Abstract

This paper examines the differential impacts of early childhood exposure to air pollution on children’s health care use across parental income groups using French administrative data. We isolate a quasi-experimental air pollution shock based on changes in local thermal inversion exposure between birth cohorts within municipalities.

We provide causal evidence that increases in air pollution exposure early in life affects the likelihood of emergency admissions and drug consumption related to respiratory issues for young children. We uncover large health inequalities related to air pollution between children by estimating the treatment effect heterogeneity with machine learning. We find that the health impacts are concentrated in about 10% of the infant population, which we characterize.

Practical information