Of fat-tailed catastrophes
Martin Weitzman, of Harvard's economics department, has now countered that device with a paper considering the possibility (however low) of a genuinely catastrophic event resulting from a failure to act now. Such events are usually ignored in standard CBA, mainly because they're a bit difficult to work with. Weitzman's model does include the possibility of extreme events (which he terms 'fat-tailed catastrophes'), and the results significantly shift the balance of costs. When applied to the current knowledge relating to climate change and emissions, Weitzman's analysis shows that mitigation investment makes a hell of a lot of sense in minimising expected future costs.
As Weitzman concludes:
Even just acknowledging more openly the incredible magnitude of the uncertainties that are involved in climate-change analysis - and explaining better to policy makers that the artificial crispness conveyed by conventional IAM-based CBAs here is especially and unusually misleading compared with more-ordinary non-climate-change CBA situations - would in my opinion go a long way towards elevating the level of a reality-based public discourse concerning what to do about global warming.
The fairly technical paper is available in draft as a PDF here. For the less technically inclined, New Scientist gives a good summary.