RWCG


Why do economists think they understand climate models?
January 31, 2014, 9:54 am
Filed under: Uncategorized

David Friedman responded to Greg Mankiw on a carbon tax:

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. [...] 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.

It is an important broader point Friedman makes about this ‘philosopher-king model of government’, and I am clearly more on Friedman’s side of this debate than on Mankiw’s. But the reality is that they are both wrong.

It is not ‘very likely’ that an optimally-implemented ‘carbon’ [sic] tax would produce net benefits, unless the sign of the externality produced from ‘carbon’ [sic] is known (or very likely) to be negative.

DEAR ECONOMISTS: You do not know the sign of the externality produced from ‘carbon’ [sic]. That is not among your areas of expertise. Please stop acting as if it is. Thanks.

What a fair-minded economist could say is something like:

If it is true that, as we are told by a different scientific field whose validity we have NO capacity or ability to evaluate, ‘carbon’ [sic] produces a negative externality, then an optimally-implemented carbon tax would produce net benefits. If not, then not.

Economists should highlight when their pronouncements rely crucially on results handed to them from external sources whose validity they are not qualified to evaluate. You know, like any other scientific field would.


UPDATE: My thanks to Friedman who kindly & patiently explains in comments that when he says ‘optimal carbon tax’ he means a carbon tax calibrated to the externality whatever, and whatever sign, it is (under the assumption that it could be known). So I’ve mischaracterized his position above and I was wrong to do so. He is also right to point out that while economics doesn’t play a role in predicting the climate, it of course plays a role in calculating the costs-given-climate under whatever climate scenario. Of course, I never meant to imply otherwise, this was just bad communication on my part.

About these ads

7 Comments so far
Leave a comment

“a different scientific field whose validity we have NO capacity or ability to evaluate”

Anyone with an understanding of the scientific method and an ability to think critically and rationally has the capacity and ability to evaluate the validity of any scientific field.

Any such person who does so to climate science will be shocked (shocked!) at how little validity there is behind the headlines written about it.

Comment by eddie

Just to be clear (“let me be clear”), I’m trying to distinguish between the validity of a field and the correctness of its results. Laymen can assess the first but not the second.

But that’s just a terminology quibble (wrapped around a substantial but tangential point). I get what you were getting at in your post, and you’re correct. As always. Of course.

(mostly always)

Comment by eddie

You are right and I stand corrected.

Yes, they have the capacity and ability. But it is presumably not what they have spent their time thinking about. What I should say is that they do not have the expertise, as drawn from study, to know whether/to what extent the claims of climate science they rely on are valid.

(Sure, there are exceptions.)

Comment by Crimsonic

Importantly, they skip (or fudge) the valuation step, as Robert Murphy discussed:

http://consultingbyrpm.com/blog/2013/08/see-the-knots-the-federal-government-is-in-regarding-social-cost-of-carbon.html

Comment by RPLong

True, there can be shenanigans and even cherry-picking with regard to how they discount future costs. (It sounds like what Murphy described there is more a case of fudging/searching under-the-streetlight using the data they had.)

What I am saying here is one step before that step: that the costs themselves are not known with sufficient precision to make such a calculation meaningful (at whatever discounting). The thing they are discounting, the ‘social cost of carbon’, comes mostly from some climate model’s estimate of what carbon [sic] does to the future climate (then multiplied by some estimate of what that future-climate ‘costs’). But those estimates could be (are likely to be) wrong, and economists wouldn’t know one way or the other, because at the end of the day it’s a string of numbers entirely handed to them by folks from a different field. They must be taking it as a given – by necessity. That calculation by itself is not part of economics.

Which is ok! It’s just that they should acknowledge and highlight this fact more prominently, instead of saying unqualified things like ‘…whether an optimal carbon tax designed and enforced by wise and benevolent economists would produce net benefits—very likely it would…’.

The really interesting thing that Murphy brings up though is that per their calculation, apparently the social cost of ‘carbon’ [sic], is *negative*, at least near term:

“climate change is actually beneficial, according to the government’s own suite of models, for the next few decades”

I hadn’t known this. Or rather, I might well have claimed it off the cuff by hunch, but fully expecting to be shot down by Smart People in blog-comments. So that’s pretty funny, but then it’s even harder to square with a claim like carbon-tax being ‘very likely’ to be beneficial.

Comment by Crimsonic

I specified an *optimal* carbon tax. In order to have an optimal carbon tax you obviously need to know the size of the externality. As you can check by looking at other posts on my blog, I have pointed out repeatedly that we do not know either its size or sign. I have made the same point in several public lectures and one of the draft chapters for the new edition of my first book.

http://www.daviddfriedman.com/Machinery_3d_Edition/The%20Misuse%20of%20Externality%20Arguments.htm

It follows that an optimal carbon tax might be zero or negative. If zero it produces neither benefits nor costs, but I think zero is unlikely—and I said “very likely” not “certain.”

I made the same point—the uncertainty of the sign of the externality—more than forty years ago in the context of arguments about population:

http://www.daviddfriedman.com/Academic/Laissez-Faire_In_Popn/L_F_in_Population.html

I should add that it is not true that what we are told about the externality comes from “a different scientific field whose validity we have NO capacity or ability to evaluate.” Even if it were true that economists have no ability to evaluate the work of climate scientists—a considerable exaggeration, as I think I have demonstrated on my blog—climate science by itself can at most tell us what will happen to climate. Getting from there to an estimate of the sign and size of the externality requires the use of several other fields, one of which is economics. My own doubts on the subject are mostly based not on my conclusions about the climate science but on what I see as the implications of the most widely accepted climate projection—roughly speaking, three degrees Centigrade of warming and a foot or two of sea level rise by the end of the century.

For more of my views on these issues, see:

http://daviddfriedman.blogspot.com/search?q=global+warming

Comment by David Friedman

Thanks for the clarification and right you are on both of these points. I’ll update the post.

Comment by Crimsonic




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s



Follow

Get every new post delivered to your Inbox.

Join 455 other followers

%d bloggers like this: