The Shotgun Blog
Sunday, August 21, 2005
My review of Freakonomics
My review of Steven Levitt and Stephen Dubner's Freakonomics: A Rogue Economist Explores the Hidden Side of Everything was published in today's Halifax Herald (it can also be read at Sobering Thoughts). Freakonomics has garnered a lot attention because of Levitt's "finding" that Roe v. Wade reduced America's crime rates despite the fact that his treatment of this issue takes up only seven pages in a book of 242 pages. Like other reviewers, I focused on the abortion-crime link:
"Leave aside two serious problems - the correlation equals causation fallacy, and the fact that Levitt doesn't prove that unwanted children are more likely to commit crime - Levitt's thesis fails an important methodological test.
As journalist Steve Sailer has demonstrated (and debated with the author), Levitt ignored the crime wave of the early 1990s by examining crime rates only in the years 1985 and 1997, thus missing a spike in crime that correlated with the crack epidemic in the intervening years.
Another problem is that Levitt looked only at crime rates, not crime rate by age. If he had done so he would have found that for the first post-Roe generation, the rate hadn't fallen at all. Furthermore, it grew significantly among the black population despite the fact that blacks are three times more likely to abort a pregnancy than whites.
In other words, Levitt only examined some of the data. Whether or not he did so intentionally one cannot know, but it seems strange that Levitt makes no effort to engage these criticisms in his book. The fact that he didn't may demonstrate that he can't refute them."
Levitt's examination of the abortion-crime link is indicative of both his interests (addressing non-economic using economic analysis) and his style ("Levitt has written a provocative book and like many provocateurs, he asks questions and finds answers that are always counterintuitive but not always correct (the abortion-crime link) or not necessarily demonstrably correct").
Overall, the book never rises above the level of being really neat. I conclude the review:
"Freakonomics is an interesting book. As the English say, it's a good read. Whether or not it is useful is another issue entirely.
If mere entertainment is an incentive to buying a book on economics, by all means purchase Freakonomics. But if utility is the motivation for your reading, you might be better off skipping Levitt's first offering."
I say that in all seriousness. It was, at times thought-provoking, but rarely persuasive. It argued for asking questions and answering those questions with economic analysis, but did not instruct the reader on how to do this. If any economics book can be considered beach reading, this is it, and indeed I have seen more than one person reading Freakonomics at the beach.
TrackBack URL for this entry:
Listed below are links to weblogs that reference My review of Freakonomics:
Freaksonomics: The Librano$$$$$$$$$$$$$$$$$
Wiped by a $5 bill........ to not coin a periphrasis. (Gigi is learning them voodoo economics (from Hatie)). Includes those old f*cus groups, again....
Link from Bourque.com>>>>>>>>>>
By DEAN BEEBY
Monday, August 22, 2005 Page A7
OTTAWA -- The federal government has backed off from a proposal to replace Canada's $5 paper currency with a more economical $5 coin and use the savings to help fund the country's Olympic athletes.
The offbeat suggestion, first raised by the Royal Canadian Mint, was taken seriously enough by the Finance Department that it commissioned a polling firm last spring to conduct a focus-group study.
Environics Research held sessions in early May with small groups of Canadians in Halifax, Hamilton & etc. [Governance by polls]
Posted by: maz2 | 2005-08-22 5:14:32 AM
What? Try again.
Freakanomics is a great read. It is an interesting book in an ocean of boring, boring wasted paper.
Posted by: Brock | 2005-08-22 5:44:33 AM
A (repetitive) suggesion: you will learn far more about economics by reading anything by Murray Rothbard (most of whose works are available for free online) than you will learn from reading statistical wanks like Freakonomics.
The lessons of economics are not to be learned from letting your government "do it to you", then spending hundreds of hours poring over government statistics trying to correlate things like dead fetuses and burglaries. The essentials of economics are to be found in everyday common sense, coupled with a clear view of the fundamental human rights of life, liberty and property.
Statistics can be useful, but only those numbers which are less prone to political and other biases. Such as: payroll (as in, government vs. private), money supply (e.g. M3), savings rate, government revenues and spending, and maybe a few others. To be avoided as much as possible (unless making allowances for their bias) are statistics such as GDP, CPI and anything having to do with unemployment.
Posted by: Justzumgai | 2005-08-22 9:15:34 AM
Posted by: EzrasLuvChild | 2005-08-22 11:44:50 AM
Doesn't a person get utility from entertainment? Perhaps you could have put it in terms of a more rigorous read. Anyone with economic knowledge is likely going to find the book falling short in one or more areas. It's an econ book for newbies.
Posted by: Aaron | 2005-08-22 3:15:18 PM
Whatever happened to post-positivism? Do we really still believe that economics is based on empiricism? Terms like causation fallacy read like a 1952 Econ textbook fumbling to find a link between tail finned Cadillacs, TV dinners and duck hunting. In other words - when it comes to casaulity, one must consider the possibility that everything is spurious.
And let's not forget the central tenets of the discipline itself: It's theoretical, working to provide approximations and simulations, not measurements. That's what finance and accounting are for. These are not only largely consensual within the academy, but directly manifested through the efforts of economists working in the for-profit sector.
Thanks for missing the point in entirely, kitten.
Posted by: GrammarMuffin | 2005-08-22 11:06:05 PM
I actually liked the part on naming patterns. Sure, it didn't get much attention in the press, but it was the most interesting part (maybe because my wife is pregnant).
For example, the girl named Sh*thead (pronounced Shatheed). The the Lane brothers were also interesting to read about.
Posted by: Half Canadian | 2005-08-23 12:36:42 PM
The comments to this entry are closed.