Choosing Congestion Pricing Policy

Abstract

Marginal cost pricing has been long advocated as an efficient way of distributing scarce road resources. In practice, however, policy makers have to retreat to second-best pricing schemes that are associated with lower transaction costs and are simpler for potential users to understand than the first-best marginal cost tolls. To date, the majority of practical applications and theoretical models in Europe and Asia are represented by cordon or area pricing mechanisms, while in North America, variations of link-based tolls have become dominant. This paper compares welfare effects of two second-best cordon pricing schemes with those of second-best link-based tolls for the Washington, D.C., transportation network. START, a strategic and regional transport planning model that features elastic travel demands as well as mode, time period, and route choice, is used to analyze the impacts of the two pricing approaches. Distributional effects of cordon and link-based tolls are also examined in the hope of understanding why one scheme might be preferred over another. Because Washington, D.C., in many respects resembles a European city, cordon policies are more likely to be effective there than in more typical North American cities. Although overall net welfare benefits achieved by the three schemes are found to be similar, their dependence on revenue recycling and distributional impacts are quite different. Although the small cordon puts a higher share of costs on low-income travelers than do other pricing schemes, in absolute terms these costs are lower. The exact distributional impact of a larger cordon is uncertain because it depends on the revenue recycling method employed.