Traffic Impacts of Variable Pricing on the San Francisco–Oakland Bay Bridge, California

Abstract

The raising of bridge tolls in the peak period on the San Francisco-Oakland Bay Bridge in California in mid-2010 provides a rare opportunity to assess traffic impacts. Carpoolers who previously traveled for free during peak hours were charged an electronic toll under this variable pricing scheme. On the basis of 29 months of time series data, the introduction of carpool charges had a stronger impact on traffic volumes than did peak period pricing of regular traffic. The estimated short-term elasticity of about −0.30 suggests significant numbers of peak hour carpoolers did not travel, switched routes, shifted to public transit, or opted to drive alone. More than half the loss in carpool traffic was estimated to be attributable to the toll increase, a far stronger influence than factors such as rising gasoline prices and unemployment. The estimated elasticity of regular traffic in response to variable pricing was −0.23, an indication that peak period motorists were fairly insensitive to pricing and a reflection of the nondiscretionary nature of many peak hour journeys. It will be important to track trends over time to gain insights into the longer term sensitivity of motorists to variable pricing on natural corridors such as the San Francisco–Oakland Bay Bridge.