Degrees of Freedom: The last thing I will ever write (today) about Pigovian taxes on gasoline: Politics, Philosophy and Economics from a libertarian point of view.

The last thing I will ever write (today) about Pigovian taxes on gasoline Wednesday, December 06, 2006

Greg Mankiw's Pigou Club has gotten still more publicity, with a fairly thoughtful article on Bloomberg.com. Degrees of Freedom has covered this before, (see here and here) but this is worth going over again if for no other reason than that so many people who claim to value individual freedom and who claim to believe that governments are the primary usurpers of that freedom somehow wind up thinking that an increased tax on gas is something that they should support.

I can imagine several reasons to disagree with Professor Mankiw, but here are a small few:

1. The motivation for a Pigovian tax is based on the premise that negative externalities are something to be avoided. So far, so good. As I suspect Professor Mankiw's textbooks would indicate, a negative externality exists when some people's decisions adversely affect the belongings of others. I'll certainly agree that such states of affairs are undesirable, but it's worth pointing out that while carbon emissions may be causing some negative externality as a side effect of beneficial uses, Professor Mankiw's preferred instrument for remedying the externality caused by carbon emissions, the federal government, creates negative externalities by design. Indeed, the distinguishing feature of any government is the ability of its agents to make decisions which adversely affect the belongings of others without penalty. I share Mankiw's opposition to externalities, but a Pigovian tax on gasoline would diminish a smaller externality and add to a larger one in the process. This is not a solution to the problem of externalities.

2. It makes very little sense to advocate a new or increased tax to discourage something that is encouraged by existing subsidies and regulations. This is largely an extension of the first point. Presently, there are taxpayer subsidized roads which people perceive as being costless. There are regulatory price controls on energy utilities which diminish the incentive for restraint in consumption. There are agricultural subsidies which create an incentive for fuel to be used in transporting exports which would not be profitable in a free market. (See here for more. HT: Jay Jardine) And then there is the elephant in the living room: the largest producer of pollution is the federal government which enjoys sovereign immunity to continue to pollute. It's hard to imagine how funneling more money to the government would be the best solution for problems that are in many ways problems of the government's own creation.

3. It's impossible to defend the morality of a Pigovian tax on gasoline without relying on a very permissive view of theft or some form of ethical particularism. If I were to advocate systematic private theft against gasoline purchasers, proportionate to the amount of gasoline that they purchased, I doubt that anyone in Mankiw's "Pigou Club" would be in favor of the idea. I'd be advocating theft, after all, and most people oppose theft. Presumably, Mankiw opposes private theft as well and I'll even bet that Mankiw's opposition to private theft is based on reasons. If, as is the case with most economists, those reasons are based on the positive consequences of theft, then those reasons pertain to taxation in just the same way as they pertain to private theft. Whether the person taking your money works has been authorized by the government to do so or not, the end result of such taking is the same: you are that much poorer. If those reasons are based on some ethical consideration other than the consequence of theft, then they should still pertain to taxation in just the same way as they pertain to private theft. To draw an exception when the persons doing the taking have been authorized by the government to do so is garden variety ethical particularism. Although I suspect Professor Mankiw would oppose ethical relativism in general, I fail to see how he might favor taxation and simultaneously oppose private theft against gasoline purchasers unless he is taking a relativist position on this issue in particular.

4. Taxes make it harder to get rich. While I'm all in favor of getting rich in any case, one of the motivations for a Pigovian tax on gasoline is the belief that carbon emissions will contribute to global climate change, which many believe will have bad consequences. I'm not qualified for or interested in a discussion of the merits of the models behind this prediction, but so far as I can tell, either there will be significant climate change as a result of carbon emissions or there will not be significant climate change as a result of carbon emissions. In the latter case, people are better off being rich for all of the usual reasons that it's good to be rich. In the former case, people are better off being rich for all the usual reasons plus one: the need to adapt to whatever climate change may occur. Those who favor a Pigovian tax as a way to deal with global climate change might take the view that the relative loss of wealth caused by the tax is a fair price for the benefits that would come from reduced carbon emissions. Theoretically, that might be possible, but in this world, the more likely outcome is that other developing economies will free ride on the pollution discipline that a Pigovian tax would create in the United States and political factors will make it impossible for the United States government to impose measures severe enough to actually prevent climate change from occurring. The choice becomes climate change with higher taxes or climate change with lower taxes. I can't imagine why the former could be better.

5. The incentives faced by politicians make it very unlikely that a Pigovian tax on gasoline will be implemented as a substitute for some other tax. If a tax on gasoline could be implemented as a dollar-for-dollar replacement for the tax on capital gains, even I would be in favor of that. But the problem here is politics, or rather, politicians. By every indication, politicians like to spend money. To do so, they have to keep getting elected and reelected to office. Changing the basis of tax assessment in a way that leaves politicians with the same amount of money to spend is certainly no improvement as far as the goal of spending money is concerned and implementing a tax on gasoline is hardly a good strategy for getting elected or reelected. In order for a Pigovian tax on gasoline to be acceptable to politicians, it would have to actually increase the amount of money that they get to spend. That is, it could be revenue neutral, or it could be implementable, but it can't be both.

11 Comments:

At Wed Dec 06, 10:34:00 PM EST, Taylor (sologue) said...: Devastating.
At Fri Dec 08, 05:54:00 PM EST, David said...: On 1: Surely, that is a quantitative question, not one of principle. There is probably an efficiency cost when the government imposes a pigouvian tax (in terms of how much it costs to administrate it etc), but whether that is greater or smaller than the benefit obtained from eliminating the externality surely must depend on the size of the externality.

2. Your examples don't make sense. Roads don't have the same externality that fuel consumption has. Roads can be used without contributing at all to climate change. There might be problems with all the examples you cite, but they are not related to the fact that fuel consumption has an externality which ought to be dealt with. I cannot follow your argument that the government is the largest polluter at all. As far as I read the stats, most greenhouse gas emissions originate from the private sector.

3: I don't understand your point. I believe there are couple of ways to look at this. One is: The polluter is the theft. He is stealing from the "clean atmosphere" by not paying for the externality he is producing. In that sense, the gov imposing a pigouvian tax is no different from the police preventing theft. In more general terms: Pollution is a classical problem of a public good. How do you suggest to deal with those? Of course, if we could assign property rights, all things would be well. I just haven't heard of any practicle proposol of how to do that with air.

4: Sorry, read a textbook in econ. A pigouvian tax will make a society richer, period. If you don't deal with externalities, you will not reach the maximum "richness" point, or the efficient solution.

5: Fair point.
At Sat Dec 09, 01:10:00 AM EST, James said...: david,

1. Mankiw hasn't made clear whether he actually accepts some norm to the effect of "do away with externalities" or not. Supposing that such a norm drives his thinking, governments, not gas use, should be his first target as they are a much bigger generator of externalities. But in any case, Mankiw is arguing that the government should implement a Pigovian tax. That's a matter of principle, not quantity.

2. My examples make perfect sense. If you disagree, quote the first sentence that you disagree with. You are correct that everyone else as an aggregate, i.e. the private sector, generates more pollution than the government. However, the largest single producer of pollution is the federal government.

3. I certainly wasn't claiming that pollution is not a form of trespass and my argument doesn't rest on any position on this.

If anyone other than the government were to take your money by force every time you bought gas, would you object even if they claimed that they were doing so for the sake of alleviating externalities? If so, is your objection based on reasons? Do those reasons pertain also to the government or not? If not, why not?

4. I can restate my position and spell out punctuation too. So what?
At Sat Dec 09, 04:27:00 AM EST, David said...: 1: Well, I am sure Mankiw is arguing that the gov should implement a pigouvian tax in those cases, in which the administration costs don't outweight the good done by the tax. I don't know his position on other measures (like shrinking the gov) but if a pigouvian tax passes your cost-benefit test, I don't see why you should then not do it only because there are other things that could be done that would improve the overal situation as well.

2: I don't think that subsidized roads are an argument that you should not have a tax on gas because those would be contradictory. Someone who proposes a tax on gas does NOT think that the externality comes from driving in general, it comes from using gas. A tax on gas would most likely lead to more use of substitudes for gas or more fuel efficient cars, which also drive on roads. I am not saying that subsidising roads is good, but I am saying that it is not in direct contradiction to a gas tax.

3: I would very much object to anyone but the police locking up thiefs. Equally, I would very much object to anyone but the government dealing with externalities.

4: Once again you are talking about the issue as if it was a yes/no question, while it really is a quantitative question. Those that favour a pigouvian tax are of course NOT saying that it should be so high that no climate change happens. They are saying that it should exactly be so high that one dollar more tax would be a bad deal because putting that dollar in a bank and later use it for adaptation would be a better deal and a tax one dollar lower would not be a good deal, because the extra damage that will happen because of the lower tax will be higher than the benefit obtained from putting that dollar in a bank and saving it for the time when the damage happens. The point is: A pigouvian tax will make a nation richer. You would only impose a pigouvian tax in a situation where you want to avoid a damage that is higher than the increase in wealth you would get by not imposing the tax.

The freeriding problem is a real one. But do not forget that in the past international environmental agreements were set up succesfully to prevent that sort of thing and it worked. The one thing that is for sure is that NO such agreement will ever come about if the US is not taking part. Plus, if this is your argument, i.e. if the only thing you have against a pigouvian tax is that for global externalities the freeriding problem is significant, I would suspect that you are not that far away from Mankiw's position (not sure though).
At Sat Dec 09, 11:53:00 AM EST, James said...: david,

You write, "I would very much object to anyone but the police locking up thiefs. Equally, I would very much object to anyone but the government dealing with externalities."

Can you answer the other questions I asked: Is your objection based on reasons? Do those reasons pertain also to the government or not? If not, why not?
At Sat Dec 09, 12:46:00 PM EST, Taylor (sologue) said...: I'm sorry, David. Can you explain how taxes make people richer? I'm not sure I get that.

Or by "nation" do you mean the government?
At Sun Dec 10, 11:30:00 AM EST, Francis St-Pierre said...: David, do you really believe that no private actors can resolve externalities, only government can?

The main reason a Pigovian tax is a bad idea is the calculation problem.

From the Mises Institute (http://www.mises.org/story/1360

A more modern Austrian approach to externalities is to show that they are impossible to calculate on a meaningful scale. Rothbard showed that traditional welfare economics was flawed because it is impossible to make an interpersonal comparison of utility. In other words, happiness cannot be measured on a quantitative scale in the same way voltage can be. This means that it is impossible to rationally calculate the utility gained or lost through government intervention.

Since the tax or subsidy proposed to correct an externality must be accomplished through some sort of government coercion, it is clear that not everyone will expect to benefit from the policy. How then are we to decide whether the policy's results add to net social utility or not? No number can be calculated, even in theory, to provide the net benefit from intervention, or even to say if the net benefit is positive or negative. One can only expect an increase in net benefit through the voluntary actions of people; an act indicates a preference for that chosen action over all other available options. The result is that the declaration of an externality is purely arbitrary.

The social justification of Pigovian taxes is definitely ambiguous.
At Tue Jan 02, 06:20:00 PM EST, Patrick Crozier said...: I am sorry if I have misread this but you don't seem to be offering an alternative. If pollution causes global warming (or whatever) then there are victims and so, shouldn't those victims receive some compensation?

The market will then decide whether pollution is (all things considered) worth it or not.

The thing is that unless it is a tax, I don't see how polluters can be made to pay the full cost of their pollution.

The alternative would be via the courts - but, surely, the sheer number of potential perpetrators and victims would make the whole process impossible?

That is why I have reluctantly concluded that a tax, albeit with very strict rules on how it is calculated and used is the only solution.
At Tue Jan 02, 07:20:00 PM EST, James said...: Patrick,

The alternative to increasing the gas tax is to not increase the gas tax.

If you were asking for an alternative means to achieve Mankiw's ends, I don't see any need to offer one, as I do not share most of his ends. The one end I do share is welfare. I see governments as a bigger threat to welfare than pollution.

Suppose, however, that I proposed the following alternative: Change all laws concerning private theft to permit theft in cases where the incident takes place at a gas station and the victim is a gas purchaser who is forced to pay the loser $1 for each gallon of gas purchased. The only variable I've changed here is whether or not the party collecting the money is acting on behalf of a government. Does this lone difference change your reaction to the proposal?

To repeat myself, if anyone other than the government were to take your money by force every time you bought gas, would you object even if they claimed that they were doing so for the sake of alleviating externalities? If so, is your objection based on reasons? Do those reasons pertain also to the government or not? If not, why not? Would you feel that your objection required that you propose an alternative?

Concerning your question, "... shouldn't those victims receive some compensation?"

In principle, sure. But (1) the burden of proof is on the complainant to establish the magnitude of damages and (2) since any individual carbon emitter is likely to be guilty of only a tiny portion of the damage to a particular complainant, we're stuck choosing between unideal alternatives. I see doing nothing as the least bad alternative, given the track record of governments and the track record of weather forecasts 100 years into the future.
At Tue Jan 09, 09:42:00 PM EST, Patrick Crozier said...: A couple of points.

Now I accept that there's a lot of hype over global warming but what if it were true? What if it were really cataclysmic? What then?

You may have written this down somewhere else and if so, I'm sorry you haven't read it but if you're not going to have a state how are you going to enforce the law in cases such as theft or, seeing as we are on the subject, pollution - where there is a polluter and the damage he causes is significant and provable?

Because, if so, you presumably have some sort of organisation that is permitted to use violence. If you are happy for such an organisation to rectify big-scale pollution why not the small-scale?
At Tue Jan 09, 10:36:00 PM EST, James said...: Patrick,

Please answer the questions I've put to you in the comment I made on Tue Jan 02, 07:20:00. We can proceed from there if you are still interested.