Freakonomics
A Rogue Economist Explores the Hidden Side of Everything
Steven D. Levitt and Stephen J. Dubner
William Morrow, 2005
What do schoolteachers and sumo wrestlers have in common? Well, it has to do with incentives, a central concept in economics. The incentive is the economist's favorite tool for modifying behavior. Create the right incentives, and suddenly it's in a person's best interests to engage in the right behavior. But every incentive has a dark side — alongside good behavior, they also motivate cheating. In chapter one of Freakonomics, Steven Levitt explores mechanisms to discover cheaters. See, No Child Left Behind and programs implemented in some states and school districts have led many, many teachers to cheat in many different ways on standardized tests. In one particularly egregious example, a fifth-grader went home and told her mother how nice the teacher was to write the answers to the test questions on the blackboard.
Most teachers chose more subtle means, however. And anonymized data gathered in Chicago allowed Levitt to see which teachers had changed answers or filled in blank answers on tests. Levitt reasoned that the only practical way to do so would be to take some tests — not all, because that would be too suspicious — and memorize a sequence of correct answers in order to quickly boost their classes' scores, thus earning salary bonuses or merely preserving their jobs. He used a computer to look for strange occurences — students who tested noticeably worse on previous and future tests, easy answers missed while hard ones were correct, and strings of similar answers on a portion of the test.
Even this simple analytical technique, detecting only flagrant changes to students' tests, found that about five percent of the teachers cheated to improve their students' scores. Spikes in cheating occurred when the stakes on testing were raised. In short, standardized testing and incentives to improve students' mastery of academic material motivated some teachers to commit fraud. In a completely different field — sumo wrestling — the incentive to cheat is also clear: the very best rikishi earn high salaries and lavish lifestyles, while slightly worse wrestlers earn very little — and they have to help their betters soap up hard-to-reach places in the shower.
Sumo matches always take place in the context of tournaments; each player faces fifteen opponents, and if they win at least eight fights of the fifteen, their ranking improves. So Levitt analyzed matches in which a player with eight wins and six losses went up against a player with seven of each. The player with eight victories could afford to lose without hurting his ranking, while the match was vital for the place of the player whose wins and losses were even. And a stunning percentage of the time, the tied player won the match — and in subsequent battles between the same two players, the previous victor almost always lost. Some sumo wrestlers must have agreed to throw a match in exchange for a guaranteed victory next time.
So what's the deal with the book?
Each chapter in Levitt and Dubner's book asks a similar unlikely question: How is the Ku Klux Klan like a group of real estate agents? (It has to do with the control of information.) Why do drug dealers still live with their moms? (This chapter discusses the fascinating story of a young sociologist who ended up spending years in the company of drug dealers in Chicago learning about the drug business and the surprising parallels between a gang of crack dealers and the McDonald's corporation.) Where have all the criminals gone? (What caused the dramatic drop in violent crime in the mid 90s? I won't say, but Levitt's answer will trouble some people.) What makes a perfect parent? (It's surprising to learn all the things you can do as a parent that, statistically, won't make a lick of difference in your child's future.) And, would a Roshanda by any other name smell as sweet? (Do obviously African-American names hurt a child's chances for success in life?)
The questions that Levitt explores in the book are fascinating, and the answers even moreso. The writing (from what I can glean, it was written by Dubner, a journalist) is engaging and liberally sprinkled with fascinating anecdotes and digressions — why does a new car lose so much value as soon as it's driven off the lot? The authors freely admit in the introduction that there is no central theme to the work — it's merely a collection of interesting discussions about things rarely analyzed quantitatively.
The book is a breeze to read, but thick with information. The style is not academic but the ample appendix provides scholarly sources for the data used and the claims made. Summarizing this book is a challenge, because it's quite true that there simply is no central organizing argument to it — it's just an immensely readable collection of interesting facts and astonishing claims. Which is why it's so disappointing that the book is so short. Not including the appendix, the book is only 207 pages long, and it took me a day to finish it. It's like a bit of mental candy — compelling to be sure, but it was over too fast and afterwards I didn't really feel like I'd gained any new insights into the world. The answers to the questions that Levitt asks are surely fascinating, but with no overarching theme to bind them together and help understand the world, it's ultimately just a collection of very, very interesting trivia.
The book is popular and it's sure been talked about a lot; the style brings to mind Malcolm Gladwell's recent bestsellers, The Tipping Point and Blink. But unlike Gladwell's work, Levitt makes few grandiose claims about how the world works, and those he makes are well-supported. His genius is in asking questions that aren't often asked, and then using the tools he has — he states often that economics is about measurement — to answer them. He constantly questions conventional wisdom and makes the point over and over that correlation does not equal causation. He's also a shameless self-promoter; each chapter starts with a quote from Dubner's 2003 New York Times Magazine story about him, which led to their collaboration. The article appears to have been entirely composed from collected praise from other economists and stories to illustrate just how cool Levitt apparently is. The "explanatory note" at the beginning of the book starts with another such quote, one that begins: "The most brilliant economist in America — the one so deemed, at least, by a jury of his elders — brakes to a stop at a traffic light on Chicago's south side." Yeah.
The verdict? I'd say get it from the library, and do it as soon as you can. Don't cough up the money for it — especially in hardcover — it's short and you probably won't find yourself poring over the pages. It's more like a really interesting magazine article than a book, an article that just happens to last 200 pages. Like a good magazine article, it's entertaining and informative. There's no doubt you'll enjoy it — but it's also gone in a flash and his queries are surely insightful but without any central idea, they might not leave you with all that much to think about.
Thanks to Two Sheds for reminding me about this: The book has its own website at http://www.freakonomics.com/, which includes articles about Steven Levitt and the Freakonomics column, which is more writing along the lines of the book's content. | 
