Curb Your Enthusiasm for Pigovian Taxes

I. INTRODUCTIONLaw professors have a tendency to act as if we are philosopher kings, descending into the cave to educate the prisoners.1 We identify the ideal, and, embracing our role as guardian of the republic, we sketch out a plan to engineer the best social policy to reach that identified goal. In the twentieth century this tendency was most apparent in the form of the command-and-control model of regulation, an approach that has fallen out of grace.2 For the newest generation of Platonic guardians, a Pigovian tax is a tempting gadget.3Corrective taxes are taxes that are designed primarily to change behavior rather than raise revenue. These taxes are often called "Pigovian" taxes in reference to Arthur Pigou, the British economist who pioneered the approach.4 The idea is that by placing a small tax, equal to marginal social cost, on each unit of an activity to be discouraged-environmental pollution is the most common example- prices will rise, forcing polluters to internalize the social cost of the harmful activity. As a result, production will decrease, leading to an allocation of economic resources that reflects the true cost of the activity causing the pollution.5Policy advocates have often inferred, erroneously,6 that using a Pigovian approach means that one need not know who is causing harm, where it is occurring, or how much it would cost each firm or individual to reduce the harmful activity. Indeed, if that were the case, one would only need an estimate of the total amount of an activity and the total social harm that results. While making such estimates would be challenging, it would be less challenging than the aggregate cost-benefit analysis required of many agency decisions under current law.7These seemingly relaxed design specifications make Pigovian taxes a tempting instrument of social engineering, especially when compared to traditional command-and-control regulation.8 One finds considerable academic support for Pigovian taxes on a wide range of products and activities, including carbon, gasoline, fat, high fructose corn syrup, guns, financial transactions, executive pay, excessive zoning, and sport utility vehicles.9 Law professors and economists of all political stripes, led by such luminaries as Louis Kaplow, Greg Mankiw, Gary Becker, and Robert Frank, are mystified by the public's inability to see the merits of using Pigovian taxes more frequently to address serious social harms.10 Most recently, Jonathan Masur and Eric Posner have issued a "Pigovian call to arms" on the grounds that not only do regulators have the legal authority to implement Pigovian taxes, they should replace any instance of command-and-control regulation with a tax.11This academic exuberance for Pigovian taxes should be tempered. My goal is not to defend command-and-control regulation. Rather, I wish to highlight some often-overlooked weaknesses in the Pigovian instrument. In the circumstances where a Pigovian tax is not the right instrument, the right answer may be command-and-control regulation, or it may be some other approach, such as information disclosure, behavioral nudges, or ex post tort liability.12 Or it may be best to let sleeping dogs lie.I understand the temptation of a Pigovian state. Externalities are all around us. Your neighbor's lawnmower is too loud. You can smell the garbage from the restaurant downstairs. You take your daughter to Disneyland and worry about the unvaccinated kids running around alongside. And a Pigovian tax is easy to design-as a uniform excise tax-if one simply assumes uniform marginal social cost across all individuals and firms.13This assumption of uniform marginal social cost pairs well with the limited information and enforcement capacity of government institutions.14 The problem is that when marginal social cost varies, average cost does not equal marginal cost, and Pigovian taxes may not lead to an optimal allocation of economic resources. …