The Spirit Level: why inequality matters

As I mentioned in a previous post (and also here), I hope that, whoever becomes leader, Labour will start to take more seriously the issue of inequality.

Even New Labour never claimed that inequality was a good thing. However, Labour in government preferred to focus on issues of poverty and social exclusion rather than inequality as such, and the assumption seemed to be that inequality didn’t really matter provided that those further down the income ladder received proper help.

Richard Wilkinson and Kate Pickett’s book The Spirit Level challenges this assumption, by arguing (on the basis of vast amounts of international data) that it is inequality, not poverty or average income levels, that lies behind most social problems in “developed” nations. Their data and arguments can be found on the website of the Equality Trust, including slides showing key charts from The Spirit Level.

The essence of their argument can be captured in just two of those slides, the first showing how health and social problems are associated with inequality in developed countries:

However, those same social problems are not related to average levels of income in those countries:

As set out in their book, and as summarised in the other slides, a relationship with inequality is found (to a greater or lesser degree) in relation to each of the social problems contained within the overall measure used in the above charts. (And yes, before anyone says it: they do address issues such as correlation and causation.)

To give just one example: the relationship between inequality and health, as shown by life expectancy between and within societies:

What the second graph in the above slide shows, the authors argue, is that inequality affects everyone in society, not just those at the bottom. At almost every point in the income scale, statistically you will tend to be healthier than those below you and less healthy than those above you.

This is only a very brief summary of The Spirit Level‘s argument – and those of you still muttering about “correlation” and “causation” are referred to the book, or at least the Equality Trust’s evidence section and FAQs – but it helps demonstrate the essence of Wilkinson and Pickett’s central claim: that inequality matters, matters perhaps more than any other factor, and that it matters to everyone, not just the poorest.

In the posts that follow I propose to look at a couple of questions which flow out of this: first, what is the UK’s current record on inequality, and second, how is that inequality has these negative effects across such a range of social issues?

Advertisements

13 thoughts on “The Spirit Level: why inequality matters”

  1. At almost every point in the income scale, statistically you will tend to be healthier than those below you and less healthy than those above you.

    In other news, at almost every point in the income scale, statistically you will tend to have a nicer car than those below you and a shabbier car than those above you. You will also tend to live in a bigger house than those below you and a smaller house than those above you.

    You should read Murray Rothbard’s Egalitarianism as a Revolt Against Nature. it will infuriate you.

    1. Josh: the fact that my boss has a nicer car than me and lives in a bigger house than me is a consequence of inequality that I can live with.

      The fact that I can expect to live at least five years longer than someone living in a poorer area (indeed a good ten to fifteen years longer than those in the very poorest) isn’t.

  2. A few years from now, people will be embarrassed to admit they ever believed in Richard Wilkinson’s absurd ‘theory of everything’. The graphs you show are based on an index of health and social problems of Wilkinson’s own devising. It’s a fiction; a bunch of simple statistical tricks and cherry-picking.

    Believe in equality all you like, but don’t put your faith in it because of this garbage.

    1. Snowdon: I’m always suspicious of monocausal “theories of everything”, and can quite believe that inequality is not the only cause of every problem. That’s not the same thing as saying it isn’t a significant factor in many problems – more so than most politicians have recognised in recent years.

      That said, I’ve not studied this in the detail you clearly have…

    2. One other point: I’ve not yet reached the section dealing with Wilkinson and Pickett’s proposals as to how to remedy inequality. What I will say is that I don’t believe that it’s as simple as saying “hike up top tax rates to Scandinavian levels and paradise will follow”, and the statement on their website about how reducing inequality will “halve the murder rate” etc is pretty crass.

      What I value about the book so far is that it has the potential to make equality a subject for political debate and reflection again (rather than either silence or populist envy) – and your own book itself demonstrates that. This is a good thing, whatever conclusions one reaches about what (if anything) needs to be done about it.

      Though selective and partial choices of data can obviously cut both ways. I noticed this point on your website:

      The trend towards more equal countries giving more in state aid is counter-balanced by the tendency for less equal countries to give more in individual, voluntary donations.

      Maybe so, but is the difference in voluntary donations enough to make up for the difference in state aid? For example, DFID’s 2009/10 budget in the UK was £9.1bn, but charitable giving for overseas development and emergency relief for 2004-5 (most recent figures I could dig up on a quick Google) was only £1bn (src pdf). Even assuming that amount increased between 2004 and 2009, that’s still a big difference.

      A halving in UK foreign aid (to US levels in terms of percentage of GDP) would need to be met by a four- or fivefold increase in charitable giving for overseas development causes.

      Similarly, I haven’t run the numbers, but I would guess that giving for overseas development is only a small proportion of US charitable giving, and that the amount given for such causes is small compared with levels of government aid: so not really much of a “counter-balance”.

  3. What I will say is that I don’t believe that it’s as simple as saying “hike up top tax rates to Scandinavian levels and paradise will follow”, and the statement on their website about how reducing inequality will “halve the murder rate” etc is pretty crass.

    Taxes are complicated things to compare, but the top marginal tax rate in Sweden, which is the usual Scandinavian example, is about 57% (src). So once you account for employee NI, we don’t actually have much headroom between us and Sweden!

    One of the questions that occurred to me is the extent to which the inequality figure is a proxy for social programmes. For example, a healthcare programme funded out of progressive taxation would reduce inequality, but it is not immediately clear that it isn’t, in fact, the healthcare spending which generates the improvements in longevity. Did Pickett and Wilkinson control for the relevant spending component? I have seen a suggestion that if you do this, the correlation with inequality gets swamped: I’m not expert enough to assess one way or the other.

    For example, DFID’s 2009/10 budget in the UK was £9.1bn, but charitable giving for overseas development and emergency relief for 2004-5 (most recent figures I could dig up on a quick Google) was only £1bn.

    One of the big questions about what we do is marginal: is a pound extra of DfID going to do more good, or less good, than a pound extra of private charitable giving? What we need to know isn’t so much where we might be best located, but which direction of travel given where we are will achieve the most good. It may be that DfID would do more good by spending its money trying to encourage the rest of us to give more!

    Actually, that goes for the stuff about inequality, too. You need to ask what happens to social indicators when inequality shifts, not simply what the levels are like; and you need to ask whether using scarce resources to tackle inequality is going to do more good than using them to tackle problems directly.

    I’m not expert enough to know the answers to those questions, but they are the sorts of question which need asking.

    1. Phil, I agree with what a lot of what you’re saying (and thanks for the correction re Swedish tax rates!). The book does address some of the questions re effects of changes in inequality, though the picture is complicated by the fact that the effects often lag behind the causes by many years.

      Similarly the question of whether one tackles inequality or its effects, given that ceasing to tackle its effects will harm people now, whereas reducing inequality will have more gradual effects. As I said to Snowdon, though, it would be an advance just to get these questions being openly considered – heck, it’d be an advance just to get them openly considered by the Labour leadership.

      I’m not sure I agree though re the marginal effects of extra giving vs extra aid. Given the gap between charity and aid is so great, you’d have to be getting a lot more bangs for buck from the former than the latter for that to make much difference. In any event, as many on the right will often bemoan, is there really such a huge functional difference between the large aid charities and government bodies?

      1. I think the interesting point about Scandinavia is that it tends to suggest that we should focus on making Anglo governments as efficient as Scandinavian ones, because clearly we have a lot to learn from them.

        I’m not sure I agree though re the marginal effects of extra giving vs extra aid. Given the gap between charity and aid is so great, you’d have to be getting a lot more bangs for buck from the former than the latter for that to make much difference.

        I’m not sure you’ve got the point about marginals. It’s not a percentage change in the relative budgets, but a fixed change of a (small-ish) number of pounds. So say you’ve got an extra £10mn to spend on development from somewhere. Where would you squeeze the most juice out of it: DfID, or private? It’s a question for econometricians to chew on, but my suspicion is that diminishing marginal returns have set in and made DfID far less attractive for the extra money.

      2. OK, understand your point re DFID. But I’m not sure that “a pound more to DFID or a pound more to Oxfam (or whoever)” is the question we’re actually faced with in practice.

        The context of this discussion is the claim that charitable giving “counter-balances” lower levels of aid. My point was that a significant reduction in levels of government aid would not be compensated for adequately by any realistically-contemplatable increase in charitable giving.

  4. John H,

    Certainly in the USA, charitable giving more than makes up for its relatively stingy state aid. A lot of this goes to religious groups and schools, but even if you exclude them, the States give more to the developing world than the ‘more equal’ countries for which we have data. There is the added benefit that this money is much less likely to be misspent and misappropriated, which is unfortunately the fate of a lot of foreign aid (see Dead Aid by Dambiso Moyo, which lost out in the Bristol Book Prize last night to you-know-who).

    This is all important stuff but my main point (discussed in chapter 6 of my book) is that it is highly misleading to use state spending on foreign aid as a proxy for how community-minded, altruistic and philanthropic the population is. If anything, the opposite is the case. Now, you can argue that people in ‘more equal’ countries elect governments who will use higher taxes to take care of these things. You can also argue, sensibly I think, that if you are taxed more, you will have less money left to give to good causes. What can’t be seriously argued is that people in Britain and America are selfish and suspicious evil capitalists who give less money to charity. But that is what Wilkinson and Pickett do. By only showing state spending, they only show half the picture.

  5. Those statements are contradictory. Living in a nicer house in a nicer neighborhood increases one’s life expectancy, because the nicer neighborhoods are less prone to crime.

    In general, the more money you make, the better quality of life you are able to enjoy. That’s almost the definition of “wealth.” (And why focus on arbitrary political boundaries? What about the inequality between Great Britain and Haiti? Between Japan and Guinea-Bissau?) It’s not surprising, then, that “wealth” translates into “longer life.” A large part of it, of course, is because wealth means having a better quality of life, but I suspect a non-negligible part is that, in a market economy, becoming and remaining wealthy requires refraining from colossally stupid decisions, such as failing out of high school because you enjoy drinking so much.

    People do not need to be “equal” to enjoy long life–Haitians are mostly equal–they need to be wealthy. Of course, wealth has this peculiar way of defying socialist attempts to spread it around equally by simply disappearing.

    1. Living in a nicer house in a nicer neighborhood increases one’s life expectancy, because the nicer neighborhoods are less prone to crime.

      Er… Are you trying to tell me the difference in life expectancy is attributable to crime rates? The homicide rate isn’t /that/ high!

      As for comparing Britain with Haiti: if you read the book (or even glance at the slides) you’ll see the authors’ argument is aimed at rich, developed countries. They are not denying that poor, undeveloped countries do worse than rich countries: their point is that once wealth reaches a certain level, however, the effects of increased average wealth in society cease to be significant and the effects of inequality within that society become more pronounced.

      You may (like Mr Snowdon) wish to take issue with that thesis, but that’s what their argument is: so saying “people in Britain live longer than in Haiti” is completely irrelevant.

      As for “wealth makes you live longer”: life expectancy in the UK (and indeed pretty much all other rich countries in the authors’ sample) is higher than in the wealthier US.

      Finally, I’m not arguing for absolute equality, but for reduced inequality: for the UK (not my place to comment on the US) to have levels of inequality that are closer to those impoverished socialist hellholes, Canada, Germany, Sweden and so on.

      1. I don’t know about your country, but here, murder is a statistically significant contributor to the life expectancy of poor, black males.

        the effects of increased average wealth in society cease to be significant and the effects of inequality within that society become more pronounced.

        This sentence contains within it a false assumption–that averages affect individuals. Averages are aggregates of individual pieces of data. The individuals determine the aggregates, not the other way around. This is a basic epistemological error.

        As for “wealth makes you live longer”: life expectancy in the UK (and indeed pretty much all other rich countries in the authors’ sample) is higher than in the wealthier US.

        The discrepancy can be accounted for wholly by non-poverty-related factors:
        1. We have a lot more black people (black people have a much higher incidence of heart disease and negatively affect the life expectancy).
        2. We have a lot more traffic deaths.

        When using statistics, you need to compare like with like by adjusting for discrepancies (like the fact that Americans drive more than the British). And life expectancy, like “average wealth,” is yet another aggregate. For the idea that wealth affects life expectancy, compare Haiti with Britain. Or compare Britain today with Britain 100 years past–nationally, you are quite a bit wealthier, possessing means of sustenance and medical care far beyond what your average London resident of 1850 possessed.

        You also need a coherent idea of what “wealth” is. It’s not money; it’s things that people desire and need. Wealth includes things like medical care, food, shelter, and so on. An aggregate composed by a statistician does not increase your life; real goods and services do.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s