Saturday, October 18, 2014

A Case of Posthumous Conscription

A recent Forbes article is headlined "What Would Milton Friedman Do About Climate Change? Tax Carbon." It reports on a forum at the University of Chicago at which several economists, including Michael Greenstone, described as the "Milton Friedman Professor of Economics at the University of Chicago," argued that Friedman would have supported a carbon tax. The evidence for that claim was a 1979 clip from the Phil Donahue show where Milton Friedman argued that if the government is going to do something about emissions, they should use an effluent tax rather than direct regulation. He does not actually say that government should do something about emissions, only that there is a case for doing so and, if it is done, the best way to do it is by a tax on emissions.

To get from there to the conclusion that he would have favored a carbon tax requires at least one further step, a reason to think that he would have believed that global warming due to CO2 emissions produced net negative externalities large enough to justify doing something about them. The problem with that claim is that warming can be expected to produce both negative externalities such as sea level rise and hotter summers and positive ones such as longer growing seasons and milder winters. The effects will be spread out over a long and uncertain future, making their size difficult to estimate. My own conclusion, defended in past posts here (one example), is that the uncertainties are large enough so that one cannot sign the sum, cannot say whether the net effect will be positive or negative.

I do not know if my father would have agreed but I have at least a little evidence on the subject, more than offered in the Forbes article. The same issue arose in the earlier controversy over population. Just as it is now routinely assumed that warming is bad, it was then routinely assumed that population increase was bad. Forty years ago I wrote a piece on the subject for the Population Council in which I attempted to estimate the  externalities associated with population. I concluded that they were too uncertain for me to tell whether the net effect was good or bad. My father read the piece and commented on it. If he had disagreed he would have said so, and he did not. It is possible that he would have felt differently in the case of climate change, but I can see no reason to expect it.

The article quotes professor Greenstone on the uncertainty:
Estimating the cost is tricky, Greenstone said, but scientists and economists have models for projecting the cost of each added ton of carbon on agricultural losses, mortality, sea-level rise, storm surge, and other climate effects.

It’s a complicated task but I think the best evidence suggests that it’s probably around $40 a ton,” he said. The U.S. government has projected the cost of carbon emissions at $37 per ton.
Current estimates of climate sensitivity, the effect on temperature of an increase in CO2, vary by more than a factor of two. One would expect the size of the externality due to an additional ton of CO2 to increase with the temperature increase. A further uncertainty, reflected in the various scenarios of the IPCC report, is the amount of CO2 that will be emitted over the next century. Lockheed Martin has recently claimed that it will have a working fusion reactor in the near future. I have my doubts that it is true, but if it is, the result should be to reduce CO2 emissions over the course of the next few decades to between half and a quarter of what they would otherwise be. That would sharply reduce warming and thus the cost of additional CO2.

One would expect similar effects from any substantial reduction in the cost of other alternatives to fossil fuels, such as nuclear or solar power, or from a substantial increase in the cost of fossil fuels due to the exhaustion of the more readily accessible sources. Additional uncertainties are associated with the relevant climate science. The IPCC, for example, claimed in its fourth report that warming increased drought, retracted that claim in the fifth report.

Whether or not my view that we cannot sign the externality is correct, I would be very surprised if Professor Greenstone could justify his confidence in the specific number he offered—which happens to be close to the official government estimate. I would be equally surprised if he could offer evidence that Milton Friedman would have taken seriously a government estimate of an uncertain number offered in support of a policy the current administration favored.

Before they died, my parents created a foundation to promote the idea of school choice. One of the terms on which they created it was that the foundation was to end a fixed number of years after the last of the founders died. The reason for that was my father's concern, possibly based on the examples of the Ford and Rockefeller Foundations, that once the founders were no longer around their names would be used in support of policies they themselves would not have supported.

Of all my father's accomplishments, I believe the one he was proudest of was his role in ending military conscription. I do not think he would be happy to be conscripted, posthumously, for someone else's cause.


P.S. Robert Murphy points at evidence against the claim that my father would have supported a carbon tax. In a 1999 comment to a recently published book, he wrote:
This encyclopedic and even-handed survey of the evidence of global warming is a welcome corrective to the raging hysteria about the alleged dangers of global warming. Moore demonstrates conclusively that global warming is more likely to benefit than to harm the general public.
It is possible that between then and now he would have reversed his view, but I can see no reason to expect it.


Anonymous said...

Love the ending to this post!

Anonymous said...

The science is too uncertain to act upon, in essence.

You're quite wrong, David.

Despite what Judy Curry now says.

David Friedman said...

GM: Two different questions

1. Was the claim about what my father's view would have been justified? I think I show it is not, hence that it should not have been made.

2. Is my view about the situation justified? That's a separate issue, one I have posted on a number of times.

One point you may be missing, and many of the people commenting on the subject are missing, is the implication of uncertainty. It isn't sufficient to calculate the expected cost of warming and then take action as if that was the certain cost. The more uncertain the cost of warming is, the stronger the argument for waiting until more information is available before taking any costly action to prevent it.

Anonymous said...

Your argument, David, rests on the assumption that the science is too uncertain for anything that necessarily follows from it. Unless we're willing to make decisions about policy without input from the science, that is - we're free to do that, but that's unwise.

My argument is that the science is certain enough to accept it as given and then go from there. Continually going back to square one and raising doubts about the science is fruitless, because such doubts aren't legitimate.

The other issue is that if one wants seven opinions on a subject involving costs and benefits, one can ask two economists. Given the inability of economics to model even the simplest application of its theories in the real world, the fail here isn't in climate science, it's in economics, and within economics, the market model. Markets simply cannot handle negative externalities well and market proponents often handwave them away, obviously or more subtly, presuming that such handwaving suffices for empirical evidence.

David Friedman said...

GM: I don't know if you have read my earlier posts. Some of the uncertainty comes from known uncertainty in the science, most obviously the range of sensitivity values recognized by the IPCC. But most of it comes from other sources.

Anonymous said...

David, here's an article you may want to consider:

Anonymous said...

"Your argument, David, rests on the assumption that the science is too uncertain for anything that necessarily follows from it. Unless we're willing to make decisions about policy without input from the science, that is - we're free to do that, but that's unwise.

My argument is that the science is certain enough to accept it as given and then go from there. Continually going back to square one and raising doubts about the science is fruitless, because such doubts aren't legitimate."

GM: Scientists make predictions about what might happen. Uncertainty means that many different things might happen, certainty means few things might happen. If the science indicates that many things might happen then by believing the future is uncertain you aren't throwing away science you are recognizing science. If science indicates that only a few scenarios are possible then using science is planning for that small set of scenarios.

David's argument, as best as I can tell from having read some of his posts, is that using mostly standard scientific sources you get predictions about a wide-range of possibilities. A second claim that he's alluded to is that outside of scientific sources but in editorials and political documents you find claims about climate where there are only a few possibilities.

The linked article (on the second page) claimed that the climate science literature recognizes uncertainties--so far no particular dispute with David on that point.
The next paragraph compared climate science to Newtonian mechanics: that like Newtonian mechanics was good enough to build bridges, climate science is good enough to act now. Its an odd comparison because Newtonian mechanics is extremely accurate for predicting all sorts of phenomena whereas climate science gives fairly wide error-bars for predicting future temperature.

The question of whether it has been "settle enough" for action right now is not strictly a scientific question. Doing so requires weighing the potential cost of missing a crucial window of action versus waiting longer and collecting more observations so that we can better target the problem. Obviously in a certain regime considering action right now is reasonable. But, I don't know that the linked article excludes the possibility that a strong scientific case can be made for uncertainty.

Daublin said...

GM, you appear to have not read the post. The post is about what Milton Friedman would have said about CO2 controls, and you are filling up the comment space with what *you* would like to say about CO2 controls.

Anonymous said...

@Anonymous, why do you presume the uncertainty is one-sided, towards no problems or easily adapted-to problems, from manmade climate change?

If anything climate science is conservative in its projections.

BC said...

Another area where your father seems to often be posthumously conscripted is on immigration. Immigration opponents often claim that your father believed that a generous welfare state implied that immigration needed to be severely restricted. Based on this YouTube video [], however, I believe his argument was that a generous welfare state should still allow a lot of *illegal* immigration to gain the benefits of immigration without allowing too many immigrants to receive welfare. He presented this as an argument against generous *welfare states* because allowing illegal immigration would undermine morality and respect for the law. His argument, though, implied that the social benefits of illegal immigration outweighed the undermining of respect for law. (Otherwise, why allow more illegal immigration?) It seems especially pernicious for immigration restrictionists to cite your father since they also often advocate that loosening legal immigration restrictions can occur only after we "secure the border", i.e., reduce illegal immigration. That seems to be the opposite of his argument!

This conscription seems to occur very frequently and is far too infrequently challenged.

David Friedman said...

GM: The simplest test of whether climate science, as represented by the IPCC reports, is conservative in its projections or the opposite is to see how past projections did. The answer is that they have consistently overpredicted warming. I provided the details in an old post here—search the blog for "IPCC Predictions."

David Friedman said...

BC: I agree with your general point. I've argued that you can run the causation the other way. The freer immigration is, the less politically popular generous welfare payments will be, so people who disapprove of the welfare state should support freer immigration.

Which may explain why many people who approve of the welfare state don't.

In The Machinery of Freedom I proposed free immigration combined with a long waiting period before becoming eligible for welfare—and a tax reduction to compensate new immigrants for the fact that they were not getting one of the things taxes were paying for.

Gordon said...

Given his bio and publications, Greenstone does not seem the most apt choice for the "Milton Friedman" chair. I take it that the family did not play a role in the establishment of the chair? Perhaps the law should recognize moral rights in names, as it does for artwork. ;-)

David Friedman said...

Gordon: You are correct.

Anonymous said...

Reducing CO2 emissions to combat climate change is a public good which rich countries buy into as an indulgence or luxury - as soon as it causes real hardship it is quickly abandoned. For evidence of this look to Germany which was a leading green industrial power that is going to expand coal mining to reduce energy prices: nobody is going to let granny freeze to death because Mauritians that have yet to be born may be better of for it assuming everybody also lets their grannies freeze.

The total cost of global warming is also quite uncertain: the IPCC estimates are unpleasant not catastrophic and a healthy global economy powered by cheap energy is more likely to provide developing countries with the tools they need to weather any changes in the climate (to say nothing of alleviating day to day poverty).

More practical goals to pursue for the climate concerned, rather than hoping people will give up cheap energy, are looking to increase efficiency, funding research into fusion, supporting nuclear powers, and encouraging governments to embrace open borders so that populations may be redistributed as is necessary with minimum fuss.

As somebody who believes that climate change is real, and probably quite unpleasant, I've never got how lowering emissions was ever supposed to be viable for anything short of a strong global government.

Anonymous said...

Nevertheless, for those who are convinced that climate change needs to be slowed down (it's apparently too late to reverse it in this century), they do have Milton Friedman's endorsement of a tax as opposed to direct regulation. Good enough for me. I care not that David wishes to ignore scientific consensus on the matter.

David Friedman said...


What is your evidence for scientific consensus on the claim that warming will have very bad net effects? All of the papers I have seen which provide figures on what percentage of researchers or articles agree are looking at a much weaker claim, typically that humans are one cause of warming.

That then gets misrepresented, in the public argument, into consensus on much stronger claims.

BillO said...

GM posits that the science is certain enough. He is wrong. The IPCC has consistently revised downward its estimate of climate sensitivity. The IPCC relies on about 50 models, each of which describes how the climate system is supposed to work. It the science was certain or settled, model builders would not need that many models, all of which have over predicted the increase in temperatures over the past two decades.
The 17 year pause, that we are now experiencing, is not what climates said would happen.

David Friedman said...


One disagreement. I don't think there has been a 17 year pause. While I'm not sure, I think that meme comes from the claim that if you fit a straight line to the past 17 years of data, the slope is not significantly greater than zero. That implies not that we know there has been a pause for that long but that we cannot be sure there has not been, a much weaker claim.

Looking at the data, I think the actual pause only starts about 2002. Earlier claims ultimately depend on one anomalously high year.

Bob Murphy said...


(1) I think M. Friedman was talking about generic pollution on Donahue, and not about CO2 right? I think you are conceding too much to the other side the way your post stands (if I'm right).

(2) Did you see his blurb for the back of a CATO book in 1999? I relay it here.

Bob Murphy said...

One last thing: You can't take the economics out of "the science" when trying to assess whether the net impact of manmade climate change is a positive or negative externality. I would think just a simple reflection on the meaning of the terms involved should drive home that point.

But yes, I agree economists are annoying and can't give a straight answer on anything. (See, I did it myself in this very comment.)