The Spirit Level

For some time we’ve been hearing about The Spirit Level,[1] a book first published in 2009 advancing a case for equality of income. Many enthusiasts for the book have followed the authors in promoting it as an evidence-based, ‘scientific’ statement of the case for equality, as if no right-thinking person could disagree with it.  What’s going on?

The book differs from previous literature on equality in that it doesn’t argue principally that equality of income is the only just and fair distribution of income. The authors actually do believe this, but to get the necessary political consensus for equality they set out to prove that, as the book’s subtitle puts it, “equality is better for everyone”. They want to make the case for equality universally irresistible as well as irrefutable.

The authors, Richard Wilkinson and Kate Pickett, are epidemiologists at British universities, and when they say that equality is ‘better’ for everyone they mean that it promotes our physical, mental and social well-being.  They don’t mean that it promotes ‘welfare’ in the typical economist’s sense of maximising the consumption of goods that individuals define and choose for themselves.  They are quite clear that economic freedom must be curtailed if it prevents more well-being from being realised through the equalisation of incomes.  But they are confident that the promise of greater well-being will persuade us to prefer to equalise incomes rather than to increase them.

The authors set out by asserting what they take to be two established facts.  The first is that in developed countries economic growth no longer much increases individual economic welfare.  The ‘happiness curve’ that people in poor countries ride up as they become richer eventually flattens out, as it already has in the richest countries. The second established fact is that man-made global warming threatens a global environmental catastrophe, and so economic growth will have to be ended anyway in order to control carbon emissions.  But we can still improve our well-being by moving towards equality of incomes.  The core of The Spirit Level is its presentation of research results to back up this claim.

The authors argue that a range of pathologies are all several times worse in the most unequal developed countries than they are in the most equal ones. The same pattern is evident in US states: the most unequal states do much worse than the most equal ones.  Of the wide range of indicators used, the central ones are tenfold: level of trust; mental illness (including drug and alcohol addiction); obesity; children’s educational performance; teenage births; homicides; imprisonment rates; and social mobility (this last indicator is not available for US states). Using internationally recognised definitions and comparisons of these ten indicators, and drawing on data from 23 developed countries and the 50 US states, the authors show that the problems are worst in the most unequal countries, which include the United States, Portugal, the United Kingdom, New Zealand and Australia, and the most unequal US states, typically those of the South.  They are least bad in the most equal countries, which include Japan and the Scandinavian countries, and in the most equal US states, which are geographically dispersed and include Alaska, New Hampshire, Utah and Iowa.

The authors find only a weak relationship between the indicators and average incomes among rich countries (and the US states), which suggests that it’s the inequality of incomes that’s significant; and they believe that cultural explanations of the relationships don’t stand up to scrutiny. They conclude that inequality of incomes actually causes or exacerbates health and social problems. The more unequal a society is, the less people trust one another; those at the bottom of society are more likely to feel stressed and demoralised, to engage in self-destructive behaviour and to fall ill; and everyone is more likely to seek solace in wasteful and futile consumerism.  More equal societies, in contrast, are less atomised, more socially mobile, more socially relaxed and healthier.

In the final part of their book the authors canvass policies to lessen the extremes of income inequality in the most unequal societies.  They want to raise the lowest incomes with higher minimum wages, but they’re more concerned with dismantling concentrations of wealth by, for example, taxing the very rich heavily (they entertain a top income tax rate of 60 percent – see how serious they are about ending economic growth!) and by promoting alternatives to the standard corporate model with schemes like employee share-ownership and control of companies, and, in finance, by encouraging credit unions and mutual funds as alternatives to orthodox banking.   They have no doubt that the outcome of these changes would be a great advance in well-being. They have set up the Equality Trust (www.equalitytrust.org.uk) to help get a social movement for equality going.

But they haven’t had a completely clear run. Already at least two books have appeared challenging the central claims of The Spirit Level.[2]  Their authors argue that the analysis of social problems and income inequality suffers from many defects.  For example, there is the omission of countries like South Korea, Slovenia and the Czech Republic, which are relatively equal but don’t achieve high levels of well-being, and also Hong Kong, which is, conversely, highly unequal but scores highly on the well-being criteria. The authors of The Spirit Level are responding to criticisms on the Equality Trust website.  The debate will go on. It’s quite possible that the entire Spirit Level thesis will eventually disintegrate under the weight of such criticism, and be forgotten. It’s equally possible that some of the book’s claims will survive more or less intact.  But to my mind the central problem with the book is not its possibly defective analysis of the link between income inequality and health and social problems. It is its opening assertion that in the rich countries economic growth no longer promotes welfare.  (I cannot deal here with the second assertion, about climate change, but claims about the scientific irrefutability of the fact of man-made climate change have waned as legitimate doubts have emerged both about it and about the appropriate policy response to it.)

The claim that in the rich countries economic growth no longer promotes welfare indicates how irrelevant The Spirit Level is to the central political issue now confronting all developed countries, which is how to cope with the huge and growing demand for government spending brought about by demographic change. The global fiscal crisis is not just a short-term matter of closing structural budget deficits and stopping the rise of national debt; it is a long-term one of financing the pensions and health care of populations that are ageing as life expectancy rises, fertility rates are low, and the ratio of dependants to workers grows.  Economic growth is an indispensable part of the policy mix that governments are devising to meet the challenge. This has nothing to do with satisfying a taste for ‘consumerism’ or vainly pursuing more ‘happiness’; it’s about preserving existing standards of living and well-being which, without economic growth, would soon start to slip back, threatening vulnerable groups with poverty. This is well understood in Sweden, which the authors of The Spirit Level consistently find to be among the most equal and healthiest of countries. The election there on 19 September 2010 confirmed the conservative government in a second term of office after it implemented a programme of tax cuts, which beat the recession by reducing unemployment and creating about 100,000 jobs.[3]  Sweden’s economy is now growing at the impressive rate of about 4 percent a year.  The country may be somewhat less equal as a result of the tax cuts; but who apart from the authors of The Spirit Level would deny that the cuts were good policy?

Another relatively equal and healthy country is Japan. It faces perhaps the biggest demographic challenge of all the developed countries. Unlike Sweden, it allows in very few immigrants to boost the workforce.  A sizable immigration programme would doubtless increase inequality.  But if the Japanese nevertheless implemented one, how many observers would think they’d made a mistake?

By dismissing economic growth so brusquely from the outset, the authors of The Spirit Level exclude from their analysis any possible links between continuing economic growth and well-being even in relatively unequal developed countries. But such evidence exists. Writing recently in The Australian, Michael Stutchbury summarised the Australian Bureau of Statistics’ latest snapshot of Australia’s national progress during a ten-year period in which Australia’s per capita gross domestic product grew by 23 percent in real terms. He wrote,

A key indicator of health – life expectancy at birth – confirms Australians are living longer. In the decade to 2008, life expectancy for girls rose 2.2 years to 83.7 years and by 3.3 years to 79.2 years for boys. The suicide rate, which the ABS classifies as an indicator of social cohesion, has fallen even amid concerns about worsening mental health. The suicide rate for men aged 20 to 24 halved during the same period. And volunteering, seen as an indicator of social networks, has increased, up from 24 per cent in 1995 to 35 per cent in 2006.[4]

This last item in particular – the growth of volunteering – shows how utterly misguided the authors of The Spirit Level are to dismiss economic growth as the search exclusively for ever more material wealth.  On the contrary, economic growth makes it possible to pursue ever more non-material goods.  Higher productivity progressively reduces the time that has to be devoted to meeting material needs, so allowing more scope for other activities, including those of a social nature that the authors of The Spirit Level value.

I share much of the aspiration of The Spirit Level for more equality although I worry more about poverty and hardship and how to alleviate it.  A simple thought experiment shows that promoting income equality per se is not a sensible social goal.  Consider what would happen if Bill Gates and Steve Ballmer decided to move Microsoft’s Seattle headquarters, and all its US million dollar employees, to New Zealand.  Income inequality would ‘worsen’.  But how many New Zealanders would regard that as a bad thing?

Experience suggests that the compulsory imposition of economic  equality can eventually be self-defeating and impoverishing – which is why Sweden is rowing away from it.  Yet economic and social reform often promotes equality as an unintended but desirable side effect of promoting freedom and efficiency. Thus, educational reform to enhance national skills could improve the performance of the poorest social groups most, since they have the most to gain. Welfare reform would free the poor to take advantage of opportunities to rise up the income ladder and to escape the physical and mental health risks of dependency. At the other end of the scale, the global financial crisis has exposed the enormous incomes of some bankers as symptoms of barriers to entry into finance (allowing bankers to consume the implicit government subsidy arising from banks being ‘too big to fail’) and also of management domination of financial institutions.  Improvements to financial regulation, competition policy and corporate governance would help to spread the profits of finance more widely, crucially to shareholders, which include superannuation funds.  A shift away from the prevalence of bank credit towards other, possibly safer forms of finance might well see the growth and spread of mutual funds, credit unions and still other kinds of financial institution, like the internet bank Zopa.  And where economic freedom prevails, non-corporate forms of economic governance can be tried.

For a further international perspective on income inequality see this paper by Will Wilkinson of the Washington-based Cato Institute.  The summary reads as follows:

Recent discussions of economic inequality, marked by a lack of clarity and care, have confused the public about the meaning and moral significance of rising income inequality. Income statistics paint a misleading picture of real standards of living and real economic inequality. Several strands of evidence about real standards of living suggest a very different picture of the trends in economic inequality. In any case, the dispersion of incomes at any given time has, at best, a tenuous connection to human welfare or social justice. The pattern of incomes is affected by both morally desirable and undesirable mechanisms. When injustice or wrongdoing increases income inequality, the problem is the original malign cause, not the resulting inequality. Many thinkers mistake national populations for “society” and thereby obscure the real story about the effects of trade and immigration on welfare, equality, and justice. There is little evidence that high levels of income inequality lead down a slippery slope to the destruction of democracy and rule by the rich. The unequal political voice of the poor can be addressed only through policies that actually work to fight poverty and improve education. Income inequality is a dangerous distraction from the real problems: poverty, lack of economic opportunity, and systemic injustice.

My bottom line is that income equality may or may not promote well-being in the manner claimed in The Spirit Level.  Economic growth certainly does promote it. Attempts to promote well-being by engineering equality via the compulsory curtailment of economic growth and economic freedom can fail, by gradually undermining prosperity.  But the authors of The Spirit Level have more scope than they realise to promote their egalitarian ideals within the framework of economic freedom.

Some useful New Zealand data and perspectives on income inequality were contained in this recent presentation by Ben Gleisner of the Treasury.  His conclusions on trends are:

1.    Income inequality in New Zealand grew sharply between mid 80s and mid  90s, but stabilised recently

2.    Including government support, growth is less

3.    But still high inequality relative to other OECD countries

The first point is crucial.  The strongest driver of income inequality is usually unemployment.  This rose steeply in the 1980s with the refusal of the Labour government to free up the labour market.  It fell sharply from the early 1990s following labour market reforms.  There has to be a worry that the recent rise in unemployment, especially the loss of employment opportunities for many young people with the abolition of youth wage rates by the last Labour-led government, will see unemployment rise again.  The youth wage decision has been estimated by Eric Crampton of the University of Canterbury to have cost at least 9000 jobs.

Ben Gleisner also makes the point that educational outcomes and benefit dependency (as well as wealth inequality) matter for future equality.  An interesting slide shows that New Zealand scores the worst of all OECD countries for the variance in educational outcomes – the so-called ‘long tail’ of under-achievement.  Another slide highlights the problems of benefit dependency, with 25 percent of the Maori working age population on welfare benefits (2006 figures).  Typically people are better off – in income, health and other terms – by being in the workforce, even at low initial wages, rather than being on benefits.  This point is especially important from a long-term and intergenerational perspective.


  1. Richard Wilkinson and Kate Pickett, The Spirit Level: Why Equality is Better for Everyone. Allen Lane, 2009; Penguin Books, 2010.
  2. Christopher John Snowdon, The Spirit Level Delusion: Fact-checking the Left’s New Theory of Everything, Democracy Institute/Little Dice, 2010; see also http://spiritleveldelusion.blogspot.com. Peter Saunders (ed. Natalie Evans), Beware False Prophets: Equality, the Good Society and The Spirit Level, Policy Exchange, 2010; see http://www.policyexchange.org.uk/images/publications/pdfs/Beware_False_Prophets_Jul_10.pdf
  3. Fraser Nelson, ‘Swedish conservatives bucked the recession by lowering taxes – and won re-election’, The Spectator, 25 September 2010, p. 12.
  4. Michael Stutchbury, ‘Plain old economic growth is good for society’, The Australian, 21 September 2010.

 

Advertisements

4 thoughts on “The Spirit Level

  1. The Spirit Level seems to be driven by the social agenda of climate change, just as the IPCC dismisses from the outset the effect of the sun on climate, you note the authors dismiss economic growth.

    The socialist have a new bandwagon, it is climate change. It is the driver they can use to show that consumption/growth will destroy the climate and all of mankind.

    Equality could be improved by providing clean water to everybody, helping to eradicate common diseases in less developed countries etc. But lets just focus on redistribution via taxes or hidden taxes called the scarcity pricing of carbon!

  2. This is the most nonsense research I’ve ever seen. Their analytical methods were just simply kindergarten. They used simple uni-variate descriptive statistics in their analysis, where as in the real world, it is multi-variate (i.e., there are more factors that @ play). This is what Physics Nobel Laureate, late Prof. Richard Feynman called, cargo cult science. They (useless academics) have become organized (& published their researches in academic peer review journals) and think that they’re doing intellectual work (or researches) but actually it is all nonsense, which means they (so called humanities academics) should just stick to rubbish collection, because that’s actually what they’re good at. They’re useless in doing proper analytical-based researches.

    I had emailed one of the co-authors of the The Spirit Level to challenge her about her findings, i.e., their use of uni-variate statistics, which was inappropriate, and she never replied back to me. In fact I quoted her some modern multivariate statistical algorithms/methods (that she should have used), but I’m pretty much sure that she never heard of them. I cited the peer review research journals that she could find those papers.

  3. Pingback: Antipodean responses to the Wilkinson-Pickett thesis at Catallaxy Files

  4. I do not find either Neil’s or Falafulu Fisi responses convincing. Neil simplistically suggests it is all tied up with a socialist (or maybe communist) conspiracy along with Climate Change and God knows what else.I have newspaper cuttings going back 50 years that raise that pathetic claim. Discuss the evidence and argument Neil. Falafulu Fisi tries to bamboozle us with mathematical terms and ends up sounding like Penny Brights and her “Cost benefit analysis” mantra.
    Anyone would grant that the detailed dynamics are complex and multivariable but that does not mean there isn’t a strong causal relationship between wealth extremes and negative statistics such as health and crime. There is ample evidence on the psycho-social effects of poverty and deprivation. What is complex is the way dispair and hopelessness plays out.
    In the absence of time and space I will only make two points:
    1. Game theory played iteratively strongly suggests players will continue to play if outcomes are fair. This is consistent with life experiences and commonsense.
    2. Countries that have extremes of wealth such as the US have a deeply compromised political process because officials and politicians can be bought or intimidated. (Wisconsin)

    On the question of Climate Change I suggest sceptics like Neil should browse the Brookings Insitutions own publications, all 1040 of them!

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