Nonadditive public transit fare pricing under congestion with policy lessons from a case study in Toronto, Ontario, Canada

With increasing urbanization and the development of technologies that support automated fare collection, policy makers need decision-support tools to evaluate differentiated public transit fare pricing policies. However, the state-of-the-art tools that consider congestion effects account only for additive fares. A stochastic user equilibrium model with elastic demand was extended to handle nonadditive station-to-station-based fares and was solved by using a method of successive averages. In this paper, an illustrative example is used to show how simple price elasticities alone are not enough to predict the effects of a fare on demand within even a simple eight-node congested network. The first case study of a fare pricing policy was conducted in Toronto, Ontario, Canada; in this case, a distance-based policy was used for the Toronto Transit Commission subway system with respect to downtown and nondowntown subpopulations. The analysis found that compared with the base scenario of a Can$3 fixed fare, there are Pareto-improving fare policies (e.g., fixed rate of Can$2 and variable rate of Can$0.06/km), but the same policy might not be Pareto-improving for all subpopulations. These findings call for more sophisticated fare pricing policies for Toronto (e.g., zone-based) that can cater to specific needs of subpopulations.