Say NO to Pigou!

Greg Mankiw has been encouraging pundits and policy wonks to join his Pigou Club. Membership only requires dedication to Pigovian taxes, particularly a Pigovian tax on driving, such as a higher retail gas tax. Mankiw has publicly inducted a few big names including Summers and Posner. Additionally, a recent paper in JPE (free link) describing the costs of each additional driver on the road would appear to support the case for a Pigovian tax. I believe Pigovian taxes are not a good answer. Pigovian taxes have a knowledge problem and only internalize one half of the cost-benefit trade. Additionally, there are other, and I think better alternatives for solving the problem.

The first problem with Pigovian taxes is alluded to by Pigou himself in "Some Aspects of the Welfare State" (1954):

It must be confessed, however, that we seldom know enough to decide in what fields and to what extent the State, on account of [the gaps between private and public costs] could interfere with individual choice.

There is in fact a knowledge problem, what is the socially optimal level? The evidence is against any possibility of philosopher-kings, or politicians and electorate wise enough to consult the economic literature. So the Pigovian tax rate gets set by those who have concentrated interests in the tax rather than those who have a diffuse interest. We are back to the problem Pigovian taxes are supposed to solve.

Another problem with Pigovian taxes is one of where the benefits are going. There seems to be great confusion between "society", "government", and the set of individuals actually bearing the costs of the externality. In the aforementioned paper the costs only apply to other drivers in congested areas, not society as a whole. Additionally, government is not "society". If California enacts such a tax, the benefits will most likely go to the public education system without any increase in the education output, thus leaving most Californians without any benefit from the tax.

There are other alternatives. For the social costs listed in the JEP paper, more and better highways and/or alternative transportation may decrease the externality. Tradable property rights in road use could match benefits and costs. My favorite is private highways with congestion variable tolls. California has experimented with private highways and variable tolls already, perhaps California and other places need to find ways to encourage more of them.

For the pollution and global warming externalities, marketable property rights in carbon emissions and sequestration internalizes both costs and benefits and could allow the market to determine the "socially optimal" level.

While Pigovian taxes are the "traditional" economic approach to negative externalities, and were a step in the right direction, I think there are now better alternatives to Pigovian taxes.

Share this