Steven D. Levitt, Stephen J. Dubner – Freakonomics

What has the legalization of abortion to do with the crime drop rate in the late 1990s? Nothing, you might think at first. The authors of the book Freakonomics would argue against your thought. Steven Levitt is a professor of economics at the University of Chicago. Stephen Dubner is a journalist working for ABC News and The New York Times. They met at an interview and decided to work together on several books. This book was a result of their productive cooperation.

The aim of the book is to look at everyday problems from the perspective of a scientist. Economics tend to deal with problems that are rather abstract for the masses. The authors of the book take a different approach and are asking questions like: What do schoolteachers and sumo wrestlers have in common? Why do drug dealers still live with their moms? What makes a perfect parent? Those questions will be answered by looking at data and evaluate it with statistical methods like regression analysis.

To answer the question of similarities between schoolteachers and sumo wrestlers, the authors identify first what both have in common: That they react to incentives. Incentives can lead to desirable behavior or to bad behavior (cheating). Schoolchildren are frequently tested to monitor their progress and to compare their results with other kids. But the test results also influence the teachers. If classes systematically achieve bad grades or made low progress, the responsible teachers will be blamed. Therefore, teachers have incentives for their classes to get good scores. Sometimes the teacher even fakes the result of tests (oftentimes without the knowledge of the children), to put himself/herself in a better light. By evaluating data of the Chicago Public School system, the authors could identify teachers that purposely alter the result of several test results. After their findings were published, the Chicago Public School system decided to fire teachers who cheated provably. As a result, cheating done by teachers fell in the following year by more than 30 percent.

But how is there a connection to sumo wrestlers? The system of professional sumo wrestling is highly hierarchical. The top players are millionaires, while the seventieth-ranked player in Japan only earns $15,000 a year. Therefore, your ranking as a player matters a lot. The ranking is determined by your performance in tournaments, in which every player usually fights 15 matches and tries to achieve a positive end score, because this leads to a higher rank. Consequently, the eight victory in a tournament is very valuable. So, what might happen if in the last match a wrestler with an 8-6 record is facing an opponent with a 7-7 one? The first player has a guaranteed higher rank because of his positive record while the other player needs to win badly. Is there a secret arrangement between the components which let the better player loses on purpose? The date affirms this question. When looking at the chance of a 7-7 contestant to win against an 8-6 one, it is only 48.7 percent. When looking at the actual numbers, the 7-7 contestant wins in 79.6 percent of the games. The same pattern can be seen when a 7-7 player faces a 9-6 one. The predicted win percentage is 47.2, while the actual win percentage is 73.4. Thus, both schoolteachers and sumo wrestlers have (for different reasons) incentives to cheat (but not all of them do it).

At the end of the 1980s, crime peaked in the United States. Violent crime had risen 80 percent and experts predicted even higher rates for the 90s. Surprisingly, crime in the 90s went drastically down, leaving those experts puzzled about the reasons for the unexpected drop. The authors of the book state several reasons that were discussed back then, among them explanations like innovative policing strategies, tougher gun-control laws, a stronger economy or an increased number of police. Surprisingly, they debunk every popular reason. For example, studies have shown that an unemployment decline of 1 percentage point will lead to a 1 percent drop in nonviolent crime. During the 1990s, the unemployment rate fell by 2 percent while the rate of nonviolent crime fell by 40 percent. Likewise, the changing of gun-control laws didn’t have any influence on the crime drop. Switzerland has one of the highest gun per capita rate in the world, because males are allowed to keep a gun at home after serving military duty. Still, the crime rate was much lower than in the US.

But what was the real reason for the sudden drop? In 1973, the Supreme Court legalized abortion in the famous Roe v. Wade case. Before that, abortion was illegal and expensive to get. Thus, mostly woman from middle- or upper-class families were able to afford it. After the Court’s decision, more unmarried or poor woman were able to afford an abortion leading to less children born in poor or single-parent homes. Infanticide fell dramatically, as did the actual birth rate. Since children from a poor background tend to commit more crimes, the authors conclude that the legislation of abortion in 1973 had the biggest influence on the dropping of the crime rate 20 years later.

Freakonomics has many more interesting cases, in which everyday questions were answered by methods used in economics and other social science. It is a well written book full of provoking ideas with tons of bonus material like an interesting Q&A with the authors. Sadly, the book has no bigger overall theme and is jumping from one case to another. Furthermore, some of the concepts used in the book (like incentives) could be explained more in detail.

Levitt, Steven D./Dubner, Stephen J. (2005): Freakonomics. A rogue economist explores the hidden side of everything, HarperCollins, New York.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s