On the political side of the Stern Review, my most-charitable interpretation of its urgent tone is that the report is an essay in persuasion that is more about gut instincts regarding the horrors of uncertain rare disasters whose probabilities we do not know than it is about economic analysis as that term is conventionally understood. Although it is difficult enough to analyze peoples motives, much less the motives of a 600-page document, I cant help but think after reading it that the strong tone of morality and alarm is mostly reecting a fear of (in my ponderously esoteric Bayesian jargon) what is potentially out there in the thick left tail of the reduced-form posterior-predictive probability distribution of g under greenhouse warming. I have argued here that this thick left tail of g is an important aspect of the economics of climate change that every analyst Stern and the critics of Stern might do well to try to address more directly.
I think that Marty's intuition about global warming is exactly right. For the policy objective of mitigating climate change within the usual range of forecasts, the costs exceed the benefits for any sensible set of discount rates.
The real issue, as Marty expresses very well, is what to do about the risk that climate forecasting models under-predict global warming, particularly in the near term. It could turn out that the best policies for addressing this turn out to be the sort of economic sacrifices proposed in the Stern report--as Marty says, the Stern report is "right for the wrong reasons." However, as Marty also says, what we really would pay a huge premium for is an option that allows us to address climate change on a "just in case" basis. To me, that favors investment in climate engineering as a better tool than trying to manage CO2 emissions.
Read his whole paper. Thanks to Greg Mankiw for the pointer.