Arnold Kling  

Carbon Tax Club?

Competitive Manipulation... Fixed Costs: The Moderate Case...

Greg Mankiw asks everyone to join what he calls The Pigou Club, by which he means economists who favor carbon taxes or gasoline taxes. The idea is that instead of taxing stuff that we want more of--work, saving, risk-bearing--we should tax stuff we want less of--carbon dioxide emissions.

I would join, but I also think that the Pigou Club ought to be looking at subsidies to eliminate that would have the same effect. Examples might include local regulations that distort land use in favor of parking places, subsidies for agriculture or other energy-intensive industries, and so on.

Also, I do not think that economists ought to be bullied into believing in human-caused global warming. Paying a big carbon tax "just in case" is sort of like an agnostic making a big contribution to a church "just in case."

Comments and Sharing

COMMENTS (20 to date)
Matt writes:

Ignore global warming for the moment.

Who pays for the enormous cost of policing the oil lanes across the globe? Who benefits?

Our transportation infrasturcture includes policing the trade routes, building roads, oil refineries, and gasoline deliveries; in which the government, local and national, takes on a significant piece of the expenses. The government share of expenses are covered by income taxes, pay as you go gas taxes (which barely cover half of the road expenses), sales taxes for local infrastructure investment, and property taxes. The remainder is covered by private sector gas puchases.

Where is the rate of return spread out for these government activities? Start down the pyramid. On the top we have investors and management of the oil, steel and automobile industry whose return on onvestment from this government activity is larger than their tax rates. Drop down the pyramid a couple of levels and your reach a point in which the individual gets a small rate of return from this government investment, smaller than his tax rate.

If libertarians ever wonder why government takes on such a burden of supporting private sector costs just examine the varying rate of return from government activities versus tax rates. A lot of this gets cloaked in patriotism and national security, but these concepts start with corporate ideals.

kristof writes:

So if we want less poor people, we should tax the poor?

Arnold Kling writes:

Instead of "stuff," I should have used "activities." We should think of taxes as taxes on activities, not as taxes on people. Taxing activities that lead to poverty is something that makes sense. Taxing poor people per se does not make sense.

Roy Lawson writes:

Question regarding Smith, Ricardo, and Mill.

Adam Smith discussed inequalities amongst labor in the Wealth of Nations.

These inequalities include restraining competition, increasing competition beyond what it naturally would be, and finally obstructing the free movement of labor from employer to employer and place to place.

If a trading partner is engaged in such practices, doesn't that give them the "monopoly of the home market"? I have some dots to connect so to continue that thought...

If countries have unfair comparative advantages derived from disparities which are deemed unfair, would Smith, Ricardo, or Mill support a temporary duty to be imposed until such disparities are corrected?

kristof writes:

We could have lots of fun with this train of thought... if we consider dying an activity we wish to discourage, then an estate tax seems like a great idea. Perhaps it will encourage people to invest in longevity reseach...

Matt writes:

kristof has it right. Absent the estate tax, death would be slightly less appalling because you could leave the family estate intact to the heirs.

When a nation applies a VAT tax for socialized medicine, the appeal is to have fewer children and live forever on goverment medicine.

The carbon tax is a pay per use, not a mandatory purchase, making it more akin to a free market purchase of services. There is nothing right or wrong with government providing services for the transportation industry as long as it is pay per use.

JohnDewey writes:

"Taxing poor people per se does not make sense."

Arnold, I don't understand this sentence. It doesn't make sense to me for poor people to not be taxed. They enjoy much the same government benefits as wealthy people. Why shouldn't they pay for those benefits?

IMO, the biggest government problem we face is that non-taxpayers can vote to tax others. If everyone had to directly pay for their favorite program, we'd have fewer programs.

Nathan Smith writes:

Forget carbon emissions. I think cars have major negative externalities because they make cities less pleasant.

Also, isn't it sort of, er, socialist that roads are provided by the government and are available to the public for free? I realize that transactions costs and land-use issues complicate the creation of competitive markets in road supply. Still, this is a basic gap in the capitalist nature of our institutions. And it creates the problems you'd think. Huge traffic jams are the American equivalent of the notorious lines and shortages of the late Soviet Union: a good that is systematically over-priced is systematically under-consumed.

What this really argues for is not a carbon tax per se, but rather a system of charging people for the use of roads. My uncle thinks it would be feasible to create sensors that would take snapshots of people's license plates and charge them for road use. This sounds Big-Brother-ish to me... but still, I'd support it. After all, we have to get driver's licenses anyway, and that's a little bit Big-Brother-ish. Road charges, rather than gas taxes, would ease traffic in city centers without penalizing people who want to burn lots of gas on rural toll roads.

I think the City of London has done something like this. Might be worth seeing whether it's something libertarians should support.

JohnDewey writes:

"What this really argues for is not a carbon tax per se, but rather a system of charging people for the use of roads."

I believe the gasoline tax is a user fee. Most highways are paid for using gasoline tax receipts. Politicians have distorted the process somewhat by diverting a small percent to mass transit.

"I realize that transactions costs and land-use issues complicate the creation of competitive markets in road supply."

I oppose eminent domain in general, but I can easily support the practice for road-building. IMO, the benefits of unrestricted transport far outweigh the costs of government-designed highway systems. Without eminent domain, we would never have developed the straight-line interstate highways we have now.

" My uncle thinks it would be feasible to create sensors that would take snapshots of people's license plates and charge them for road use."

Perhaps, but such a system could be easily defeated. The gasoline tax is far simpler and far less costly to implement and police.

"and that's a little bit Big-Brother-ish."

The privacy implications will prevent implementation of such a system in our lifetimes.

Matt writes:

Two arguments against government nationalization of the road system are that it is uncompetitive and not "pay per use" financed.

The carbaon tax takes care of the second problem.

To argue the first problem from a libertarian viewpoint is too little too late. The goverment as gotten pretty good at building and maintaining roads and most of our national highway system as been built. Competitive private sector roads is sort of a non-starter for libertarians, we have better dogs to kick.

Regarding te taxing of the poor for goverment services. The poor generally get a raw deal on government services, in my opinion.

Roger M writes:

Who pays for the enormous cost of policing the oil lanes across the globe?

A lot of people seem to think the last two wars in the Persian Gulf were about protecting our access to oil, and advocate a tax on oil to pay for it. I'm not convinced of the war for oil reasoning, but just for the sake of arguement, let's say it' accurate. In that case, I would ask were they necessary? Do we need to protect access to Middle East oil when oil is fungible and the producers need to sell it more than we need to buy it? Did the latest invasion of Iraq succeed in making oil more accessible and cheaper?

If it's true that we fought two wars to maintain access to Middle Eastern oil, then our response should be to stop it, not tax it. Middle East oil doesn't need protecting.

JohnDewey writes:

"Regarding te taxing of the poor for goverment services. The poor generally get a raw deal on government services, in my opinion."

How is that?

How do the poor benefit less than the non-poor from defense against terrorism, highway funding, and food and drug testing?

Given that the poor suffer more from violent crime than the non-poor, doesn't apprehension of criminals by the FBI help them more?

Isn't it only the poor who benefit from subsidized housing and food stamps and Earned Income Tax Credits?

I think every state has government programs to help the "needy" with the cost of healthcare. I don't think the non-poor have such programs.

The U.S. Department of Education has established Educational Opportunity Centers all over the nation to assist low-income families in paying for higher education. How is that a raw deal?

Do you mean that the poor get a raw deal because they become dependent on government welfare? and lose the ability to take care fo themselves? If that's what you mean, then I can agree.

Kevin Feasel writes:

Here's my beef with the Pigou Club and therefore with Pigouvian taxation in general: it posits a social welfare function which does not exist. The entire concept of Pigouvian taxation is based around the idea that (due to externalities) the marginal cost of a good for each individual equals the marginal revenue, but the social marginal cost is not equal to the individual's marginal cost. But in order for this to be so, this implies that a social welfare function exists, and this is a highly problematical notion. There are two ways that I can see developing a social welfare function: positing as an external norm, or through contractarian agreement.

An externally-imposed social welfare function is entirely arbitrary and useless for purposes of scientific argument, is impossible to effectively refute (because a social welfare function can neither be observed in practice nor directly imputed from choice behavior, given that "society" is not an actor) and also smacks of paternalism.

A contractarian agreement would be superior in that it is endogenously determined through the voice competition of the participants in a society rather than as a result of the thoughts and arbitrary implementation of some intellectual (or politician), but even it has major flaws. Once again, the social welfare curve is neither directly measurable nor directly imputable from the decisions of the citizenry. As such, we can't measure exactly where to tax what by how much to match up the marginal revenue to the social marginal cost, even if you know marginal revenue. If you can't measure the marginal cost, you can't know where to set the price in order to get your social welfare function to match up at the right point. And there's also the fact that things like preferences, technology, and circumstances are constantly changing, so an effective Pigou tax would have to be constantly adjusted to keep in line with this pre-defined social welfare point.

There is admittedly the inability to measure directly marginal costs on the individual level, yet there exists a well-functioning (within reason) mechanism that deals with matching up marginal revenue and marginal cost. But the answer there is that this is possible only because nobody _tries_ explicitly to do that, and is the cornerstone of the Hayekian division of knowledge: all those individual cost curves are subjective knowledge and could not be aggregated into an ex ante market-clearing price in a dynamic world. Markets work precisely because there isn't some grand external price setter who uses the powers of his mind to determine what all the marginal costs are and where they will meet up with the marginal revenues of firms.

However, this is definitionally impossible within the social welfare function. The entire idea of externalities is that aggregated individual choices end up at a point different than the "social optimum," and there is no entreprenuerial knowledge or mechanism which allows an effective matching to that point.

So in the end, I don't think Pigouvian taxation could be a quality solution to any externality issues, and I find them quite paternalistic--the government is, after all, telling me what I should purchase, when I should buy it, and how much I should buy; I personally believe that I and most others are fairly competent and can make sound decisions. If an externality exists, the best way of resolving the issue that I can see is to, on the constitutional level, extend the scope of property rights to encompass this externality. This would be extremely difficult with things like air pollution, but then again, it's easier than calculating a non-existent figure...

Nathan Smith writes:

Now hold on, Matt McIntosh. We shouldn't accept the "pay per use" function of carbon taxes as a substitute for user fees for roads quite so easily, because it isn't really a very good one.

Cars use land. The rental value of land varies radically. If I'm driving in downtown Washington, DC, or downtown LA, the rental value of the land under me is huge. If I'm on a highway in the boondocks, cruising at 80 miles per hour, the rental value of the land under me is negligible.

A carbon tax charges me equally for cruising on rural highways and for city driving. This is a major failure in the encouragement of rational land-use.

As for political feasibility, this is a case-- like civil liberties-- where there's a natural alliance between libertarians and the left. That can be an opportunity to build cross-partisan credibility... while risking the libertarian-conservative alliance, of course. Anyway, if you're a libertarian, you have to kind of enjoy a certain unpopularity, don't you? Wouldn't it be fun to walk around all the time calling free public roads "socialism"?

JohnDewey writes:

Nathan Smith,

That's an interesting idea: including land rental costs in the user fees paid by automobile users.

Consider these thoughts:

1. The amount of land consumed by urban highways is often far less than that used by rural ones.

2. Traffic counts on urban freeways are many times those of rural ones, indicating there are at least dozens of times as many vehicle owners to share the land rental cost.

3. In most cases, the value of land alongside a freeway increases by fourfold and sometimes tenfold as a direct result of the freeway itself. It is the existence of the automobile traffic that drives up the value of the land. New highways that take land off the tax rolls actually increase property taxes overall. There are few land uses other than highways that provide as high an increase in land values.

4. Land owners were already compensated by the public for the real value of their land when eminent domain was exercised.

I can't think of a reason to tax highway users for use of highway land. Can you help me understand this issue a little better?

[N.B. JohnDewey: Your posting privileges have been suspended on EconLog for rude behavior on other threads. You've been notified in various ways. Please email us at to reinstate your posting privileges.--Econlib Ed.]

Max writes:

My question would be, why Economists tend to leave free market principles and liberty, whenever they believe that something has to be done to set "the market right", whatever that means.
Thinking like this has brought us the Soviet Union and European's totalitarian social-democracies...

Don't economists have a spine or do they just tend to forget about history?

Nathan Smith writes:

John Dewey:

Sure, happy to help.

re: 1. Even if urban roads use less land, the land is orders of magnitude more valuable.

2. Even if traffic counts are dozens of times higher, I'm not sure this offsets the difference in land value. Land in downtown Manhattan or Washington, DC, is-- this is an educated guess based on real estate prices-- worth millions of dollars an acre.

3. That rural highways increase land value actually reinforces my claim that urban drivers, but not rural drivers, should be charged rent for their road use. If the land (rural) highways are built on is valuable, it is valuable mainly due to the highway. Drivers are making the land more valuable-- if anything, they should be charged negative rent! (Which is another reason that a carbon tax is a bad substitute for a land-use tax; driving on rural highways may be something we want to encourage, from some points of view.) Urban land, on the other hand, does not need roads to make it valuable.

4. That land owners were compensated is like arguing that there's nothing wrong with nationalizing the restaurant industry and give food away free, if we compensate restuarant owners. A socialized restaurant industry would not operate efficiently. Likewise I question whether mass driving in urban areas represents rational land use, as opposed to an economic distortion induced by the government's provision of free road space.

My own view is that pedestrians, bicycles, trolleys and some other forms of transportation create positive externalities (community, cuteness) while cars produce negative externalities (traffic jams, danger, isolation). Government provision of free roads is like smoke-belching Soviet factories-- it's a socialist policy with large negative side-effects for society.

JohnDewey writes:

Nathan Smith: "That rural highways increase land value actually reinforces my claim that urban drivers, but not rural drivers, should be charged rent for their road use."

All new highways and highway expansions, rural and urban, increase total value of surrounding land. Urban and suburban highways increase land values even more in absolute dollars than do rural highways, with a few exceptions.

JohnDewey writes:

"That land owners were compensated is like arguing that there's nothing wrong with nationalizing the restaurant industry and give food away free, if we compensate restuarant owners. "

I don’t think that’s a valid argument. Restaurants already possess their value as restaurants. Government would not add anything through nationalization.

Government did not nationalize any existing highways, at least not in my lifetime (55 years). Government paid market value for existing land, built highways on it, and increased the value of all other surrounding land. Furthermore, it did so:

- using vehicle drivers' funds;
- at vehicle drivers' request, through their elected officials;
- for the primarily benefit of those very same vehicle drivers.

The land has already been paid for by the vehicle drivers. Maintaining the highways is funded by fuel taxes plus other taxes on the general public - almost all of whom are vehicle drivers.

Why should government force vehicle drivers to pay again for land they've already purchased and they continue to maintain through their taxes?

JohnDewey writes:

"Government provision of free roads is like smoke-belching Soviet factories-- it's a socialist policy with large negative side-effects for society."

What are the negative side effects of highways? Highways prevent congestion. Reduced congestion leads to less pollution, not more. Highways everywhere lead to faster commutes and faster transport of goods. Highways thus provide a higher quality of life and increased standard of living.

Highways are not free. Every vehicle owner pays for the highways every time he fills up at the pump.

The only free ride is being given to rail transit users. They steal from the highway fund to heavily subsidize ineffective light rail and commuter rail lines.

Those who favor mass transit should advocate bus rapid transit (BRT). Not only is it cost effective, but BRT shares at very low cost the same highways (and HOV lanes) that transit advocates criticize.

Comments for this entry have been closed
Return to top