Pascal asks the question: What is the expected value of a very small chance of an inﬁ nite loss? And, he answers, “Inﬁ nite.” In this example, what is the cost of lowering CO2 output and having the long-term effect of increasing CO2 turn out to be nominal? The cost appears to be equal to foregoing, once in your life, six months’ to one year’s global growth – 2% to 4%, or less. The beneﬁ ts, even with no warming, include: energy independence from the Middle East; more jobs, since wind and solar power and increased efﬁ ciency are more labor-intensive than another coal-ﬁ red power plant; less pollution of streams and air; and an early leadership role for the U.S. in industries that will inevitably become important. Conversely, what are the costs of not acting on prevention when the results turn out to be serious: costs that may dwarf those for prevention; and probable political destabilization from droughts, famine, mass migrations, and even war. And, to Pascal’s real point, what might be the cost at the very extreme end of the distribution: deﬁ nitely life changing, possibly life threatening.
That's Jeremy Grantham of investment firm GMO, from his summer essay "Everything You Need to Know About Global Warming in 5 Minutes."
Cost/benefit analyses like this are what conservatives have to come to grips with on the issue of global warming. Jim Manzi has made a similar case, but come to the conclusion that the benefits are not worth the cost at this point.
Given the seeming proliferation of black swans of late, Grantham's argument is certainly worth noting.