A sample text widget

Etiam pulvinar consectetur dolor sed malesuada. Ut convallis euismod dolor nec pretium. Nunc ut tristique massa.

Nam sodales mi vitae dolor ullamcorper et vulputate enim accumsan. Morbi orci magna, tincidunt vitae molestie nec, molestie at mi. Nulla nulla lorem, suscipit in posuere in, interdum non magna.

Retain carbon pricing

I was one of the 50 economists who signed a letter urging the retention of carbon pricing. Keeping limits on carbon emissions is the most severe environmental problem the world has ever faced. Not controlling greenhouse gas emissions possibly threatens the survival of human and non-human life on our planet but, at the minimum, will change our lifestyles in drastically costly ways – these costs will increase the longer action to address climate change is delayed.  Australia is one of the world’s wealthiest countries and cannot ignore its obligations to address this issue.

Every basic economics text – even those written by those on the right-wing of politics, such as Gregory Mankiw – endorse carbon pricing as the cheapest way to address climate change.  Over the last week Ross Garnaut has proposed a compromise which, while not ideal, would keep the architecture for pricing in place by setting a very low price (40 cents per ton CO2)  and by enabling international purchases of emissions permits.  This was promptly rejected by the Coalition but it should not be.   The price of 40 cents is very low but would rise as other countries practice carbon emissions control.  This is one way of meeting a key (though misleading) objection to carbon pricing that Australia is “going it alone”. Pricing would only become significant when other countries act.

Of course I would prefer a much higher price than this and a firm commitment on the part of Australian policy makers to enforce a switch away eventually from the use of all carbon-based fuels but at a minimum the Garnaut proposal should be considered.¹-OPEN-LETTER-SUPPORTING-A-PRICE-AND-LIMIT-ON-CARBON-POLLUTION.pdf

21 comments to Retain carbon pricing

  • I am an economist, and I do not support a carbon tax.

    I agree that it is a more efficient way of reducing CO2 output than the obvious alternatives. But I am not convinced that CO2 has net negative externalities, which is the implicit assumption—an assumption almost everyone seems to make for which I can find very inadequate justification.

    The current climate was not designed for us nor we for it, so there is no good reason to suppose in advance that a climate a few degrees warmer would be much worse for us—especially given that humans already exist and prosper across a range of climates much wider than the projected change due to CO2. It is true that we are currently optimized against our current environment, making rapid change costly. But we are talking about change measured in tenths of degrees per decade and millimeters a year. Over a century farmers will have changed crops multiple times for other reasons, much of the housing stock will have been replaced or modified for other reasons.

    There are at least two reasons to think the next externalities might be positive. One is that, for well understood physical reasons, AGW tends to increase low temperatures more than high, resulting in more warming in cold places and cold seasons, where and when warming is on the whole a good thing, less in hot when it is, on the whole, a bad thing. The other is that human habitation at present is limited by cold not by heat—the equator is populated, the poles are not. So warming will tend to increase the amount of the earth’s surface usable by humans, an effect orders of magnitude larger than the loss of area from sea level rise, even assuming (implausibly) that that is not prevented by diking.

    The alternative to looking for general arguments is trying to add up costs and benefits, as people such as Tol and Norhaus have tried to do. Both of them conclude that net externalities are negative, but much smaller than the catastrophist rhetoric implies. I am not convinced even of that. The costs and benefits will be spread over a long and uncertain future, making their size very hard to estimate. Someone who expects the sum to be negative will tend to make generous estimates of the costs, conservative estimates of the benefits, perhaps leave out benefits because he is not looking as hard for them as for costs—a pattern I have pointed in some of Nordhaus’ work. It’s worth noting that Tol’s survey article in the JEP suggested net positive externalities for about the first two degrees of warming, negative only thereafter. The IPCC scenario (8.5) that puts warming above that by 2100 depends on assuming a continued exponential growth of CO2 emissions, which I think unlikely.

    If we are not sure whether the externalities will be positive and negative, let alone whether they will be negative and large as much of the discussion assumes, the sensible strategy is to wait until we know more before taking expensive actions to reduce CO2 emissions. Nordhaus, a few years ago, in a piece arguing against a WSJ op ed that had argued that AGW was not a crisis requiring immediate response, gave an estimate of the net cost of postponing action for fifty years as opposed to taking the optimal actions immediately. While he did not say so, his figure worked out to an annualized cost (present value) of under a tenth of a percent of world GNP.

    Finally, it is worth noting that if action is taken it is quite unlikely to be the optimal action that an economist would recommend, hence more costly for whatever reduction it produces. I offer the cap and trade bill that passed the House but not the Senate a few years ago as evidence.

  • hc

    Then we are in substantive agreement on the need for carbon pricing if climate change is a dangerous externality. The disagreement rests in the part of the statement after the “if”. There is a possibility that the thousands of climate scientists around the world are wrong in forecasting the effects of continued greenhouse gas emissions and that those few discredited scientists attached to the Heartland Institute have i right. But I’d prefer to back the mainstream science as the alternative view seems incredibly risky and the costs of getting it wrong negligible relative to these risks. You mention human life which can adapt to some degree – I am also concerned with all non-human life on earth and the implications for biodiversity seem severe. Nordhaus seems to have changed is tune recently and is urging prompt action – I think you are referring to much earlier views. I don’t think we can wait given the irreversibilities, the uncertainties and the high levels of emissions now occurring in China and about to occur in the sub-continent and in parts of Africa and South America.

  • hc writes: “There is a possibility that the thousands of climate scientists around the world are wrong in forecasting the effects of continued greenhouse gas emissions and that those few discredited scientists attached to the Heartland Institute have i right.”

    I think you are confusing the results of climate science with a well organized publicity campaign. If you actually read the scientific portions of the latest IPCC report, the catastrophe looks rather like a wet firecracker. The link to droughts in the previous report has been retracted. No link to worse hurricanes now or with moderate levels of warming. Costs of sea level rise to island nations and low lying coastal states may in some cases be as much as a few percent of GNP. The only extreme weather events that they actually predict increasing are temperatures highs, hardly surprising, and heavy rainfall.

    A lot of talk in the summary for policymakers, which is a political document, is about the costs of bad weather, not restricted to weather bad due to warming. Scare rhetoric, but not a lot of substance.

    Nordhaus argues in favor of action. But if you look carefully at his piece in the NYT Review of books, attacking a WSJ op ed, you discover that the substance of his results is a lot weaker than the rhetoric.

    There are at least three sources of uncertainty. One is the rate of CO2 emissions, with most of the scare talk assuming the RCP 8.5 scenario, the one that burns twice the current estimated coal reserves by 2100. One is the consequences of higher CO2 for climate. One is the consequences of those consequences for humans. The talk about almost all climate scientists agreeing is only supported for the claim that temperatures are trending up and humans are an important cause, not for the claim that the results will be very bad or that the best response is to try to prevent the change rather than dealing with it.

    Do you have evidence to the contrary? Can you point me at a serious piece of research that finds that almost all climate scientists agree with the strong claim—which is what Obama asserted? Cook et. al. 2013 provides some evidence that most climate scientists think humans contribute to climate change, but not the rest of it. Anderegg 2010 gives its 97% figure for the 100 climate scientists, among those who have publicly taken a position, who have published most, doesn’t point out that in their whole sample of those who have taken a position (on the IPCC reports) about a third criticized them. Do you have something else and better, and if so have you actually read the paper to check on what it really says, rather than what people claim it says?

  • One further point I should have made … . The whole global warming scare reminds me of the population scare of the sixties and early seventies—I don’t know if you are old enough to remember it. Then too, we were told that terrible things would happen if we did not take prompt action—and, by some, would happen even if we did. Those who rejected the conclusion, most notably Julian Simon, were attacked in very much the same terms that you and others attack critics of the current campaign. My (published) conclusion at the time was that if you actually tried to estimate the net externality from population increase, you could not sign the sum, since there were positive externalities and negative externalities, both of uncertain size. I think the same is true in this case. It’s easy to prove that the net is negative (or positive) if you start out believing it, hard to sign the sum if you start with an open mind.

    That particular scare turned out to be wildly wrong, although, as with the more common religious version of the phenomenon, believers could always revise their dates to move the catastrophe into the future. What actually happened was not mass famine but a continued increase in calorie consumption per capita in the third world. Taiwan, Singapore, Hong Kong, South Korea demonstrated that the source of poverty was not overpopulation but bad policies—and China is now repeating the demonstration on a larger scale.

  • Bob Beale

    David, in my view your logic is back to front in the case of global warming. Uncertainty in the science is widely misunderstood and misinterpreted. It means the case for preventive action now is greater, not lesser. It is apparent that you have made your mind up on this, and I have learned to my cost that there is no point in enjoining you in debate because it will only entrench you in your views. I do hope, however, that as one intelligent person to another you might reconsider the logic of your position. The do-nothings and we’ll-just-adapts on this issue invariably latch onto the possibility that things won’t be as bad as feared. They never seem to consider that uncertainty cuts both ways, and that things could turn out much worse than feared. The latest research out today, for example, only adds to concern that melting of ice sheets in Antarctica could be far more serious and happen much faster that previously thought. New knowledge on this issue comes along every day, and none of it – none of it – suggests that projections of climate change and it’s effects on the biosphere, atmosphere and geosphere have been too extreme. On the contrary, it increasingly looks as though the IPCC in particular has been too conservative. If you are confident enough to think and project otherwise, you must have an excellent grasp of climate science and a truly exceptional degree of skill in forecasting global climate sensitivity.

  • Bob: You are assuming that uncertainty means uncertainty about the size of an effect known to be negative, hence that the expected utility loss is more than the disutility from losing the expected value. At least, that’s the only sense I can make of your “it means the case for preventive action now is greater, not lesser.” If you have a different explanation of that phrase, feel free to explain.

    As should be clear from what I wrote, I am uncertain not only about the magnitude of the externality but about its sign. If we don’t know whether the effects will be good or bad, that’s an argument against adopting costly policies to prevent it. Global warming means longer growing seasons, milder winters, more land usable by humans. Increased CO2 tends, ceteris paribus, to increase agricultural productivity, CO2 being an input to photosynthesis. Those effects have to be balanced against the negative effects. Doing so is difficult since we do not know how much CO2 will be produced, how sensitive temperatures are to CO2 concentration, what the climate response to increasing temperature will be, or what humans will be doing by the time the changes occur—a century is a long time.

    Lots of people think they know the answers to those questions, but if you pay attention it becomes clear that most of it is assertion. The Fourth IPCC report linked drought to warming, the fifth retracted that link. Various people have been saying for a long time that warming increases the frequency, force, or both of hurricanes. Chris Landsea some years back resigned from the IPCC in protest against such claims, on the basis that they were not supported by any scientific evidence, and as of the latest report the IPCC agrees with him.

    So far as the IPCC being too conservative, the obvious test is to go back to the old reports, see what you would expect from them, and compare that to what happened. I did that experiment some time ago—you can find the result on my blog:

    The short summary is that the IPCC has done worse at predicting the rate of warming than a straight line extrapolation from 1910, when the warming started, to the date of the first IPCC report would have done—and the error has consistently been in the direction of overestimating future warming.

    I did that experiment because people arguing about these things can easily make claims in whichever direction they prefer and support them by a suitable selection of evidence. I am guessing, for instance, that you are not aware of the recent study finding that the shrinkage of Caribbean corals was due to a cause unrelated to warming. You probably haven’t read Richard Tol’s survey article in the JEP, which suggested that warming has positive net externalities up to about two degrees, negative thereafter. You may or may not even know that the IPCC retracted the drought claim—an example of a case where the evidence showed that they had been too pessimistic.

    The evidence that temperatures have risen over the past century is very strong. The reasons to believe that humans are a sizable part of the cause are pretty good. But there is no similarly strong evidence for the central claim that your argument depends on—that the net effects of the change will be large and negative.

    Do you remember the population scare of the sixties and early seventies, the arguments between Julian Simon and everyone else, with everyone else making much the sort of arguments you and HC are making?

  • Tom

    You may or may not know that Richard Tol’s JEP article has recently been corrected, after ‘gremlins’ (his words) mucked up some of his data entry. Combined with new studies released since Tol’s initial meta-analysis, the estimated economic impact is negative for all warming.

  • Is your account of the effects on Tol’s paper what he says they were or what someone else says they were? I think if you take a look at Tol’s blog, you will conclude the latter.

    You might start with

    There are more.

    If I am correct about the situation, you might want to think about what approach to evaluating information leads you to jump from “A says that B’s paper is wrong” to “B’s paper is wrong” without checking on what B says on the subject.

  • Bob Beale

    “there is no similarly strong evidence for the central claim that your argument depends on—that the net effects of the change will be large and negative”
    Come on David, stop this bullshit misrepresentation. Like I said, it’s a waste of time arguing with people like you because you only dig in deeper. For the record, what I said is plain and in writing above. Uncertainty cuts both ways. You seem unable or unwilling to entertain that idea openly. That’s it from me. No more.

  • Tom

    There’s no need to be quite so patronising. I’ve followed the back and forth with Tol and others, and have looked at his original paper and the corrections myself. I was referring to Figure 2 in the corrigenda here:

    This impact curve, negative everywhere, includes two estimates overlooked in Tol’s original paper and a subsequent five papers that appeared after the original JEP article came out, as I said above.

    As Tol himself has pointed out, it’s likely that the range over which the original impact curve was positive was irrelevant to policy anyway, since those benefits are already sunk.

    As it happens, I’m not particularly convinced by his methodology anyway, especially the use of a quadratic to fit the estimates.

  • Tom:

    I apologize for being patronizing, but I’ve encountered people who took the criticisms of Tol’s work for gospel without having seen his responses.

    You wrote “the estimated economic impact is negative for all warming.” Given the context, I assumed that was a statement about Tol’s corrected version. Looking at Figure 2 in the corrected version that you linked to, I observe that the corrected line is positive from zero to about two degrees, although lower than in the original version.

    I now take it that your statement is about your, or someone else’s, conclusion from adding in additional estimates from other sources. Does Tol agree? If not, your account is, as I suggested, of what someone else thinks the effects on his paper were, not what he thinks they were, where the someone else may possibly be you.

    I have no particular opinion on his methodology. What I observe is that Tol and Nordhaus both argue for action against climate change, hence presumably any bias in their calculations are in that direction. Both of them come up with results much weaker than the usual rhetoric about climate change implies. After watching the population arguments some decades back as well as other public policy cum publicity campaign projects in my lifetime, I read that as “it is possible to do the analysis in a way that ends with negative externalities,” not as “the analysis shows there will be negative externalities.”

    For one example of why I take it that way, see my post on Nordhaus some time back.

    For further evidence wrt Nordhaus, see:


    Whether the range over which net effects are positive was or is irrelevent depends on what sensitivity turns out to be and what happens to CO2 emissions over the rest of the century. Checking the current IPCC report, I find:

    “Increase of global mean surface temperatures for 2081–2100 relative to 1986–2005 is projected to
    likely be in the ranges derived from the concentration-driven CMIP5 model simulations, that is, 0.3°C to 1.7°C (RCP2.6), 1.1°C to 2.6°C
    (RCP4.5), 1.4°C to 3.1°C (RCP6.0), 2.6°C to 4.8°C (RCP8.5).”

    For all except RCP 8.5, 2° is within the possible range. Given the rate of technological (and other) change in the world, I think basing policies on estimates of effects a century or more into the future rarely makes sense. And I think RCP 8.5, which assumes a continued exponential growth in the use of fossil fuels over the rest of the century, is quite unlikely even if nothing at all is done to control climate change.

  • Tom


    I think you’re reading the corrected figure wrong. The line that is positive to 2 degrees (ie the top, dashed line) is the upper bound of a 95% confidence interval: the actual impact curve is the solid green line below it, which is negative everywhere. Tol himself says in the correction piece: “First, unlike the original curve (Tol 2009, Figure 1) in which there were net benefits of climate change associated with warming below about 2°C, in the corrected and updated curve (Figure 2), impacts are always negative, at least in expectation.”

    The argument that “Tol and Nordhaus both argue for action against climate change, hence presumably any bias in their calculations are in that direction” is patently silly. There is no strict intellectual dichotomy between ‘action’ and ‘inaction’: someone can believe (prior to commencing a study) that some action is necessary, but not a lot. This is a perfectly legitimate prior to hold, but it could also conceivably bias the research towards such a conclusion. There are other biases that might also affect a study: for example, a bias towards confirming one’s own previous results, which might themselves have been completely unaffected by ideological biases.

    I’m not saying, by the way, that Tol’s use of a quadratic fit is *necessarily* wrong, but it seems an odd choice, and wasn’t justified in the JEP article. I haven’t played around with the data, but it would be interesting to see whether a non-parametric fit yields the same positive impacts over low warming ranges.

  • Tom: I was reading Figure 1. You are correct about Figure 2. My error.

    On the question of bias, you might take a look at:

    When I observe Nordhaus, in the NY Review of books, attacking a WSJ editorial whose claim is consistent with his work as he reports it, I think I am entitled to reach a conclusion about what results he wants to reach. For how he reaches them—specifically, the way in which the prior influences an attempt to add up externalities—see:

    On the more general issue … . Tol is basing his calculations on estimates by lots of people. If there is some reason why those people mostly are biased in the same way, that will show up in his results. Assuming he is calculating his confidence interval in the conventional fashion, he will understate the range, because the errors will be correlated.

    Is there such a bias? I don’t know if you have been following Dan Kahan’s interesting work on why people believe what they do, with examples in both climate and evolution. He offers quite a lot of evidence that, where a belief has become linked to group identification, as in both of those cases, it is that rather than the evidence that mostly determines beliefs. The argument is straightforward, and analogous to the standard rational ignorance argument. Whether I believe in evolution (or AGW) has very little effect on what happens in the world. But if I am in a social environment where everyone knows that believers in X are stupid or evil, the effect on me of believing in X is large and negative.

    The argument applies to both sides of such disputes. I suggest a simple experiment in introspection. Imagine that you are presented with evidence or arguments that lead to the conclusion that AGW is not a problem. Is your response “wonderful. If it turns out to be right I can stop worrying. I’ll check the evidence and if it holds will be delighted to tell all my friends.”

    Or will it be “I hope I can find something wrong with this, because announcing that I am now on the other side of that controversy is going to cost me a lot of friends”?

    If you get the result I think you will, generalize it to academics in general and think about what the implications are for the sort of research, such as adding up estimates of externalities spread over most of a century, that involves a lot of judgement calls.

  • rog

    Writing in the FT Martin Wolf deals with the so called “debate”

    “Climate sceptics are losing their grip.

    We do not have a Chinese or an American atmosphere. We have a global atmosphere. We cannot run independent experiments upon it. We have instead been running a joint experiment. This was not a conscious decision: it happened as a result of the industrial revolution. But we are consciously deciding not to stop.

    Conducting irreversible experiments with the only planet we have is irresponsible. It would only be rational to refuse to do anything to mitigate the risks if we were certain the science of man-made climate change is bogus. 

    Since it rests on well-established science, it would be ludicrous to claim any such certainty.

    On the contrary, any reasonably open-minded reader of the Summary for Policymakers from the Intergovernmental Panel on Climate Change would reach the conclusion that any such certainty on the science would be ludicrous. It is rational to ask if the benefits of mitigation outweigh the costs. It is irrational to deny the plausibility of man-made climate change…

  • Tom

    Of course, you are entitled to reach a conclusion about what Nordhaus or Tol might like their results to show. This may very well impact on their research, but without actually looking in detail at their research design/methodology (which, I see, is what you’ve done on your blog) you can’t simply assume that any bias in their method reflects their ideological priors or that there haven’t been other unconscious or conscious biases, which is how I read your previous comment (“hence presumably any bias in their calculations are in that direction”). There is an extent to which this becomes a bit circular, too: your critiques of others’ work and the extent to which their ideological priors are embedded in them are ALSO affected by your own ideological priors.

    As to your point about his confidence intervals and correlation, I completely agree. More broadly, the assumption that each of the data points in his meta-analysis come from the same underlying curve plus an iid error seems kind of shaky.

    The other troubling thing about Tol’s paper and the correction is the relabelling of his axes. In his original JEP paper, the x-axis was said to be warming ‘relative to today’; whereas in the corrected figures, this has changed to ‘relative to preindustrial times’. There was no discussion of this in the correction and as far as I can tell from Google searching (but happy to be corrected) Tol hasn’t explained this change at all, even though it has been pointed out to him. If we assume that the version in his corrected figures is accurate, then this might explain why his view differs from yours when it comes to sunk benefits and IPCC-predicted warming: we’ve already had warming since preindustrial times.

  • Tom: I certainly agree about my own biases. I discuss that a little in a recent blog post wrt Kahan’s work. I discussed it at greater length in an older post:

    But I am offering arguments to you, and I have no reason to expect you to share my biases. Do you disagree with my claim that Nordhaus’ behavior is evidence that he would rather conclude that AGW is a problem which requires a response than conclude that it does not—as illustrated in my blog comments? Do you know of any evidence in the other direction?

    I don’t have similar evidence for Tol, beyond my general familiarity with the academic world that he and I both inhabit. But I think I do have reason to think that researchers will tend to have a bias towards overestimating net costs, again given what I know about that world. And the Tol piece we have been discussing is basically a summary of work by multiple people in the field.

    You don’t say whether you performed the introspective experiment I suggested, or if so what the result was. If you did and the result was what I expected, do you agree that that would be a reason to expect academic estimates of net externalities from warming to be biased upwards?

  • Tom

    I’m afraid that to be able to give a competent response to your claim about Nordhaus’s behaviour I’d have to know an awful lot more than I do about his work, although your supposition doesn’t seem a priori unreasonable. I wasn’t making any actual claim either way, I was just saying that your initial statement–that his preferred conclusions are the only thing that might influence the direction of any bias in his methodology–seems like an overly strong statement. I certainly don’t disagree that it might have an effect on the size of the net externalities he proposes, and I don’t object to your claim that it does: I only require that such a claim be made as a result of a study of his methodology, rather than by assumption.

    Your proposed experiment of introspection, while philosophically interesting, is a little hard for me to run in good faith, since I don’t really mix in circles that discuss AGW much or in which it functions as any kind of shibboleth, so the social cost of me changing my mind wouldn’t be very high.

  • Tom: For my evidence on Nordhaus as revealed by his writing, see my blog posts on his NY Review of Books piece. So far as the methodology, I only offer one example, in an earlier post. For biases of researchers more generally, all I can offer is extended first hand experience of American academia.

    If you can’t do my experiment wrt AGW, try it with regard to whatever issues do work for the circles you mix in. Without knowing your environment I can’t offer any good suggestions. The best one I could think of for myself was minimum wages:

  • David, you ask if we remember the population scares of the 60’s and 70’s. I do. And we are living the disaster now. Why do you think there is such a huge number of “economic” refugees?

  • John:

    By “economic refugees” do you mean people trying to migrate from poor countries to rich countries? That’s not something new. In the years before WWI, the U.S. was accepting about a million immigrants a year.

    But if you look at conditions in the poor countries, they are better, not worse, than they were fifty years ago. More calories per capita, higher real incomes.

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>