Radical findings or radical propaganda?



More junk science from the Left

By Peter Saunders

Last year, two socialist academics, Richard Wilkinson and Kate Pickett, published a book called The Spirit Level which has had a huge impact in ‘progressive’ circles. In Britain, The Guardian’s Polly Toynbee likened Wilkinson to Charles Darwin, and The Independent was so impressed by the book it wanted to make ‘free-marketeers memorise it cover to cover.’

The book claims that the case for radical income redistribution in rich countries can be defended scientifically. It suggests that countries where incomes are more equally distributed do better on a range of social indicators – health, homicides, literacy, trustworthiness, teenage births, bullying – than more unequal countries. It concludes that even affluent people in unequal countries would benefit if incomes were spread more evenly.

Not everybody on the left has endorsed the book. Andrew Leigh at the Australian National University is one who has expressed caution, pointing out that life expectancy and infant mortality (two of Wilkinson and Pickett’s favourite indicators) have actually improved most rapidly in countries where incomes have widened the furthest.

But the real problem lies in Wilkinson and Pickett’s own data. I recently published a critique of their work available for free download here

There is, for a start, a problem in their choice of countries. They say they selected the 50 richest nations, but they only included 23, and they excluded several unequal countries with strong social profiles that would have undermined the patterns they wanted to find.

They were similarly selective in their choice of indicators. Imprisonment gets in, but not crime (except homicides). Drugs are in, but alcoholism is out. Murder is included, suicide is excluded. Social capital is measured by whether people say they trust each other, but membership of voluntary organisations is ignored. Using a different set of indicators, we could demonstrate the opposite of their hypothesis, that social problems appear to be worse in more egalitarian countries.

Then there are their graphs, like their plot of international homicide rates. It shows that 22 countries have similar rates but one (the United States) has a much higher rate. They happily fit a straight trend line, which is inevitably distorted by this single outlier. The line seems to show that homicides rise as inequality widens, but it’s nonsense. Britain has a lower homicide rate than Sweden, yet according to their trend line, homicides in the United Kingdom are much higher than in Sweden. They even suggest that, if Britain reduced income inequality to Swedish levels, the number of murders would fall by three-quarters! Utter rubbish.

As well as making spurious international comparisons, The Spirit Level analyses the 50 US states, arguing that more equal states have better social outcomes. But what’s driving this is not income distribution but ethnicity. The proportion of African-Americans in a state is, for example, 18 times more powerful than income inequality in predicting its infant mortality rate. But Wilkinson and Pickett never control for ethnicity, and in The Guardian, they even called me a ‘racist’ for daring to suggest that they should.

This is shabby stuff, propaganda masquerading as social science. But that won’t stop true believers using the book to push an egalitarian agenda. Two weeks ago, I debated with Wilkinson and Pickett in front of a sell-out audience at the Royal Society of Arts. Many in the audience had been marshalled by Wilkinson’s ‘Equality Trust’ (a pressure group formed to drive forward the book’s agenda). At the end, the chairman asked if anyone had changed their mind as a result of the debate. One brave soul in an audience of 200 raised his hand.

The above is a press release from the Centre for Independent Studies, dated Aug. 13. Enquiries to cis@cis.org.au. Snail mail: PO Box 92, St Leonards, NSW, Australia 1590.

No comments:

Post a Comment

All comments containing Chinese characters will not be published as I do not understand them