RealClimate logo

Who ya gonna call?

Filed under: — gavin @ 22 August 2007

Gavin Schmidt and Michael Mann

Scientific theories gain credence from successful predictions. Similarly, scientific commentators should gain credibility from whether their comments on new studies hold up over time. Back in 2005 we commented on the Bryden et al study on a possible ongoing slowdown in the North Atlantic overturning circulation. In our standard, scientifically cautious, way we said:

… it might be premature to assert that the circulation definitely has changed.

Our conclusion that the Bryden et al result ‘might be premature’ was based on a balance of evidence argument (or, since we discussed this a few days ago, our Bayesian priors) for what the consequences of such a slowdown would be (a (unobserved) cooling in the North Atlantic). We also reported last year on some data that would likely help assess the uncertainty.

Well, now that data has been properly published (reported here) and it confirms what we thought all along. The sampling variability in the kind of snapshot surveys that Bryden et al had used was too large for the apparent trends that they saw to be significant (which the authors had correctly anticipated in the original paper though).

Score one for Bayesian priors.

107 Responses to “Who ya gonna call?”

  1. 101
    Andrew says:

    “That said, we ignore economists at our peril.”

    Or, for those of us who have ended up working in finance, to our profit.

  2. 102
    Nick Gotts says:

    Re #100 John,
    Thanks for these interesting references. They come from yet another non-neoclassical research program (in Lakatos’s sense): ecological economics. It’s not one I know much about, though I should know more. It’s mainly concerned with macroeconomics. This aims to explain or more ambitiously predict things like GDP growth rates, inflation, unemployment rates, business cycles etc., as opposed to microeconomics, which concentrates on the behavior of individuals, or quasi-individuals such as firms – though obviously the two are related. The basic premise of ecological economics is that economics needs to take more account of biophysical processes and constraints, particularly thermodynamics. I was rather turned off it because one of the few papers I’ve read in the area (which I can’t now identify), made the crass error of claiming that the second law of thermodynamics implies that the more economic activity there is, the faster the biosphere will necessarily degrade to a point where no useful work is possible. I think Jeremy Rifkin may be a current proponent of this belief. However, the two papers you point to don’t appear to make any similar error. I don’t feel qualified to comment on them further than that, but they certainly appear highly relevant to the question of how far economic growth can be decoupled from the grwoth of energy use.

  3. 103
    Ray Ladbury says:

    Re: economists. I fear we are in danger of descending into the type of debate we had about engineers vs. scientists a few months ago. The economic aspects of climate change will assume increasing importance as the public begins to understand the unassailable nature of the science of climate change and as vested interests begin to understand the futility of attacking such an unassailable position.
    Steve Reynolds has pointed out that many economists have estimated the costs of climate change to be in the single-digit percentage points of the global economy. This does not seem credible to me for several reasons–ranging from my understanding of the potential impacts to my experience with tropical agriculture and invasive species and diseases. I think it would be interesting and important to examine the assumptions made by economists in these assessments. I also think it is important consider the potential positive impacts of technological developments to mitigate climate change (just as it is important to include potential positive impacts of a warmer world in economic assessments). As pointed out by Nick, technological innovation is not something neoclassical economics treats very well.
    Now I realize that we have a tendency to see any problem through the lens of our own education and training, but I think that a risk analysis/management approach could pay significant dividends for such a complicated analysis. In such a system, the costs of an adverse outcome are weighted by its probability. The probability can be, but is not necessarily Bayesian. This allows risks to be considered even if there is substantial uncertainty over their probability and impact.

  4. 104
    Nick Gotts says:

    RE #103 [Now I realize that we have a tendency to see any problem through the lens of our own education and training, but I think that a risk analysis/management approach could pay significant dividends for such a complicated analysis. In such a system, the costs of an adverse outcome are weighted by its probability. The probability can be, but is not necessarily Bayesian.]

    I see at least three problems with that:
    1) Costs to whom? The cost to the global economy could be quite small while still causing enormous amounts of suffering and tens of millions of premature deaths.
    2) Costs over what period of time? If you’re integrating over time, what discount rate do you use?
    3) How do you cost, say, mass extinctions, loss of cultural treasures, or the end of industrial civilisation? Unless we think the probability of these things happening would be unaffected by climate change, they need to be taken into consideration, but they don’t appear commensurable with costs as economists reckon them. I think this points to one of the limitations of neoclassical economics: the idea that all values can be reduced to a common “currency” (whether you call it money, utility, or whatever).

  5. 105
    Ray Ladbury says:

    Nick, very reasonable caveats. I, too, am doubtful that all outcomes can be expressed in monetary terms. However, if we simply sit back and refuse to play this game, we will never convince the majority of politicians, businessmen, etc. that the risk is serious. Just as many “denialists” have sidelined themselves from the process by attacking the science (which they don’t understand), we will preclude ourselves from injecting our values into the debate by trying to remain above it. And as distasteful as we may find the process of assigning a “value” to human life and suffering, the idea of allowing those who value it less to do so is more distasteful.
    In short, I view these factors as caveats and challenges rather than obstacles. And I don’t see that we have much choice but to phrase our arguments in terms of the dismal science while trying to make it less dismal in the process.

  6. 106
    Nick Gotts says:

    Ray, I should make clear I don’t object to risk analysis or management approaches provided their limitations are understood. Are there current attempts within those approaches to use multi-criteria analysis? Suppose for simplicity we just consider two criteria, expected reduction in economic output (with some discount rate), and probability of a disaster of sufficient scale to make further monetary-based calculation pointless (again, perhaps, discounted over time). You might then come up with a range of policy options with different possible trade-offs. The choice between these would then be obviously a moral/political one, on which everyone’s opinion would be equally valid, not a question with a scientifically justifiable answer. In practice, there would be more than two dimensions, quantification would be difficult, and there’s still the matter of selecting a discount rate – probably also a political/moral choice rather than one where there’s a scientifically justifiable answer. But maybe something along these lines is a way of using economists’/risk analysts’ relevant expertise, without handing over to them matters on which we all (worldwide) should have an equal say?

  7. 107
    Timothy Chase says:

    I am not sure how much you can separate the human from the economic. The fact is that climate change poses a seious threat both in human lives and economic impact and there is enough uncertainty and risk to both.

    I know that many here are primarily concerned with the human dimension, elsewhere oftentimes primarily the economic. But given the uncertainties, risks to human life will necessarily also involve economic risks, and risks to the world economy involves a potential reduction in the resources, including those that are available for dealing with disasters. One figure I ran across (“Modern Times” by Paul Johnson, who at least seemed to get his facts if not his judgments right) was that presumably the oil shocks of the late nineteen seventies had through its economic repercussions in all likelihood resultied in the deaths of over a million people.

    What I would stress is trying to frame things in terms of the values which are important to a given audience. Then if you can open a dialogue – perhaps by showing how various values are related – all the better.