Do city lockdowns effectively reduce air pollution?

Abstract

Severe air pollution remains a problem in developing countries such as India and China owing to unfavourable winter climate conditions that encourage the diffusion of air pollutants. Although city lockdowns can alleviate serious air pollution, they incur significant social costs. This study explored the effect of China's COVID-19 lockdown on air pollution. Machine learning was utilised to eliminate weather-related fluctuations in air pollutant concentrations, and two difference-in-difference models were employed to estimate the effects of general epidemic prevention and enhancement measures on air pollution and evaluate the total impact of city lockdown on air pollutant concentrations. By carefully examining the proportion of these effects in the overall impact, this study analysed the cost-effectiveness of city-level lockdown measures on air pollution. Results indicate that city lockdowns significantly reduced the concentrations of five air pollutants: NO2 (−28.3 %), PM10 (−21.41 %), PM2.5 (−15.83 %), CO (−8.66 %), and SO2 (−4.2 %). The contributions of the enhancement measures are 41.82 % for PM2.5, 42.66 % for PM10, −4.04 % for CO, and 31.83 % for NO2. This suggests that city lockdowns may not be cost-effective compared to general epidemic prevention measures. Thus, this study recommends limited city lockdown measures for severe air pollution, similar to the general epidemic prevention measures.