Externalities of driving

Thursday's "Economic Scene" column in The New York Times by Hal Varian, available here, summarizes an article in the Journal of Political Economy by Aaron Edlin and Pincar Karaca-Mandic.  The article, "The Accident Externality from Driving," is available at the BE Press website, here. 

The article identifies an externality or activity-level effect of driving: simply by getting on the road, one increases the likelihood of an accident with other drivers, but people, in deciding whether to drive, almost never take those external costs to others into account.  Edlin and Karaca-Mandic seek to estimate the value of this accident externality.  They recognize that it must be greater in jurisdictions where the density of cars is greater.  That is, the more congested an area is, the greater the externality of driving.  So, in North Dakota, where congestion is minimal, the value of the externality is $10 per driver, while in California, the value ranges between $1,725 and $3,239. Nationwide, the authors estimate that the value of this accident externality may run to more than $200 billion per year. 

Clearly, in densely populated states these externalities can be substantial, and, as a result, the fact that these externalities are not being internalized could lead to far too much driving and, therefore, far too many and too severe accidents.   

The question then is, "How do we induce drivers to internalize this accident externality?"  Edlin and Karaca-Mandic suggest a gasoline  tax.  As Professor Varian points out, one of the shortcomings of the gasoline tax is that it applies equally to good and bad drivers.  One might take account of that by having a separate tax on accident-insurance premiums, hoping that experience rating would have adjusted those premiums to reflect the quality of the insuree's driving. 

Nonetheless, there is, as Edlin and Karaca-Mandic and Varian note, no stomach among politicians for these taxes. 

TSU