Taxing vehicle miles traveled: The Traffic Choices Experiment in the Puget Sound region Most recent version: Download

Driving creates a number of externalities, including congestion, emissions of conventional air pollutants and greenhouse gases, road wear and tear, and accidents. In principle, these externalities could be corrected by charging drivers a tax for each mile driven, but the effectiveness of such measures when implemented in practice is unclear. This paper analyzes the implementation of a tax on vehicle miles traveled in the Seattle, WA, metropolitan area between 2005 and 2006. Prior to the beginning of the trial, the travel behavior of 276 volunteer households was recorded through devices installed in their vehicles. During the trial period participants had to pay a toll for each mile driven. The rate changed with the hour of the day, the day of the week and the type of road. The results from my analysis show that on average in the first week of the trial period participants reduced their miles driven by 6.83 % for each 10 cents/mile toll. The effect of the toll system disappears after the first week, with the exception of miles driven in highways. This calls for caution when inferring about taxes and other economic incentives or disincentives based on short-lived programs and on behaviors within a narrow window of the implementation of the program itself. There is also a strong heterogeneity in household response to the toll, with higher-income households and households with children being less responsive whereas households used to drive often or during high-toll times and roads show a higher reduction in miles