outcomes at province level, that is lockdown decisions, demographic structure, economic activity, temperature and quality of air (proxied by particulate matter and nitrogen dioxide). We find that poor quality of air is negatively correlated with adverse outcomes of the epidemic, with lockdown being strongly significant and more effective in reducing deceases in more polluted areas. Results are robust to different methods including cross-section, pooled and fixed-effect panel regressions (also controlling for spatial correlation), as well as difference-in-differences estimates of lockdown decisions through predicted counterfactual trends. They are consistent with the consolidated body of literature in previous medical studies suggesting that poor quality of air creates chronic exposure to adverse outcomes from respiratory diseases. The heterogeneity of diffusion does not seem to depend on other controls such as temperature, commuting flows, quality of regional health systems, share of public transport users, population density and the presence of Chinese community. We also find that adverse COVID-19 outcomes are significantly and positively correlated with the share of small (artisan) firms. Our findings provide suggestions for investigating uneven geographical distribution patterns in other countries, and, if preliminary evidence is corroborated by causation links, have relevant implications with respect to environmental and lockdown policies.

Keywords: COVID-19, coronavirus pandemic, particulate matter, lockdown, economic activity

JEL Classification: I18, Q53, J18, H12