Corrective taxes are favorable.

We’ve already seen that ex post instruments have some unique costs, so I’ll just add here that abandoning the tax system in favor of “command and control” regulation also may be costly. Proponents of price instruments point out that Pigouvian taxes are superior to regulation in that they generate resources that can be used for other socially useful projects, such as lowering the income tax.177 I’ve argued that claim isn’t inevitably true, but it is accurate at least in the case of traditional regulation (as opposed to “nudges”), and where the government’s alternative is to use sticks and set aside that money for public, rather than private, use.178 In other words, if we follow Fleischer’s suggestion, we may get more accuracy, but at the cost of the revenue benefits we might have gained from a tax. If, however, government’s alternative was to use a subsidy or “carrot,” then Fleischer’s proposal is more appealing in comparison, since regulation saves the government on the cost of paying out rewards. 179

On the other hand, it is also possible that Fleischer intends to argue not against all Pigouvian taxes, but rather only against those that are part of the traditional tax system. For example, some industries must pay licenses in order to operate. Often these licenses vary depending on the location and intensity of the industrial operation.180 From an economic perspective (and sometimes a legal one), licenses work very similarly to “taxes.”181 The difference would be that they are administered by a regulatory agency, rather than the tax collector. This may seem a minor difference, but a number of scholars point to differences in expertise and institutional norms in the two bureaucracies, differences that are not coincidental but instead are the structural result of the underlying goals and functioning of each.182

If this is Fleischer’s argument, then the lost-revenue concern is mitigated. License revenues may be less useful than taxes, however, to the extent that they are dedicated to some specialized purpose, as that purpose may not always be the most socially beneficial. For example, banking examination fees often go to pay for the operation of bank examiners, even if the bank examiners’ need for additional funding is minimal.183

Whatever the costs of Fleischer’s alternatives, the core of his case against the tax system overlooks the possibility that our traditional tax system is flexible enough to get most of the benefits of flexible pricing. I’ve argued that two or three prices usually are adequate for many purposes, and our tax system can typically achieve that level of detail. Our income tax, for instance, routinely offers multiple prices for the subsidies built into it. Taxpayers who itemize can claim deductions for many activities not available to non-itemizers.184 Itemizing taxpayers who suffer medical expenses or unexpected loss of property can claim deductions, but only if their costs exceed a floor.185 Charitable contributors can no longer claim a deduction once their contribution exceeds half their adjusted gross income.186 The Alternative Minimum Tax reduces the value of the federal subsidy for state and local governments.187 Judicial interpretation of rules might add some flexibility not evident on the face of a rule or statute.188 And the value of all deductions varies depending on the taxpayer’s marginal rate.189

This is not to say that all of these existing price variations are accurately matching the value of the government’s subsidy to the marginal benefit of those who face a given price.190 Some existing rules are defensible on that ground; for example, evidence suggests that wealthier households are (to a point) more responsive to tax incentives for charitable giving, and so it makes some sense that top-earners’ higher marginal rates give them a larger subsidy for each dollar they donate.191 My goal is not necessarily to defend all the particulars of our existing system, however, as much as it is to show that matching prices and marginal damage or benefit through the tax code is not pure science fiction. Indeed, in recent work, Jim Hines and Kyle Logue set out in greater detail how the model of the Alternative Minimum Tax could be used to do a better job of matching within our existing tax structure.192

Finally, as other authors have noted, Pigouvian taxes can also be combined with other instruments, 193 which adds yet another layer of flexibility. For instance, in some states drivers who explore the top end of their sports car’s performance on public roads can receive jail time as well as a speeding ticket.194 For extreme outliers like the man in the Mazzerati, a linear price doesn’t fit very well. But for most of the population, the usual speeding ticket does a reasonable job of approximating the social harm, and has the added benefit of bringing in some revenue.

All of this is to say that corrective taxes likely can play an important role in regulatory policy. While they aren’t perfect instruments, they are less bad than Fleischer maintains, and none of the alternatives are without flaws, either.