Friday, August 01, 2008

Frakonomics - thoughts

If you've ever wondered about the link between abortion and crime, the motivations of sumo-wrestling cheats, or how the Superman radio show helped to bring down the Ku Klux Klan, then Freakonomics might just be the book for you. A collaboration between a prize-winning economist (Steven Levitt) and a journalist (Stephen Dubner), it is a very readable trip through some of the odder aspects of human nature, and an attempt to explain them by way of the application of microeconomic theory.

If that sounds like the dullest possible premise for a book, then you might be in for a bit of a shock. It reads less like a text book, and more like the thoughts of someone who is fascinated by people and seeks to understand them. I enjoyed this book for many of the same reason that I love The Hitch Hiker's Guide to the Galaxy. In it, Douglas Adams manages to pick out little parts of how we behave and make them obvious to us. (for example, the editor of the eponymous guide's decades-long lunch break remind us all of a colleague or two!) It's that kind of insight which this book provides: it takes tiny things that people do and shows their relevancy on a much larger scale.

It is an odd book in many ways, sometimes careering from topic to topic with wanton abandon. As the authors state openly in their introduction, they do not have a unifying theme, which may not appeal to every reader. However, a theme is not necessary for every book, and can sometimes be an unnecessary restraint. Unless you have a particularly deep-seated grudge against short stories or magazines, you shouldn't have too many problems here. When I was reading it, I felt the chapters were like six very well written and accessible journal articles.

As it turns out, that's almost exactly what they are. At its most basic, the book is a popularisation of Levitt's academic work, and interesting work it is to. The most eye catching claim he's made is that the legalisation of abortion is the reason that crime dropped in the mid 90's. The theory goes like this: the people who had abortions after its legalisation tended to be young single mothers who didn't want that child, the very children who are most likely to become criminals. It's quite a compelling thesis, although it has come in for some questioning from other economists. I should point out that they in no way endorse that drop in crime as a reason for allowing abortion (they take no position), but rather just state that it exists. At its most basic the thesis makes a lot of sense. If you abort all the children, there will be a drop in crime. That doesn't mean that you should abort all the children, just that the two have a causal relationship. It's those relationships which the book attempts to explore. If it has any message, it's that the conventional wisdom is often wrong, and that just because a conclusion is startling doesn't mean it shouldn't be discussed.

I had a couple of niggles with the book, the main one being the last chapter which asks whether the names that black parents give their children are economically disadvantageous in comparison with the names white parents give their children. Although it's an interesting question, it is more so in an academic than a general situation, and it reads like that, with lots of tables and not a lot of text. I also think it might have been of more interest to the authors than me, because between them they had six children under the age of six at the time of writing! [as a personal disclaimer I read this chapter while waiting for a blood test, so I concede that might have influenced my mood too!] The other problem that I had was that they sometimes exaggerated just how 'rouge' Levitt's theories are. He won what is effectively the Nobel prize for economists under 40 and is a tenured professor at the Chicago School of Economics, not really the hallmarks of a maverick outsider!

That said, as I try to get my head around Economics before I start my masters, I can't help but think that it could do with more people who write like this. Science has succeeded in bringing its stories to a wider audience in a way that economics certainly hasn't. You don't need to know the intricate details of Newtonian Mechanics to find the story of an apple falling on young Isaac's head interesting. Nor should you need a Ph.D in advanced econometrics to find it fascinating that most drug dealers still live with their mothers, swimming pools kill more kids than guns in the U.S., and that the school a child attends makes only a tiny difference to their academic performance.

Image: Penguin celebrations cover, Freakonomics site


Noel Rock said...

Fair point about economics making its case in a more appealing way to the masses Stephen. Freakonomics is my favourite book and what made me consider studying economics (particularly social economics) in the first place.

You might want to pick up either of Tim Harford's books or have a look at the website

Keep up the good blogging!

Noel said...
This comment has been removed by the author.
Noel Rock said...

Perhaps I should scroll down in the future - you've already read The Undercover Economist indeed! Check out the website in that case, should be up your newfound street.

Darryl said...

You may be interested in the book "Predictably Irrational" which took a similarly narrative approach to behavioral economics and stormed the blogosphere a few months ago. I've yet to read it, but I gave it to my father who loved it.

You may also be interested in the Freakonomics blog It covers some of Levitt's new research, similar theories, and occasionally follows up on the book's research.

randombassist said...

Thanks guys, I'll have to check that stuff out alright!

LindieNaughton said...

Must read this book Stephen - have picked it up more than once, but my one year of economics in UCD back in the Dark Ages scarred me for life. Macro economics at 10am on a Saturday - ARRRGGH!

randombassist said...

On a Saturday?! That definitely doesn't sound like the cure for a Friday night out! I can lend it to you next time you're over if you want.