A few things that inequality causes

Of the endless stream of papers that flash in front of my eyes every week, a large number are ‘Spirit Level style’ – that is, they look at the relationship of inequality and a ‘bad thing’ between countries/areas. If I blogged about each of these then there would be no room for anything else on the blog, but I thought this week I’d summarise four that particularly caught my eye. They variously cover crime, the family burden of caring for children with special needs, self-perception, and intergenerational mobility – which if nothing else, tells you that people with a lot of different interests are doing this kind of research…

Inequality and Crime in English areas

The first is by Adam Whitworth, and looks at inequality WITHIN AREAS of England.  He tests whether high-inequality areas have greater levels of crime than low-inequality areas, controlling for a number of area characteristics. An unusual feature of this study is how it defines within-area inequality – this is not about individual incomes, but instead is income inequality between different (small) areas within the bigger area. The main regression results are shown below.

Put simply, Whitworth shows that high-inequality areas have higher crime levels for ALL these types of crime than low-inequality areas. I’ll have a final word on what these results actually mean below.

Inequality and the burden of caring for kids with special needs, in US states

Now this is an outcome I’ve never seen before: does inequality raise the burden of caring for kids with special needs? Parish et al 2012 measure this outcome in multiple ways, and – controlling for child’s condition, ethnicity etc – find that high levels of state inequality are associated with (i) no help in arranging care, (ii) stopping working due to child’s health; (iii) increases in the financial burden of care, as a % of income. They present the results unusually too, in terms of how the financial burden would be different in each state if they had average levels of income inequality.

Inequality and seeing yourself as better than average, internationally

OK, so this is perhaps my favourite of the four (and not just because it was led by a colleague at the University of Kent – I don’t even know Steve Loughnan!).  Loughnan et al (in press) looked at whether high-inequality countries had higher levels of self-enhancement – that is, seeing yourself as better than average. Note that (like most psychological research…), this was a study done on university samples.

In this study, self-enhancement is measured by asking people “How much do you possess this characteristic compared to the average person” across 80 personality traits, and then “This characteristic is desirable, it is a characteristic that people generally want.”  Self-enhancement is shown if people are more likely to say they possessed a characteristic if they saw it as desirable (p1256) – which people in all countries did.

Self-enhancement has reliably been shown to differ between countries, but – according to Loughnan et al – people previously thought this was due to differences in Western and Eastern ways of viewing the world. Their argument is that it’s actually due to inequality – as the chart below would suggest. When they test this more closely, they found inequality is a much better predictor of self-enhancement than individualism vs. collectivism, or power distance, the two cultural variables that are usually used to explain it.

Inequality and social mobility, internationally

Finally, a graph that is already well-known, and deservedly so: do high-inequality countries have lower levels of social mobility? The answer is shown below (taken from Jared Bernstein’s blog, using Miles Corak’s work) – and basically, is ‘yes’. Apparently this is known by some as ‘the Great Gatsby curve’ (!). I discuss this a bit more in the final section below.

So what does this all mean?

Longstanding readers will know that I think there are problems in reading too much into these analyses (see these two posts of mine, plus Rob’s post). They show an association – but does this really show the causal effect, or is there something else going on behind them?

  • Is it that inequality causes the ‘bad thing’, or the other way round? Loughnan et al are suitably cautious about drawing causal conclusions from their work; they say there ‘may’ be an effect, but don’t present their results as definitive. They also say that the link between the two (even if causal) is likely to be complex; it “is unlikely that economic inequality directly leads to biased self-perception.”
  • Is there something else that causes BOTH inequality and the ‘bad thing’? This seems especially likely for the study on the burden of caring for kids with special needs – more right-wing states are likely to tolerate (encourage?) more inequality, and to have less generous social welfare budgets that provide lower support. So it seems unlikely to me that the link with inequality is causal.

That said, Miles Corak offers an excellent justification for a causal interpretation of his graph on social mobility. These relationships give us invaluable, interesting clues about the possible effects of inequality – but they need (i) further empirical research + (ii) a theoretical idea of what’s going on, if they are to really provide evidence of the effects of inequality.

About these ads

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 Blog posts and tagged , , , , , , . Bookmark the permalink.

6 Responses to A few things that inequality causes

  1. mel bartley says:

    Hi Ben –
    You say something interesting about the theoretical approach taken to understanding the relationships of income inequality to all these outcomes. Very good point, but then not developed, do you have more thoughts on this? Having a strong welfare state ‘de-commodifies’ more people’s labour power, so that the penalty for not working full time in a well paid job is less. This might give more energy for community involvement, caring and so on. In terms of social mobility, though, I have long thought that one problem in a very high-inequality society is that parents will struggle harder to avoid downward mobility (because the relative wage penalty is greater). So young people who might be much happier and healthier doing a craft or technical job are either forced into a high-income occupation or live in a twilight world where their occupation is nominal only, and their income actually comes from their parents.

    • Ben Baumberg says:

      Thanks Mel – getting more detail on the theoretical aspects is really important (chats with Rob de Vries are always really useful around this). That theoretical account seems defensible: strong welfare state -> lower penalty for not working full-time -> more time given to caring. But that doesn’t to me seem a consequence of inequality, it’s a consequence of a strong welfare state (which itself will also reduce inequality, i.e. the effect of inequality is due to confounding).

      Whereas the second chain you describe is a GENUINE (if indirect) causal effect of inequality: high inequality -> greater incentives to avoid downward mobility -> kids with special needs receive additional pressure. (Although I’m not quite sure the link works all the way down).

      But this is the kind of thing that needs doing, and I don’t think (from skimming the paper) that the authors of the article do this enough.

  2. Daniel Sage says:

    Hi Ben,

    Really interesting post. I actually did my MSc thesis at Stirling on the relationship between inequality and various indicators of social cohesion for the same sample that Wilkinson and Pickett use in The Spirit Level. In short, I found that inequality was significantly associated with social cohesion across certain dimensions, but not others. The difference (I thought) was between (a) social relationships that exist ‘between’ different groups in society and (b) those relationships ‘within’ more tight-knight groups (where social status is less important). I found that the latter type of relationships were not associated with inequality, while the former generally were.

    All in all, although I think it seems clear that there is some sort of link between inequality and various social ills, the problem is that income inequality is perhaps a poor way to measure what the theory really states is important: the scale and qualitative nature of differences in social status. I think this is where theoretical work is most urgent.

    Daniel

    • Ben Baumberg says:

      Will email you in a second Daniel, but before then – thanks for the comment, can I persuade you to write a post on your MSc thesis?!

      And I couldn’t agree with your second point more. We’ll be returning to this on the blog.

  3. Ben Baumberg says:

    And via the Equality Trust – more detail on the debate about intergenerational mobility, by Justin Wolfers on Freakonomics – http://www.freakonomics.com/2012/01/19/is-higher-income-inequality-associated-with-lower-intergenerational-mobility/

  4. Pingback: Health Promotion Headlines from Robyn & Penny 2012/03/20 | Health Promotion Today / Promotion de la santé aujourd'hui

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