Competitive Spatial Pricing for Urban Parking Systems: Network Structures and Asymmetric Information

Abstract

Problem/Methodology. Empowered by new technologies to monitor parking occupancy (availability) and process market signals, we aim to expand the application of revenue management in the parking industry. In this paper, we consider competitive spatial pricing in parking systems under endogenous asymmetric information structure. We consider a general graphical Hotelling model wherein each garage has information (forecast) for its own incoming parking demand. Results/Academic Relevance. We focus on the impact of urban network structure on the incentive of information sharing. Interestingly, our analyses suggest that the garages are always better off in a circular-networked city, while they could be worse off in the suburbs of a star-networked city. Nevertheless, in both scenarios, the overall revenue for garages and the aggregate utilities for customers are improved under information sharing. When inaccurate demand forecasting is further considered, information sharing can be undesirable. From an information system designer’s perspective, we identify the optimal market information assignment as a quadratically constrained quadratic program. We also consider an information exchange platform where garages form a two-stage game, and find that complete information sharing is a Nash equilibrium in any market. Practical Relevance/Managerial Implications. Using San Francisco data, we empirically confirmed the value of information sharing. In particular, garages with higher price-demand elasticity and lower demand variance tend to enjoy larger benefits via information sharing. Intuitively, these garages represent the parking lots located in busy downtown areas which are sensitive to price competition, and information sharing helps these garages better predict their competitors’ rates.