Unconventional Wisdom

By Amanda Ripley

Imagine a whip-smart economist with a sprawling imagination. Now imagine he's 9 years old and wants to know everything. That is the basic profile of Steven Levitt. A University of Chicago economist, Levitt, 37, is in fact an adult. But he has built his name by asking questions packed with curiosity and devoid of judgment: If drug dealers make so much money, why do they still live with their moms? Did crime in the 1990s go down because the number of abortions in the 1970s went up, or is that just a coincidence? Does parenting actually matter when it comes to kids' test scores?

In his first book, co-authored by the writer who profiled him for the New York Times Magazine in 2003, Stephen Dubner, Levitt has compiled a greatest-hits mix of his research. The book, unfortunately titled Freakonomics (William Morrow; 242 pages), is broken into six chapters, each posing a different social question. Levitt and Dubner answer them using empirical research and statistical analysis. And unlike academics who usually address these matters, they don't clutter the prose with a lot of caveats. They just show you the goods.

For example, it turns out that drug dealers don't really make so much money after all. Levitt and a colleague who had obtained copies of a Chicago gang's accounting books found that street-corner crack dealers in the 1980s made less than minimum wage. They stayed in the job because they aspired to rise through the ranks and make six figures--which only a few top leaders ever achieved. In other words, the authors explain, "the gang's wages [were] about as skewed as wages in corporate America."

As for the crime rate, Levitt makes a convincing case that more abortions did lead to fewer poor, unwanted kids, which led to fewer criminals. Levitt has faced scorching criticism for that claim, which he has been making since 2001. But here and throughout the book he remains blissfully devoted to facts, even unpleasant ones. He calmly displays evidence that abortion has had a greater effect on crime than have broken-window strategies of policing or economic prosperity. But as a crime-control measure, it is "terribly inefficient," he writes, as only an economist could, since the number of aborted fetuses is so exponentially greater than the number of lives saved through less violence.

And yes, don't worry. Parents do influence children's test scores--but what they do is less important than who they are. If parents are high earners, well educated and out of their 20s, for example, their children will do better, according to Department of Education statistics. Staying home with the kids and carting them off to museums, however, has little effect.

Levitt has published most of this material before in academic journals. But Freakonomics is accessible to people who don't understand regression analysis, the procedure statisticians use to sort through data. And its authors knit in research by other scholars. Each chapter is an enlightening field trip, like the investigations into human nature in Malcolm Gladwell's books, The Tipping Point and Blink. But in Freakonomics, attempts to link the chapters together fall flat. There is no unifying theory here, which is a shame.

Still, Levitt does his own number crunching, unlike Gladwell. And he forms some different conclusions. While Blink argues that our subconscious judgments are often correct, Levitt finds human error everywhere. We aren't very good at judging risk, assessing causality and avoiding temptation, he learns. But he relishes studying us anyway. --By Amanda Ripley