Strategic investment and pricing decisions in a congested transport corridor

This paper studies pricing and investment decisions on a congested transport corridor where the elements of the corridor are controlled by different governments. A corridor can be an interstate highway or railway line, or an inter-modal connection. We model the simplest corridor: two transport links in series, where each of the links is controlled by a different government. Each link is used by transit as well as by local traffic; both links are subject to congestion. We consider a two stage non-cooperative game where both governments strategically set capacity in the first stage and play a pricing game in the second stage. Three pricing regimes are distinguished: (i) differentiated tolls between local and transit transport, (ii) one uniform toll on local and transit traffic, and (iii) only the local users can be tolled. Numerical analysis illustrates all theoretical insights. A number of interesting results are obtained. First, transit tolls on the network will be inefficiently high. If only local traffic can be tolled, however, the Nash equilibrium tolls are inefficiently low. Second, raising the toll on transit through a given country by one euro raises the toll on the whole trajectory by less than one euro. Third, higher capacity investment in a given region not only reduces optimal tolls in this region under all pricing regimes but it also increases the transit tolls on the other link of the corridor. Fourth, capacities in the different regions are strategic complements: when one country on the corridor increases transport capacity, it forces the other country to do the same. Fifth, we find interesting interactions between optimal capacities and the set of pricing instruments used: capacity with differentiated tolls is substantially higher than in the case of uniform tolls but overall welfare is lower. Finally, if transit is sufficiently important, it may be welfare improving not to allow any tolling at all, or to only allow the tolling of locals.(This abstract was borrowed from another version of this item.)