The Social Affairs Unit

Print Version • Website Home • Weblog Home

Use the buttons below to change the style and font size of our site.
Screen version     Print version:   
April 20, 2006

How to write a best-seller about economics - first win the John Bates Clark Prize and then don't write about what most people think of as economics: Freakonomics - Steven D. Levitt and Stephen J. Dubner

Posted by Tim Worstall

Freakonomics: A Rogue Economist Explores the Hidden Side of Everything
by Steven D. Levitt and Stephen J. Dubner
Pp. 256. Penguin, 2006 (Allen Lane, 2005)
Paperback, £8.99

Take a series of academic papers, add journalist and republish. Doesn't sound like the basis of a bestseller really, does it, but that's the basic conceit of Freakonomics, take a few years of the research of Steven Levitt, add Stephen Dubner and watch the royalty cheques roll in. A word or two of caution for those hoping to repeat the success and have their own runaway hit of an economics book (Macroeconomics: Its Big and Its Bouncy! perhaps?): Levitt has just been awarded the John Bates Clark Prize, something the profession gives to the best American economist under forty each year so the competition is quite stiff and more importantly, he doesn't actually publish papers on what most people think is economics at all.

Rather, what he does is take the economist's tool kit of insights and mathematical techniques and apply them to interesting questions. Economists think this is economics but those outside the hallowed halls seem to think that economics is all about currencies, interest rates and GDP. None of these are discussed in the book at all. The Guardian's spoof review closed with:

What is the probability that a collection of often trivial and obvious data will be passed off as brilliance? Regrettably high.
Which really only goes to show that The Guardian (Quelle Surprise!) hasn't quite got the point of economics. To misquote Paul Samuelson it’s all obvious or trivial (he actually meant all results in all of the social sciences) except for David Ricardo on trade.

The results reached may indeed seem trivial, that, to take one chapter, both school teachers and sumo wrestlers cheat. They may also, once they've been explained, seem obvious as well. If teachers' pay or continued employment as a teacher depends upon their class getting certain exam results then a simple knowledge of human nature will lead us to assume that some will cheat to achieve those results. But how do you actually go about proving which ones? Or how they've done it? This is where Levitt's specific talents come to the fore. He looked at a mass of data (30,000 students per grade per year, 700,000 sets of answers and nearly 100 million individual answers) and added the basic insight that incentives matter before doing his number crunching. He was able to show that cheating did indeed occur, design a trap for those suspected of it and while only twelve teachers were fired, the following year such cheating fell by 30%.

It's the combination of these two things, the ability to manipulate and interrogate those large data sets plus basic economic insights that marks Levitt's work. As Glenn Whitman notes (in his entertaining little game, The Two Things About Everything) the first thing that you need to know about economics is that incentives matter. Levitt uses this, as I say very basic, insight in two ways. What are the incentives that are leading people to act in this specific manner? Secondly, if we suspect that the incentives are likely to make people operate in a specific manner, how would we go about constructing a data set to find out whether they are?

An example of the latter would be whether real estate agents work quite as hard on behalf of their clients as they really ought to. In the US they'll get 6% of the transaction (less given commission shares and expenses) so we would suspect that they might be willing to accept a lower price in order to get the sale done rather than advertise or negotiate more to get the very finest terms possible. Levitt looks at the way in which said agents sell their own houses as opposed to those of their customers and finds that yes, they do take a longer period of time and do get better prices. You can call this result either obvious or trivial at your leisure for once you've understood that a higher price of $10,000 for a house means, if the house is the agent's own, $10,000 in the bank but if it is a customer's perhaps $150, well, the world works just as we think it would. What Levitt has done is work out a way to test the proposition and then prove its truth.

The same techniques are used to investigate a number of other conundrums. If drug dealers make so much money why are most of them still living at home with their mothers? The answer being to do with the economics of tournaments where only the winners make fortunes, the new or failing players getting little or nothing. In the drug dealing world of course the returns can be highly negative: in the gang and period studied the death rate was higher than that of those already convicted and sitting on Death Row in Texas.

An investigation into the power of information amongst the KKK leads to the terrific story of Stetson Kennedy (related to the hat family not the drink-driving one) who infiltrated the Klan and then leaked all of the passwords and "secrets" to the script writers of The Adventures of Superman radio show. Within a week children all over the land were playing Superman against the Klan and one Kladd (well, these titles are sort of ridiculous aren't they) said:

Our sacred ritual being profaned by a bunch of kids on the radio!
The Klan never really recovered from this episode and most certainly never wielded as much power again. This should have some resonance to UK readers as I'm convinced that one of the reasons Mosley didn't do all that well here was P. G. Wodehouse's invention of Roderick Spode and the Blackshorts. Pompous and preening egomaniacs tend not to know how to deal with people laughing at them.

The most contentious of his claims is the one about the fall in crime rates in the late 80s and early 90s in the US. Various ideas have been put forward, changes in policing, demography, higher incarceration rates but none of these, when analysed, seem to quite explain the precipitous drop. Levitt looks at the implications of Roe v Wade, that Supreme Court ruling that outlawed restrictions on abortion. Taking as his basic idea the thought that it is poor women carrying unwanted children who took advantage of the change, and that unwanted children of poor women are most likely to become criminals in their late teenage years, he concludes that crime rates dropped because the criminals weren't born. Gruesome as the explanation is it does seem valid, the killer fact for me being that the crime rates dropped earlier in those States which had legalised abortion before Roe.

There might be a paper or two in it for some bright spark in the UK actually. Roe was in 1973 and our own Abortion Act in 1967. Our own precipitous drop in violent crime should therefore have been in the mid 80s: I can't say that I actually recall that happening. Several theories occur - from my own failing memory to perhaps that our changes in policing were so flawed as to actually increase crime so much as to mask the effect? Perhaps even, it isn't the poor that actually have the abortions in the UK? Isn't fertility highly skewed towards the chav classes?

Of more immediate interest to UK readers might be the discussion of what actually affects the life prospects of children. We find that Polly Toynbee is wrong. (Gosh, the world is full of surprises today.) One of the things that does not make a difference to the education, career or future of a child is whether it attends Head Start or not. When that programme makes it to our shores it is called Sure Start and as it makes no difference we don't have to worry about spending 2% of GDP on it, the money can continue to fructify in the pockets of the populace.

No non-fiction release is complete these days without a blog to push it so here is the Freakonomics blog. Reading Levitt's entries you can see (as he admits in the book) why he took on a co-writer.

One final point. American spelling is used throughout in the UK edition of the book. Come along now Penguin, if you're going to do that why bother to have a separate edition at all? Do buck up!

Tim Worstall graduated from the LSE and immediately went into small business where he has remained for twenty odd years, working in the US, UK and Russia in fields as diverse as newspaper distribution, offshore programming and exotic metals. He is the author of 2005 Blogged: Dispatches from the Blogosphere and blogs at

Comments Notice
This comments facility is the property of the Social Affairs Unit.
We reserve the right to edit, amend or remove comments for legal reasons, policy reasons or any other reasons we judge fit.

By posting comments here you accept and acknowledge the Social Affairs Unit's absolute and unfettered right to edit your comments as set out above.
Isn't fertility highly skewed towards the chav classes?

I have often pondered the why and wherefore of this. The wherefore, in terms of a biological or sociological explanation, evades me. However, when I go into Pagan mode, to my mind the why is only too clear.

I have a strong Darwinian background, and I have no difficulties with the theory of evolution, nor do I need to resort to “Intelligent Design” or any other such hypothesis coming from across the pond. However, I do observe that one apparent consequence of Darwin’s theory is that people became even wiser in their own eyes. Thus, the fertility phenomenon mentioned above, and the concomitant increasing difficulty of finding native scientists in this country, may be viewed as a Darwinian Nemesis punishing the hubris of our nation.

Posted by: Robert H. Olley at April 21, 2006 09:56 PM
Post a comment

Anti-spambot Turing code

Creative Commons License
Except where otherwise noted, this site is licensed under a Creative Commons License.

The Social Affairs Unit's weblog Privacy Statement