Pricing vehicle emissions and congestion using a dynamic traffic network simulator

Road traffic is a major contributor to air pollution which is a serious problem in many large cities. Experience in London, Milan, and Stockholm indicates that road pricing can be useful in reducing vehicle emissions as well as congestion. This study uses the dynamic traffic network simulator METROPOLIS to investigate the effectiveness of tolls to target emissions and congestion externalities on a stylized urban road network during a morning commuting period. The concentration and spatial distribution of four pollutants are calculated using a Gaussian dispersion model that accounts for wind speed and direction. Single and double cordon tolls are evaluated, as well as flat tolls that do not change during the simulation period and step tolls that change at half-hourly intervals. The presence of emission externalities raises optimal toll levels, and substantially increases the welfare gains from tolling although the advantage of step tolls over flat tolls is lower than if congestion is the only externality. The individual welfaredistributional effects of tolling vary strongly with residential and workplace locations relative to the cordon, and also differ for the windward and leeward sides of the city.