Estimating the social cost of congestion using the bottleneck model

Abstract

This paper uses the bottleneck model of Vickrey (1969) to empirically measure the social cost of traffic congestion in the US. Using a detailed trip-level data, we estimate extra travel time over and above hypothetical free-flow travel time, which we call “queuing time”, for each average commute trip. The estimated individual queuing time implies that the annual cost of congestion borne by all US commuters is about 29 billion dollars. We find that a higher level of congestion in a city may be attributed to a smaller per capita road stock in the city. This paper also empirically quantifies a toll that depends both on the commuter's arrival time and trip distance.