I finished SuperFreakonomics the other night and give four-fifths of it a thumbs up (chapter 5 gets a thumbs down). My guess is that if a reader picked up SuperFreak first it would give the same positive jolt to utility that Freak gave when read first. I enjoyed SuperFreak, but not as much as the first. Maybe that is due to diminishing returns? Most of the stories are good but the highlight of SuperFreak is the epilogue where we're told how monkeys learn to use fiat money, in a variety of ways. Besides the lack of economics in the climate change chapter, I can take issue with several things, including:
- Prostitutes aren't patriotic because they work on July 4th.
- The drunk walking stuff is an assertion.
- Volcanoes don't produce positive externalities.
- The disconnect between two statements repeated more than once: "it is hard to change people's behavior" and "people respond to incentives." Huh?
- A 25% reduction in the number of "minor" injuries suffered by kids in car accidents due to car seats and relative to seat belts is shrugged off as a small reduction. It sounds like a lot fewer injuries to me. Well worth the cost of the car seat.
Yet, the strangest thing to me about the book is the unnecessary dissing of lab experiments.
I've been through this stuff before as a young researcher trying to use the contingent valuation method. After the Exxon Valdez oil spill, the stakes were raised and revealed preference types and lab experimentalists (here is how I put it recently) dumped on surveys that elicited hypothetical behavior. Researchers eventually saw revealed preference and stated preference data as complements and not substitutes. Most recently, choice experiment types are dumping on contingent valuation (er, behavior). Come on, guys, these methods aren't necessarily competing!
Anyway, who cares what I think? Here is how Falk and Heckman conclude their recent piece, which seems to be motivated by the lab vs field experiment debate, in Science (abstract):
In this context it is important to acknowledge that empirical methods and data sources are complements, not substitutes. Field data, survey data, and experiments, both lab and field, as well as standard econometric methods can all improve the state of knowledge in the social sciences. There is no hierarchy among these methods and the issue of generalizability of results is universal to all of them.
Hard to argue with that.
Now that I'm finished I can get back to Dan Brown and my bag of cheetos.
Update: Add this to the things I can take issue with:
- "Eyeballing" a too-short length of the
global temp time series data and concluding cooling should not have led
to a global cooling subtitle.
Here is a fairly comprehensive post detailing the negative reviews of the climate change chapter (including one of mine). And, yikes, apparently a link to it was deleted from the Freakonomics blog comments section.