Thursday, January 30, 2014

Me vs Mankiw on Global Warming

A commenter on a old post of mine about global warming points at a response by Greg Mankiw. Both are from 2007. The issues have not changed since then and, while I responded to Mankiw at length by email, I do not think I ever did so publicly.

Mankiw supports a carbon tax. I argued that while a carbon tax might make sense as the answer to the question "what is the best policy for dealing with negative externalities due to global warming," it did not make sense as the answer to the question "what policy should economists support to deal with global warming" since there was no good reason to believe that, if a carbon tax was implemented, it would take the form economists would recommend. Interested readers should probably read both my post and Mankiw's before going on to my response below, which I have copied from my email correspondence with him.


It seems to me that you are making the error that was the norm in textbooks and the profession fifty years ago, before public choice theory. You are evaluating proposals for government policy on the basis of what they could do if optimally implemented not on what one can expect them to do given the incentives of the people making the decisions—what used to be referred to as the philosopher king model of government. It makes no more sense than evaluating the market alternative on the assumption that all the decision makers in that case will act to maximize social welfare rather than in their own interest. The question is not whether an optimal carbon tax designed and enforced by wise and benevolent economists would produce net benefits—very likely it would. It's whether passing a carbon tax designed and implemented as we can best expect it to be would produce net benefits.

Two further points with regard to your original blog post:

1. I wasn't making a slippery slope argument. If I had been, I would have argued that carbon taxes would initially be a good thing but would set the precedent for other bad things. In fact I argued that, as implemented, they would probably be a bad thing. As I made explicit in my post, it was a public choice argument—completely ignored in your response. I plan to send in my complaints to the Society for the Protection of Straw Men just as soon as I can find their email address.

2. My argument is  consistent with my father's views. For evidence, take a look at the discussion of professional licensing in Capitalism and Freedom. The argument is not that professional licensing, applied by wise and benevolent officials, could do no good. It is that we can expect, on grounds of both theory and evidence, that professional licensing will usually be controlled by the profession and used to restrict entry and raise prices.

If you look again at the quote from him you link to, he isn't saying that one should recommend policies independent of how one thinks they will be implemented—consider the "in light of what can be done." Professional licensing that isn't captured can't be done, or at least not reliably done, on the evidence. He is saying that one shouldn't refrain from making a proposal merely because you think it can't be passed.

In this context, the implication is that one might argue that a specified form of carbon tax would be a good thing and simultaneously that any carbon tax that could be passed would be a bad thing. I don't believe that is your position.

And, again:

Can you see any hint of evidence that the people proposing cap and trade have made any effort to estimate marginal cost of reduction of carbon dioxide, optimal level of emission, or any of the information necessary for a scheme designed to actually produce net benefits?

Isn't that question relevant to how one can expect a carbon tax (or cap and trade) to be implemented, and isn't that relevant to whether one ought to be in favor of it?

Labels: , , , , ,


At 5:01 AM, January 31, 2014, Blogger RJM said...

Ha, there's some irony:

Mankiw says, D. Friedman makes a 'slippery slope' argument. Now I have heard of this Slippery-Slope-Fallacy. I would describe it as "argument X is wrong, because thinking that way leads to argument Y which is DEFINITELY wrong (and bad and evil)".

In this sense neither Friedman's argument nor "Road to Serfdom" qualifies as slippery-slope-fallacy. Opposing measure X based on the consequences of it does not qualify as fallacy.

And in the end, Mankiw quotes a guy called Volokh with something very close to a slippery slope fallacy. "When you accept argument X (government should not do this), you will ultimately accept argument Y (government should not do anything), which is definitely wrong".

Or is there some sort of sarcasm I missed?

At 5:42 AM, January 31, 2014, Blogger kirill predko said...

To see why David is correct after all see "specific models" section of wikipedia story on negative income tax... In particular the following part "...The negative income tax has come up in one form or another in Congress, but (Milton) Friedman eventually opposed it because it came packaged with other undesirable elements antithetical to the efficacy of the negative income tax..."

At 4:47 PM, February 01, 2014, Blogger Will McLean said...

Consider a carbon tax imposed by real world humans, and so suboptimal. It will either be too low or too high. If too low, it is better than no carbon tax at all. And what is the chance that the tax will be too high, given the concentrated interests who will try to make it as low as possible?

At 1:15 AM, February 02, 2014, Blogger Bill Friedman said...

@RJM: I thought there was some irony; Mankiw is basically saying, "If you subscribe to the slippery slope argument, you must believe it when it applies to yourself."

I mean, it isn't a slippery slope argument that my father is making, but other than that Mankiw's logic is still sound, if ironic.


What is the chance it will be too high? Very good, given the concentrated interests who try to make it as high as possible.

It would be a compromise - or a war - between two parties, one a coalition of religious environmentalists and manufacturers of competing energy sources, and the other a coalition of manufacturers of carbon producers and "If it ain't broke, don't fix it" conservatives. I can't say who'd win the war between them as to the level of the carbon tax, but I'm not betting on either side.

But you're missing two things. The first of which is that Coase + Pigou leads to inefficient outcomes, as my father has explained elsehwere on his blog, and the second is that all taxes have a cost to administer, just because of the additional forms that need to be printed out and the additional bureaucrats you need to hire and the additional forms that need to be filled out. Even if a perfectly Pigouvian tax is enacted, it still has some costs.

So a suboptimal tax will still not necessarily be better than nothing.

Bill Friedman

At 10:38 AM, February 02, 2014, Blogger David Friedman said...

Will: To begin with, if the net externality is positive, any carbon tax will be too high.

Further, if you look at the actual cap and trade bill that passed the House, it combined the equivalent of a carbon tax with a good deal of direct regulation plus sizable subsidies to various groups in the form of allocations of permits.

It's not at all clear that interest groups will push it down rather than up. If it takes the form of cap and trade, it might increase profits of the energy companies, since it raises marginal cost more than average cost--work through the logic for yourself. If it's a direct tax, that provides revenue which lots of interest groups would like an opportunity to spend. And it benefits various interest groups that gain by high energy prices--producers of electric cars, producers of housing insulation, probably others I haven't thought of.

At 11:29 AM, February 03, 2014, Anonymous Beliavsky said...

Thanks for your post. Typo in your label -- "Mankiw" not "Makiw".

At 6:10 PM, February 09, 2014, Anonymous Mark Bahner said...

"To begin with, if the net externality is positive, any carbon tax will be too high."

Indeed. Consider that most people would probably agree that the world climate during the Little Ice Age was worse than at present. So most people would probably agree that the ~1 degree Celsius that the world has warmed since the 1880s is probably a good thing.

Now, what are the odds that somehow we have warmed exactly the best possible amount, and that all warming from here onwards has a net negative effect?

At 11:01 PM, May 16, 2014, Anonymous macsnafu said...

I like the invocation of Public Choice in the argument against the carbon tax, and David said some other good things, too, but I think he merely hinted at an argument that could have been stronger.

Even without Public Choice, I question whether a "benevolent dictator" could impose a beneficial tax. David merely hints at the real problem when he says "they would not be based on a realistic estimate of the marginal costs;"

A fuller argument would say that governmental policies such as a carbon tax are not merely motivated by political reasons, but that they lack the proper economic incentives and feedback to guide the policy makers into fine-tuning such a tax to do as much good as possible.

It is precisely because they are coercive policies, not voluntary, that they lack the proper economic incentives and feedback.

A clearer example would be the difference between licensing and certification. Since a license is a mandatory requirement by government, there is no reason to believe that the licensing requirements would be proper and appropriate for the circumstances involved. More likely, they will be too difficult (and not necessarily just because of a desire to restrict entry), or too easy, and thus of little value.

Worse, because licensing is a political creation, making changes to it, either to fine-tune it or to adapt it to changing conditions and cicumstances, is also a political matter, and difficult to do.

Certification, on the other hand, involves persuading people to voluntary seek out and pay for certification, because said certification offers them some perceived value or benefit.

Making certification too easy dilutes the value of certification; making it too difficult limits the certifier's clients--there are economic incentives and feedback to make certification as useful and beneficial as possible, moreso because it is voluntary, and there could be competing certification agencies.

Admittedly, while this points out the flaw of the carbon tax, or any "social engineering" tax, I must admit I'm not quite sure exactly what the voluntary alternative to the carbon tax is. Maybe someone could help me out on that point.

I don't think this is a moral argument. While I certainly hold the libertarian view that the initiation of force is immoral, saying that coercion short-circuits the economic incentives of the market is a statement of fact, not morality.


Post a Comment

<< Home