‘Statistical catfights’ on the effects of inequality

Front Cover of 'The Spirit Level'However hard we try to communicate our work, inequality research is often tucked away on the unread pages of newspapers – if it even gets that far. So it’s slightly shocking that a book presenting evidence that ‘more equal societies almost always do better’ has become the must-read political book of the year in the UK.  Richard Wilkinson and Kate Pickett’s The Spirit Level has not only been endlessly debated in the media, but has  been cited approvingly by the British (Conservative) Prime Minister, and as I write is even at #15 in the amazon.co.uk bestsellers list (just behind the third Twilight book…).  But the inevitable backlash has been public and fierce, and we’re now at a point that we can stop to think:  has the book helped change the terms of debate in British politics?  And more controversially: has the book actually done more harm than good in tackling inequality?

(We’ll return to the separate question of how far the claims are actually *true* in later posts…)

In thinking about the effects of the book, it’s helpful to think about two completely different audiences.  Firstly, there are the left-wingers who already thought that the currently high level of inequality is a ‘bad thing’ because the inequalities are intrinsically unjustifiable.  The Spirit Level has galvanized debate around inequality within the Left, but it hasn’t really changed anyone’s political position because they were already convinced.

Secondly, though, there is the rest of country, who *don’t* believe that there is anything wrong with inequality per se.   The Spirit Level has the potential to change these people’s minds because it shows that the things that everyone cares about – death, mental illness, crime, the wellbeing of children – are linked to inequality.  For example, in a report by The Fabian Society,1 their qualitative sample drawn from a wide range of the ‘moderate’ British public found that the most convincing argument to tackle inequality was the idea that it caused crime and reduced trust.  For those on the right, as the  ‘progressive Conservative’ Max Wind-Cowie has argued:2

“[For conservatives], concern [around inequality] is founded not, as it is for many on the left, on a sense of injustice but rather on an evidence-based assessment of the problems that inequality causes.”

But has it helped?

The book matters, then – but to actually change people’s minds, people  have to be *convinced* that The Spirit Level is true.   Max Wind-Cowie feels that “the evidence for the impact of inequality is now irrefutable” (p21), yet other Conservatives have been more sceptical.  2010 has seen very public critiques launched by the influential centre-right think-tank Policy Exchange, the seriously right-wing pressure group The Taxpayers Alliance,  and a book by Chris Snowden.  Their arguments take us back to the way that Wilkinson and Pickett try and convince us that they’re right.

Their central tool of persuasion is a series of two-way graphs showing the relationship of inequality with various social ills.  Most prominent among these is the graph below, which shows how inequality relates to a combined index of health and social problems.  The relationship is indeed very strong, and supported by a statistical significant correlation, and it’s easy to think that this is a perfect example about how we should present research results.

Relationship of Inequality to Health & Social Problems (graph taken from The Spirit Level)

Unfortunately, simplicity comes at a price.  Graphs like this are easy to for non-specialists to understand – but they’re almost as easy for the non-specialist critics to take apart. Some of the graphs look very sensitive to particular countries – what happens if we leave out ‘atypical’ countries that look like outliers?  The analysis is simple and the data are publicly available – so why don’t I try repeating the research myself, but using slightly different indices or combinations of countries? More importantly, it often seems like the results are being driven by the Nordic countries and Japan at one extreme, and the US at the other.  How do we know that it’s not some other difference between these countries – culture, politics etc. – that is causing the worse outcomes?

These are the main (although not the only) arguments of the right-wing critics of The Spirit Level.  Wilkinson and Pickett have energetically responded, leading to a very public ‘statistical catfight’ across the UK media, as one blogger put it.   (That phrase was triggered by a debate I attended at the Royal Society of Arts – ‘catfight’ is the right word for a fascinating debate that was marred by an aggressive, patronising tone on all sides…).

In these debates, the crudeness of the two-way graphs has become a major handicap because they are *obviously* not very convincing. Wilkinson and Pickett have instead defended themselves through their academic credentials and the large volume of peer-reviewed articles on inequality.  This begs the question about why these simplistic graphs should been given such a prominent role in the first place? Richard Wilkinson’s comment at the end of the RSA debate left me incredulous here:

“All of these attacks are based on a complete misunderstanding of what we were about. We took relationships that had been largely established by other people’s work on other other groups of countries or states, and we looked to see if they could be demonstrated …amongst this group of rich market democracies.”3

So the graphs were there to demonstrate relationships, but not to try and prove them. This is frustrating: Wilkinson and Pickett do discuss the much larger and more robust research on the effects of inequality, but it has been marginalised by the attempt to convince people through these simple graphs.   I can’t help feeling that the cat’s now been let out of the bag; once we’ve suggested that we believe inequality is harmful because of these graphs, it’s then difficult to say that we never thought the graphs were much use anyway.

The future UK debate on inequality

There are three things that I take from this.  Firstly, researchers shouldn’t sacrifice rigour when they’re presenting data to the public.  It would easily have been possible to display two-way graphs that already adjust for other possible explanations (GDP, culture, ethnicity etc),  as Bob Putnam did for Bowling Alone.  I think Andrew Gelman is a master at this: slides 34 and 36 in the linked presentation show complex underlying models in an intuitive way.

Secondly, The Spirit Level genuinely did manage to spur debate on inequality among left-wingers in the UK.  Perhaps this would have happened anyway in the post-Blair years (given that no progress was made in reversing the rise in inequality under Thatcher, although they had to work hard to stop it getting any worse), but the book probably helped a little too.

Finally, after all the media furore about The Spirit Level, it seems as if we’re now back where we started: left-wingers believing that inequality is damaging, and right-wingers vehemently disagreeing.  In my darker moods, I sometimes wonder if research will ever convince anyone of anything, in these highly politicised fields where conclusive evidence is hard to come by…

…but then I finish my blog post and turn back enthusiastically to my own research, so perhaps I have a more optimistic/naive side hidden deep within me too…


Footnotes

1. The Fabians are a intellectual left-wing think-tank with a long history.   This research itself was funded by the Joseph Rowntree Foundation, which seeks “to understand the root causes of social problems, to identify ways of overcoming them, and to show how social needs can be met in practice,” and is an important UK funder of social research.  The report can be accessed at http://www.jrf.org.uk/publications/attitudes-economic-inequality, see p44-45.

2. See http://www.demos.co.uk/files/One_Society_Con-web.pdf?1265674614 , page 14.  It’s worth noting that the Demos report was part-funded by The Equality Trust, the advocacy body set up by Wilkinson and Pickett.

3. This quote starts 53 minutes and 50 seconds into the audio file, which you can download from http://www.thersa.org/events/audio-and-past-events/2010/the-spirit-level.

About Ben Baumberg

I am currently a Lecturer in Sociology and Social Policy at the School of Social Policy, Sociology and Social Research (SSPSSR) at the University of Kent. I also helped set up the collaborative research blog Inequalities, where I regularly write articles and short blog posts. I have a wide range of (too many...) research interests, at the moment focusing on disability, the workplace, inequality, deservingness and the future of the benefits system, and the relationship between evidence and policy. You can find out more about me at http://www.benbaumberg.com
This entry was posted in Articles and tagged , , . Bookmark the permalink.

6 Responses to ‘Statistical catfights’ on the effects of inequality

  1. Ben Baumberg says:

    The question of whether Wilkinson and Pickett’s claims are actually true is covered in a separate post by Rob de Vries, which will be published this time next week (29th Sep).

  2. Pingback: Wilkinson & Pickett: Are they right? | Inequalities

  3. Pingback: Winner-Take-All Politics: Why the Ultra Rich are Getting Ultra Richer | Inequalities

  4. Pingback: Highlights so far… | Inequalities

  5. Pingback: A few things that inequality causes | Inequalities

  6. Pingback: Crossing the divide | thenextwave

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