Beware False Rebuttals – Our Response to Peter Saunders’ Policy Exchange Report

The Equality Trust Media Release – 8 July 2010 for immediate release

Beware False Rebuttals – A response by the authors of The Spirit Level to a report by Peter Saunders (Beware False Prophets), published by Policy Exchange.

Responding to the new report by Peter Saunders, published today by Policy Exchange, Professor Richard Wilkinson & Professor Kate Pickett said:

“We welcome open debate of our findings that more equal societies do better, but Peter Saunders’ analysis contains serious methodological errors. There are many peer reviewed analyses of relationships with inequality carried out by other researchers which support The Spirit Level’s conclusions. In particular there is substantial evidence elsewhere that infant mortality, life expectancy, violence, trust, social capital and school bullying are all worse in more unequal societies. The evidence for the benefits of greater income equality remains compelling.”

  • 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. (1)
  • 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 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. (2)
  • 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. (3)
  • 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 thehighest 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. (4)
  • 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.

Richard Wilkinson & Kate Pickett are available for interview. Please contact 020 3637 0324 or 

Notes for Editors

1. The first paper showing that more equal societies have better health was published in 1979 (Rodgers) and since then it has been tested several hundred times on different data, using different methods in many different settings. (See Wilkinson & Pickett, Income Inequality and Health, a review and explanation of the evidence, 2006)

There have also been a number of other peer reviewed analyses published showing more unequal societies do worse on measures of trust, social capital, bullying, obesity, infant mortality. (Wilkinson & Pickett, 2009, Income Inequality & Social Dysfunction

The tendency for violence to be more common in more unequal societies has, similarly, been shown in 30-40 different research papers in peer reviewed journals. (Wilkinson & Pickett, 2006, op.cit)

2. In The Spirit Level analysis the authors took countries among the 50 richest in the world with populations of more than 3 million, for which there was comparable income distribution data. They did this because they wanted to look at the countries where life expectancy and other outcomes have ceased to be related to economic growth. Peter Saunders adds in Chile, Argentina, Mexico, Venezuela, Turkey, Trinidad & Tobago, Malaysia, Russia, Estonia, Lithuania, Latvia, Poland, S. Korea, Romania, Slovenia, Hungary, Croatia, Czech Republic, Slovakia. In Figure 1.1 (in The Spirit Level) it can be seen that all these countries are on the rising part of the curve indicating that for them, unlike the richest countries, economic growth remains beneficial. Saunders’ later demonstration that economic growth remains beneficial is entirely a result of including these poorer countries. Whilst it would have been methodologically mistaken for the authors to control for Gross National Income per capita because it has no effect on outcomes among the richest countries, Saunders should have controlled for it among his countries because it does have a powerful effect. Numerous analyses published in peer reviewed journals have shown the importance of inequality in poorer countries after controlling for GNIpc. It is extremely illogical to include such disparate countries in the analysis while saying that the Scandinavian countries or the USA should be removed in order to compare like with like.  

3. Ram, R., Income inequality, poverty, and population health: evidence from recent data for the United States. Soc Sci Med, 2005. 61(12): p. 2568-76. Ram, R., Further examination of the cross-country association between income inequality and population health. Soc Sci Med, 2006. 62(3): p. 779-91. Ash, M. and D.E. Robinson, Inequality, race, and mortality in U.S. cities: a political and econometric review of Deaton and Lubotsky (56:6, 1139-1153, 2003). Soc Sci Med, 2009. 68(11): p. 1909-13; discussion 1914-7. Subramanian, S.V. and I. Kawachi, Income inequality and health: what have we learned so far? Epidemiol Rev, 2004. 26: p. 78-91. For the empirical argument please see here.

4. This is what is shown in The Spirit Level figures 13.2 – 13.5 and is supported by a number of research papers using multilevel models in the academic journals (discussed in The Spirit Level, chapter 13). Saunders suggests that figs 13.2 – 13.5 suggest only that the average is better in more equal than less equal countries. They actually show that every class, educational and income group would do better than the equivalent groups if they were in a more equal society.

  • In May 2010, The Spirit Level: Why More Equal Societies Almost Always Do Better (Allen Lane, 2009) won the book prize at the Bristol Festival of Ideas. The prize is awarded for the best book presenting new, important and challenging ideas, and which is engaging, accessible and rigorously argued.
  • Richard Wilkinson has played a formative role in international research on the social determinants of health and on the societal effects of income inequality; his work has been published in many languages. He studied economic history at the London School of Economics before training in epidemiology. He is Professor Emeritus of Social Epidemiology at the University of Nottingham Medical School, Honorary Professor at University College London and a Visiting Professor at the University of York. Richard co-wrote The Spirit Level with Kate Pickett and is a co-founder of “>The Equality Trust.
  • The international data is available from here and statistical sources and methodology from here.