Intertemporal Mahalanobis distance matching for heterogeneous-treatment-effect evaluations of air-pollution-abatement policies

This paper develops a novel Intertemporal Mahalanobis-Distance Matching (IMDM) model for evaluating air-pollution-abatement interventions, such as limited-traffic and congestion-charge zones, bans of fossil-fuel traffic and switches to low-emission public transportation. These policies have a strong potential for heterogeneous impacts based on prevailing daily meteorological conditions, determining different average baseline pollution levels, and they pose quite unique causal impact identification conditions. Under these conditions, our IMDM model offers a number of relevant advantages with respect to the existing alternatives. An empirical analysis on data from the covid-19-lockdown traffic abatement in Northwestern Italy illustrates the applicability of the model and shows the policy-relevance of the heterogeneous impacts.