I read Freakonomics last week. My brother Sean has already provide a great synopsis, so I'm going to content myself with impressions. I found the book a quick and sometimes enjoyable read, but less so as the book went on. Very interesting info on the bagel man, cheating teachers, and real estate agents, but down hill aftwards. Not what I expected from a brilliant eoncomist writting a best seller -- and the cheating teachers section has nothing to do with economics.
I remember when I was in my one and only Economics class (yes, I loved the subject that much) the textbook, probably Economics by Samuelson, or possibly the professor went on about how economics was the study of scarcity. At the time I thought this a logical error, as economics is the study of the production, distribution, and consumption of wealth. Relative scarcity is an important component of wealth, but it isn't the actual subject matter. I bring this up because there seems to be a trend with economists who confuse techniques used in economics with economics itself. The use of regression analysis (or any statistical analysis for that matter) isn't performing an economics analysis, it's the use of a mathmatical tool widely used by many disciplines, including science and engineering. And while human behavior is an integral part of economics since we're the ones doing the producing, distributing, and consuming, economics isn't just about human behavior nor is all of human behavior of interest to economists. Once again the examination of incentives in areas outside traditional economics isn't economic analysis, it's human behavior analysis. So while I thoroughly agree with the importance of incentives to understanding human behavior, I suppose only in academic circles is the application of common sense thought to be revolutionary. (Yes, I know
The book is made up of chapters that don't have anything to do with each other than Dr. Levitt finds the subjects interesting. Of itself this isn't a problem for me, but there are times when the chapters are contradictory. For instance, Dr. Levitt claims in one chapter that abortion lowered the crime rate, and we can tell this because maternal characteristics such as education, age at birth, and marriage status -- none of which are genetically determined -- has a significant effect on whether their children become criminals. And let me be clear - as confirmation of his thesis on crime and abortion, he says that parents who have children they don't want understandably do a worse job at parenting than those that do so it's reasonable that such children are more likely to commit crime. In the next chapter on how much influence parents have on their children, the conclusion is that it's pretty much genetic. Parents matter because of their genetic contribution, not their parenting contribution. He arrives at this conclusion by looking at educational achievement. So if smart parents have smart kids, can we then generalize that parenting doesn't have much effect on their children? In one chapter parenting does, and in the other parenting doesn't.
And there are a couple of big threads left dangling that I would think somebody like Dr. Levitt would just love to pull. One is if the crime rate has returned to the rate of the first part of the century because of abortion, what else has changed that is keeping it there despite abortion's lowering affect? The other is if parenting (as opposed to the genetics of parents) has no effect on children, is crime genetic?
