To much economic comment on climate change is focused on cost-benefit analysis, as if we can just count up the cost of losing ecosystems etc. Dr. Weitzman points out that the economic question isn't about accounting and discount rates, its about uncertainty. We don't know exactly what's going to happen and we need to mitigate that risk. If you consider a probability distribution of temperature increases, it looks like a typical "bell" curve, with 2-4 degrees being most probable and cases on either side being lower and lower the further you get away from that center. The problem is that the probabilities for the really bad cases don't fall off fast enough, the tails of the bell curve are "fat". From Krugman's column:
It’s true that scientists don’t know exactly how much world temperatures will rise if we persist with business as usual. But that uncertainty is actually what makes action so urgent. While there’s a chance that we’ll act against global warming only to find that the danger was overstated, there’s also a chance that we’ll fail to act only to find that the results of inaction were catastrophic. Which risk would you rather run?
Martin Weitzman, a Harvard economist who has been driving much of the recent high-level debate, offers some sobering numbers. Surveying a wide range of climate models, he argues that, over all, they suggest about a 5 percent chance that world temperatures will eventually rise by more than 10 degrees Celsius (that is, world temperatures will rise by 18 degrees Fahrenheit). As Mr. Weitzman points out, that’s enough to “effectively destroy planet Earth as we know it.” It’s sheer irresponsibility not to do whatever we can to eliminate that threat.
This mention was actually part of a larger, and good, point about the recent lying about off-shore drilling which is subject for another post.