Research Lead: James Sallee
Research Team: Matthew Tarduno
UC Campus(es): UC Berkeley
Problem Statement: How can congestion pricing be implemented without creating undo harm to vulnerable populations? The economically efficient solution to congested roads is congestion pricing. But imposing efficient pricing schemes can lead to inequitable outcomes. Lower income households spend a larger share of their income on travel and often have fewer transit options and less flexibility in their commuting schedule. There are several options for fostering greater equity in road pricing schemes, ranging from exemption of some users, targeted investments of program revenue in transit, or modifications of cordon areas. These methods suffer from dual problems in terms of both politics and economics. Targeted transit investments and other expenditures aimed at vulnerable populations, even if properly executed, do not provide transparent and salient benefits that engender community buy in. Other schemes that modify prices or exempt some users sacrifice program efficiency. Targeted rebates—schemes that return a portion of program revenue back to households via lump sums as a function of their demographic characteristics—can be simultaneously salient, well targeted, and economically efficient. But little attention has been given to such schemes, and policymakers may not understand relevant strengths and limitations.
Project Description: This study used data from a congestion pricing experiment in the Seattle metro area to examine the feasibility of using revenue from congestion pricing to compensate those harmed by the policy. Results indicate that the initial burden of congestion pricing is highly inequitable, with the lowest income drivers paying an average of 7 percent of their weekly income in congestion charges. There are also considerable differences in burdens within income groups. Results also show that policymakers face a tradeoff in ameliorating these two types of unequal burdens. Returning an equal fraction of the toll revenue to all drivers can make a policy progressive on average, but doing so leaves many drivers either overcompensated or under-compensated. Additionally, while compensation packages based on basic demographic information could improve targeting, many low-income drivers would be left with large proportional burdens because of the fundamental difficulty in predicting individual-level tax burdens. Survey data on travel behavior from Seattle and California metro areas show that the difficulty of designing equitable transfers would be similar in the California metro areas most likely to consider adopting some form of congestion pricing.