In a new paper, RFF Fellow Elena Safirova, Research Assistant Sébastien Houde, and Senior Fellow Winston Harrington evaluate a range of policies, including three types of cordon-pricing (or area-based congestion tax) schemes, a distance-based toll charged on all freeways, a distance-based toll charged on all metro-area roads, and a tax on vehicle miles traveled (VMT), based on how well they address the full menu of societal damages from driving. Safirova and her colleagues used the LUSTRE model to test the various approaches for metropolitan Washington, D.C.

The authors find that targeting the fuller set of problems not only changes the appropriate level of fees charged to drivers, but it can also have major implications for the relative attractiveness of different policy alternatives. The key variable, they conclude, is the impact on daily vehicle miles traveled.

Marginal Social
Cost Pricing on a Transportation Network:
Comparison of Second-Best Policies

"Using full social cost pricing in place of congestion pricing makes an important difference when the transportation policies substantially affect vehicle miles traveled," says Safirova. "For cordon tolls, for example, the transportation outcomes of congestion pricing and social costs pricing are very similar."

But, she notes, while "comprehensive variable time-of-day pricing on the entire road network turns out to be, by far, the most effective and efficient policy when it comes to reducing congestion alone, when other social costs are factored in, a VMT tax performs almost as well."

Any policy to allay the social costs of driving will be controversial, the authors note, because all involve some measure of high tolls. Moreover, the authors admit that predicting trends in the cost of traffic accidents is difficult because of factors such as the rising cost of health care and safer vehicle technology.

Their paper will be presented in January 2008 at the annual meeting of the Transportation Research Board in Washington.

####