The apt name would have been F**konomics, for the book hovers around the passing of abortion bill in USA. How can somebody write a book of 200 pages out of nothing is a mystery to me. What intrigues me more is that many newspapers had wrote great things about this book, a perfect case of hype creating a best seller.
All gas no substance. And nothing to do with economics rather than some stupid black-white demographics and some obvious facts. Thank God, I'm through with this.

what's wrong with this book?

1. this book has nothing to do with economics. Something like "freakostatistics" would make much more sense as the name of the book.

2. in the introduction, the writers warn us that correlation does not imply causality. OK, agreed. But the rest of the book is just that: THEY act as if correlation and causality are the same thing (but they just have more "fun" explanation for stuff). What exactly is their proof that they are not "inventing explanations" based on correlation, without any real evidence of causation?


This book is just a pile of interesting nonsense.

EDIT (2016): There is another thing in the authors' arguments that really bother me (besides the "correlation vs causality" debate). The authors' approach to economics is that economics is all about how incentives play out. While incentives are no doubt important in human behaviour, this is an extremely shallow way of talking about economics. It is also very consistent with "unfettered free market" ideology, which says that the best way to organize society is through profit incentive. This is a dangerous proposition (as we have seen in 2008) and scientifically very doubtful.

You know, I really enjoy economics and I used to listen to the Freakonomics podcast so I figured I'd enjoy this book more than I did. I think it's lack of a theme (which the authors clearly warn readers about ahead of time) contributed to my overall "meh" attitude. Additionally, I found myself yearning for more than just Levitt's observations and findings. For example, simply stating that data shows higher abortion rates yield lower crime rates and then not delving too deeply into the reasons that underlie this phenomena just didn't do it for me. That's an extremely weighty thing to state, regardless of if you have data to back you up or not. I guess I wanted some sociology mixed in with my economics. Alas, Levitt is an economist not a sociologist or political scientist.

I also want to add that as someone with an MBA (thus, I took classes on data in grad school and would be fine to never run a regression again despite how interesting it can be) and who works with data at her job, I know firsthand how easy it is to manipulate data. I want to assume that Steven Levitt was completely neutral in his research, but who can really know except for Levitt himself? And, who funded all of his studies? I realize Harvard comped him for some work, but what about his studies on Chicago Public Schools? I find it interesting that a school system would turn over their data to him whenever it was inevitable that he find some questionable (and illegal) things going on. I don't know. Again, I want to assume that he was completely neutral, but spinning data is pretty darn easy. And even assuming he completed his research with no bias, what about the actual data he was working with. Some of the "perfect parent" data (I think it was still Chicago Public School data) seemed very subjective. For example, how exactly was "a lot" quantified in the the data point regarding "parents who have a lot of books in their home"*? I have ~200-300 books and don't really consider that "a lot." You might disagree which is my entire point. Won't different people quantify and thus answer that question differently? Ok, I'm going too deep into this. Basically, I know first hand that you can tell whatever story you want to with data, so I wonder what approach Levitt took in order to remain unprejudiced and unbiased.

And finally, I imagine I wasn't the only one holding their breath while reading over the low income baby names. Not that it means anything, but it was still a bit jarring to hear how a person's name reflects their parents' income. Some of the names were very obviously high or low income names, but it was interesting to hear how many common names have moved back and forth on the different lists over time. That portion of the book was undoubtedly the most interesting, and seemingly the most objective--both from sociological and data standpoints.

*might not be the exact wording used, but I listened to the book and thus can't reference the print.