Elisabeth Kolbert of the New Yorker takes apart Levitt and Dubner's claim that climate change would be 'easy' to solve (the economist and journalist believe that shooting sulfur into the atmosphere via a 18 mile hose would do the trick):
Neither Levitt, an economist, nor Dubner, a journalist, has any
training in climate science—or, for that matter, in science of any
kind. It’s their contention that they don’t need it. The whole conceit
behind “SuperFreakonomics” and, before that, “Freakonomics,” which sold
some four million copies, is that a dispassionate, statistically minded
thinker can find patterns and answers in the data that those who are
emotionally invested in the material will have missed.
Kolbert also delves into what she believes is the major flaw in the book: That Levitt and Dubner are a little too pleased with themselves:
But what’s most troubling about “SuperFreakonomics” isn’t the authors’
many blunders; it’s the whole spirit of the enterprise. Though climate
change is a grave problem, Levitt and Dubner treat it mainly as an
opportunity to show how clever they are. Leaving aside the question of
whether geoengineering, as it is known in scientific circles, is even
possible—have you ever tried sending an eighteen-mile-long hose into
the stratosphere?—their analysis is terrifyingly cavalier.
I did enjoy their first book 'Freakonomics', but I thought much of it was highly speculative and slightly far fetched. For example, Levitt and Dubner ascribed the falling crime rate in the U.S in the 90's to the legalization of abortion in 1967 (all the would be criminals weren't born). While it's a clever idea, the fact is, the exact opposite happened in the United Kingdom. Abortion was legalized in the same year in the U.K, but was hit with a major crime wave in the late 90's that has continued till today.
The 'cold hard numbers' may work on paper, but human societies and the environment are far to complex to reduce to statistics. After all, it's why economics is referred to as the 'junk science' - because it all too often fails.