Counterfactual Evaluation of Traffic Restrictions on Air Quality in Milan's Congestion Charge Zone Using Matrix Completion

Urban traffic is widely recognized as a major contributor to air pollution in densely populated areas. To address this issue, many cities have implemented traffic restriction zones to reduce pollutant concentrations. Area C, a congestion charge and traffic restriction zone in central Milan, introduced in January 2012 and still in operation today, is a prominent example of such an initiative. This study investigates the causal impact of Area C on air quality by applying advanced statistical learning techniques, specifically Matrix Completion. These methods enable robust counterfactual analysis while relaxing traditional econometric assumptions, such as parallel trends. Using monthly pollution data from 2008 to 2019 in Lombardy and incorporating meteorological variables to control for confounding influences, we find a statistically significant reduction in PM10$$ {\mathrm{PM}}_{10} $$ concentrations within Area C following the policy's implementation. However, no consistent effect is observed for nitrogen oxides (NOX$$ {\mathrm{NO}}_{\mathrm{X}} $$), suggesting that additional or alternative interventions may be required to address gaseous pollutants. Our findings underscore the effectiveness of targeted traffic restrictions in reducing particulate pollution and highlight the value of statistical learning methods for the evaluation of environmental policy.