The Spirit Level

by THE CHRISTIAN INSTITUTE

March 2011

Contents

• Reducing inequality by coercion?
• What the book’s supporters say
• Technical deficiencies
• Influence of culture dismissed
• Fairness does not mean uniformity
• Leaping to conclusions
• True equality allows diversity
• New duty threatens more unfairness
• Innocent inequality or guilty discrimination?

Achieving true equality 

Christians are being increasingly marginalised in our society. One reason for this marginalisation is the current discrimination law framework, which has undermined diversity in the name of equality. 

But why has something with apparently admirable aims proved so damaging? The answer lies in the equality of outcomes approach underpinning our equality laws. 

This approach is exemplified by an influential book entitled The Spirit Level, published in 2009. Written by academics Richard Wilkinson and Kate Pickett, the book has been hailed as a “sweeping theory of everything” which proves that key social problems are caused by inequality.[1] 

According to the authors, the more unequal a society, the worse it does in a range of areas such as mental illness, teenage births and life expectancy. The book has repeatedly been cited in support of the existing equality agenda, including by politicians backing the recent Equality Act. Yet critics point out that, as well as containing significant inaccuracies, The Spirit Level focuses on equality of outcomes rather than equality of opportunity – the flawed approach which has caused so many problems. 

Outcome vs. opportunity 

At the heart of the problem with The Spirit Level is its focus on ‘equality of outcomes’ as opposed to ‘equality of opportunity’. The two approaches are poles apart, with radically different implications for society. 

Equality of opportunity seeks to ensure that everyone has the same chance to be the best they can be. It holds that talent should be allowed to flourish regardless of a person’s background. However, equality of outcomes seeks to ensure that everyone achieves the same results in life. If life were a race, equality of outcomes would ensure that everyone crossed the finishing line at exactly the same time. 

Equality of opportunity promotes liberty within a society, rewarding endeavour and encouraging the freedom for a person to make the most of their talents. But equality of outcomes presumes that difference must mean injustice, so it promotes top-down bureaucracy in a bid to iron out difference. It advocates positive discrimination and creates a tendency towards the lowest common denominator, levelling down achievement. 

All major political parties support increased social mobility (improving the social position of people from less advantaged backgrounds). In 2009, Alan Milburn chaired a Panel on Fair Access to the Professions. He said: “social mobility has slowed down in our country. Birth, not worth, has become more and more a determinant of people’s life chances”.[2] All agree there is a problem. The question is whether the equality of outcomes approach helps or hinders. 

Reducing inequality by coercion? 

The Spirit Level has been described as “a sweeping theory of everything”[3] and “big enough to change political thinking”.[4] It tries to sell the idea that people in more economically equal countries are healthier, happier and more successful. 

To compare the inequality between countries the authors choose to look at the difference in incomes between the richest 20 per cent and the poorest 20 per cent of a population.[5] They also compare the 50 US states. 

Wilkinson and Pickett use nationwide averages like death rates to argue that greater equality is better for everyone, not just the poor.[6] The authors call for power to be taken from individuals and redistributed more widely in order to achieve greater economic equality.[7] The book labels rich people as “damaging” for “the social fabric”[8] and endorses “substantial redistribution of wealth” in employment contexts.[9] 

The worldview that inequality is the cause of almost all social problems could have a massive effect on government policy. Followed to its logical conclusion, it would mean the coercive and radical redistribution of wealth by the state. 

What the book’s supporters say 

Politics 

There are numerous examples of politicians endorsing The Spirit Level. It has been heralded in the House of Commons as proof that “inequality is a major cause of practically every ill that afflicts our society”.[10] 

Using the book in the House of Lords to justify part of the Equality Act, the then Cabinet minister Baroness Royall said it was “much quoted in this Chamber”. She also praised the book as showing “that societies that are more equal in terms of income distribution tend to be better societies in every way”.[11] 

The Spirit Level has even been called “arguably one of the most important books written in recent times”.[12] 

David Cameron cited the book’s research in a speech before becoming Prime Minister.[13] Ed Miliband openly supported its hypothesis when he became Labour leader.[14]

Despite its many fans, others query whether The Spirit Level deserves such accolades. 

Media 

Some social and media commentators have strongly endorsed The Spirit Level. One article backed the idea that “everyone fares better in a more equal society”.[15]

Writer Johann Hari (below) praised the book’s ideas as “a compass to rebuild our societies”.[16]

Published rebuttals 

The Spirit Level Delusion by Christopher Snowdon and Beware False Prophets by Peter Saunders argue that The Spirit Level’s conclusions lack empirical support and are implausible. Saunders says: “The Spirit Level has little claim to validity. Its evidence is weak, the analysis is superficial and the theory is unsupported.”[17] 

Snowdon recognises that the book made an “astonishing claim” and, if true, would have “offered a whole new way of looking at politics”. But he highlights the problem with the claim: “It wasn’t true”.[18] The TaxPayers’ Alliance has also strongly criticised the book, saying that it promotes “utterly absurd ideas” and that its findings “do not withstand scrutiny”.[19]

Technical deficiencies

The Spirit Level has many statistical flaws. 

The findings are based on data from 23 of the world’s richest countries and 50 US states. The authors excluded countries with a population of under 3 million on the basis that they did not want to involve tax havens. 

But this amounts to cherry picking the data. Tax havens could have been eliminated far more precisely, since they are so wellknown. 

Rebuttals of The Spirit Level have reinstated the data for many missing countries and found that their inclusion weakens the findings.[20] 

The book uses graphs to look for relationships betwen inequality and individual social problems. However, there may in reality be other factors at work. Taking these into account undermines the association with inequality supposed by the authors. 

When analysing their data, Wilkinson and Pickett handle ‘outliers’ incorrectly. Outliers are extreme results which significantly differ from the trend and expected outcome. 

Outliers can significantly skew results, so standard statistical procedure dictates that they should be excluded for accurate analysis to be made.[21] They can be found easily by running statistical tests. But The Spirit Level’s analysis does not exclude outliers. 

What the detractors say 

A number of commentators have expressed extreme caution about The Spirit Level. 

Writing in The Spectator, Toby Young questions the book’s conclusions. He argues that even if The Spirit Level were true, which he does not believe, it “doesn’t mean we ought to equalise incomes”. 

Young says there are other factors to consider such as “social and economic freedoms”.[22] 

Ed West, a blogger for The Daily Telegraph, has described the book’s hypothesis as “complete rubbish”, highlighting the statistical flaws in both its methodology and conclusions.[23]

Influence of culture dismissed 

The Spirit Level rejects history and culture as having a significant influence on the social outcomes of a nation. 

The comparison between Sweden and Japan is a good example. Saunders says: “They are not cohesive societies because their incomes are equally distributed; their incomes are equally distributed because they evolved as remarkably cohesive societies”.[24] There is an emphasis on national belonging in Sweden and it has had low levels of immigration. 

Japan, however, has a very different culture to Sweden yet shows similar outcomes. The Spirit Level argues that this proves it is not culture that is causing the outcomes. 

But the two cultures do have crucial similarities. Saunders points out that both have a strong ‘folk’ tradition and historically they have been relatively ethnically homogeneous. In contrast, the Englishspeaking countries are very individualistic and have been settler nations with an emphasis on open borders.[25] 

These different cultural foundations surely have an influence on the types of outcomes measured by the authors of The Spirit Level. 

Correlation not the same as causation

It is a basic principle of statistics that correlation does not always imply causation. When faced with two trends that seem to coincide it is easy to think one causes the other.

However this common assumption is often mistaken. It is possible that a change in one thing accompanies a change in a second, but that both changes are due entirely to a separate factor. 

For example, it was found that women on HRT had a lower incidence of coronary heart disease. The premature conclusion was drawn that HRT gave protection against heart disease. 

But further investigation revealed that women using HRT were of a higher socioeconomic group.[26] 

Likewise, The Spirit Level is far too quick to infer causal links between inequality and social ills. 

Fairness does not mean uniformity 

Equality of outcomes tends to look only at results. It does not go below the surface. When results differ between groups of people there is an assumption that there must be something inherently unfair in the system. But the reality is that a fair system will frequently yield unequal results. A fair system offers the same opportunities for everybody. But given those opportunities, a complex combination of many factors may determine the outcome. 

For example, it is a well known fact that there are more female than male primary school teachers, men making up less than 20 per cent of the profession.[27] Yet there is no inequality in the recruitment system. Men are just as free to apply and be appointed. But it may be the case that more women than men want to work with small children. It might even be the case that women are more likely to be promising applicants. This does not mean the recruitment system needs to change to boost male recruits into the profession. It is already fair. 

Leaping to conclusions

Culture 

The Spirit Level claims that Japan and the Scandinavian countries tend to perform better than the English-speaking countries. A very likely explanation for this lies in the history and cultures of these countries. They have a resilient sense of collective identity. Historically, they have also been relatively ethnically homogeneous. The Spirit Level disregards these factors and erroneously attributes the findings to income inequality.

Suicide 

According to The Spirit Level there are some areas where more equal societies apparently do worse. The book notes that suicide rates are higher in more equal countries while murder rates are lower. The authors amazingly suggest this is because people in more equal countries take their frustrations out on themselves rather than killing other people![28]

Camaraderie 

The Second World War is lauded by the authors as a time when greater equality resulting from “deliberate government policies” caused a sense of “camaraderie and social cohesion”.[29] However, Snowdon points out that this was more likely “a result of fighting for national survival”.[30]

Educational achievement 

The authors hold that educational achievement is based on income inequality. However, a stronger correlation with educational achievement can be found by putting the countries in alphabetical order or looking at the geographic latitude of the capital cities.[31] 

Crime levels 

The Spirit Level argues that there was little economic discontent during the Second World War and that, with greater equality, crime rates fell. Perhaps that was the cause. But, as Snowdon points out, “a far more plausible explanation is the conscription and export of several million young men”.[32] 

Recycling 

The Spirit Level claims more equal countries recycle more. A graph shows two categories of countries: one recycles vast proportions of rubbish, the other does not. The reason is simple: compulsion. Some countries have tough recycling laws.[33] Recycling levels reflect government requirements, rather than community spirit. 

True equality allows diversity 

Christians believe that men and women are created in God’s image and are consequently of equal worth in God’s sight.[34] The New Testament teaches that Christians “are all one in Christ Jesus”, regardless of race, sex or social status.[35] Church members come from a very diverse background in terms of their ethnicity, culture and socio-economic status. 

Whilst people are equal in value, they can have different characteristics, beliefs or conduct which justifies a difference of treatment. 

For example, women have fewer car accidents than men. It is morally acceptable for insurance companies to charge them lower insurance premiums. 

Atheist teachers do not have to attend school assemblies and pacifists are not compelled to fight in war. Synagogues can reject non-Jews as members. A resident of a Hindu old people’s home cannot demand that beef is served on the menu. 

It is freedoms like these which are absolutely essential for democratic society. 

There are limited areas when basic freedoms can be denied because of a person’s conduct. Many see stopping prisoners from being able to vote as legitimate. 

But using equality law to coerce Christians to live and work as though they were atheists is not.

New duty threatens more unfairness

The Government is about to use regulations to implement the Public Sector Equality Duty under the Equality Act. These regulations will require public authorities to publish equality objectives. This target-setting emphasis is driven by an equality of outcomes approach, and will inevitably lead to further injustice against Christians. 

The regulations come despite equality minister Theresa May distancing the Government from equality of outcomes. She has said that the Government should be encouraging equality of opportunity. She pointed out that, under an equality of outcomes approach, the word ‘equality’ had become a “dirty word”, associated with “political correctness and social engineering”.[36] 

But the focus on outcomes indicated by the requirement to publish targets goes far beyond what is actually required by the Equality Act and will be a distraction for public bodies. 

Furthermore, there is a risk that continuing this same failed approach will further damage the religious liberty of Christian groups who work in or with the public sector. Recent years have shown that public bodies have been all too willing to interfere with religious liberty at both an organisational and personal level in the name of equality (see examples below). Increased pressure on public bodies to set and meet equality objectives threatens to accelerate the marginalisation that Christians are already experiencing.

Example 1 

Pilgrim Homes 

A care home for elderly Christians, including retired missionaries, was stripped of £13,000 of local council funding. Among other things, Brighton and Hove Council wanted the home to ask the residents about their sexual orientation every three months. The funding was restored after legal action was taken.

Example 2 

Caroline Petrie 

Christian nurse Caroline Petrie was suspended on 17 December 2008 by North Somerset Primary Care Trust after she offered to pray for a patient. She was accused of breaking nursing guidelines by failing to “demonstrate a personal and professional commitment to equality and diversity”. Mrs Petrie was later reinstated.

Innocent inequality or guilty discrimination? 

• In 2010, 57.9% of boys and 71.8% of girls gained A*-C for GCSE English, in line with the consistent pattern for many years.[37]
• None of the 8 contestants in the 2008 men’s 100m Olympic final were Caucasian.  
• 60% of foster carers are female.[38] 
• Less than 11% of nurses are male.[39] 
• There are far more male than female bodybuilders.[40]

References

1. The Guardian, 12 March 2009; The New Statesman, 26 March 2009
2. Unleashing Aspiration, Cabinet Office, July 2009, page 5
3. The Guardian, 12 March 2009 
4. The Sunday Times, 8 March 2009 
5. Wilkinson, R and Pickett, K, The Spirit Level, Penguin Group, 2009, page 18 
6. Ibid, pages 186-187 
7. Ibid, page 253 
8. Ibid, page 262 
9. Ibid, page 253
10. House of Commons, Hansard, 19 October 2009, col. 687 
11. House of Lords, Hansard, 11 January 2010, col. 327 
12. House of Lords, Hansard, 11 January 2010, col. 316 
13. Conservatives Speech, David Cameron: The Big Society, 10 November 2009, see http://tinyurl.com/yjeq72f as at 17 February 2011 
14. BBC News Online, 28 September 2010, see http://tinyurl.com/4gslxvk as at 17 February 2011
15. Guardian Unlimited, 30 September 2010 
16. Johann Hari, Blog, 15 April 2009, see http://tinyurl. com/4lp5hyf as at 24 February 2011
17. Saunders, P, Beware False Prophets, Policy Exchange, 2010, page 8 
18. Snowdon, C, The Spirit Level Delusion, Little Dice, 2010, pages 9 and 12 
19. Sanandaji N, Malm, A and Sanandaji, T, The Spirit Illusion, The TaxPayers’ Alliance, pages 3 and 27
20. Snowdon, C, Op cit, pages 13-15 and 26 
21. Saunders, P, Op cit, page 30
22. Spectator.co.uk, 13 August 2010, see http:// tinyurl.com/37dcozd as at 17 February 2011 
23. Ed West, Blog, 9 July 2010, see http://tinyurl. com/38qt5e7 as at 16 December 2010
24. Saunders, P, Op cit, page 8 
25. Saunders, P, Op cit, pages 117-121
26. Lawlor, D A et al, ‘The Hormone Replacement– Coronary Heart Disease Conundrum: Is This the Death of Observational Epidemiology?’ International Journal of Epidemiology, 33 (3), 2004, pages 464-467
27. The Daily Telegraph, 7 August 2008
28. Wilkinson, R and Pickett, K, Op cit, page 175 
29. Wilkinson, R and Pickett, K, Op cit, page 85
30. Snowdon, C, Op cit, page 112 
31. Snowdon, C, Op cit, pages 140-142
32. Snowdon, C, Op cit, page 112 
33. Snowdon, C, Op cit, page 101 
34. Genesis 1:27
35. Galatians 3:26-28
36. Home Office, Equality Strategy Speech, 17 November 2010, see http://tinyurl.com/3and6fn as at 24 February 2011
37. BBC News Online, 24 August 2010, see http://tinyurl.com/5sxkm8y as at 4 March 2011; Ofsted Evaluation, English at the Crossroads, see http://tinyurl. com/4tcuph3 as at 1 March 2011; The Guardian, 22 August 2002 
38. The Fostering Network, Press Release, Men Should Not Be Put Off Fostering, Says Leading Charity, 18 November 2010 
39. Nursing and Midwifery Council, Statistical Analysis of the Register, see http://tinyurl.com/4cqo4b7 as at 17 February 2011 
40. Sunday Telegraph – Stella Magazine, 14 November 2010