Last year, Richard Wilkinson and Kate Pickett published The Spirit Level, which caused quite a stir as it appeared to offer empirical evidence that egalitarian societies benefit rich and poor alike.
The report argued that income inequality harms not only the poorest in society but almost everyone, no matter how prosperous they are. Wilkinson and Pickett based their claims on statistical evidence, which they argued showed that more unequal countries (and within the US, more unequal states) suffer from, among other things, higher crime, lower average life expectancy, higher infant mortality and less social mobility. The authors concluded that we would all benefit from a more egalitarian distribution of income. The Spirit Level was lauded by politicians and commentators on the left who believed they finally had empirical proof that economic inequality leads to a wealth of social ills.
However, a new report by Peter Saunders, published by Policy Exchange, finds significant flaws in Wilkinson and Pickett's analysis that fatally undermines their arguments and conclusion. Beware False Prophets re-examines the empirical claims made in The Spirit Level and finds that of the 20 statistical claims made in it, 14 are spurious or invalid and in only one case (the association internationally between infant mortality and income inequality) does the evidence unambiguously support their hypothesis. Contrary to Wilkinson and Pickett's claims, income inequality does not explain international homicide rates, childhood conflict, women's status, foreign aid donations, life expectancy, adult obesity, childhood obesity, literacy and numeracy or social mobility rates. Nor does it explain variations among US states in homicide, infant mortality or imprisonment rates.
Saunders identifies a couple of key analytical failures that run through The Spirit Level.
First, there are many instances where the authors draw conclusions from data where one or two extreme cases are used to support unwarranted generalisations. These "outliers" totally skew the data but give the conclusion the authors seek – for example, the claim that there is an association between a country's homicide rate and its level of income inequality depends entirely on the high murder rate in the US, while the claim that average life expectancy is linked to income inequality rests entirely on the long lives of the Japanese. Across the other 22 countries analysed, there simply is no association.
Second, many of Wilkinson and Pickett's claims rest on the influence of "clusters" of countries or states that do not hold true across the wider sample of countries. So the (more equal) Scandinavian countries routinely appear at one end of their graphs, and the (less equal) Anglo nations often appear at the other. Yet beyond these clusters the search for evidence that links inequality and negative social outcomes is in vain, as other more egalitarian countries such as Japan and Austria do not appear to support the hypothesis any more than less egalitarian ones such as Portugal or Singapore.
Peter Saunders contends that the single-minded focus of Wilkinson and Pickett to prove their hypothesis led them to resist consideration of the impact of factors such as history and culture, which in fact are often one of the key factors behind their findings. It also meant they were selective in their choice of evidence, so The Spirit Level ignores a range of social indicators such as suicide rates, HIV infection rates, alcohol consumption and divorce rates, which are in fact worse in more equal countries.
Beware False Prophets is a hard-hitting critique that shines a powerful spotlight on the flaws in the analysis, assumptions and conclusions of The Spirit Level. We all want to improve people's quality of life and tackle deep-rooted social ills, but as Saunders clearly identifies, the case for radical income distribution to achieve this is no more compelling now than it was before The Spirit Level was published.
