So I just finished reading Freakonomics. It was an entertaining and fast read - I bought it Monday night and finished it Tuesday night.

The authors Steven Levitt, an economist at the University of Chicago, and Stephen Dubner, New York Times Magazine, claim there is no unifying theme in the book. I have to disagree. The unifying theme is the the nature of causality and correlation and how we commonly confuse the two.

Each of the chapters asks a question, some more absurd than others:

  • What do schoolteachers and sumo wrestlers have in common?
  • How is the KKK like a group of Real-Estate agents?
  • Why do drug dealers still live with their moms?
  • Where have all the criminals gone?
  • And what makes a perfect parent?

Besides poking holes in the conventional wisdom about certain topics, the book is an interesting journey into different subjects. They describe how one man helped destroy the KKK by publicizing all their silly little secrets and rituals on the Superman radio show. Once people realized how infantile the KKK was their membership started dropping like a rock. The point was to show the power information has. Once the KKK’s secrets became public, the group lost its mystery - and much of its ability to generate fear. It is also why Real-Estate Agents, along with lawyers, doctors and insurance salesmen, have an advantage over the rest of us in their areas of expertise.

In “Why Do Drug-Dealers Still Live With Their Moms?” they tell the story about another economist who went to live with drug dealers for a few years. Levitt describes in detail the organizational structure and business model of the Black Gangster Disciple Nation of Chicago. It operates as a franchise. There are the owners, called “the board of directors”, the local franchisee, his top two lieutenants, the foot soldiers (the ones who sell the crack), and the members (who pay dues and one day hope to be chosen as foot soldiers). While there is a lot of money to be made in illegal drugs, like a McDonalds franchise the low end foot soldier earns next to nothing ($3.30/hr) by the time everyone gets their various cuts. Poor inner-city kids don’t become drug dealers for the money - they become drug dealers in the hopes of getting promoted to boss. For every millionaire rock start there are hundreds of musicians that never make it.

One of Levitt’s most controversial claims is that the dramatic drop in crime from the 1980s to the 1990s had nothing to do with the conventional wisdom (he debunks all the usual suspects: strong economy, stricter gun-control laws, better and more police, larger prisons, etc). It had to do with a Supreme Court decision made 20 years earlier - and still evokes controversy today: Roe v Wade. The logic is that the mothers most likely to get an abortion were also the most likely to raise children that would become criminals. By legalizing abortion those children were never born and never became criminals. 16-18 years later, when those kids would be hitting their prime criminal years, the crime numbers started dropping. He uses various techniques to prove causality rather than correlation including watching how states that legalized abortion before Roe v Wade saw their crime drop begin before the other states. He finishes the chapter with an economic case that shows, via cost-benefit analysis, that abortion is not an effective crime fighting tool.

The last two chapters discuss parenting, and the various different things parents do to help their kids succeed. By looking at the data Levitt shows that while there is a correlation between reading to your child and their success in school, that the reading is not the cause. By comparing adopted children with natural born children, he shows that genetics is most often the cause of success, not parenting, and that when parenting does affect success it has more to do with what the parents have done prior to having children, rather than what they do as they are raising the child.

The book as some annoyances. Between each chapter they quote one of Dubner’s NYT Magazine articles about Levitt. These citations are incredibly self serving and really distract from the quality of the book.

All in all, I recommend it for two reasons - it helps people understand the differences between causality and correlation - and it is a fascinating insight into several diverse subjects.