Uncertainty and climate change
Posted by Daniel Hall on January 6, 2008
Jim Manzi, who has commented here before, has a long post criticizing Marty Weitzman’s view that the uncertainty about potentially catastrophic outcomes from climate change justifies much stronger action to prevent climate change than more conventional benefit-cost analysis might suggest.
I’ll let the succinct and insightful Tyler Cowen do most of the heavy lifting in rebutting Manzi:
I am not persuaded by Jim Manzi’s major point of rebuttal, namely: “the heart of Weitzman’s paper revolves around the first point: the probability of extreme disaster is larger than current models assume.” My reading of Weitzman (which may not exactly be Weitzman’s own view) is the following: raising the discount rate doesn’t choke off our worries of future dangers. In many plausible models, a higher discount rate means a higher degree of risk aversion as well, and thus we are back to worrying.
Manzi also argues that: “There is No Good Reason to Think That the Probability Distribution for Estimates of Climate Sensitivity Fits Any Functional Form.” Fair enough, there is only one world and ultimately the meaning of probability is murky, Bayesian or not. But we still have to act on probability estimates in an “as if” way and indeed we all do in a personal context.
I’ll note that Cowen puts more emphasis on the discount rate than Weitzman seems to in his own paper, although he is correct that Manzi does not address this issue at all.
I would add that Manzi refutes a misinterpretation of Weitzman’s argument, stating that the “claim that Weitzman makes [is] that we can now usefully quantify the loss in utility that extreme AGW impacts may cause.” But it doesn’t appear to me that Weitzman is claiming at all that we know what the potential damages are; it’s that they could be so catastrophically large as to be unacceptable. Those who would ignore such outcomes must justify why they aren’t possible or important. In Weitzman’s words:
The take-away message here is that the burden of proof in the economics of climate change is presumptively upon whoever wants to model or conceptualize the expected present discounted utility of feasible trajectories under greenhouse warming without considering that structural uncertainty might matter more than discounting or pure risk per se. Such a middle-of-the-distribution modeler should be prepared to explain why the fat bad tail of the posterior-predictive distribution is not empirically relevant and does not play a very significant role in the analysis.
Manzi fails this test — he disputes Weitzman’s specific choices (about the severity of expected losses, or the functional form), but fails to address why we should not be primarily concerned about the very worst outcomes. Even if they remain highly unlikely, the extremity of their consequences means that they should receive more attention that the median (most likely) outcomes.