Spirit Level authors hit back at Policy Exchange


6:10 pm - July 8th 2010

by Newswire    


      Share on Tumblr

The right-wing thinktank Policy Exchange today published a report titled Beware False Prophets – which takes aim squarely at the book Spirit Level and tries to question many of its assumptions.

In an article for Guardian Comment is Free, Natalie Evans writes: ‘A new report fatally undermines the authors’ claims about the link between income equality and social problems’.

But the authors of the book Spirit Level, Professor Richard Wilkinson & Professor Kate Pickett, responded today with a full rebuttal.

We publish their press release here:

——

  • The
    Spirit Level
    is based on many decades of research by its authors and other respected academics – it represents a synthesis of research and critical thought that has been subjected to stringent and robust quality control before being widely disseminated.

  • All analyses of income inequality and health and social problems in The Spirit Level have been either: (a) replicated by other researchers, in some cases hundreds of times, or (b) published in peer-reviewed academic journals This is fully referenced in The Spirit Level, but Peter Saunders is either unaware of this very large body of evidence or has chosen to ignore it.

  • The selective removal of countries suggested by Peter Saunders does not have the effect of removing the relationship between inequality and health & social problems. The Index of Social Problems (http://www.equalitytrust.org.uk/images/index-graph-inequality.jpg) remains statistically significant even if those countries suggested for removal – Japan, Norway, Sweden, Finland, USA and Portugal – are
    disregarded.

  • Peter Saunders analysis includes much poorer countries. The Spirit Level explicitly restricts analysis to rich, developed market democracies, where average levels of income are no longer related to average life expectancy, happiness or quality of life. Confining the analysis to the richest countries very clearly demonstrates the effects of relative income (Fig 1.4 in The Spirit Level) which contrast so clearly with the lack of effect of absolute income (Figure 1.3 in The Spirit Level). By including poorer countries the sharp distinction between relative and absolute income is lost.

  • Saunders is wrong to claim, in analyses of the US states, that many of the associations are explained by the proportion of African Americans in each state. There is a detailed, empirical argument against Saunders’ claim and other researchers also show his analysis is incorrect.

  • Saunders misunderstands the evidence that shows that almost everyone does better in more equal societies. The Spirit Level does not say that everybody in a more equal society does better than the highest social class and income groups in a less equal country. It shows that for any given social class or income level, people do better than their class or income counterparts who live in a less equal society.

  • The Spirit Level is sometimes called a ‘theory of everything’ but the book makes it clear that it is a theory of problems which have a social gradient – that is, problems which become more common further down the social and income ladder. Saunders ignores this and chooses counter examples such as suicide rates which do not have this social gradient.

——–

    Share on Tumblr   submit to reddit  


About the author

· Other posts by


Story Filed Under: News

Sorry, the comment form is closed at this time.


Reader comments


Why is the Guardian permitting a Deputy Director of Policy Exchange to plug a report issued by her own think-tank and in which she has a direct political interest?

2. Ken McKenzie

@1

Because the Guardian likes a bit of a row to bump up the hits, and printing a piece from the reliably dreadful Reform is essentially offering out their entire readership, especially that part that cares about education, evidence, and not just making things up to fit a political agenda.

3. Luis Enrique

here is some more thoughtful analysis of The Spirit Level.

There’s a difference between the general idea that inequality is bad, for a range of reasons, and the explanatory weight that book gives to inequality. It’s possible to broadly agree with the authors, and yet be unconvinced by the specifics of what they claim.

@ 1

because comment is free?

Sorry, but as a piece of social science The Spirit Level is simply dreadful. Pretty good as a piece of political polemic, I’ll give you that, but as a piece of science it’s awful.

Yes, I have read it, when it first came out, and I got raised eyebrows about five paragraphs in. They quite correctly state that there is a diminishing marginal utility to wealth and then in the next para state that, having shown that there is no utility to increased wealth therefore we don’t have to worry about economic growth.

To move from diminishing to none is a simple rhetorical trick, a political one, not the mark of careful and concerned scientists.

That they’re also ignorant of the economic research into this point doesn’t help: the level of wealth may have diminishing marginal effect upon happiness but the rate of growth does not. Societies without economic growth (or, heaven forfend, shrinking economies) are notably less happy than societies with economic growth.

By the time I got to their look at inequality of income and working hours I was pretty cynical. When I looked at their data set for that I actually laughed out loud. For, as you might surmise, they showed that low income inequality was associated with low working hours. In this data set though, they manage to drop Japan….something of an outlier, yes, but a country with extremely long working hours and low income inequality (although, absolutely not, low social inequality).

So, they drop outliers when they don’t confirm their prejudices and include them when they do.

Great political polemic, not good social science.

6. Luis Enrique

I don’t know what I’m doing wrong trying to get html links to work – here’s the one I tried above:

http://lanekenworthy.net/2010/01/18/inequality-as-a-social-cancer/

the author shares the same political sympathies as The Spirit Level, but still has doubts about the quality of some of their results. Really worth reading.

8. John Meredith

The response from Wilkinson and Pickett is pretty poor, just a lot of hand waving, really. It doesn’t bode well for their defenders.

I’ve just read the Policy Exchange critique right through and it’s pretty devastating. The ‘rebuttal’ doesn’t even begin to address the main points. I’m afraid the Spirit Level is toast. Shame, because it was quite an attractive notion.

The Guardian site published this rather shabby hatchet job because it’s desperate for hits and knows something controversial like this will attract comments and wider attention. It’s got nothing to do with admirable notions that ‘comment is free’.

The authors’ response is far from poor. It is a straight-to-the-point, dignified, factual rebuttal of accusations thrown at them. I’ve read their book and it’s – contrary to Tim Worstall’s claims – much more than just stirring political polemic. It presents an enormous wealth of evidence, and outlines the methodology responsible for getting the evidence and reaching their conclusions. It remains unanswerable.

@10 – Luis’s link @6 however is not shabby

12. John Meredith

“The authors’ response is far from poor. It is a straight-to-the-point, dignified, factual rebuttal of accusations thrown at them. I”

Except that it isn’t. They don’t address any of the specific points. All they do it say: there’s loads of other research that says the same as us, peer review, peer review, so there!!!!

13. John Meredith

” It presents an enormous wealth of evidence, and outlines the methodology responsible for getting the evidence and reaching their conclusions. ”

I should say that this is true. The Spirit Level does have a lot of evidence and the methodology is clearly outlined. But that is why it is demolished by the Policy Exchange article: they use its own methodology to show why its conclusions are shallow or wrong (except in, I think, two areas).

I agree with Jerry that it is a pity because if the book had been robust it would have changed everything, but I guess that is why we should have been suspicious.

@ John Meredith

You should read the linked press release, which includes footnotes giving the detail of the arguments on race in the US and the literature. People commenting on the Guardian article have noted that this report is flawed mainly because it tries to take down The Spirit Level – which is one popular science book summarising the wider academic literature with some original research – and thinks that this does the job on its own. You can’t “fatally undermine” a theory by going after what is bound to be the least rigorous treatment of the theory, and ignoring the rest of the field. That is cherry picking.

Also, the Policy Exchange report (which I haven’t read in full, but I have skimmed the first 50 pages or so) doesn’t actually use the methodology of TSL, it flits between doing that, using different methodologies, and standard non-empirical argument depending on which is most convenient. If you really think that TSL would “change everything” were it true, I think you are underestimating the ability of people to block out inconvenient evidence.

I should also point out that there is at least one error in the report which suggests malice – on p. 25 he states that the TSL authors never explain why they use the Gini coefficient for US state comparisons and a different measure for international data. In fact they do, quite clearly in a quite short and comprehensible segment, from which he takes quotes.

Oh I should say that I don’t really know whether the Policy Exchange report is wrong or right generally in its conclusions. The “outliers” point seems a fair one.

However I am still a bit sceptical, based partially on the fact that he wants to erase the Scandinavian countries and frequently Japan from the data, as well as the US and Portugal, and then seems shocked that removing the most equal and most unequal countries from their sample weakens some of the relationships (not the broadest relationship though, as they point out in their press release). And of course partially because the think tank has an obvious agenda.

It should also be noted that both of the authors have day jobs, and so it’s a bit weak to say “oh, their response doesn’t bode well”. Not everyone works for a thinktank who can fund a rapid-rebuttal squad. They point to the literature, which apparently supports their argument, which seems reasonable. What do you want them to do Jerry, knock out a rebuttal report in a spare 25 minutes?

16. John Meredith

“However I am still a bit sceptical, based partially on the fact that he wants to erase the Scandinavian countries and frequently Japan from the data, as well as the US and Portugal”

No, you have misunderstood. What he does is standard practice in statistical regressions and what should have been done by the authors of The Spirit Level. He checks to see what effect outliers and groupings have on the outcomes and finds that is is very large. Imagine, for example, that the government produced a graph to show that average income in, say, Hackney showed a steep upward trend since the coalition government came to power and made important claims based on that fact. Now imagine that you checked the data and found that Bill Gates had recently moved to Hackney but that his income was an outlier when you ran a box graph. Remove Bill Gates and the upward trend in income vanishes. You would want to remove the outlier, wouldn’t you? (And in fact you must if you want the trend to mean anything.)

According to the right-wing Policy Exchange the Spirit Level is flawed on every account except for the link between inequality and infant mortality. So… do the Policy Exchange think dead babies are a happy price for inequality? Is that not a good enough reason for reducing inequality? [Assuming that their report is correct, and on first reading it seems more flawed than what it purports to crtique…]

anon @ 17

The lowest infant mortality rate is in Singapore and the second lowest in Bermuda. I don’t know how these countries rate for income equality, but neither strikes me as obviously egalitarian.

meanwhile, Cuba (which makes a big point of its egalitarianism) is at 44th place…. only two places away from the stubbornly unequal United States.

So it looks as if the Spirit Levellers may be wrong about that too.

@18

Smaller countries are not included in the SL research for reasons given within the text. The graph showing correlation is here: http://www.equalitytrust.org.uk/images/infant-mortality.gif and the Policy Exchange agrees with it showing that more inequality = more infant death. So the question remains: why does the Policy Exchange think that is a fair trade-off?

Anon @ 19

Sorry, still not convinced. I was talking to a friend earlier who works in a big hospital NICU and says there are only about 3000 infant deaths per year. Most are in the first 28 days. About 70% are attributable either to genetic factors (usually malformations of key organs) or to problems associated with prematurity. He reckons some of those attributable to prematurity might, at a pinch, be caused by poor diet in pregnancy but the numbers are so small as not to be statistically significant. He can’t see how social factors like income inequality would have much effect here with an NHS free at point of use. In countries where the poor have to pay for neonatal care – there’d be a big difference.

21. Luis Enrique

Jerry,

it could be down to variation in the quality of hospitals. Perhaps countries with more inequality have geographic concentrations of poverty that are associated with unusually bad hospitals.

[I’m not putting forward this argument – I have no idea. But there are statistically significant differences in mortality across hospitals. Here’s an unrelated example to illustrate the possibility: ]
http://www.bristol.ac.uk/cmpo/publications/papers/2010/abstract237.html

22. Richard W

5. Tim Worstall

‘ They quite correctly state that there is a diminishing marginal utility to wealth…’

If there is a diminishing marginal utility to wealth the further up the income scale one goes, which is clearly correct. There should be no problem taking income from those where it has least marginal utility and redistributing it to those where it has the greatest marginal utility. In those circumstances societal utility is raised because the increase in utility for the poor is not offset by the loss of utility for the rich. Clearly an efficient use of resources.

Wonder if Policy Exchange’s original research this time is as strong as the time it turned out some of their evidence was faked.

I’m still waiting for them to sue the BBC!

24. Rhys Williams

Isn’t this an irrelevance.
Even if the report was correct and it may be. Would not most right wing posters (Jerry and John B ) say “tough”. That is life.
Inequality is never going to stop.
In so called communist country there massive inequality between members of the politburo and a worker in tank factory.

25. Rhys Williams

Sorry I meant to say John M not John B

And I wouldn’t call Jerry (me) right wing. As I said earlier in this thread, I was initially attracted by the ideas in the SL. But the refutation by PE (actually, it’s only published by PE – the work is written by a prof from Sussex Uni) really is so damning. To argue to change society on the basis of the SL is to argue on the basis of a false prospectus. There are surely enough good reasons to do good things as to have no need for iffy propaganda.

27. gwenhwyfaer

What struck me most forcefully when I heard about the Policy Exchange’s refutation, criticising the methodological and analytical flaws in The Spirit Level, is the article I’d read recently which talked about the effects of confirmation bias – that when people who believed X were shown two pieces of work, one supporting X and one refuting it, they were far more willing to point to all the methodological problems in the work refuting X. And this held true for both left and right X.

I’ve not read either TSL or the refutation, so I can’t comment on the individual case. On the other hand, the fact that nothing here is surprising doesn’t encourage me to take it seriously. Show me a left-wing rebuttal of TSL, on the other hand, and I’ll get interested.

28. Rhys Williams

And I wouldn’t call Jerry (me) right wing. As I said earlier in this thread, I was initially attracted by the ideas in the SL. But the refutation by PE (actually, it’s only published by PE – the work is written by a prof from Sussex Uni) really is so damning. To argue to change society on the basis of the SL is to argue on the basis of a false prospectus. There are surely enough good reasons to do good things as to have no need for iffy propaganda.

My point is that even if the evidence was true and it might be.
Do you honestly think it would change the mind of the right or government policy or even your views.
You either believe inequality is inevitable or even desirable or it is something to change and eradicate.
The work by two academics isn’t going to change most peoples mind one way or another.
As for statistical evidence of any sort can be manipulated to support any view.
Disraeli summed it well “Damn statistics”.
As for you being right wing , I don’t know if you are or not but most of your comments like John M, tend to be of that ilk.
Also what is wrong by being called right wing. It is certainly the flavour of the month and most English people, are quite happy with being called right wing

Rhys Williams @ 28

Yeah well, the only party I’ve ever been a member of is Fianna Fail – twinned with the Lib Dems in Europe.

30. Richard P

Fianna Fail can be regarded as rightwing:
– Until they joined the Liberal group in the European Parliament, they were part of the hard-right UEN group.
– Since joining the Liberal group, they have annoyed their new colleagues by refusing to back gay rights ( http://euobserver.com/9/28912 ), by rejecting a motion critical of Berlusconi, and by having different views on civil liberties from the rest of the group ( http://www.votewatch.eu/blog/wp-content/uploads/2009/09/votewatch_report_voting_behaviour_in_the_new_european_parliament.pdf )
– FF are notorious for the cuts programme they are carrying out in Ireland.


Reactions: Twitter, blogs
  1. Liberal Conspiracy

    Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  2. charlotteahenry

    RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb >> @One_Society

  3. sunny hundal

    Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  4. Sarah Ditum

    Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb (via @sunny_hundal)

  5. One Society campaign

    Can't say it better than @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  6. Joshua Fenton-Glynn

    RT @libcon Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb <thorough rebuttal hope the guardian prints

  7. Oxford Kevin

    RT @libcon Spirit Level authors hit back at Policy Exchange report http://bit.ly/ay0Soc

  8. mikeblakeney

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  9. Chris Goulden

    RT @libcon Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb < this really is a bandwagon

  10. Emma Burnell

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  11. Mark Pack

    RT @sarahditum: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb (via @sunny_hundal)

  12. Louise Pennington

    RT @sarahditum: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb (via @sunny_hundal)

  13. Jonnie Marbles

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  14. Fran Bury

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  15. Michael Hanley

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  16. RupertRead

    Well done @EqualityTrust ! : RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange http://bit.ly/dfGkBb

  17. no

    RT @sarahditum Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb (via @sunny_hundal)

  18. The Equality Trust

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  19. SYJ

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  20. Tony Dyer

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  21. FPH

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  22. Joseph Bush

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  23. Joseph Bush

    Mind you, I could have done it for them so poor was the 'critique': Spirit Level authors hit back at Policy Exch report http://bit.ly/dfGkBb

  24. Karen Varga

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  25. Oxford Kevin

    RT @sunny_hundal: Important RT @libcon: Spirit Level authors hit back at Policy Exchange report http://bit.ly/dfGkBb

  26. Peter McColl

    This Policy Exchange report demonstrates how desperate the UK right are to *promote* inequality: http://bit.ly/cIe1sM (via @equalitytrust)

  27. blogs of the world

    All analyses of income inequality and health and social problems in The Spirit Level have … http://reduce.li/dusgdi #level





Sorry, the comment form is closed at this time.