Inframarginal Travelers and Transportation Policy

Half of all travelers have such inflexible schedules that they strictly prefer their ex-ante arrival times to all others; however, existing models of traffic congestion implicitly assume that no such inframarginal travelers exist. This leads these models to predict travel times nearly seven times greater than those observed. Accounting for these travelers significantly improves these models’ ability to fit the data and changes policy prescriptions. In the case of congestion pricing, it reduces the socially optimal road toll by up to 72%.