Managing bottleneck congestion with tradable credits under asymmetric transaction cost

Tradable credit schemes (TCS) have been promoted as an alternative to congestion pricing in recent years. Most existing TCS studies assume a frictionless trading market that incurs zero transaction cost. In this study, we propose to examine how transaction cost, taking the form of brokerage fee charged by a TCS operator for the deal-matching service, impacts the performance of TCS in the context of morning commute. Unlike the existing studies, the brokerage fee is assumed to be proportional to the transaction value and asymmetrically split between buyers and sellers. Using the bottleneck model, the optimal TCS design is first obtained for the case of ho-mogeneous commuters and asymmetric transaction cost. We also derive the conditions that ensure Pareto-improving for commuters and financial self-sufficiency for the operator. The latter means the brokerage fee can cover the operator's cost. These analytical results are then extended to cases considering user heterogeneity-which allows commuters to have different values of times (VOT) and desired arrival times at the workplace-and coarse charging design. Among other things, we find that an asymmetric fee structure is better for system performance when buyers bear a higher share of the transaction cost.