A Casual Analysis of FlexPass: Incentives for Reducing Parking Demand

A parking incentive program named FlexPass have been conducted in University of California, Berkley. The causal structure underlying employee parking behavior is examined in this study by a randomized controlled trial, where participants receiving treatment were offered incentives for parking less and taking other modes. This field experiment lasted for three months and recruited 392 staff and faculty members. Practicable problems encountered during the study were non-random differential dropout after the group assignment and non-ignorable missing data. Missing data were measured by follow-up emails and estimated utilizing a mixed latent factor model, which outperformed traditional feature based models. Dropout bias was corrected by sample selection model. During the study, control participants, served as baseline, parked 4.3 days per week and the FlexPass induced an average treatment effect of 4.2% parking demand reduction. A heterogeneity treatment effect has been discovered. Participants who claimed to be interested in the pricing scheme, accounted for 77% of the enrolled population. There is a larger treatment effect of 6.0% in this group. For the rest, most of whom are regular drives, there is no significant treatment effect. The finding suggests that instead of building new parking structures, increasing the parking prices and providing incentives at the same time could reduce parking demand. It also brings significant rewards to those who choose to travel by other modes.