This paper studies the impact of a prefecture’s lockdown measure in China (Out-side Hubei province), in response to the outbreak of Covid-19, on its air pollution and Covid-19 related health outcomes. In doing that, we use a difference-in-difference identification strategy. To address the potential endogenous problem that the decision on a prefecture’s lockdown and its timing might not be random, we employ the bilateral population flow from "Baidu" Migration index to first predict a prefecture’s probability to undertake a lockdown measure. Our results of difference-in-difference with instrumental variable show that a prefecture’s lockdown measure reduces both its air quality index (AQI) and PM10 significantly by around 30%, and yet the result for difference-in-difference with OLS is only around 10%. Our difference-in-difference with instrumental variable results for health outcomes show that a prefecture’s lockdown measure increases its number of daily new recovered patients by around 46%, and yet this result for difference-in-difference with OLS is only around 27%. The sharp difference between these two approaches on both the environment and Covid-19 related health outcomes implies there exits substantial misallocation of lockdown measures across different prefectures. We also find that the induced improvement on air quality contributes positively and significantly to the effects of lock-down measures on the number of daily new recovered patients.
Lead investigator: |
Linyuan Huang |
Affiliation: |
Hunan University |
Primary topic: |
|
Region of data collection: |
|
Country of data collection |
|
Status of data collection |
|
Type of data being collected: |
|
Unit of real-time data collection |
|
Frequency |
