Freakonomics by Steven D. Levitt and Stephen J. Dubner

The Blurb On The Back:

Levitt and Dubner’s Freakonomics is the cult hit that turns economics on its head, using surprising information about the world to understand what’s really happening under the surface of everyday life.  From how your name affects how successful you are, to why drug dealers live with their mothers (and the unexpected links between estate agents and the Ku Klux Klan), this book unravels the secret codes behind, well … everything.

You can order FREAKONOMICS by Steven D. Levitt and Stephen J. Dubner from Amazon UK, Waterstone’s or UK.  I earn commission on any purchases made through these links.

The Review (Cut For Spoilers):

Steven D. Levitt teaches economics at the University of Chicago.  Stephen J. Dubner is a writer for the New York Times and The New Yorker.  This book resulted from a profile that Dubner wrote on Levitt and was a phenomenon when first published in 2005, offering explanations for a variety of questions.  It’s a page-turning read that tells a good story but some of the statistics are questionable and its reliance on racial assumptions very telling.

I read the revised version of FREAKONOMICS that was published in 2006.  A further revision was published in 2020 and I want to make clear that this may address some of the questions/concerns I had about this book.  In the 2006 revision, Levitt and Dubner make some corrections of material that had come to light following publication of the first edition. The corrections are highlighted in a foreword and mainly relate to the questions about Stetson Kennedy’s role in publishing Klu Klux Klan secrets and the extent to which he overstated his involvement.  There is also bonus material at the back, consisting of Dubner’s original profile of Levitt together with some articles and blog posts that they wrote.  The authors also use the end notes for each chapter to answer criticisms raised about the work cited, most notably in relation to the argument that increased access to abortion was linked to the reduction in crime in the 1990s.

From the outset, the authors state that they wrote the book from a specific worldview based on the ideas that: 

– incentives are the cornerstone of modern life;

– conventional wisdom is often wrong;

– dramatic effects have distant, even subtle, causes;

– experts use their informational advantage to serve their own agenda; and

– knowing what to measure and how to measure it makes a complicated world much less so.

Each of the following chapters then aims to answer a specific question while weaving in each of the above principles. 

WHAT DO SCHOOLTEACHERS AND SUMO WRESTLERS HAVE IN COMMON? looks at incentives and cheating.  It is specifically focused on how some teachers in Chicago were incentivised to cheat on the standardised tests administered to their students and the similarities this has with the Japanese sumo league, where the statistics suggest that some bouts are fixed because of the benefits this creates for wrestlers.  Along the way, the authors look at incentive structures generally (including a famous study about an Israeli childcare centre, which decided to fine people who were late picking up their children, only to find that this resulted in more parents turning up late as they priced the fine in), how a change to the tax forms to make parents provide the social security numbers for their children resulted in 7 million children apparently disappearing from the population and how the owner of a bagel company delivering to office sites discovered that the middle classes and senior executives are perhaps the worst cheats of all.  

It’s an engaging read and an interesting introduction to the economics of incentives and disincentives and I was interested in reading about the reaction of the Japanese authorities and media to suggestions that sumo wrestling was in any way rigged and their deliberate reluctance to investigate further.  However, I wanted more consideration of the role that low pay has in encouraging cheating behaviour because it is something that the authors do not really address.  For example, in the case of the Chicago school teachers, they risked losing their jobs if the children in their class performed badly and given that these were generally schools in poor neighbourhoods, there was a real risk of the school being shut down.  Similarly, it’s the sumo wrestlers at the bottom of the system who benefit from any cheating because they could potentially gain a place in the ranking that improves their pay and conditions of work.  The authors use the Bagel Man study to suggest that executives cheat because it’s how they’re used to working but there’s no real examination of necessity versus privilege and I think that’s a bit of a blind spot here.

HOW IS THE KLU KLUX KLAN LIKE A GROUP OF REAL-ESTATE AGENTS? looks at the role of expertise in incentivising performance and in dissemination of information.  However, I have to say that I found the sections on real estate agent performance to be a lot more convincing in terms of the authors’ conclusions than their analysis of the KKK.  To be fair, the authors do a good job of setting out the history of the KKK and their activities and then bringing in the work of Stetson Kennedy, who together with another man infiltrated the KKK and then started to share the group’s secrets with anti-KKK media and radio shows.  However, there are some questionable statistics and viewpoints within the chapter.  

For example, the authors assert that the KKK’s threats of violence seldom went beyond the threat stage and they then produce statistics on the number of lynchings that took place to make the point that there were apparently more lynchings when the KKK were dormant following its becoming a banned association than when it became active again.  Those lynching statistics are doing a lot of heavy lifting there – for starters, there’s no consideration of how accurate they are or whether there was any incentive on police and the authorities in the South to cover up or minimise these crimes – certainly I’d want to take a look at corresponding “suicide by hanging” statistics available for this time.  Secondly, there’s the fact that public lynching was not the only act of violence perpetrated by the KKK and I’d also want to know what the general murder, bodily harm and property crime rates were at this time.  

The fact that neither author seems to consider this really caused me to raise my eyebrows because it seems like such a basic point for what is a highly emotive topic.  I’d expect them to make very sure that they are broadly correct rather than pick on something that fits their narrative.  It also made me question the use of statistics more broadly in the remaining chapters because I began to wonder to what extent they were cherry picking figures that supported their view rather than delving into the topic.  This worry wasn’t helped by the fact that Levitt is frequently described as not being very good at numbers or econometrics in terms of being an economist and you know, I’m not either, but when this is your actual job and you’re offering up figures and arguments to make people view questions in a different way, it’s somewhat important to make sure that you’ve locked it all down.

WHY DO DRUG DEALERS STILL LIVE WITH THEIR MOMS? was, for me, the most interesting and informative chapter in the book.  It’s drawn heavily from sociological research carried out by Sudhir Venkatesh who accidentally stumbled on a Chicago drug gang while carrying out surveys for another piece of work.  Venkatesh and Levitt collaborated on papers relating to the economics of the drug trade as a result of this research and it makes for fascinating reading, especially in the context of the impact of crack on drug dealing and the reasons for the corresponding high levels of violence that it generated.  

Unfortunately, the chapter does have the same racial blind spot as the preceding chapter, e.g. it suggests that crack cocaine was correlated with high levels of violence and incarceration in black communities, the increasing educational gap between black and white pupils and high levels of child abandonment in the black community.  This description comes right after a paragraph talking about the role civil rights played in improving black life expectancy, school outcomes and a lowering of the black-white income gap but at no point does it look at how crack cocaine hit black communities in the 1980s, which came with a corresponding crime crackdown (which disproportionately affected black communities) and a reduction in welfare programmes – most notably food stamps and Reagan’s fight against “welfare queens”.  You can’t talk about how civil rights and welfare rights improved outcomes and then not even take a look at whether there may have been a corresponding reduction in welfare rights at the same time as crack cocaine made an appearance.  The effect is to suggest that crack cocaine was responsible for the terrible effects on black communities when it may well have been part of a wider societal problem.

WHERE HAVE ALL THE CRIMINALS GONE? is the chapter that probably caused the most controversy within the USA because it’s the chapter where the authors look at the various explanations put forward for why the crime levels fell during the 1990s even though many experts had predicted that they would rise.  The authors put forward statistics to say that the fall was not due to improvements in the economy (although they provide interesting details on how the collapse in crack prices reduced violent crime) and also discount the rise in police numbers.  Their argument is that the fall is due to the impact of Roe vs Wade being felt as women who did not want to have children were able to procure abortions.  Although I found the evidence for a correlation between abortion and the drop in crime to be interesting, I can’t say that I was fully convinced.  This is partly because the same racial blindspot that affects previous chapters is present here and actually more overt as they quote Gary Becker’s assertion that “African Americans and Hispanics commit a disproportionate share of felonies” without any consideration of arrest and conviction statistics and how they are leaked to economic circumstance and lack of good legal representation.  Also, while their consideration of drug activity is interesting, they again fail to consider crime against social security provision plotted against the conviction statistics.  As such, although I do find this interesting, the deficiencies stop me from believing it.

WHAT MAKES A PERFECT PARENT? starts by looking at how bad parents are at judging threats and the economics of fear, specifically how some threats get more money thrown at them to tackle them from others, regardless of the actual risks that they represent and how fears are manipulated by marketing companies and lobbyists.  The authors then move on to look at what constitutes good parenting, which they assess in terms of school performance.  Again, race gets brought into it but it does look at other factors like parents’ education and corporeal punishment.  It’s fine but again, there’s no interrogation of the statistics and although they look at the impact of people moving to good schools, they claim that it doesn’t impact the scores of the children who move, which is a huge generalisation to make and fails to consider the performance of those children against the “richer” contemporaries attending the same school.

PERFECT PARENTING PART II picks up on the impact of parents on performance from the previous chapter but this time looks at the impact of names in terms of how they impact on life.  The chapter kicks off with an interesting story about a man who named one of his children “Winner” and one “Loser”.  From there, unfortunately, it goes downhill as the authors decide to look at Fryer’s work on whether distinctive black culture is a cause of the economic disparity between black and white communities or whether it’s a reflection of it.  The blindspot that the authors have on race really comes out here.  For example, they say that there’s a flaw in the data about whether people with “black” sounding names are discriminated against when they submit CVs compared with those with “white” sounding names.  It’s one of the few instances where the authors actually question the date, saying that even though Black people are less likely to be called, it’s not clear whether this due to racism or not because it could be because the interviewer thinks the person is from a low income home and I’m sorry but, come the heck on, guys.  Seriously?  

They then go on to argue that the studies don’t go on to show what would have happened at interview and again, come the heck on – the whole point is that if you have a “Black” name then you don’t get to that stage to find out.  And then to close it out, they argue that having the “Black” name doesn’t make a difference if you’re Black but have a “white” sounding name because it’s your motivation and support that matters and honestly, by this point I was about to throw the book against the wall because there are so many layers of assumption and complacency there that beggars belief, not least given the impact of expectations on all stages of life.  The chapter is more interesting in how baby names percolate through the social and economic strata and how parents use them to signal their expectations of their children.

The book finishes with an epilogue that decides to draw a comparison between Ted Kaczynski and Roland G Fryer, Jnr, which I’m sure the authors think prove that your background is no guarantee of future prospects and yet cuts across basically everything that’s in the previous chapters.  There’s also some bonus material, which is basically the original article that Dubner wrote for The New Yorker on Levitt and some other articles and blog posts, some of which reinforce what’s in the chapters or are what the chapters drawn on (in particular they look at the accusations about Stetson Kennedy’s work on the KKK).  Again, I had problems with these, notably with one about whether fear of AIDS can change sexual preference, which is pretty offensive in and of itself, which draws on one self-reported study about sexual preference carried out in 1992 when people were still deep into the AIDS panic and it was duelling anti-gay feeling and policy.  Also distasteful is a blog post reproduction where Dubner decides to out a restaurant that served him rancid chicken because he was too weak to ask for the recourse he wanted and instead wanted them to guess and offer it themselves.

Where I will give the authors credit is that this is a very easy book to read and the focus on stories and quirky questions means that it holds your interest and on that basis, I can completely understand why it was such a phenomena when it was first released.  However, given that this is a book that is aimed at ordinary members of the public rather than specialised economists or sociologists, the lack of openness about possible issues with underlying data is a big glaring problem and it particularly stuck in my craw given that one point the authors explain how correlation does not equal causation.

I’m going to conclude by repeating is that once more it is possible that the authors address some of these points in the 2020 edition of the book and they may well have tackled them in their blog.  However, on the basis of this book I can’t say that I would rush to read any more in the Freakonomics series.   

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s