Monday, January 21, 2008

Freakonomics, Steven D. Levitt and Stephen J. Dubner

Title: Freakonomics
Authors: Steven D. Levitt and Stephen J. Dubner
Rating: Good

Yes, I finally got around to Freakonomics. I know, I know. I'm way, way, way late in doing so. It seems as if everyone has already read this book. What can I say... I'm slow.

The best brief description of Freakonomics I can give would be be something like: "A book of short narratives discussing human behavior in interesting and possibly controversial ways, and that manages to aggrandize the authors while simultaneously claiming that they aren't seeking aggrandizement."

Let me be a bit more specific. Levitt is a highly respected economist with an interest in human behavior. He's won all kinds of awards and so on. Dubner is a writer working for (among others) the New York Times. He wrote a piece about Levitt that eventually lead to their collaboration on this book. Somehow, despite the text indicating that the authors are very modest and self effacing, that isn't what comes across. You can only read that someone is quiet and reserved so many times before you start wondering if the writer isn't trying a bit too hard.

As for the discussion of economics, it's interesting, though it's pretty brief in reality. The major items are discussions of how information is used by some to their advantage, how much money one branch of one Chicago street gang made selling drugs, the possible connection between legalized abortion and the drop in crime rates in the 1990's, and a few items on parenting, like nature vs. nurture, risk assessment and name selection.

Most of that content is just fine and not too controversial as far as I can tell. Note, however, that I'm not an economist (nor a mathematician), and even if I was qualified in some way, the backing data supporting the conclusions isn't given here. (Though, apparently, it is available somewhere else. Read on.)

On the controversial side there is a discussion about the relative safety of children around guns and swimming pools that some who want to limit gun access don't like. And then there's the biggie: the question of whether or not one impact of Roe v. Wade was a reduction in crime 20 years later.

I cannot answer that question. I think it is possible - I buy the arguments as presented in the book - but I've also done a tiny bit of web research and there is (or was) a serious argument over this. It turns out Levitt made some sort of error in his analysis and admitted it publicly. He claims that it doesn't substantially change his conclusions. His detractors - who may or may not have an ax to grind or be staunch abortion foes, I really don't know - seem to disagree.

I suspect I could do many hours of research trying to resolve the dispute to my satisfaction. I note, however, that Levitt did the right thing by letting others examine his methodology and data, and by admitting the error that was found. He gets points for that as far as I am concerned.

A major point of the book is that there are always unintended consequences of any economic action. If you provide an incentive to do (or not do) something, you will almost certainly cause other effects that you did not predict. That is a truism, of course, but it's nice to see some concrete examples laid out for review.

Levitt and Dubner are apparently writing a blog for the NYT now, capitalizing on the fame that came with Freakonomics. Maybe some of the articles there are as interesting as those in the book. If they cause some to question things a bit more - rather than just accepting the "obvious" answer - that'd be just fine with me.

In the unlikely event that you haven't read Freakonomics I can suggest you do so. But get it used. I think Levitt and Dubner have made a lot of money off the concept, and while I usually don't care about that, the possibility of false modesty here bugs me a bit.