Abstract
India faces severe air pollution; nine out of the ten most polluted cities in the world are located here.
This study estimates the causal effect of long-term exposure to PM2.5 on health outcomes in India, viz., early life mortality (neo-natal, infant, under-five). We utilise the “valley effect” owing to the peculiar topography and location of the Indo-Gangetic Plains as a basis for a natural experiment. The lower boundary of the Indo-Gangetic Plains drawn from the geographical literature acts as the exogenous threshold for regression discontinuity design (RDD). Applying RDD and two-stage least square method to household level data, the study provides the first causal estimates of the health impact of long-term exposure to PM2.5 using health and pollution data from India. We find PM2.5 exposure is 47% higher, and life expectancy is 2.1 years lower in the Plains relative to other districts in the sample. Early life mortality is positively and significantly affected by sustained exposure to PM2.5.