The central message in ‘the spirit level’ is that most of our important health and social problems become more common in direct proportion to the degree of social inequality that exists – amongst the so called developed parts of the world Japan and Scandinavia are at one end of the equality spectrum (most equal) while the likes of the USA, Australia and the poor old UK languish at the other (unequal) – the authors are epidemiologists.

But, in the interest of balance (admittedly from a tainted source) …………..
http://www.taxpayersalliance.com/spirit-level-debate-summary

You can visit the campaign page at indiegogo.com/spiritlevelfilm! Please help & support us.

The TPA is a bunch of rich people, some of whom don’t live here and don’t pay tax, they are right-wing shills.

You will, I hope, correct some of the more egregious errors in the book?

[5] “You will, I hope, correct some of the more egregious errors in the book?” – oh, I love a challenge – perhaps you would be good enough to tell us what the worst error in the book is – put another way do you deny the central message – ordinary people far less well in unequal societies?

One that comes immediately to mind is that they claim that more unequal countries have longer working hours.

However, their data set inexplicably seems to exclude Japan, a country with very long working hours and not much income inequality.

Strange that really.

On a more conceptual level they waver between different measures of inequality. They are always measuring income inequality. But quite how this actually affects happiness etc seems more often to be described as status inequality. This is something of a problem: for if you use status inequality then Japan is a grossly unequal country, if income, quite an equal one.

And on a theoretical level I was spluttering by about page 3. They start with hte Easterlin Paradox (which is wrong), then talk about diminishing marginal utility of income (which is correct).

Three paras later they’re saying that more income has no effect at all: that’s a very different statement indeed from diminishing marginal utility and one that is simply wrong.

It’s simply extremely shoddy work, run up out of prejudices more than anything else.

[6] ‘One that comes immediately to mind is that they claim that more unequal countries have longer working hours – However, their data set inexplicably seems to exclude Japan, a country with very long working hours and not much income inequality’ – yes, working hours are discussed (on p228) – and yes, Japan is not included on this particular measure.

The spirit level authors (Wilkinson & Pickett) cite data published by Bowles & Park who claim “Data on work hours in ten countries over the period 1963–98 show that greater inequality is indeed associated longer work hours. These ‘Veblen effects’ are large and the estimates are robust using country fixed effects and other specifications”.
(The Veblen effect is named after economist Thorstein Veblen, who first pointed out the concepts of conspicuous consumption and status-seeking).
http://onlinelibrary.wiley.com/doi/10.1111/j.1468-0297.2005.01042.x/abstract

Isn’t rather churlish to highlight a minor discrepancy (working hours in a single country) then portray this as some sort of fatal flaw in the central hypothesis? Obviously Wilkinson & Pickett are dealing with aggregates of complex phenomena to demonstrate a wider pattern.

Maybe you think people tend to have FEWER health and social problems when countries have higher levels of social inequality?

Isn’t rather churlish to highlight a minor discrepancy (working hours in a single country) then portray this as some sort of fatal flaw in the central hypothesis?

I offered something that was off the top of my head: it’s one of the things that really buged me when reading the book. For a more detailed analysis try reading the Spirit Level Delusion.

You see, the problem is, so much of thier data suffers from these sorts of problems that it doesn’t (and agreed, people argue about this) in the end support their very strong claim.

“Maybe you think people tend to have FEWER health and social problems when countries have higher levels of social inequality?”

That isn’t their claim. Someone who said that those at the bottom of the inequality pile suffer more the greater the inequlity? Sure, no one at all would argue with hte conclusion.

They make a much stronger claim. That the more unequal the society the more even the rich suffer such things. That’s a very strong claim indeed and not one that their rather ramshackle data manages to support.

You might want to check out The Spirit Level Delusion by Christopher Snowdon, which details the many inaccuracies, non-sequitors and outright lies contained in the The Spirit Level. Of course, if you correct all those then you’re not going to have a very powerful film, and if you cut the relevant parts then the film will last for about two minutes.

A&E Charge Nurse,

I haven’t read the book, but it sounds like you’re equivocating slightly.

Are you saying that if we took two people, identical in terms of a set of background variables (like income, education, etc), and put one in an unequal society and one in an equal society, they would have different outcomes on some measure of welfare?

If that’s the finding of the book, what are the factors we are holding constant and what are the outcome measures that are affected by inequality?

Thanks.

[9] isn’t Snowden the, err, analyst who said “I don’t think people outside the intelligensia worry about inequality”?
http://www.guardian.co.uk/books/2010/aug/14/the-spirit-level-equality-thinktanks

On closer inspection we find Snowden’s riposte, ‘the Spirit Level Delusion’, was published through the Democracy Institute, a rightwing thinktank in Washington DC.

A rightwing thinktank in the USA ……. mmm, strange bedfellows for a yorkshireman?

[10] ‘Are you saying that if we took two people, identical in terms of a set of background variables (like income, education, etc), and put one in an unequal society and one in an equal society, they would have different outcomes on some measure of welfare?’ – no, not for individuals, but for populations, there are always exceptions to the general rule.

Inequality is a variable in a range of social ills (but not a sole cause) if you accept what the spirit level is saying – obviously right wing think tanks might provide a different interpretation?

@12,

Maybe this is a little bit technical, I don’t know. Presumably, the author of the OP or someone who has read the book and has a quant background could tell me…

If inequality per se is bad, then this suggests that then there should be some measurable effect on otherwise identical people.

In cross-sectional social science research the basic idea is to measure the effect of some variable (like, say, a measure of inequality) on some other variable (like say a measure of health), holding everything else constant. By doing this you can approximate a controlled experiment where the treatment is assigned randomly and so is independent of all other factors.

So when a social scientist says that something like inequality has a causal effect on something else, like health, I interpret that to mean that if you could take one person from the unequal group and put him into an more equal group, and hold everything else the same, his health would improve.

Here’s a review of the work of Wilkinson and Pickett from Lane Kenworthy. He’s a sociologist and certainly left-of-centre: http://lanekenworthy.net/2010/01/18/inequality-as-a-social-cancer/

11. the a&e charge nurse

On closer inspection we find Snowden’s riposte, ‘the Spirit Level Delusion’, was published through the Democracy Institute, a rightwing thinktank in Washington DC.

A rightwing thinktank in the USA ……. mmm, strange bedfellows for a yorkshireman?

If you’re going to close your mind to anyone who disagrees with you, you will never learn if your opinions are wrong or right. Let’s apply this same test to the actual book in question. It was produced by two people who work at the Equality Trust.

And therefore you will agree is worthless?

It is actually a good test of who knows sh!t from shinola in the journalism world. Richard Reeve may work for the Guardian but even he recognised it was crap. As did the Scotsman. Needless to say Will Hutton thought it was great.

[15] ‘If you’re going to close your mind to anyone who disagrees with you’ – well, if you look back a bit further, you will see I link to the TPA [1] – hardly a bunch LC are generally sympathetic towards?

If your point is that my take on a complex set of questions is biased from a left/liberal perspective then yes, I freely admit that – but I’m not sure it equates to a ‘closed mind’ neither does it stop alarm bells ringing when pro-market groups pronounce on the same phenomena.

16. the a&e charge nurse

If your point is that my take on a complex set of questions is biased from a left/liberal perspective then yes, I freely admit that – but I’m not sure it equates to a ‘closed mind’ neither does it stop alarm bells ringing when pro-market groups pronounce on the same phenomena.

Quoting the TPA while calling it tainted is precisely the problem isn’t it? It does not demonstrate your openness to other points of view. Remember that someone like the TPA may publish something just because no one else would. You can see that with Peter Saunders – who also did not like the book. A well respected sociologist at Sussex University who happened to have some doubts about the evilness of Thatcher. And was foolish enough to say so. Grants soon dried up. People declined to sit on panels with him. Invitations ceased. Pretty soon he was forced out of academia and into a group like the TPA just because no one else would employ him.

You don’t need to rely on pro-market groups. Even the Guardian ran critical articles. Try this:

http://www.guardian.co.uk/books/2009/apr/05/spirit-level-richard-wilkinson

Which considering its source is about a robust condemnation as you are likely to read.

[17] the guardian article hardly constitutes a ‘robust condemnation’ for example Reeves says ‘The Spirit Level is strongest on Wilkinson’s home turf: health. The links between average health outcomes and income inequality do appear strong, and disturbing – so, disturbing and strong links, according to the, err, critic.

By the way if you don’t think the TPA are biased then you are the one suffering with ‘a closed mind’.

Good project, because equality remains important, even at a relatively high level of average wealth: http://andreasmoser.wordpress.com/2012/05/21/equality-versus-sufficiency/

I would tend to agree with the Lane Kenworth critique of The Spirit Level. I don’t think many people would seriously dispute that those lower down the income distribution will tend to on average have worse health outcomes and life expectancy than those further up the income distribution. For that there are causal mechanisms that are convincing. Making bad choices or not having enough money income to make good choices is something that most people can accept. However, the claim in The Spirit Level that inequality in money income distribution can worsen the average outcomes in health etc are deeply unconvincing. So-called status competition as a causal mechanism in unequal societies seems to me to be making stuff up to fit a desired conclusion.

What makes Sweden have good social and health outcomes is not just their policies but the fact that Sweden is full of Swedes. Copying the policies of a society with good outcomes probably would allow the copier to improve their outcomes, but without replicating the culture where the good outcomes occurred would leave the copier short because other factors other than inequality in money income are just as relevant. Do we seriously believe that if the whole of Sweden decided to adopt a sedentary lifestyle, eat a bad diet, smoke too much, drink too much, that their health and life expectancy would be held constant if they held the money income distribution constant? What would be interesting is some study of the outcomes of people from a culture of low income inequality living in a high inequality society. For example, those from a Japanese and Scandinavian background who emigrated to the high inequality US. If there was a reversion to the US mean in their health and life expectancy then The Spirit Level has some merit. If the health and life expectancy outcomes for Japanese and Scandinavian Americans more mirror their homeland than the US average then the dominant factor surely is culture and not the money income distribution.

@8. Tim Worstal

They make a much stronger claim. That the more unequal the society the more even the rich suffer such things. That’s a very strong claim indeed and not one that their rather ramshackle data manages to support.

In some ways is that not common sense. I mean, an increasing gap between the haves and the have nots, is arguably not so good for the haves either in many ways. And is poisonous for society as a whole.

Why? Well, the haves become increasingly paranoid about the have nots due to crime etc, and increasingly separate themselves from the rest of society, by living in exclusive places, sending their children to exclusive schools etc, all of which serves to isolate them from the realities of the lives of the have nots. And as a society becomes more polarised this becomes a self reinforcing vicious cycle.

In the extreme version of this, in places such as parts of South America and South Africa, the haves isolate themselves entirely from society and live within heavily guarded gated communities. Only venturing out with bodyguards because they are constantly at risk from crime and kidnapping etc. This is hardly great for their psychological health living in fear and under virtual siege now is it?

Furthermore, this is poisonous for society. As the haves are isolated from the realities of life for the have nots, they become increasingly resentful of paying taxes for the common good. Making it far more difficult to achieve a political consensus to solve social problems.

18. the a&e charge nurse

the guardian article hardly constitutes a ‘robust condemnation’

Is robust a word I used? I am not sure that is the point. Surely the point is that even on the most friendly turf the authors could find, the problems with the work is so obvious that the reviewer was critical.

By the way if you don’t think the TPA are biased then you are the one suffering with ‘a closed mind’.

Biased in an interesting word. Do I think they have an opinion? Sure. So does the Guardian. But I would not close off evidence from either.

19. Andreas Moser

Good project, because equality remains important, even at a relatively high level of average wealth

I admire the way you get a link to your blog into every post, but why on Earth do you think that equality is important?

20. Richard W

I don’t think many people would seriously dispute that those lower down the income distribution will tend to on average have worse health outcomes and life expectancy than those further up the income distribution.

But that is not what we should care about. The issue is not how the poor compare to the rich, but what the average life expectancy is. We may have more unequal health outcomes now, but on the other hand people live a hell of a lot longer than they did 60 years ago. That is what matters. And you can prove it by just asking people – if you go into a hospital ward and say that the relatives can have a choice, six months more with their loved one if that rich bastard over there gets another year and that poor sod outside in the corridor only gets six weeks, or they can all have two days each and then will be put down, I don’t think many people are going to choose the more equal outcome.

What makes Sweden have good social and health outcomes is not just their policies but the fact that Sweden is full of Swedes.

And indeed people of Swedish descent in the US do not have radically different health outcomes. What is noticeable about the Spirit Level is that virtually all the effects they find are strongly correlated with race. But they chose not to mention it.

21. Graham

Why? Well, the haves become increasingly paranoid about the have nots due to crime etc, and increasingly separate themselves from the rest of society, by living in exclusive places, sending their children to exclusive schools etc, all of which serves to isolate them from the realities of the lives of the have nots. And as a society becomes more polarised this becomes a self reinforcing vicious cycle.

This is interesting mainly for the patronising view it has of poor people. You are right that high crime levels are bad for society but crime is not a proxy for inequality. It just means that crime is bad for society. We have had more crime the more that we have spent on welfare. The problem is not inequality. The problem is a pro-crime culture.

Jimmy Hill,

Thanks for that link–that was exactly what I was after.