Low Income Drivers Gain from Congestion Pricing
….there is disagreement about the distributional effects of highway toll lanes. On one side, policymakers refer to dynamic tolling as “value pricing” and emphasize that it provides choice to drivers (Samdahl et al., 2013). On the other side, opponents are concerned that “Lexus lanes” enrich the wealthy at everyone else’s expense (Astor, 2017; Rosendorf, 2018). Evaluation of these perspectives depends on two empirical objects: the distribution of driver preferences and what we call the “road technology”—the relationship between traffic quantities and travel times. When one lane becomes tolled, drivers substitute from the newly priced lane into the remaining unpriced ones, increasing travel times in the unpriced lanes. High peak-hour prices may also induce drivers to substitute toward driving off-peak (or not at all), which can increase average speeds when the road technology is convex. Finally, since tolling changes the predictability of travel times, having the option to take the priced lanes can serve as insurance against worse-than-expected traffic conditions.
In this paper, we study the aggregate and distributional impacts of dynamic tolling. To do this, we bring together data on toll transactions, historical traffic conditions, and driver characteristics from the I-405 Express Toll Lanes in Washington State. We begin by presenting two sets of descriptive facts: first, aggregate speed and throughput increased after the introduction of tolling on this highway, and second, low-income drivers face advantageous trade-offs between price and travel time savings in the toll lanes. Next, to quantify the equilibrium effects of tolling, we build and estimate a model of driver demand, the road technology, and the pricing algorithm. In particular, the demand model incorporates the features of dynamic tolling highlighted above: choices of where and when to drive, as well as uncertainty about prices and travel times. Using the estimated model, we find that low-income drivers in fact gain the most from status-quo tolling, and we explore how equilibrium outcomes would change under counterfactual pricing policies.
Pearl Li from Stanford is on the job market.