Lights out? COVID-19 containment policies and economic activity

Epidemics and pandemics have large impacts on human life and livelihoods. During a pandemic, governments typically intervene to slow its spread, as well as to mitigate the economic impact and revive activity. Thus, understanding the impact of government restrictions on reviving or hindering economic activity is important.

In our study, we estimate the impact of differential containment policies implemented by the Government of India (GoI) during the COVID-19 pandemic on aggregate economic activity. The first COVID-19 infection in India was reported at the end of January 2020. In March, the GoI implemented one of the most stringent lockdowns globally. After five weeks of nationwide lockdown, uniform restrictions were replaced with targeted measures that introduced variation across districts in May. The differential relaxation of restrictions across districts permits us to examine the impact of the heterogeneous containment measures. Districts were classified into three zones: those with the most severe restrictions (Red), those with intermediate restrictions (Orange), and those with the least severe ones (Green). Using a difference-indifferences approach, we compare the speed of recovery in the three zones after the uniform national lockdown in May, June and July.

To conduct this analysis, we combine pandemic-era, district-level data from a range of novel sources like monthly nighttime lights from global satellites, Facebook’s mobility data from individual smartphone locations, and high-frequency, household-level survey data on income and consumption. Nighttime light intensity contains information about economic activity at high spatial granularity and have also been used in another study to assess the economic impact of the national lockdown in India. 

Nighttime light intensity in May was 12.4 percent lower for districts with the most severe restrictions and 1.7 percent lower for districts with intermediate restrictions, as compared to districts with the least restrictions. The differences were largest in May, when the different policies were in place, and slowly tapered in June and July. These estimates point to large short-run costs of containment policies and especially of the mobility restrictions that differentiated Red zone and Orange zone districts. 

Facebook mobility data from cellphone locations indicates that restricted mobility is a plausible channel causing lower household income and consumption, which in turn explains the aggregate impact. Some districts were affected more by the restrictions than others. Stricter containment measures had larger impacts in more developed districts characterized by a greater population density with older residents, as well as more services employment and bank credit.