Welfare Effects of Stockholm Congestion Charges Using Dynamic Network Assignment
M Borjesson, I Kristoffersson, KTH, SE
In this paper the welfare effect of the Stockholm congestion charges is analyzed using the dynamic assignment. It is found that the welfare effects are considerably larger compared to analyses with a static model.
It is generally known that congestion pricing could be an effective measure to solve environmental and congestion problems in urban areas, but there is still low political and public acceptability in many urban areas. One possible reason for low acceptability is the notion that drivers paying the toll are worse off if not rebated (which if often unclear). That drivers are worse off before return of revenues is also the result of most static models, but not the case in the dynamic single bottleneck model where drivers neither loose nor win if not rebated (De Palma, Kilani and Lindsey, 2005).
A dynamic setting taking into account travelers are heterogeneous, however, opens for the possibility that some users become better off even if not rebated for several reasons. First, in a dynamic model it is possible to take into account that the optimal toll is a time-varying toll, which gives drivers the possibility to adjust their departure times in such a way that departure times are socially optimal. Second, in a dynamic assignment model, network effects are captured. Through network effects drives not paying the toll may get shorter travel times because the problem of blocking back of upstream links and signal plans at intersections will be smaller. Third, ignoring heterogeneity of value of time in a system with a free parallel road, leads to great underestimation of benefits by ignoring the efficiency gains due to separation of traffic (Verhoef and Small, 1999).
In this paper the welfare effect of the Stockholm congestion charges is analyzed using the dynamic network assignment model Contram and a demand model taking account of heterogeneous users. It is found that the welfare effects are considerably larger compared to earlier analyses with a static assignment model with homogenous users. In fact, the analysis shows that drivers as a group are not being worse off even if not compensated with revenues. We analyze to what extent this is due to network effects, dynamics allowing travelers to adjust their departure times and to allowing for heterogeneous users.
Previous analyses of congestion charging systems implemented in Stockholm and London have been analyzed with static assignment models. The present study indicates that the welfare effects of the congestion charges are likely to be underestimated in these analyses. We argue hence that it is desirable to move towards more dynamic models in analyzing the welfare effects of congestion charging. The result that the benefits for drivers even without or with little rebate is important and can have effects on political and public acceptability of congestion charges. In fact, this could be one reason for the increasing acceptability for the congestion charges in Stockholm.
Association for European Transport