HOW ECONOMIC INEQUALITY HARMS SOCIETIES – RICHARD WILKINSON

https://www.ted.com/talks/richard_wilkinson_how_economic_inequality_harms_societies/transcript

You all know the truth of what I'm going to say. I think the intuition that inequality is divisive and socially
corrosive has been around since before the French Revolution. What's changed is we now can look at
the evidence, we can compare societies, more and less equal societies, and see what inequality
does. I'm going to take you through that data and then explain why the links I'm going to be showing
you exist. 

00:34

But first, see what a miserable lot we are. (Laughter) I want to start though with a paradox. This shows
you life expectancy against gross national income -- how rich countries are on average. And you see the
countries on the right, like Norway and the USA, are twice as rich as Israel, Greece, Portugal on the
left. And it makes no difference to their life expectancy at all. There's no suggestion of a relationship
there. But if we look within our societies, there are extraordinary social gradients in health running right
across society. This, again, is life expectancy. 

01:15

These are small areas of England and Wales -- the poorest on the right, the richest on the left. A lot of
difference between the poor and the rest of us. Even the people just below the top have less good
health than the people at the top. So income means something very important within our societies, and
nothing between them. The explanation of that paradox is that, within our societies, we're looking at
relative income or social position, social status -- where we are in relation to each other and the size of
the gaps between us. And as soon as you've got that idea, you should immediately wonder: what
happens if we widen the differences, or compress them, make the income differences bigger or
smaller? 

02:04

And that's what I'm going to show you. I'm not using any hypothetical data. I'm taking data from the
U.N. -- it's the same as the World Bank has -- on the scale of income differences in these rich developed
market democracies. The measure we've used, because it's easy to understand and you can download
it, is how much richer the top 20 percent than the bottom 20 percent in each country. And you see in
the more equal countries on the left -- Japan, Finland, Norway, Sweden -- the top 20 percent are about
three and a half, four times as rich as the bottom 20 percent. But on the more unequal end -- U.K.,
Portugal, USA, Singapore -- the differences are twice as big. On that measure, we are twice as unequal as
some of the other successful market democracies. 

02:51

Now I'm going to show you what that does to our societies. We collected data on problems with social
gradients, the kind of problems that are more common at the bottom of the social
ladder. Internationally comparable data on life expectancy, on kids' maths and literacy scores, on infant
mortality rates, homicide rates, proportion of the population in prison, teenage birthrates, levels of
trust, obesity, mental illness -- which in standard diagnostic classification includes drug and alcohol
addiction -- and social mobility. We put them all in one index. They're all weighted equally. Where a
country is is a sort of average score on these things. And there, you see it in relation to the measure of
inequality I've just shown you, which I shall use over and over again in the data. The more unequal
countries are doing worse on all these kinds of social problems. It's an extraordinarily close
correlation. But if you look at that same index of health and social problems in relation to GNP per
capita, gross national income, there's nothing there, no correlation anymore. 

04:03

We were a little bit worried that people might think we'd been choosing problems to suit our
argument and just manufactured this evidence, so we also did a paper in the British Medical Journal on
the UNICEF index of child well-being. It has 40 different components put together by other people. It
contains whether kids can talk to their parents, whether they have books at home, what immunization
rates are like, whether there's bullying at school. Everything goes into it. Here it is in relation to that
same measure of inequality. Kids do worse in the more unequal societies. Highly significant
relationship. But once again, if you look at that measure of child well-being, in relation to national
income per person, there's no relationship, no suggestion of a relationship. 

04:55

What all the data I've shown you so far says is the same thing. The average well-being of our societies is
not dependent any longer on national income and economic growth. That's very important in poorer
countries, but not in the rich developed world. But the differences between us and where we are in
relation to each other now matter very much. I'm going to show you some of the separate bits of our
index. Here, for instance, is trust. It's simply the proportion of the population who agree most people
can be trusted. It comes from the World Values Survey. You see, at the more unequal end, it's about 15
percent of the population who feel they can trust others. But in the more equal societies, it rises to 60 or
65 percent. And if you look at measures of involvement in community life or social capital, very similar
relationships closely related to inequality. 

05:54

I may say, we did all this work twice. We did it first on these rich, developed countries, and then as a
separate test bed, we repeated it all on the 50 American states -- asking just the same question: do the
more unequal states do worse on all these kinds of measures? So here is trust from a general social
survey of the federal government related to inequality. Very similar scatter over a similar range of levels
of trust. Same thing is going on. Basically we found that almost anything that's related to trust
internationally is related to trust amongst the 50 states in that separate test bed. We're not just talking
about a fluke. 

06:34

This is mental illness. WHO put together figures using the same diagnostic interviews on random
samples of the population to allow us to compare rates of mental illness in each society. This is the
percent of the population with any mental illness in the preceding year. And it goes from about eight
percent up to three times that -- whole societies with three times the level of mental illness of
others. And again, closely related to inequality. 
07:06

This is violence. These red dots are American states, and the blue triangles are Canadian provinces. But
look at the scale of the differences. It goes from 15 homicides per million up to 150. This is the
proportion of the population in prison. There's a about a tenfold difference there, log scale up the
side. But it goes from about 40 to 400 people in prison. That relationship is not mainly driven by more
crime. In some places, that's part of it. But most of it is about more punitive sentencing, harsher
sentencing. And the more unequal societies are more likely also to retain the death penalty. Here we
have children dropping out of high school. Again, quite big differences. Extraordinarily damaging, if
you're talking about using the talents of the population. 

08:05

This is social mobility. It's actually a measure of mobility based on income. Basically, it's asking: do rich
fathers have rich sons and poor fathers have poor sons, or is there no relationship between the
two? And at the more unequal end, fathers' income is much more important -- in the U.K., USA. And in
Scandinavian countries, fathers' income is much less important. There's more social mobility. And as we
like to say -- and I know there are a lot of Americans in the audience here -- if Americans want to live the
American dream, they should go to Denmark. 

08:46

(Laughter) 

08:48

(Applause) 

08:52

I've shown you just a few things in italics here. I could have shown a number of other problems. They're
all problems that tend to be more common at the bottom of the social gradient. But there are endless
problems with social gradients that are worse in more unequal countries -- not just a little bit worse, but
anything from twice as common to 10 times as common. Think of the expense, the human cost of that. 

09:18

I want to go back though to this graph that I showed you earlier where we put it all together to make
two points. One is that, in graph after graph, we find the countries that do worse, whatever the
outcome, seem to be the more unequal ones, and the ones that do well seem to be the Nordic countries
and Japan. So what we're looking at is general social disfunction related to inequality. It's not just one or
two things that go wrong, it's most things. 

09:47

The other really important point I want to make on this graph is that, if you look at the bottom, Sweden
and Japan, they're very different countries in all sorts of ways. The position of women, how closely they
keep to the nuclear family, are on opposite ends of the poles in terms of the rich developed world. But
another really important difference is how they get their greater equality. Sweden has huge differences
in earnings, and it narrows the gap through taxation, general welfare state, generous benefits and so
on. Japan is rather different though. It starts off with much smaller differences in earnings before tax. It
has lower taxes. It has a smaller welfare state. And in our analysis of the American states, we find rather
the same contrast. There are some states that do well through redistribution, some states that do
well because they have smaller income differences before tax. So we conclude that it doesn't much
matter how you get your greater equality, as long as you get there somehow. 

10:49

I am not talking about perfect equality, I'm talking about what exists in rich developed market
democracies. Another really surprising part of this picture is that it's not just the poor who are affected
by inequality. There seems to be some truth in John Donne's "No man is an island." And in a number of
studies, it's possible to compare how people do in more and less equal countries at each level in the
social hierarchy. This is just one example. It's infant mortality. Some Swedes very kindly classified a lot of
their infant deaths according to the British register of general socioeconomic classification. And so it's
anachronistically a classification by fathers' occupations, so single parents go on their own. But then
where it says "low social class," that's unskilled manual occupations. It goes through towards the skilled
manual occupations in the middle, then the junior non-manual, going up high to the professional
occupations -- doctors, lawyers, directors of larger companies. 

12:00

You see there that Sweden does better than Britain all the way across the social hierarchy. The biggest
differences are at the bottom of society. But even at the top, there seems to be a small benefit to being
in a more equal society. We show that on about five different sets of data covering educational
outcomes and health in the United States and internationally. And that seems to be the general picture
-- that greater equality makes most difference at the bottom, but has some benefits even at the top. 

12:33

But I should say a few words about what's going on. I think I'm looking and talking about the
psychosocial effects of inequality. More to do with feelings of superiority and inferiority, of being valued
and devalued, respected and disrespected. And of course, those feelings of the status competition that
comes out of that drives the consumerism in our society. It also leads to status insecurity. We worry
more about how we're judged and seen by others, whether we're regarded as attractive, clever, all that
kind of thing. The social-evaluative judgments increase, the fear of those social-evaluative judgments. 

13:18

Interestingly, some parallel work going on in social psychology: some people reviewed 208 different
studies in which volunteers had been invited into a psychological laboratory and had their stress
hormones, their responses to doing stressful tasks, measured. And in the review, what they were
interested in seeing is what kind of stresses most reliably raise levels of cortisol, the central stress
hormone. And the conclusion was it was tasks that included social-evaluative threat -- threats to self-
esteem or social status in which others can negatively judge your performance. Those kind of
stresses have a very particular effect on the physiology of stress. 

14:09
Now we have been criticized. Of course, there are people who dislike this stuff and people who find it
very surprising. I should tell you though that when people criticize us for picking and choosing data, we
never pick and choose data. We have an absolute rule that if our data source has data for one of the
countries we're looking at, it goes into the analysis. Our data source decides whether it's reliable
data, we don't. Otherwise that would introduce bias. 

14:39

What about other countries? There are 200 studies of health in relation to income and equality in the
academic peer-reviewed journals. This isn't confined to these countries here, hiding a very simple
demonstration. The same countries, the same measure of inequality, one problem after another. Why
don't we control for other factors? Well we've shown you that GNP per capita doesn't make any
difference. And of course, others using more sophisticated methods in the literature have controlled for
poverty and education and so on. 

15:19

What about causality? Correlation in itself doesn't prove causality. We spend a good bit of time. And
indeed, people know the causal links quite well in some of these outcomes. The big change in our
understanding of drivers of chronic health in the rich developed world is how important chronic stress
from social sources is affecting the immune system, the cardiovascular system. Or for instance, the
reason why violence becomes more common in more unequal societies is because people are sensitive
to being looked down on. 

15:55

I should say that to deal with this, we've got to deal with the post-tax things and the pre-tax
things. We've got to constrain income, the bonus culture incomes at the top. I think we must make our
bosses accountable to their employees in any way we can. I think the take-home message though is that
we can improve the real quality of human life by reducing the differences in incomes between
us. Suddenly we have a handle on the psychosocial well-being of whole societies, and that's exciting. 

16:29

Thank you. 

16:31

(Applause)