Epidemiological and Economic Effects of Lockdown
A. Arnon, John Paul Ricco, Kent Smetters
Abstract
We examine the period of national lockdown beginning in March 2020 using an integrated epidemiological-econometric framework in which health and economic outcomes are jointly determined. We augment a state-level compartmental model with behavioral responses to nonpharmaceutical interventions (NPIs) and to local epidemiological conditions. To calibrate the model, we construct daily, county-level measures of contact rates and employment and estimate key parameters with an event study design. We have three main findings: First, NPIs introduced by state and local governments explain a small fraction of the nationwide decline in contact rates but nevertheless reduced COVID-19 deaths by about 25 percent - saving about 39, 000 lives - over the first three months of the pandemic. However, NPIs also explain nearly 15 percent of the decline in employment - around 3 million jobs - over the same period. Second, NPIs that target individual behavior (such as stay-at-home orders) were more effective at reducing transmission at lower economic cost than those that target businesses (shutdowns). Third, an aggressive and well-designed response in the early stages of the pandemic could have improved both epidemiological and economic outcomes over the medium term.