Copyright © 2004 Phillips & Nelson Media, Inc.

Steven D. Levitt teaches economics at the University of Chicago; he recently received the John Bates Clark Medal, awarded every two years to the best American economist under forty.

Levitt is regarded as among the most creative thinkers in contemporary economics, gifted at drawing connections among seemingly unrelated forces. He believes, for example, that there is a relationship between legalization of abortion and the decline in crime -- more on that in a moment. Levitt is the latest winner of the American Economics Association's John Bates Clark Medal, granted biennially to the top economist under the age of 40. Clark medalists often become influential figures: Paul Krugman, an economist at Princeton and a closely read columnist for the New York Times, was a Clark winner, for instance. In typically dry economics-speak, the Clark judges declared, "Steven Levitt is the most innovative empirical researcher in his cohort." American Economics Association, just come out and say it: The guy is interesting!

Freakonomics presents Levitt's findings in accessible, non-academic terms. It is an engaging and always interesting work, rich in insights, full of surprises. Readers, though, may find themselves in a perpetual state of confusion regarding just what it is they are reading. The book is by Levitt and Stephen J. Dubner, a writer whose previous books include Turbulent Souls and Confessions of a Hero-Worshipper. You're never sure who is speaking. Sometimes Levitt or his work is spoken of in the third person, as if Dubner were writing of them in a detached way; sometimes the text sounds like Levitt addressing the reader; sometimes the book spends pages discussing work done by people other than Levitt, yet the impression given is that it's all Levitt's thinking. Often the reader must flip to the source notes to try to figure out what's going on.

The mishmash quality of Freakonomics seems to trace to its origin. In 2003, Dubner wrote a New York Times Magazine article on Levitt's quirky theories. The piece was a great read, and Levitt and Dubner began to collaborate on articles. According to Freakonomics, Levitt then agreed to write a book, but only if Dubner actually did the writing. (Writing is work, as any economist will tell you!) Yet it seems the two never resolved the question of whether Dubner would ghostwrite for Levitt or write about Levitt. The result feels too much like a magazine article padded to book length. Each chapter of Freakonomics even begins with quotations from the Times piece, as if it merited study by future historians.

Confusing structure aside, Freakonomics is packed with fascinating ideas. Consider Levitt's notion of a relationship between abortion access and the crime drop. First, Freakonomics shows that although commonly cited factors such as improved policing tactics, more felons kept in prison and the declining popularity of crack account for some of the national reduction in crime that began in about the year 1990, none of these completes the explanation. (New York City and San Diego have enjoyed about the same percentage decrease in crime, for instance, though the former adopted new policing tactics and the latter did not.) What was the significance of the year 1990, Levitt asks? That was about 16 years after Roe v. Wade. Studies consistently show that a disproportionate number of crimes are committed by those raised in broken homes or who were unwanted as children. When abortion became legal nationally, Levitt theorizes, births of unwanted children declined; 16 years later crime began to decline, as around age 16 is the point at which many once-innocent boys start their descent into the criminal life. Leavitt's clincher point is that the crime drop commenced approximately five years sooner in Alaska, California, Hawaii, New York and Washington state than it did in the nation as a whole. What do these states have in common? All legalized abortion about five years before Roe.

Other Levitt theories span the landscape. Along with sociologist Sudhir Venkatesh of Columbia University, Levitt has found that drug gangs are structured essentially like street versions of Fortune 500 corporations. Levitt studied high-tech car alarm devices and found they benefited the people who did not buy them almost as much as those who did, by discouraging all auto theft. Levitt has shown that campaign money has almost nothing to do with who wins elections. He has even looked at how parent name children. Parents who expect their children to go far in life give them classy-sounding names such as Katherine; parents who do not expect their children to go far give them names such as Brittney. (That Levitt can cite statistics for all this -- you'll have to read the book -- is impressive.) For amusement value, Levitt makes a case that academic cheating differs little from cheating in sumo wrestling.

Not all of Levitt's ideas meet the test of originality. He proffers that real estate agents serve themselves rather than their clients when they push sellers to take the first bona-fide offer, even if holding out might bring a better price. The agent wants the seller to take the first solid deal because then the agent gets the commission right away; holding out might mean weeks or months of extra work for the agent, while increasing his or her commission only slightly. Freakonomics presents the notion that homeowners and real-estate agents may have conflicting monetary incentives as big news. Memo to the University of Chicago Economics Department: Everyone who has ever sold a house already knows this.

Freakonomics proposes four basic notions: that incentives govern life, that conventional wisdom is often wrong, that "dramatic effects often have distant, even subtle causes" and that experts sometimes use their "informational advantage" to pursue private agendas. Valid points -- but is there anyone who disagrees with any of them? There is no requirement that an economist's work have any larger theme: To be a good professor and produce interesting papers is more than most accomplish. But Freakonomics leaves the reader with the sense of encountering an assortment of clever ideas that have been crowbarred together into something that doesn't really work as a book. Academic careers may not need unifying themes; books do.

Reviewed by Gregg Easterbrook
Copyright 2005, The Washington Post Co. All Rights Reserved.