Epsilon Archive for Student Projects

Abstract

This study investigates the causal impact of Stockholm’s congestion tax on urban air quality over the extended period of 2000 to 2023, focusing on the annual highest concentration levels of key pollutants: carbon monoxide (CO), nitrogen dioxide (NO₂), fine particulate matter (PM₂.₅), and coarse particulate matter (PM₁₀).
The study employs the Difference-in-Differences (DiD) approach, leveraging Stockholm as the treated city and Malmö, a comparable Swedish city that did not implement a congestion charge, as the control. The analysis controls meteorological conditions, city-center traffic flow, and demographic and socio-economic factors, to isolate the causal effect of the congestion tax. Crucially, robustness checks, including placebo tests, were conducted to assess the validity of the parallel trends assumption that air quality trends would have been similar in both cities absent the intervention.
Relative to the control city, results indicate statistically significant reductions of approximately 21.9% for PM₁₀ and 29.5% for CO. Placebo tests supported the parallel trends assumption for these pollutants, strengthening the findings. For NO₂, a significant reduction of approximately 29% was estimated, but placebo tests suggested pre-existing trends—warranting caution. PM₂.₅ showed a non-significant negative estimate.
The long-term analysis provides substantial evidence that Stockholm’s congestion tax has delivered sustained environmental benefits, particularly in reducing PM₁₀ and CO peaks. This reinforces its potential as a double dividend policy for addressing traffic congestion and emissions.