Critical Analysis and Social Care Concepts

I N T E R N A T I O N A L B E S T S E L L E R

THE SPIRIT LEVEL

------------------- M ----------------------

Why Greater Equality Makes Societies Stronger

'N, \ ■ • 4 • / • ‘A \

RICHARD WILKINSON and KATE PICKETT Foreword by Robert B. Reich

‘Might be the most important book of the year.’ G u ardian (UK)

‘Fascinating and deeply provoking . . . The Spirit L evel does contain a powerful political message. It is impossible to read it

and not to be impressed by how often greater equality appears to be the answer, whatever happens to be the question. It provides a

connection between what otherwise look like disparate social problems.’

David Runciman, L o n d o n R e v ie w o f B o o k s (UK)

‘A compass to rebuild our societies . . . A shining vision.’ Johann Hari, In dep en den t (UK)

‘A crucial contribution to the ideological argument . . . It provides a vital part of the intellectual manifesto on which

the battle for a better society can be fought.’ Roy Hattersley, N e w Statesman (UK)

‘I recognize in this book a truth that most of us know in our bones. A fair society is an essential part of our wellbeing.’

Colette Douglas Home, G la sg o w H era ld

‘Compelling and shocking. All free marketers should be made to memorize it from cover to cover.’

Yasmin Alibhai-Brown, In dep en den t (UK)

‘ An impressive body of evidence, presented in an easily digestible form . . . It raises some big questions.’

Will Kymlicka, G lo b e and M a il (Canada)

‘Brave and imaginative . . . A far-reaching analysis.’ Michael Sargent, N ature

‘M y bet to become the manifesto for the next ten years.’ Richard Gillis, Irish Times

‘A groundbreaking work and one that deserves the widest possible readership.’ Iain Ferguson, Socialist R e v ie w (UK)

‘Surprising . . . Upends the traditional debate about income inequality.’ Peter Wilson, T h e Australian

‘This is a book with a big idea, big enough to change political thinking . . . In half a page [The Spirit L evel] tells you more about the pain o f inequality than any play or novel could.’

John Carey, Sunday Tim es (UK)

‘The connection [between income inequality and dysfunctional societies] is spelt out with stark clarity in Richard Wilkinson and

Kate Pickett’s remarkable new book, T h e Spirit L e v e l. Income inequality, they show beyond any doubt, is not just bad for those

at the bottom but for everyone.’ Will Hutton, O b server (UK)

‘Richard Wilkinson and Kate Pickett put forward compelling evidence that income inequalities are at the root of a wide range

of health and social problems in society.’ Niall Crowley, Irish Tim es W eekend R e v ie w

‘The evidence, here painstakingly marshaled, is hard to dispute.’ E co n o m ist (UK)

‘ M any [N ew Statesm an] readers will be inspired as I am by a new book, The Spirit L e v e l . . . Wilkinson and Pickett compare

not only different countries, but also the 50 US states. They show that greater equality benefits not just the poor, but all

occupational groups . . . [The Spirit L e v e l has] lots of graphs but no jargon.’

Peter Wilby, N e w Statesman (UK)

‘In this fascinating sociological study, the authors do an excellent job of presenting the research, analyzing nuances, and offering

policy suggestions for creating more equal and sustainable societies. For all readers, specialized or not, with an interest in

understanding the dynamics today between economic and social conditions.’

L ib ra ry Jo u r n a l

R IC H A R D W ILKIN SO N A N D K A T E P IC K E T T

T h e Spirit Level

Why Greater Equality Makes Societies Stronger

B L O O M S B U R Y P R E S S N E W Y O R K • B E R L I N • L O N D O N • S Y D N E Y

C o p y r ig h t © 2 0 0 9 , 2 0 1 0 b y R ic h a r d W ilk in so n an d K a te P ic kett F o r e w o r d c o p y r ig h t © 2 0 1 0 b y R o b e r t B . R eich

A ll righ ts re se rv e d . N o p a r t o f th is b o o k m a y be u sed o r re p r o d u c e d in a n y m an n er

w h a ts o e v e r w ith o u t w r itte n p e rm issio n fro m the p u b lish e r e x c e p t in the c a se o f b rie f

q u o t a tio n s em bo d ied in c ritic a l artic le s o r re v ie w s. F o r in fo rm a t io n ad dress

B lo o m sb u ry P ress, 1 7 5 F ifth A v e n u e , N e w Y o r k , N Y 1 0 0 1 0 .

P u b lish e d b y B lo o m sb u r y P re ss, N e w Y o r k

A l l p a p e rs used by B lo o m sb u r y P ress are n a t u r a l, re c y c la b le p ro d u c ts m a d e fro m

w o o d g r o w n in w e ll-m a n a g e d fo re sts. T h e m a n u fa c tu rin g p ro c e sse s co n fo r m to

the e n v iro n m e n ta l r e g u la tio n s o f th e c o u n t r y o f o rigin .

L i b r a r y o f C o n g r e s s C a t a l o g i n g - I n - P u b l i c a t i o n D a t a

W ilk in s o n , R ic h a r d G . T h e sp irit level : w h y g re a te r e q u a lity m a k e s so cie ties stro n g e r /

R ic h a rd W ilk in so n an d K a te P ic ke tt.— 1 s t A m e ric a n ed.

p . cm .

In clu d es b ib lio g ra p h ic a l re fe re n ce s a n d in dex.

IS B N 9 7 8 - 1- 6 0 8 1 9 - 0 3 6 - 2 (h a rd c o v e r : a lk . p ap er)

. E q u a lity . 2 . S o c ia l m o b ility . 3. Q u a lity o f life . 4 . S o c ia l p o lic y . I. P ic ke tt, K a te . II. T itle .

H M 8 2 1 .W 5 5 2009

3 0 6 .0 1— d c2 2

2 009 03042 8

F ir st p u b lish e d in G r e a t B rit a in b y A llen L a n e ,

a d iv isio n o f the P en gu in G r o u p , in 2009

F ir st p u b lish e d in the U n ited S ta tes b y B lo o m sb u r y P ress in 2 0 1 0

T h is p a p e r b a c k e d itio n p u b lish e d in 2 0 1 1

P a p e rb a c k IS B N : 9 7 8 - 1- 6 0 8 1 9 - 3 4 1- 7

3 5 7 9 1 0 8 6 4

T y p e s e t by R o w la n d P h o to ty p e se ttin g L td , B u r y S t E d m u n d s, S u ffo lk

P rin ted in the U n ited S tates o f A m e ric a b y Q u a d / G r a p h ic s , F a ir fie ld , P e n n sy lv a n ia

F o r o u r p a ren ts D o n a n d M a r io n C h a p m a n

G e o r g e a n d M a r y G u ille m a r d

Contents

F o r e w o r d , b y R o b e r t B . R e ich ix

P re fa c e xiii

A c k n o w le d g e m e n t s x v ii

N o t e o n G r a p h s x ix

P A R T O N E

M aterial Success, Social Failure

i T h e end o f an era 3 2 P o v e rty o r in e q u a lity ? 15 3 H o w in e q u a lity gets u n d er the sk in 3i

P A R T T W O

The Costs o f Inequality

4 C o m m u n ity life an d s o cia l re la tio n s 49 5 M e n ta l h ealth an d d ru g use 63

6 P h y sic al h ealth a n d life ex p e c ta n c y 73 7 O b e sity : w id e r in co m e g a p s, w id e r w a ists 89

8 E d u c a tio n a l p e rfo rm an c e 10 3

9 T e e n a g e b irth s: re cy clin g d e p riv a tio n 1 1 9

i o V io le n c e : g a in in g resp ect 12 9

n Im p riso n m e n t an d p u n ishm en t 145 12 S o cia l m o b ility : u n eq u a l o p p o rtu n itie s 157

P A R T T H R E E

A Better Society

13 D y sfu n c tio n a l societies 17 3

14 O u r so cia l in h eritan ce 19 7

15 E q u a lity a n d su sta in a b ility 2 1 7

1 6 B u ild in g the fu tu re 235

P o s t s c r ip t - R e s e a r c h M e e ts P o litic s 273

T h e E q u a lit y T r u s t 299

A p p e n d ix 3 0 1

S o u r c e s o f D a ta f o r th e In d ic e s o f H e a lth a n d S o c ia l P r o b le m s 306

S tatistics 3 10

R e fe r e n c e s 3 1 2

I n d e x 343

Foreword

R O B E R T B. R E I C H

Professor o f Public Policy, University o f California Former U.S. Secretary o f Labor

M ost American families are worse o ff today than they were three decades ago. The Great Recession of 2 008-2009 destroyed the value of their homes, undermined their savings, and too often left them without jobs. But even before the Great Recession began, most Americans had gained little from the economic expansion that began almost three decades before. Today, the Great Recession notwithstanding, the U.S. economy is far larger than it was in 1980. But where has all the wealth gone? Mostly to the very top. The latest data shows that by 2007, America’s top 1 percent of earners received 23 percent of the nation’s total income— almost triple their 8 percent share in 1980.

This rapid trend toward inequality in America marks a significant reversal of the move toward income equality that began in the early part of the twentieth century and culminated during the middle decades of the century.

Yet inequality has not loomed large as a political issue. Even Barack Obama’s modest proposal to return income tax rates to where they stood in the 1990s prompted his 2008 Republican opponents to call him a socialist who wanted to spread the wealth. Once president, Obama’s even more modest proposal to limit the income tax deductions of the wealthy in order to pay for health care for all met fierce resistance from a Democratically controlled Congress.

If politicians have failed to grapple with the issue of inequality, few scholars have done better. Philosophers have had little to say on the subject. Some who would tax the rich to help the poor frame their arguments as utilitarian. Taking a hundred dollars from a rich person and giving it to a poor person would diminish the rich person’s happiness only slightly, they

argue, but greatly increase the happiness of the poor person. Others ground their arguments in terms of hypothetical consent. John Rawls defends redistribution on the grounds that most people would be in favor o f it if they had no idea what their income would otherwise be.

Nor have economists, whom we might expect to focus attention on such a dramatic trend, expressed much concern about widening inequality. For the most part, economists concern themselves with efficiency and growth. In fact, some of them argue that wide inequality is a necessary, if not inevitable, consequence of a growing economy. A few worry that it cuts off opportunities among the children of the poor for productive lives'—but whether to distribute wealth more equally, or what might be gained from doing so, is a topic all but ignored by today’s economic researchers.

It has taken two experts from the field of public health to deliver a major study of the effects of inequality on society. Though Richard Wilkinson and Kate Pickett are British, their research explores the United States in depth, and their work is an important contribution to the debate our country needs.

The Spirit L evel looks at the negative social effects of wide inequality— among them, more physical and mental illness not only among those at the lower ranks, but even those at the top of the scale. The authors find, not surprisingly, that where there are great disparities in wealth, there are heightened levels of social distrust. They argue convincingly that wide inequality is bad for a society, and that more equal societies tend to do better on many measures of social health and wealth.

But if wide inequality is socially dysfunctional, then why are certain countries, such as the United States, becoming so unequal? Largely because of the increasing gains to be had by being just a bit better than other competitors in a system becoming ever more competitive.

Consider executive pay. During the 1950s and ’60s, CEOs of major American companies took home about 25 to 30 times the wages of the typical worker. After the 1970s, the two pay scales diverged. In 1980, the big-company CEO took home roughly 40 times; by 1990, it was 100 times. By 2007, just before the Great Recession, CEO pay packages had ballooned to about 350 times what the typical worker earned. Recent supports suggest that the upward trajectory of executive pay, temporarily stopped by the economic meltdown, is on the verge of continuing. To make the comparison especially vivid, in 1968 the CEO of General Motors—then the largest company in the United States— took home around 66 times the pay and benefits of the typical G M worker at the time. In 2005, the CEO of Wal- Mart— by then the largest U.S. company— took home 900 times the pay and benefits of the typical Wal-Mart worker.

What explains this trajectory? Have top executives become greedier? Have corporate boards grown less responsible? Are CEOs more crooked? Are investors more docile? Is Wall Street more tractable? There’s no evidence to support any of these theories. Here’s a simpler explanation: Forty years ago, everyone’s pay in a big company— even pay at the top— was affected by bargains struck among big business, big labor, and, indirectly, government. Big companies and their unions directly negotiated pay scales for hourly workers, while white-collar workers understood that their pay grades were indirectly affected. Large corporations resembled civil service bureaucracies. Top executives in these huge companies had to maintain the good will of organized labor. They also had to maintain good relationships with public officials in order to be free to set wages and prices; to obtain regulatory permissions on fares, rates, or licenses; and to continue to secure government contracts. It would have been unseemly of them to draw very high salaries.

Since then, competition has intensified. With ever greater ease, rival companies can get access to similar low-cost suppliers from all over the world. They can streamline their operations with the same information technology their competitors use; they can cut their labor force and substitute similar software, culled from many of the same vendors. They can just as readily outsource hourly jobs abroad. They can get capital for new investment on much the same terms. They can gain access to distribution channels that are no less efficient, some of them even identical (Wal-Mart or other big- box retailers). They can attract shareholders by showing even slightly better performance, or the promise of it.

The dilemma facing so many companies is therefore how to beat rivals. Even a small advantage can make a huge difference to the bottom line. In economic terms, CEOs have become less like top bureaucrats and more like Hollywood celebrities or star athletes, who take a share of the house. Hollywood’s most popular celebrities now pull in around 15 percent of whatever the studios take in at the box office, and athletes are also getting a growing portion of sales. As the N ew Yorker’s James Surowiecki has reminded us, Mickey Mantle earned $ 60,000 in 1957. Carlos Beltran made $15 million in 2005. Even adjusting for inflation, Beltran got 40 times as much as Mantle. Clark Gable earned $ 100,000 a picture in the 1940s, which translates into roughly $ 800,000 today. Tom Hanks, by contrast, makes closer to $20 million per film. Movie studios and baseball teams find it profitable to pay these breathtaking sums because they’re still relatively small compared to the money these stars bring in and the profits they generate. Today’s big companies are paying their CEOs mammoth sums for much the same reason.

In the world of finance, the numbers are yet greater. Top investment bankers and traders take home even more than CEOs or most Hollywood stars. For the managers of twenty-six major hedge funds, the average take- home pay in 2005 was $363 million, a 45 percent increase over their average earnings the year before. The Wall Street meltdown took its toll on some of these hedge funds and their managers, but by the end of 2009 many were back.

This economic explanation for these startling levels of pay does not justify them socially or morally. It only means that in our roles as consumers and investors we implicitly think CEOs, star athletes, and Hollywood celebrities are worth it. As citizens, though, most of us disapprove. Polls continue to show that a great majority of Americans believes CEOs are overpaid, and that inequality of income and wealth is a large problem.

In short, our nation’s wealth is becoming even more concentrated at the top. It has become the financial equivalent of hydrodynamics: Large streams of income create even larger pools of wealth. The family of Wal-Mart founder Sam Walton has a combined fortune estimated to be about $90 billion. In 2005, Bill Gates was worth $46 billion; Warren Buffet, $44 billion. By contrast, the combined wealth of the bottom 40 percent of the United States population that year—some 120 million people—was estimated to be around $95 billion. Here again, the Great Recession of 2008-2009 took a toll; some of these billionaires’ fortunes were whittled down by 20 to 40 percent. But even then, they remained immense.

As citizens, we may feel that inequality on this scale cannot possibly be good for us, and Wilkinson and Pickett supply the evidence that confirms our gut sense of unease. Such inequality undermines the trust, solidarity, and mutuality on which responsibilities of citizenship depend. It creates a new aristocracy whose privileges perpetuate themselves over generations (one of the striking findings in these pages is that America now has less social mobility than many poorer countries). And it breeds cynicism among the rest of us.

This is not to say that the superrich are at fault. By and large, “ the market” is generating these outlandish results. But the market is a creation of public policies. And public policies, as the authors make clear, can reorganize the market to reverse these trends. The Spirit L evel shows why the effort to do so is a vital one for the health of our society.

Berkeley, California July 2009

Xll

Preface

People usually exaggerate the im portance o f their o w n w o rk and w e w o rr y about claim ing to o much. But this b o ok is not just another set o f nostrum s and prejudices about h o w to put the w o rld to rights. Th e w o rk w e describe here com es out o f a very long period

o f research (over fifty person-years between us) devoted, initially, to trying to understand the causes o f the big differences in life expectancy - the ‘ health inequalities’ - between people at different levels in the social hierarchy in m odern societies. The focal problem initially w a s to understand w h y health gets w orse at every step dow n the social ladder, so that the p o o r are less healthy than those in the m iddle, w h o in turn are less healthy than those further up.

Lik e others w h o w o rk on the social determ inants o f health, our training in epidem iology means that our methods are those used to trace the causes o f diseases in populations - trying to find out w h y one group o f people gets a particu lar disease w hile another group doesn ’t, or to explain w h y some disease is becom ing m ore com m on.

The same methods can, h ow ever, also be used to understand the causes o f other kinds o f problem s - not just health.

Ju s t as the term ‘ evidence-based m edicine’ is used to describe current efforts to ensure that m edical treatm ent is based on the best scientific evidence o f w h at w o rk s and w h at does not, w e thought o f calling this b o ok ‘Evidence-based Politics’ . The research which underpins w h at w e describe com es from a great m any research team s in different universities and research organizations. R eplicable methods have been used to study observable and objective outcom es, and peer-reviewed research reports have been published in academ ic,

scientific journals.

This does not m ean that there is no gu essw ork. Results alw ays have to be interpreted, but there are usually good reasons fo r fav o u r­ ing one interpretation over another. Initial theories and expectations are often called into question by later research findings w hich m ake it necessary to think again. We w ou ld like to take yo u on the journey w e have travelled, signposted by crucial bits o f evidence and leaving

out only the vario u s culs-de-sac and w ron g turnings that w asted so much time, to arrive at a better understanding o f h ow w e believe it is possible to im prove the qu ality o f life fo r everyone in m odern societies. W e shall set out the evidence and our reasons fo r inter­ preting it the w a y w e d o , so that yo u can judge fo r yourself.

A t an intuitive level people have alw ay s recognized that inequality is socially corrosive. But there seemed little reason to think that levels o f inequality in developed societies differed enough to expect any m easurable effects. T h e reasons w hich first led one o f us to lo o k fo r effects seem n o w largely irrelevant to the striking picture w hich has emerged. M a n y discoveries ow e as m uch to luck as judgement.

The reason w h y the picture w e present has not been put together until n o w is pro b ab ly that much o f the data has only becom e a v a il­ able in recent years. W ith internationally com parable inform ation not only on incomes and income distribution but also on different health and social problem s, it could only have been a matter o f time before som eone cam e up w ith findings like ours. Th e em erging data

have allow ed us, and other researchers, to analyse h ow societies differ, to discover h ow one factor is related to another, and to test theories m ore rigorously.

It is easy to imagine that discoveries are m ore rapidly accepted in the natural than in the social sciences - as if physical theories are som ehow less con troversial than theories abou t the social w o rld . But the history o f the natural sciences is littered w ith painful personal disputes, w hich started o ff as theoretical disagreem ents but often lasted fo r the rest o f peop le’s lives. C ontroversies in the natural sciences are usually confined to the experts: m ost people do not have strong view s on rival theories in particle physics. But they do have view s on h o w society w o rk s. Social theories are partly theories about ourselves; indeed, they m ight alm ost be regarded as p art o f ou r self­ aw areness or self-consciousness o f societies. W hile natural scientists

do not have to convince individual cells or atom s to accept their

theories, social theorists are up against a plethora o f individual views

and po w erfu l vested interests. In 1847, Ignaz Sem m elweiss discovered that if doctors w ashed

their hands before attending w om en in childbirth it dram atically reduced deaths from puerperal fever. But before his w o rk could have

much benefit he had to persuade people - p rin cip ally his medical colleagues - to change their behaviour. H is real battle w a s not his initial discovery but w h at fo llo w ed from it. H is view s w ere ridiculed and he w a s driven eventually to insanity and suicide. M u ch o f the m edical profession did not take his w o rk seriously until Louis

Pasteur and Jo sep h Lister had developed the germ theory o f disease,

w hich explain ed w h y hygiene w as im portant. We live in a pessim istic period. A s w ell as being w o rried by the

likely consequences o f global w arm in g, it is easy to feel that m any societies are, despite their m aterial success, increasingly burdened b y their social failings. A n d n o w , as if to add to our w o es, w e have

the econom ic recession and its afterm ath o f high unem ploym ent. But the know ledge that w e cannot carry on as w e have, that change is necessary, is perhaps grounds fo r optim ism : m aybe w e do, at last, have the chance to m ake a better w o rld . The extrao rd in arily positive

reception o f the h ard back editon o f this b ook confirm s that there is a

w idespread appetite fo r change and a desire to find positive solutions

to our problem s. W e have m ade only m inor changes to this edition. D etails o f the

statistical sources, m ethods and results, from w hich w e thought most readers w o u ld w a n t to be spared, are n o w provided in an appendix fo r those w ith a taste fo r data. C hap ter 13 , w hich is substantially a bo u t causation , has been slightly reorganized and strengthened. W e have also expan ded our discussion o f w h at has m ade societies substantially m ore or less equal in the past. Because w e conclude that these changes have been driven by changes in political attitudes, w e think it is a m istake to discuss po licy as if it w ere a m atter o f finding the right technical fix. A s there are really hundreds o f w a y s that societies can becom e m ore equal if they choose to, w e have not nailed our colou rs to one or other set o f policies. W hat w e need is not so much a clever solution as a society w hich recognizes the

benefits o f greater equality. I f correct, the theory and evidence set out in this b ook tells us h o w to m ake substantial im provem ents

in the q u ality o f life fo r the vast m ajo rity o f the popu lation . Y e t unless it is possible to change the w a y m ost people see the societies they live in, the theory w ill be stillborn. Public opinion w ill only support the necessary political changes if som ething like the perspec­ tive w e outline in this b o ok perm eates the public mind. W e have therefore set up a not-for-profit organization called The E q u ality Tru st (described at the end o f this book) to m ake the kind o f evidence set out in the fo llo w in g pages better k n o w n and to suggest that there is a w a y out o f the w o o d s fo r us all.

Acknowledgements

W e are grateful to D an n y D orlin g, Stuart Proffitt and A liso n Q uick fo r their careful reading and m any helpful com m ents on ou r m anu­

script. We also thank M o lly Scott C ato fo r her com ments on

C hapter 15 , M a jid Ezzati fo r kin dly sending us his corrected esti­ mates o f body m ass in d ex fo r US states, and Stephen Bezruchka for helpful discussions.

W e are also grateful to the Jo sep h R o w n tree C haritable T rust,

especially Stephen Pittam , fo r supporting our efforts to disseminate

this research via T h e E q u ality T ru st; and K ath ryn B usb y and Bill K erry w hose h ard w o rk has m ade it a going concern.

R ich ard W ilkinson w ou ld like to thank the U niversity o f N o ttin g ­ ham and his form er colleagues in the D ivision o f E pidem iology and

Public H ealth fo r the freedom w hich allow ed him to devote his time

to the research w hich went into this book. K ate Pickett thanks the U niversity o f Y o r k and her colleagues fo r their generous support.

Figures 3 .1 and 3.2 are reproduced by kind perm ission o f Je an Tw en ge. W e are grateful to C am bridge U niversity Press fo r perm is­ sion to reproduce Figures 4.3 and 10 .1 . Figure 6. 1 is reproduced with perm ission from B M J Publishing group; Figure 6.7 w ith perm ission from B ryan C hristie D esign, L L C ; and Figure 15 .3 w ith perm ission from the E c o n o m ic Jo u r n a l. A ll other Figures are our o w n , and can be freely reproduced w ith acknow ledgem ent.

T h e c a rto o n s on pp. 16 , 32, 64, 10 4 , 1 1 8 , 1 3 0 , 14 6 , 15 8 , 17 2 , 19 4 , 2 1 4 an d 230 are fro m w w w .C a r t o o n S t o c k .c o m . G ra te fu l a c k n o w le d g e m e n t is giv en to them an d to the fo llo w in g fo r p e r­ m issio n to re p ro d u ce c a r to o n s : p. 2, c o p y r ig h t © A n d y S in ger, 2007, p o litic a lc a r to o n s .c o m ; p . 48, c o p y rig h t © T h e N e w Y o r k e r

xvu

c o lle c tio n , 1996, Peter Steiner, carto o n b an k .co m ; p. 74, copyright © Jo sep h F arris, carto o n b an k .co m ; p .88, copyright © T h e N e w Y o rk e r collection, 2005, Lee Loren z, cartoon b an k.com .

Note on Graphs

F A C T S F R O M F I G U R E S : H O W T O L O O K A T T H E G R A P H S I N T H I S B O O K

M o st o f the graphs that w e use in this b ook are charts linking income inequality to different health and social problem s. T h ey show relationships, either: (i) internationally, com paring rich countries or (2) in the U SA , com paring different states.

In all o f these graph s, w e put incom e inequality along the horizon­ tal line at the bottom (the x -a x is), so societies w ith lo w levels o f inequality are to the left, and societies w ith high levels o f inequality are to w ard s the right o f the graph.

The different health and social outcom es are show n on the vertical line (the y-axis) on the left side o f the graph.

O n m ost o f the graphs, there are tw o features. First there is a scatter o f points, either o f rich countries, or o f US states, so that readers can see exactly h ow each society com pares to others. Second,

there is a line, called a regression lin e, w hich show s the ‘ best fit’ relationship between income inequality and the outcom e on that graph. This line is not chosen by us, but is calculated by statistical softw are to give the line w hich best fits the trend through the data points. It is also possible to calculate h ow unlikely it is that the pattern w e see could result from chance alone. W e have only included a best fit line through the points if the relationship w ou ld be very unlikely to occur by chance. W hen a graph has no best fit line it means that there is no evidence o f a relationship.

If the line slopes steeply u pw ard s from left to right, it show s that the health or social outcom e becomes m ore com m on in more

unequal societies. This pattern tends to occur w ith problem s that we think o f as bad, such as violence:

y-axis

x - a x is

I f the line slopes steeply do w n w ard s from left to right, it show s that the health or social outcom e is much less com m on in more unequal societies. W e see this pattern fo r things that w e think o f as good , such as trust:

y-a x is

x -a x is

A w id er scatter o f points on the graph m eans that there are other im portant influences on the outcom e. It m ay not m ean that inequality is not a p o w erfu l influence, sim ply that other factors matter as w ell:

y-axis

x-axis

A n a rro w scattering o f points m eans that there is a very close

relationship between inequality and the outcom e and that inequality is an excellent predictor o f the outcome:

y -axis

x - a x is

Further details o f our methods can be found at: w w w .

equalitytrust.org.uk

o

P A R T O N E

Material Success, Social Failure

The end of an era

I care for riches, to make gifts to friends, or lead a sick man back to health with ease and plenty. Else small aid

is wealth for daily gladness; once a man be done with hunger, rich and poor are all as one. Euripides, Electra

It is a rem arkab le p a ra d o x that, at the pinnacle o f hum an m aterial and technical achievem ent, w e find ourselves an xiety-rid den , prone

to depression, w o rried abo u t h o w others see us, unsure o f our friendships, driven to consum e an d w ith little or no com m unity life. L a c k in g the relaxed social con tact and em otional satisfactio n w e all need, w e seek co m fo rt in over-eating, obsessive shopping and spend­

ing, o r becom e prey to excessive alco h o l, psych oactive medicines and

illegal drugs. H o w is it that w e have created so m uch m ental an d em otional

sufferin g despite levels o f w ealth and co m fo rt unprecedented in hum an history? O ften w h a t w e feel is m issing is little m ore than time enjoying the co m pan y o f friends, yet even that can seem beyond

us. W e talk as if ou r lives w ere a constant battle fo r psych ological su rv iv a l, struggling again st stress and em otion al exh au stio n , but the truth is that the lu x u ry and extravagan ce o f our lives is so great that

it threatens the planet. R esearch from the H a rw o o d Institute fo r Public Innovation

(com m issioned by the M erc k F am ily Foundation) in the U S A show s that people feel that ‘m aterialism ’ so m eh o w com es betw een them and the satisfaction o f their social needs. A report entitled Y e arn in g f o r B a la n c e , based on a n ation w id e su rvey o f A m erican s, concluded

that they w ere ‘ deeply am bivalen t abo u t w ealth and m aterial ga in ’ . 1 * A large m ajo rity o f people w an ted society to ‘m ove a w a y fro m greed an d excess to w a rd a w a y o f life m ore centred on valu es, com m unity, and fa m ily ’ . But they also felt that these priorities w ere not shared by m ost o f their fello w A m erican s, w h o , they believed, had becom e ‘in creasin gly atom ized, selfish, and irrespon sible’ . A s a result they often felt isolated. H o w eve r, the report says, that w hen brought together in focu s grou ps to discuss these issues, peop le w ere ‘ su r­ prised and excited to find that others share[d] their v ie w s’ . R ath er than uniting us w ith others in a com m on cause, the unease w e feel

abo u t the loss o f social values and the w a y w e are d raw n into the

pursuit o f m aterial gain is often experienced as if it w ere a purely private am bivalence w h ich cuts us o ff from others.

M ain stream politics no longer taps into these issues and has aban don ed the attem pt to provid e a shared vision capable o f in sp ir­

ing us to create a better society. A s vo ters, w e have lost sight o f an y collective belief that society could be different. Instead o f a better society, the on ly thing alm ost everyone strives fo r is to better their o w n po sitio n - as in d ividu als - w ithin the existin g society.

The con trast betw een the m aterial success an d social failu re o f m any rich countries is an im portan t signpost. It suggests that, if we

are to gain further im provem ents in the real q u ality o f life, w e need to shift attention fro m m aterial standards and econom ic g ro w th to w a y s o f im p rovin g the psych o lo gical and social w ellbein g o f w hole societies. H o w eve r, as soon as an yth in g p sych o lo g ical is m entioned,

discussion tends to fo cu s alm ost exclu sively on individual remedies

and treatm ents. Political thinking seems to run into the sand. It is n o w possible to piece together a new , com pelling and

coherent picture o f h o w w e can release societies from the grip o f so much d ysfu nction al beh aviou r. A pro p er understanding o f w h a t is

goin g on cou ld tran sfo rm politics and the q u ality o f life fo r all o f us.

It w o u ld change ou r experience o f the w o rld aroun d us, change w h at w e vote fo r, and change w h a t w e dem and from our politicians.

In this b o ok w e sh o w that the q u ality o f so cial relations in a society is built on m aterial foun d ation s. T h e scale o f incom e d iffer­

*Superscrip ts refer to num bered references listed at the end o f the book.

ences has a p o w erfu l effect on h o w w e relate to each other. R ather than blam ing parents, religion, valu es, education or the penal system,

w e w ill sh o w that the scale o f inequality p rovides a p o w erfu l policy

lever on the psych o lo gical w ellbein g o f all o f us. Ju s t as it once to o k studies o f w eigh t gain in babies to sh o w that interacting with a lo vin g care-giver is crucial to child developm ent, so it has taken studies o f death rates and o f incom e distribution to sh o w the social

needs o f adults and to dem onstrate h ow societies can meet them.

L o n g before the financial crisis w hich gathered pace in the later part o f z o o 8 , B ritish politician s com m enting on the decline o f com m unity or the rise o f vario u s form s o f an ti-social b eh aviour, w o u ld sometimes refer to our ‘ broken society’ . Th e financial collapse shifted attention to the broken econom y, and w h ile the broken

society w a s sometimes blam ed on the b eh aviou r o f the p o o r, the

broken econ om y w a s w id ely attributed to the rich. Stim ulated by the prospects o f ever bigger salaries and bonuses, those in charge o f some o f the m ost trusted financial institutions th rew cau tion to the w in d and built houses o f card s w h ich could stand o n ly w ithin

the protection o f a thin speculative bubble. But the truth is that

both the broken society and the broken econ om y resulted from the gro w th o f inequality.

W H E R E T H E E V I D E N C E L E A D S

W e shall start by outlining the evidence w h ich show s that w e have got close to the end o f w h a t econom ic gro w th can do fo r us. For thousands o f years the best w a y o f im p rovin g the qu ality o f hum an life w a s to raise m aterial living standards. W hen the w o lf w a s never fa r from the d o o r, go o d times w ere sim ply times o f plenty. But for the vast m ajo rity o f people in affluent countries the difficulties o f life are no longer abou t filling ou r stom achs, h avin g clean w a ter and keeping w arm . M o st o f us n o w w ish w e cou ld eat less rather than m ore. A n d , fo r the first time in history, the p o o r are - on average - fatter than the rich. E con om ic gro w th , fo r so long the great engine

o f p rogress, has, in the rich countries, largely finished its w o r k . N o t o n ly have m easures o f w ellbein g and happiness ceased to rise w ith

econom ic gro w th but, as affluent societies have gro w n richer, there have been long-term rises in rates o f a n xiety, depression and num er­ ous other social problem s. T h e p o pu lation s o f rich countries have got to the end o f a long h istorical journey.

The course o f the jo u rn ey w e have m ade can be seen in Figure i . i .

It show s the trends in life expectan cy in relation to G ro ss N a tio n a l

Incom e per head in countries at vario u s stages o f econom ic develop­ ment. A m o n g po o rer countries, life expectan cy increases rap id ly during the early stages o f econom ic developm ent, but then, starting am ong the m iddle-incom e countries, the rate o f im provem ent slow s

do w n . A s living standards rise an d countries get richer an d richer,

the relationship betw een econom ic gro w th and life expectan cy w eaken s. E ven tu ally it disappears entirely and the rising cu rve in Figure i . i becom es h orizon tal - sh o w in g that fo r rich countries to get richer adds nothing further to their life expectan cy. T h a t has already happened in the richest thirty or so countries - nearest the

top right-hand corner o f Figure i . i .

T h e reason w h y the curve in Figure i . i levels out is not because w e have reached the lim its o f life expectan cy. Even the richest co u n ­ tries go on en joyin g substantial im provem ents in health as time goes by. W h at has changed is that the im provem ents have ceased to be

related to average living standards. W ith every ten years that passes,

life expectan cy am on g the rich countries increases by between tw o and three years. This happens regardless o f econom ic g ro w th , so that a co u n try as rich as the U S A no longer does better than Greece or N e w Z e a la n d , although they are not m uch m ore than h a lf as

rich. R ath er than m ovin g out a lo n g the curve in Figure i . i , w h at

happens as time goes by is that the curve shifts u pw ard s: the sam e

levels o f incom e are associated w ith higher life expectan cy. L o o k in g at the data, you can n ot help but conclude that as countries get richer, further increases in average living standards do less and less fo r health.

W hile go o d health and longevity are im portan t, there are other com ponents o f the q u ality o f life. B ut just as the relationship between health and econom ic g ro w th has levelled o ff, so too has the relation ship w ith happiness. L ik e health, h o w h ap p y people are rises in the early stages o f econom ic gro w th and then levels o ff. This

Li fe

ex

pe ct

an cy

(y

ea rs

)

70-

60 -

50

40-

Australia Iceland Spain Sweden Canada Switzerland

Israel Italy France Austria „ Sineaoore . ■

Chile

Albania

Malta

Portugal

Barbados

Costa Rica Cuba

U rug u ay Belize Mexico Poland _

OmanPanama v^roatia ,

M o nte n e g ro ' J j 8 S £ l o ™ , ^ “ "s” vak,a

S erb ia V«ne»ueU c h m a Nicaragua J amalca El Salvador Romania

r , n r o i ^ gyP t Philippines TurkeyGeorgia Morocco ™ j TnHnnpQia 1 * 1 reru

New Zealand G i

Slovenia Czech Republic

Bahrain

___ ______ a l<" ’ rrance Austria Cyprus Greece Germany FinlancP1̂ Netherlands

UAE , Ireland Kuwait Belgium ™ Denmark

Brunei

Qatar

----- o— morocco Iraq Indonesia Honduras Thailand

Moldova Fiji Belarus Grenada Uzbekistan Micronesia

N. Korea A*erbaiian Ukraine Mongolia a£ “ tan Maldives Kazakhstan

a " Bolivia Russia Comoros Laos jncjja Guyana

Bangladesh Mauritania Senegal Nepal

Yemen T urkm enistan H a iti

M a d a g a s c a r Gh G a m b ia Togo

I ra a c j Cambodia lracl Sudan .E ritre a Papua N ew G uineaKT- Gabon B enin N « ' r

Guinea Congo Djibouti

Mali

Hungary

Latvia Saudi Arabia Estonia Bahamas

Trinidad & Tobago

Ethiopia £ en7.a _ Tanzania Burk,n“ Fas°

Chad Uganda Cameroon

Burundi Cote d’Ivoire

Somalia Malawi Nigeria

Congo-DR Guinea-Bissau Liberia Rwanc â

C A R Lesotho Afghanistan

Mozambique Sierra Leone Zim babwe . . Swaziland

Angola

_____ Zam bia______________________

Namibia South Africa

Equatorial Guinea

Botswana

I O O O O 2 0 0 0 0 3 O O O O

National income per person ($)

4 0 0 0 0

Figure i . i Only in its early stages does economic developm ent boost life expectancy.1

Norway

USA

is a point m ade stron gly by the econom ist, R ic h a rd L a y a rd , in his

b ook on happ iness.3 Figures on happiness in different countries are p ro b ab ly strongly affected by culture. In som e societies not saying you are h ap p y m ay sound like an adm ission o f failu re, w h ile in another claim ing to be h ap p y m ay sound self-satisfied and smug.

But, despite the difficulties, Figure 1 . 1 show s the ‘happiness cu rv e’

levelling o f f in the richest countries in m uch the sam e w a y as life expectancy. In both cases the im portan t gains are m ade in the earlier stages o f econom ic g ro w th , but the richer a cou n try gets, the less getting still richer adds to the p o p u latio n ’ s happiness. In these

graphs the curves fo r both happiness and life expectan cy flatten o ff

at aroun d $ 2 5 ,0 0 0 per capita, but there is som e evidence that the

incom e level at w hich this occurs m ay rise over tim e.4 The evidence that happiness levels fail to rise further as rich

countries get still richer does not com e only from com parison s o f d ifferent countries at a single p oin t in time (as show n in Figure 1.2 ) .

In a fe w countries, such as Ja p a n , the U S A and B ritain , it is possible

to lo o k at changes in happiness o ver sufficiently long periods o f time to see w hether they rise as a cou n try gets richer. T h e evidence show s that happiness has not increased even over periods long enough for real incom es to have doubled. T h e sam e pattern has also been found

by researchers using other indicators o f w ellbein g - such as the

‘ m easure o f econom ic w e lfa re ’ or the ‘genuine progress in d icator’ ,

w hich try to calculate net benefits o f gro w th after rem oving costs like traffic congestion and pollution.

So w hether w e lo o k at health, happiness or other m easures o f w ellbein g there is a consistent picture. In p o o rer countries, econom ic developm ent continues to be very im portan t fo r hum an w ellbeing.

Increases in their m aterial living standards result in substantial im provem ents both in objective m easures o f w ellbeing like life expectan cy, an d in subjective ones like happiness. But as nations join the ran ks o f the affluent developed countries, further rises in incom e count fo r less and less.

This is a predictable pattern. A s you get m ore and m ore o f an y­ thing, each addition to w h a t you h ave - w hether loaves o f bread or cars - contributes less and less to y o u r w ellbeing. I f you are hungry, a lo a f o f bread is everything, but w hen y o u r hunger is satisfied, m any

Pe r

ce nt

‘v

er y

h ap

p y’

or ‘q

ui te

h

ap p

y’

8 0 -

Nigeria

Philippines • Venezuela*

Algeria •C olom bia Jo rd a n * * • Brazil

Morocco Bosnia

• Uganda Azerbaijan _ . . . • China • Bangladesh

Pakistan*

Uruguay

South Africa

• Argentina

Croatia

• Poland

• Lithuaina

• Macedonia • Dominican Republic

Turkey •

• S. Korea

• Czech Republic

Slovenia*

Greece*

Netherlands Canada • A u s tr a lia * ‘ . . S w i„ „ land

Sweden* Denmark • Belgium

Spain*

Germany*

Jap an * Finland

60 -

Armenia * • Zimbabwe

• Moldova Russian Federation •

Ukraine •

40- --1--------------------------- 1--------------------------- 1-----------------------

2 0 0 0 0 3 0 0 0 0

National income per person ($)

Ireland Norway

United States

4 0 0 0 0

Figure i . z Happiness an d average incom es (data for U K unavailable).5

m ore lo aves do n ’t p a rticu la rly help you and m ight becom e a nuisance as they go stale.

Sooner o r later in the long h istory o f econom ic g ro w th , countries

in evitab ly reach a level o f affluence w h ere ‘ dim inishing return s’ set in

and add itio n al incom e buys less and less add itio n al health, h ap p i­ ness o r w ellbeing. A num ber o f developed countries have n o w had alm ost continuous rises in average incom es fo r o ver 1 5 0 years and add ition al w ealth is n ot as beneficial as it once w as.

T h e trends in different causes o f death confirm this interpretation. It is the diseases o f p o verty w h ich first decline as countries start

to get richer. T h e great infectious diseases - such as tuberculosis, ch olera or m easles - w h ich are still com m on in the p o o rest countries to d ay, g ra d u a lly cease to be the m ost im p ortan t causes o f death. A s they disappear, w e are left w ith the so-called diseases o f affluence - the degenerative ca rd io v ascu la r diseases an d cancers. W hile the

infectious diseases o f po verty are p a rticu la rly com m on in childhood

and frequ en tly kill even in the prim e o f life, the diseases o f affluence are very largely diseases o f later life.

O ne other piece o f evidence confirm s that the reason w h y the curves in Figures 1 . 1 and 1 . 2 level o f f is because countries have

reached a threshold o f m aterial livin g standards after w h ich the

benefits o f further econom ic gro w th are less substantial. It is that the diseases w hich used to be called the ‘ diseases o f affluence’ becam e the diseases o f the p o o r in affluent societies. D iseases like heart disease, stroke and obesity used to be m ore com m on am ong

the rich. H e a rt disease w a s regarded as a b usinessm an’ s disease and

it used to be the rich w h o w ere fat and the p o o r w h o w ere thin. B ut from abo u t the 1 9 5 0 s o n w a rd s, in one developed co u n try after another, these patterns reversed. D iseases w hich had been m ost com m on am on g the better-off in each society reversed their social distribution to becom e m ore com m on am on g the poor.

T H E E N V I R O N M E N T A L L I M I T S T O G R O W T H

A t the same tim e as the rich countries reach the end o f the real benefits o f econom ic gro w th , w e have also had to recognize the problem s o f g lo b al w arm in g and the environm ental limits to grow th. T h e dram atic reductions in carb o n em issions needed to prevent

ru n a w a y clim ate change an d rises in sea levels m ay mean that even

present levels o f consum ption are unsustainable - p articu larly if living standards in the p o o rer, developing, w o rld are to rise as they need to. In C h ap ter 1 5 w e shall discuss the w a y s in w h ich the per­ spective outlined in this b ook fits in w ith policies designed to reduce

g lo b al w arm in g.

I N C O M E D I F F E R E N C E S W I T H I N A N D B E T W E E N S O C I E T I E S

W e are the first generation to have to find new answ ers to the question o f h o w w e can m ake further im provem ents to the real qu ality o f hum an life. W h at should w e turn to if not to econom ic gro w th? O ne o f the m ost p o w erfu l clues to the an sw er to this question com es fro m the fact that w e are affected very differently by the incom e differences w ith in ou r ow n society from the w a y w e are affected by the differences in average incom e b e tw e e n one rich

society and another. In C hapters 4 - 1 2 w e focus on a series o f health and social

problem s like violence, mental illness, teenage births and educational failu re, w hich w ithin each cou n try are all m ore com m on am ong the p o o r than the rich. A s a result, it often lo o k s as if the effect o f higher incom es and living standards is to lift people out o f these problem s. H o w ev e r, w hen w e m ake com parison s betw een different societies, w e find that these social problem s have little or no relation to levels o f a v era g e incom es in a society.

T a k e health as an exam ple. Instead o f lo o kin g at life expectan cy across both rich and p o o r countries as in Figure 1 . 1 , lo o k just at

the richest countries. Figure 1 . 3 show s just the rich countries and

confirm s that am ong them some countries can be alm ost tw ice as

rich as others w ith o u t an y benefit to life exp ectan cy. Y e t w ith in an y o f them death rates are clo sely and system atically related to incom e. Figure 1 . 4 show s the relation between death rates and incom e levels w ithin the U S A . Th e death rates are fo r people in

zip code areas classified by the typical household incom e o f the area

in w hich they live. O n the right are the richer zip code areas w ith lo w er death rates, an d on the left are the p o o rer ones w ith higher death rates. A lth o u gh w e use A m erican data to illustrate this, sim ilar health gradients, o f v a ry in g steepness, run across alm ost

every society. H igher incom es are related to lo w er death rates at

every level in society. N o te that this is not sim ply a m atter o f the p o o r h avin g w orse health than everyone else. W h at is so striking abou t Figure 1 . 4 is h o w regu lar the health gradient is right across society - it is a gradient w h ich affects us all.

W ithin each cou n try, peop le’ s health and happiness are related to

82

e0 5 1 8 0

78 ~

76

•Ja p a n

• Sweden

• Israel , S Pain Australia* * Cana<la „ France • • Switzerland

Ita ly . • Belgium

• Austria

• Norway

Greece • • New Zealand Germany % • Netherlands • Singapore

Ireland* • USA

• Denmark

• Portugal

25000 3OOOO 35000 N atio n al incom e per person ($)

40000

Figure 1.3 L ife expectancy is unrelated to differences in average income between rich countries.6

Poor Rich

Average income o f US zip codes

Figure 1.4 Death rates are closely related to differences in income within societiesP

their incom es. R ich er people tend, on average, to be healthier and happier than po o rer people in the sam e society. But co m parin g rich countries it m akes no difference w hether on average people in one society are alm ost tw ice as rich as people in another.

W hat sense can w e m ake o f this p a ra d o x - that differences in

average incom e or living standards between w h ole popu lation s or countries d o n ’t m atter at all, but incom e differences w ithin those sam e popu lation s m atter very m uch indeed? T here are tw o plausible exp lan ation s. O ne is that w h a t m atters in rich countries m ay not be y o u r actu al incom e level and living stan d ard, but h o w yo u com pare

w ith other people in the sam e society. Perhaps average standards d o n ’t m atter and w h a t does is sim ply w hether you are doing better or w o rse than other people - w here yo u com e in the social pecking order.

T h e other p ossib ility is that the social gradient in health show n in

Figure 1 .4 results not from the effects o f relative incom e or social status on health, but from the effects o f social m obility, sortin g the healthy from the unhealthy. Perhaps the healthy tend to m ove up the social ladder and the unhealthy end up at the bottom .

T h is issue w ill be resolved in the n ext chapter. W e shall see

w hether com pressing, o r stretching out, the incom e differences in a

society m atters. D o m ore and less equal societies suffer the same

o verall burden o f health and social problem s?

Poverty or inequality?

Poverty is not a certain small amount of goods, nor is it just a relation between means and ends; above all it is a

relation between people. Poverty is a social statu s. . . It has grown . . . as an invidious distinction between classes . . .

M arshall Sahlins, Stone Age Economics

H O W M U C H I N E Q U A L I T Y ?

In the last chapter w e saw that econom ic gro w th and increases in average incom es have ceased to contribute m uch to w ellbein g in the rich countries. But w e also saw that w ithin societies health and social problem s rem ain strongly associated w ith incom es. In this chapter

w e w ill see w hether the am ount o f incom e inequality in a society

m akes a n y difference. Figure 2 . 1 show s h ow the size o f incom e differences varies from

one developed country to another. A t the top are the m ost equal countries and at the bottom are the most unequal. The length o f the h orizon tal bars show s h ow m uch richer the richest 20 per cent o f

the popu lation is in each cou n try com pared to the p oorest 20 per cent. W ithin countries such as Ja p a n an d some o f the Scandinavian countries at the top o f the chart, the richest 20 per cent are less than fo u r times as rich as the poorest 20 per cent. A t the bottom o f the chart are countries in w h ich these differences are at least tw ice as big, including tw o in w hich the richest 20 per cent get abo u t nine times as much as the poorest. A m o n g the m ost unequal are Sin gapore, U S A , P ortugal and the United K in gd om . (The figures are

Miss Smith, buy up the rights to the Bible an d g e t that p a rt changed ab out the rich m an a n d the eye o f the needle

Ja p a n ..................... ' " 1

Finland 1

N o rw a y Sweden ' | ™

Denm ark 1 Belgium 1 Austria ......................................................................................................... 1

G erm any N etherlands

......................................................................................................................................................................... " 1

Spain M France

C an ad a Switzerland

Ireland Greece .....................- 1

Italy . 1 Israel 1

N e w Z eala n d A u stralia .................. ........... 1

U K - I 1 1 1 — l a a a i Portugal ....................... _ J

U SA 1 Singapore " " .................... ......... ■ . ' " ■/ ' . |

' • ' i : -------------------------------

o 2 4 6 8 10

Incom e gap

Figure z .i H o w much richer are the richest 20 per cent than the poorest zo per cent in each country?1

fo r household incom e, after taxes and benefits, adjusted fo r the num ber o f people in each household.)

Th ere are lots o f w a y s o f m easuring incom e in equality and they are all so closely related to each other that it doesn ’ t u sually m ake much difference w hich yo u use. Instead o f the top and bottom z o per cent, w e cou ld com pare the top and bottom 1 0 or 3 0 per cent. O r w e could h ave looked at the p ro po rtion o f all incom es w h ich go to the poorer h a lf o f the p o pu lation . T y p ic a lly , the poorest h a lf o f the popu lation get som ething like zo o r 25 per cent o f all incom es and the richest h a lf get the rem aining 7 5 or 80 per cent. O ther m ore sophisticated m easures include one called the G in i coefficient. It m easures in ­ eq u ality across the w h o le society rather than sim ply co m parin g the

extrem es. I f all incom e w ent to one person (m axim um inequality) an d everyone else go t nothing, the G in i coefficient w o u ld be equal

to i . If incom e w a s shared eq u ally and everyone go t exactly the same (perfect equality), the G in i w o u ld equal o. T h e lo w er its valu e, the m ore equal a society is. T h e m ost com m on values tend to be between 0 .3 and 0 .5 . A n oth er m easure o f inequality is called the R o b in H o o d In dex because it tells yo u w h a t p ro p o rtion o f a so ciety’s incom e

w o u ld have to be taken from the rich and given to the p o o r to get com plete equality.

T o av o id being accused o f pickin g and ch oosin g ou r m easures, our ap p ro ach in this b o o k has been to take m easures provid ed by official agencies rather than calcu latin g ou r o w n . W e use the ratio o f

the incom e received by the top to the bottom 20 per cent w h enever

w e are co m parin g in equality in different countries: it is easy to understand and it is one o f the m easures pro vid ed ready-m ade by the United N a tio n s. W hen co m parin g in equality in U S states, w e use the G in i coefficient: it is the m ost com m on m easure, it is favo u red by

econom ists an d it is a v ailab le fro m the U S C ensus B ureau. In m any

academ ic research papers w e and others have used tw o different inequality m easures in o rder to sh o w that the choice o f m easures rarely has a significant effect on results.

D O E S T H E A M O U N T O F I N E Q U A L I T Y M A K E A D I F F E R E N C E ?

H a vin g got to the end o f w h a t econom ic gro w th can do fo r the q u ality o f life and facin g the problem s o f environm ental dam age, w h a t difference do the inequalities show n in Figure 2 . 1 m ake?

It has been k n o w n fo r som e years that p o o r health and violence are m ore com m on in m ore unequal societies. H o w ev e r, in the course o f ou r research w e becam e a w a re that alm ost all problem s w h ich are m ore com m on at the b ottom o f the social ladder are m ore com m on in m ore unequal societies. It is not just ill-health and violence, but also, as w e w ill sh o w in later chapters, a host o f other social p ro b ­ lems. A lm o st all o f them contribute to the w id esp read concern that m odern societies are, despite their affluence, social failures.

T o see w hether these problem s w ere m ore com m on in m ore unequal countries, w e collected in tern ation ally co m parab le data on

health and as m an y social problem s as w e could find reliable figures fo r. Th e list w e ended up w ith included:

• level o f trust • mental illness (including drug and alcohol addiction) • life expectan cy and infant m ortality • obesity • children’s educational perform ance • teenage births • hom icides • im prisonm ent rates • social m obility (not availab le fo r US states)

O ccasio n ally w h a t ap p ear to be relationships betw een different things m ay arise sp u riou sly or by chance. In o rder to be confident

that ou r findings w ere sound w e also collected data fo r the same health and social problem s - or as near as w e could get to the same - fo r each o f the fifty states o f the U S A . T h is a llo w ed us to check w hether or not problem s w ere consistently related to inequality in these tw o independent settings. A s Lyn d o n Jo h n so n said, ‘A m erica is not m erely a nation , but a nation o f n atio n s.’

T o present the overall picture, w e have com bined all the health and social problem data fo r each cou n try, and separately fo r each U S state, to form an In dex o f H ealth and Social Problem s fo r each

co u n try and U S state. E ach item in the indexes carries the same

w eigh t - so, fo r exam p le, the score fo r m ental health has as much influence on a so ciety’s overall score as the hom icide rate or the teenage birth rate. Th e result is an in d ex sho w in g h o w com m on all these health and social problem s are in each cou n try and each US state. Things such as life expectan cy are reverse scored, so that on

every m easure higher scores reflect w o rse outcom es. W hen looking at the Figures, the higher the score on the In d ex o f H ealth and Social Problem s, the w o rse things are. (For in form ation on h o w w e selected countries show n in the graphs w e present in this b o o k , please see the A ppen d ix.)

W e start b y show in g, in Figure 2 .2 , that there is a v e ry strong tendency fo r ill-health and social problem s to occur less frequently in

Incom e inequality

Figure 2.2 Health and social problems are closely related to inequality among rich countries.

the m ore equal countries. W ith increasing in equality (to the right on the h orizon tal axis), the higher is the score on ou r In dex o f H ealth

and Social Problem s. H ealth and social problem s are indeed m ore com m on in countries w ith bigger incom e inequalities. T h e tw o are e x tra o rd in a rily closely related - chance alone w o u ld alm ost never produce a scatter in w h ich countries lined up like this.

T o em phasize that the prevalence o f p o o r health and social problem s in w h ole societies really is related to inequality rather than

to average living stan dards, w e sh o w in Figure 2.3 the sam e index o f health and social problem s but this time in relation to average incom es (N atio n al Incom e per person). It show s that there is no sim ilarly clear trend to w ard s better outcom es in richer countries. T his confirm s w h a t w e sa w in Figures 1 . 1 an d 1 . 2 in the first ch a p ­ ter. H o w ev e r, as w ell as k n o w in g that health an d social problem s are m ore com m on am o n g the less w e ll-o ff w ithin each society (as show n in Figure 1.4 ) , w e n o w k n o w that the overall burden o f these problem s is much higher in m ore unequal societies.

W orse - • USA

C/5

6 J J I S

0

• Portugal

a • UK

2̂ G 0 C/5

-O c as

JC

0) J= , Uh

O

Greece • • N ew Zealand

• Spain

• Ireland Australia . .

France* • Austria Italy* #Germany • Canada

* • Denmark Belgium

Finland* > Switzerland* Netherlands

Sweden*

Norw ay •

X <u

- a a Japan*

Better -

2 . 0 0 0 0 2 5 O O O 3 0 0 0 0 3 5 O O O 4O OO O

N atio n al incom e per person ($)

Figure z.3 Health and social problems are only w eakly related to national average income among rich countries.

T o check w hether these results are not just some odd fluke, let us see w hether sim ilar patterns also occur w hen w e look at the fifty states o f the U S A . W e w ere able to find data on alm ost exactly the sam e health and social problem s fo r U S states as w e used in our international in d ex. Figure 2.4 show s that the Index o f H ealth and Social Problem s is stron gly related to the am ount o f inequality in

each state, w h ile Figure 2.5 show s that there is no clear relation betw een it and average incom e levels. The evidence from the U S A confirm s the international picture. The position o f the U S in the international graph (Figure 2.2) show s that the high average incom e level in the U S as a w h ole does nothing to reduce its health and social problem s relative to other countries.

W e should note that part o f the reason w h y our in d ex com bining data fo r ten different health and social problem s is so closely related to inequality is that com bining them tends to em phasize w h a t they have in com m on and d o w n p lays w h a t they do not. In C hapters 4 - 1 2 w e w ill exam ine w hether each problem - taken on its o w n - is

Incom e inequality

Figure 2.4 Health and social problems are related to inequality in US states.

N atio n al incom e per person ($)

Figure 2.5 Health and social problems are only weakly related to average income in US states.

related to inequality and w ill discuss the vario u s reasons w h y they m ight be caused by inequality.

T h is evidence can n ot be dism issed as some statistical trick done w ith sm oke and m irrors. W hat the close fit show n in Figure 2 .2 sug­ gests is that a com m on element related to the prevalence o f all these health an d social problem s is indeed the am ount o f inequality in each country. A ll the data com e from the m ost reputable sources - fro m the W orld Ban k, the W orld H ealth O rgan ization , the United N a tio n s and the O rgan ization fo r E con om ic C o o p eratio n and D evelopm ent ( O E C D ) , and others.

C o u ld these relationships be the result o f som e unrepresentative selection o f problem s? T o an sw er this w e also used the ‘In dex o f child w ellbein g in rich coun tries’ com piled by the United N atio n s C h ild ren ’s Fund ( U N I C E F ) . It com bines fo rty different indicators coverin g m an y different aspects o f child w ellbeing. (We rem oved the m easure o f child relative po verty from it because it is, by definition, closely related to inequality.) Figure 2 .6 show s that child w ellbein g is

Incom e inequality

Figure 2.6 The U N I C E F index o f child wellbeing in rich countries is related to inequality.

U N

IC E

F

in de

x of

ch

ild

w el

lb ei

n g

Better -

W orse -

2 0 0 0 0 2 5 0 0 0 3 0 0 0 0 3 5 0 0 0 4 0 0 0 0

N a tio n a l incom e p er person ($)

F ig u r e 2 . 7 The U N I C E F index o f child wellbeing is not related to Gross National Income per head in rich countries.

stron gly related to in equality, and Figure 2 .7 show s that it is not at

all related to average incom e in each country.

S O C I A L G R A D I E N T S

A s w e mentioned at the end o f the last chapter, there are perhaps tw o w idespread assum ptions as to w h y people nearer the bottom o f society suffer m ore problem s. Either the circum stances people live in cause their problem s, or people end up nearer the bottom o f society because they are prone to problem s w h ich drag them dow n . The evidence w e have seen in this chapter puts these issues in a new

light. L e t’ s first consider the view that society is a great sorting system

w ith people m ovin g up or dow n the social ladder accordin g to their person al characteristics and vulnerabilities. W hile things such as h aving p o o r health, doin g b ad ly at school or h avin g a b ab y w hen

Sweden*

• Netherlands

Finland*

• Spain • Switzerland

Denmark* • Norway

• Greece • Portugal

Belgium Italy* • * Canada

•G e rm a n , . Ire|and Austria *

France • Australia • Japan

• Israel

• USA

• N ew Zealand

• U K

still a teenager all lo ad the dice against y o u r chances o f getting up the social ladder, sorting alone does nothing to explain w h y more unequal societies have m ore o f all these problem s than less unequal ones. Social m o b ility m ay p a rtly explain w hether problem s co n ­ gregate at the b ottom , but not w h y m ore unequal societies have m ore problem s o verall.

Th e vie w that social problem s are caused directly by p o o r m aterial conditions such as bad housing, p o o r diets, lack o f educational o pportunities and so on implies that richer developed societies w o u ld do better than the others. But this is a long w a y from the truth: som e o f the richest countries do w orst.

It is rem arkable that these m easures o f health and social problem s in the tw o different settings, and o f child w ellbein g am ong rich co u n ­ tries, all tell so m uch the sam e story. T h e problem s in rich countries are n ot caused by the society not being rich enough (or even b y being

too rich) but by the scale o f m aterial differences betw een people w ithin each society being too big. W hat m atters is w here w e stand in relation to others in ou r o w n society.

O f course a sm all p ro p o rtion o f the least w e ll-o ff people even in the richest countries som etimes find them selves w ith o u t enough m oney fo r food . H o w ev e r, surveys o f the 1 2 . 6 per cent o f A m ericans

living belo w the federal p overty line (an absolute incom e level rather than a relative stan d ard such as h alf the average income) sh o w that 80 per cent o f them have air-con d ition in g, alm ost 7 5 per cent o w n at least one car or truck and around 33 per cent have a com puter, a dish w asher o r a second car. W h at this m eans is that w h en people

lack m oney fo r essentials such as fo o d , it is u sually a reflection o f the strength o f their desire to live up to the p revailin g standards. Y o u m ay, fo r instance, feel it m ore im portan t to m aintain appearances by spending on clothes w h ile stinting on fo o d . W e k n ew o f a you ng

m an w h o w a s unem ployed and had spent a m on th ’ s incom e on a n ew m obile phone because he said girls ignored people w h o h ad n ’t got the right stuff. A s A d a m Sm ith em phasized, it is im p ortan t to be able to present on eself creditab ly in society w ith o u t the sham e and stigm a o f apparent p overty.

H o w ev e r, just as the gradient in health ran right across society from top to b ottom , the pressures o f in equality and o f w an tin g to

keep up are not confined to a sm all m in ority w h o are po o r. Instead,

the effects are - as w e shall see - w idespread in the popu lation .

D I F F E R E N T P R O B L E M S - C O M M O N R O O T S

T h e health and social problem s w h ich w e have foun d to be related to inequality tend to be treated by p o licy m akers as if they w ere quite separate fro m one anoth er, each needing separate services and remedies. W e p a y doctors and nurses to treat ill-health, police and prisons to deal w ith crim e, rem edial teachers and educational psych ologists to tackle educational problem s, and social w o rk e rs, drug rehabilitatio n units, p sych iatric services and health pro m o tio n experts to deal w ith a host o f other problem s. T h ese services are all exp en sive, and none o f them is m ore than p a rtia lly effective. F o r instance, differences in the q u ality o f m edical care have less effect on p eop le’ s life exp ectan cy than social differences in their risks o f getting som e life-threatening disease in the first place. A n d even

w h en the vario u s services are successful in stopping som eone re­ offen din g, in curing a cancer, getting som eone o ff drugs or dealing w ith educational failu re, w e k n o w that our societies are endlessly recreating these problem s in each n ew generation. M ea n w h ile, all these problem s are m ost com m on in the m ost deprived areas o f our

society an d are m any times m ore com m on in m ore unequal societies.

W H A T D O E S I N C O M E I N E Q U A L I T Y T E L L U S ?

B efore proceeding, in the fo llo w in g chapters, to lo o k at h o w the scale o f incom e differences m ay be related to other p roblem s, w e should say a fe w w o rd s abou t w h a t w e think incom e differences tell us abo u t a society. H u m an beings have lived in every kind

o f society, from the m ost egalitarian prehistoric hunting and gather­

ing societies, to the m ost plu tocratic dictatorships. A lth o u gh m odern m arket dem ocracies fall into neither o f those extrem es, it is

reason ab le to assum e that there are differences in h o w hierarchical

they are. W e believe that this is w h a t incom e inequality is m easuring. W here incom e differences are bigger, social distances are bigger and social stratification m ore im portant.

It w o u ld be nice to have lots o f different indicators o f the scale o f h ierarch y in different countries - to be able to com pare inequalities not on ly in incom e, but also in w ealth , education and p o w er. It w o u ld also be interesting to see h o w they are related to social dis­ tances, to indicators o f status like people’ s choice o f clothes, music and film s, or to the im portance o f h ierarchy and position. W hile

add ition al m easures w hich can be com pared betw een countries m ight becom e a vailab le in the future, at the m om ent w e m ust rely sim ply on incom e inequality. But w h a t is perhaps surprising is h ow m uch this m easure tells us even on its ow n .

T h ere are tw o im portan t reasons fo r interpreting incom e in­

eq u ality in this w a y . T h e first pointer is that o n ly the health and social problem s w h ich have strong social class gradients - becom ing m ore com m on further d o w n the social h ierarch y - are m ore com m on in m ore unequal societies. T h is seems to be a general phenom enon: the steeper the social gradient a problem has w ithin society, the m ore Strongly it w ill be related to in eq u ality.8 T h is not on ly applies to each problem - to teenage birth rates or to children doin g b ad ly at school, fo r exam ple - it lo o ks as if it also applies to sex differences in the sam e problem . T h e reason w h y w o m en ’ s obesity rates turn out - as w e shall see - to be m ore clo sely related to inequality than m en’s, seems to be that the social gradient in obesity is steeper am ong w o m en than men. H ealth problem s such as breast cancer, w h ich are not u su ally m ore com m on am on g the less w ell o ff, are unrelated to in eq u ality.9

T h e other pointer w h ich suggests that incom e inequality reflects

h o w h ierarch ical societies are, became clear w h en w e review ed n early 1 7 0 academ ic papers reporting different pieces o f research on the relationship betw een incom e inequality and h ea lth .10 T h e size o f the areas o ver w h ich researchers had m easured inequality varied substan tially. Som e had calculated h o w much inequality there w a s in

local n eighbourhoods and lo o ked to see if it w a s related to average death rates in those neighbou rhood s. O thers had used w h o le tow ns

and cities as the units in w h ich in equality and health w ere m easured. Still others had looked at regions and states, or done international studies co m parin g w h o le countries. W hen w e review ed all this research, a clear pattern em erged. W hile there w a s o verw helm ing evidence that inequality w a s related to health w hen both w ere

m easured in large areas (regions, states or w h o le countries), the

findings w ere m uch m ore m ixed w hen in equality w a s m easured in sm all lo cal areas.

This m akes perfect sense if w e think abou t w h y health tends to be w o rse in m ore deprived local areas. W h at m arks out the neighb ou r­

hoods w ith p o o r health - w here life exp ectan cy m ay be as m uch as ten years shorter than in the healthiest n eighbourhoods - is not o f course the inequality w ithin them. It is instead that they are unequal - or deprived - in relation to the rest o f society. W h at m atters is the

extent o f inequality right across society. W e concluded that, rather than telling us abo u t som e p revio u sly

u nknow n influence on health (or social problem s), the scale o f incom e differences in a society w a s telling us abou t the social h ierarchy across w hich gradients in so m any social outcom es occur. Because gradients in health and social problem s reflect social status differences in culture and beh aviou r, it lo o ks as if m aterial inequality is p ro b ab ly central to those differences.

We should perhaps regard the scale o f m aterial inequalities in a society as pro vid in g the skeleton, or fram ew o rk , round w h ich class

and cultural differences are form ed. O ver tim e, crude differences in w ealth g ra d u a lly becom e o verlaid by differences in clothing, aesthetic taste, education, sense o f self and all the other m arkers o f class identity. T h in k , fo r instance, o f h o w the co m p aratively recent em ergence o f huge incom e differences in R u ssia w ill com e to affect its class structure. W hen the children o f the new R u ssian o ligarch s have gro w n up in grand houses, attended p riv ate schools and travelled the w o rld , they w ill have developed all the cultural trap pin gs o f an upper class. A B ritish C o n servative p olitician w a s fam o u sly described by another as som eone w h o ‘ had to buy his o w n fu rn itu re’ . A lth ou gh there has a lw a y s been prejudice again st the n ouveau riche, w ealth does not rem ain new fo r ever: once the furniture is inherited it becom es old m oney. Even as far back as the eighteenth century, w hen

people thought that birth and breeding w ere w h a t defined the upper

echelons o f society, if you lost yo u r fortune you m ight m aintain

status briefly as ‘genteel p o o r’ , but after a generation o r so there w o u ld be little to distinguish yo u from the rest o f the poor. M o reo v e r, as Ja n e A usten show s in both M a n sfield P a rk and Sense a n d S e n sib ility , the consequences - w h atever y o u r birth - o f m a rry ­

ing fo r love rather than m oney could be serious. W hether m aterial

w ealth is made or lost, you cannot long rem ain ‘ a person o f substance’ w ithou t it. A nd it is surely because m aterial differences p ro vid e the fram ew o rk round w hich social distinctions develop that people have often regarded inequality as so cially divisive.

Q U A L I T Y O F L I F E F O R A L L A N D N A T I O N A L S T A N D A R D S

O F P E R F O R M A N C E

H a v in g com e to the end o f w h a t higher m aterial living standards can o ffer us, w e are the first generation to have to find other w a y s o f im p rovin g the real q u ality o f life. The evidence show s that reducing inequality is the best w a y o f im proving the qu ality o f the social environm ent, and so the real q u ality o f life, fo r all o f us. A s w e shall see in C hapter 1 3 , this includes the better-off.

It is clear that greater equality, as w ell as im p rovin g the w ellbeing o f the w h ole p o pu latio n , is also the key to nation al standards o f achievem ent and h o w countries perform in lots o f different fields. W hen health inequalities first cam e to prom inence on the public health agenda in the early 19 8 0 s , people w o u ld som etim es a sk w h y there w a s so much fuss abo u t inequalities. T h ey argued that the task o f people w o rk in g in public health w a s to raise o verall standards o f health as fast as possible. In relation to that, it w a s suggested that health inequalities w ere a side issue o f little relevance. W e can n o w see that the situation m ay be alm ost the opposite o f that. N a tio n a l

standards o f health, and o f other im portant outcom es w h ich we shall discuss in later chapters, are substan tially determ ined b y the am ount o f inequality in a society. If you w a n t to k n o w w h y one cou n try does better or w o rse than another, the first thing to look

at is the extent o f inequality. There is not one po licy fo r redu c­ ing in equality in health or the educational p erform ance o f school children, and another fo r raisin g n ation al standards o f perform ance. R edu cin g inequality is the best w a y o f doing both. A n d if, fo r

instance, a country w an ts higher average levels o f educational achievem ent am ong its school children, it m ust address the under­ lying in equality w hich creates a steeper social gradient in educational achievement.

D E V E L O P I N G C O U N T R I E S

B efore leaving this topic, w e should em phasize that although inequality also m atters in developing countries, it m ay do so fo r a different m ix o f reasons. In the rich countries, it is n o w the sym bolic

im portance o f w ealth and possessions that m atters. W hat purchases

say abo u t status and identity is often m ore im p ortant than the goods them selves. Put crudely, second-rate goods are assum ed to reflect second-rate people.

Possessions are m arkers o f status everyw h ere, but in poorer societies, w here necessities are a m uch larger part o f consum ption,

the reasons w h y m ore equal societies do better m ay have less to do w ith status issues and m ore to do w ith few er people being denied access to fo o d , clean w a ter and shelter. It is on ly am ong the very richest countries that health and w ellbein g are no longer related to

G ro ss N a tio n a l Incom e per person. In po o rer countries it is still

essential to raise living standards and it is m ost im portan t am ong the poorest. In those societies a m ore equal distribution o f resources w ill m ean few er people w ill be living in shanty to w n s, w ith dirty w ater and fo o d insecurity, or trying to scrape a living from inadequate

land-holdings.

In the n ext chapter w e w ill lo o k in a little m ore detail at w h y people in the developed w o rld are so sensitive to inequality that it can exert such a m ajo r effect on the psych o lo gical and social w e ll­ being o f m odern populations.

How inequality gets under the skin

’Tis very certain that each man carries in his eye the exact indication of his rank in the immense scale of men, and we are always learning to read it.

Ralph Waldo Emerson, The Conduct o f L ife

H o w is it that w e are affected as stron gly by inequality and our position w ithin society as the data in the last chapter suggest? Before ex p lo rin g - as w e shall in the n ext nine chapters - the relations between inequality and a w ide range o f social problem s, including

those in our In d ex o f H ealth and Social P roblem s, w e w a n t to

suggest w h y hum an beings m ight be so sensitive to inequality. A s inequality is an aspect o f the b road structure o f societies, e x ­

plan ation s o f its effects involve show in g h o w in dividuals are affected by the social structure. It is in dividuals - not the societies them selves

- w h o have p o o r health, are violent or becom e teenage m others. A lth ou gh individuals do not have an incom e distribution, they do have a relative incom e, social status or class p osition in the w ider society. So in this chapter w e w ill sh o w the w a y s in w hich our in d ividu al sensitivity to the w id er society exp lain s w h y livin g in m ore unequal societies m ight have such p ro fo u n d effects.

T o understand ou r vu ln erability to inequality means discussing som e o f our com m on psych o lo gical characteristics. T o o often when w e speak or w rite abo u t these issues, people m isinterpret our purpose. W e are not suggesting that the problem is a m atter o f in d ividu al p sych o lo gy, or that it is really peop le’s sensitivity, rather than the scale o f inequality, that should be changed. T h e solution to

w w w .C a r t o o n S t o c k .c o m

problem s caused by inequality is not m ass p sych otherap y aimed at

m akin g everyone less vulnerable. T h e best w a y o f responding to the

h arm done by high levels o f inequality w o u ld be to reduce inequality

itself. R ath er than requiring an ti-an xiety drugs in the w a ter supply o r m ass p sych o th erap y, w h a t is m ost excitin g abou t the picture we present is that it show s that reducing inequality w o u ld increase the w ellbein g and q u ality o f life fo r all o f us. F a r from being inevitable

and unstoppable, the sense o f deterioration in social w ellb ein g and

the q u ality o f social relations in society is reversible. Understanding the effects o f in equality m eans that w e suddenly have a p o licy handle on the w ellbeing o f w h ole societies.

Th e po w erfu l m echanism s w h ich m ake people sensitive to in­ equality cannot be understood in terms either o f social structure or

o f individual psych o lo g y alone. In dividual p sych o lo gy and societal in equality relate to each other like lock and key. O ne reason w h y the effects o f inequality have not been p ro perly understood before is because o f a failure to understand the relationship between them.

T H E R I S E I N A N X I E T Y

G iven the unprecedented m aterial co m fo rt and ph ysical convenience o f m odern societies, it m ight seem sensible to be sceptical o f the w a y everyone talks o f stress, as if life w a s b arely su rvivable. H o w ever,

Je a n T w en ge, a p sych ologist at San D iego State U n iversity, has put together im pressive evidence that w e really are m uch m ore an xiou s than w e used to be. B y review in g the large num ber o f studies o f an xiety levels in the po pu latio n carried out at different dates, she has docum ented very clear trends. She found 269 b ro ad ly com p arab le studies m easuring a n xiety levels in the U S A at vario u s times between 1 9 5 2 and 1 9 9 3 . 11 T ogether the surveys covered over 5 2 ,0 0 0 in d ivid­ u als. W h at they show ed w a s a continuous u p w ard trend throughout this fo rty-year period. H er results fo r men and w om en are show n in F igure 3 . 1 . E ach dot in the graph show s the average level o f an xiety found in a study recorded against the date it w a s undertaken. The rising trend across so m any studies is unm istakable. W hether she looked at college students or at children, T w en ge foun d the same

C3 Q <S)

o a t-OJ c <

3 1

Q S £ . I

o & o

- 3

Women

1 9 5 0 1 9 5 5 i 9 6 0 1 9 6 5 1 9 7 0 1 9 7 5 1 9 8 0 1 9 8 5 1 9 9 0 1 9 9 5

Figure 3 .1 Rise in anxiety levels among US college students 1 9 5 2 - 9 3 . Data from 269 samples covering 52,0 0 0 individuals.15 (Reproduced with kind permission from jea n M. Twenge.)

pattern: the average college student at the end o f the period w a s m ore an xio u s than 85 per cent o f the po pu latio n at the beginning o f it and, even m ore staggering, by the late 19 8 0 s the average A m erican child w a s m ore an xio u s than child psych iatric patients in the 1 9 5 0 s .

This evidence com es from the adm in istration o f standardized

a n xiety m easures to sam ples o f the p o p u lation . It can n ot be e x ­

plained a w a y by sayin g that people have becom e m ore a w are o f an xiety. The w o rsen in g trend also fits w h a t w e k n o w has been happening in related conditions such as depression. D epression and a n xiety are closely connected: people w h o suffer from one often suffer fro m the other, and psychiatrists som etim es treat the tw o con ­

ditions in sim ilar w a y s. T h ere are n o w large num bers o f studies

sh o w in g substantial increases in rates o f depression in developed countries. Som e studies have looked at change over the last h alf century o r so by co m parin g the experience o f one generation w ith another, w h ile takin g care to av o id pitfalls such as an increased

aw aren ess leading to m ore frequent reportin g o f dep ression .12

O thers have com pared rates in studies w h ich have fo llo w e d up representative sam ples o f the p o pu latio n born in different years. In B ritain, fo r exam p le, depression m easured am on g people in their m id -zos w a s foun d to be tw ice as com m on in a study o f 1 0 ,0 0 0 or

so people born in 1 9 7 0 as it had been in a sim ilar study carried out

earlier o f people in their m id -zos b o m in 1 9 5 8 . 13 R e v iew s o f research conclude that people in m any developed

countries have experienced substantial rises in a n xiety and depres­ sion. A m o n g adolescents, these have been accom panied by increases in the frequency o f b eh aviou ral problem s, including crim e, alcohol

and dru g s.u >14 T h ey ‘ affected m ales and fem ales, in all social classes

and all fam ily typ es’ . 13 It is im portant to understand w h a t these rises in a n xiety are

abo u t before their relevance to inequality becom es clear. W e are not suggesting that they w ere triggered by increased inequality. T h at

p ossib ility can be discounted because the rises in a n xiety and depres­

sion seem to start w ell before the increases in inequality w h ich in m an y countries to o k place during the last q u arter o f the twentieth century. (It is possible, h ow ever, that the trends betw een the 19 7 0 s and 19 9 0 s m ay h ave been aggravated b y increased inequality.)

S E L F - E S T E E M A N D S O C I A L I N S E C U R I T Y

A n im p ortan t clue to w h a t lies behind the mental health trends com es fro m evidence that they w ere accom pan ied by a surprising rise in w h a t at first w a s thought to be self-esteem. W hen com pared over tim e, in m uch the sam e w a y as the trends in an xiety are show n in Figure 3 . 1 , stan d ard m easures o f self-esteem also show ed a very clear long-term u p w ard trend. It looked as if, despite the rising

a n xiety levels, people w ere also tak in g a m ore positive view o f them ­ selves o ver time. T h ey w ere, fo r instance, m ore likely to say they felt prou d o f them selves; they w ere m ore likely to agree w ith statements such as ‘I am a person o f w o rth ’ ; and they seemed to have put aside

self-doubts and feelings that they w ere ‘useless’ or ‘ no g o o d at a ll’ .

T w en ge says that in the 1 9 5 0 s on ly 1 2 per cent o f teenagers agreed

w ith the statem ent ‘ I am an im p ortan t person ’ , but by the late 19 8 0 s this pro p o rtion had risen to 80 per cent.

So w h a t could have been going on? People becom ing m uch m ore self-confident doesn ’t seem to fit w ith them also becom ing much

m ore an xio u s and depressed. T h e an sw er turns out to be a picture o f increasing anxieties abo u t h o w w e are seen and w h a t others think o f us w h ich h as, in turn, produced a kind o f defensive attem pt to shore up ou r confidence in the face o f those insecurities. The defence involves a kind o f self-prom oting, insecure egotism w hich is easily

m istaken fo r high self-esteem . T h is m ight seem like a difficult set o f

issues to pin do w n , p a rticu la rly as w e are talkin g abou t general trends in w h ole p opu lation s. But let us lo o k briefly at the evidence w h ich has accum ulated since the self-esteem m ovem ent o f the 19 8 0 s , w hich show s w h a t has been happening.

O ver the years, m any research grou ps lo o k in g at in dividual

differences in self-esteem at a point in time (rather than at trends in p o p u latio n averages o ver time) began to notice tw o categories o f people w h o cam e out w ith high scores. In one catego ry, high self-esteem w en t w ith positive outcom es and w a s associated w ith h appiness, confidence, being able to accept criticism , an ability to

m ake friends, and so on. But as w ell as positive outcom es, studies

repeatedly found that there w a s another grou p w h o scored w ell on self-esteem m easures. T h ey w ere people w h o show ed tendencies to

violence, to racism , w h o w ere insensitive to others and w ere b ad at

person al relationships. T h e task w a s then to develop psych o lo gical tests w hich cou ld dis­

tinguish between people w ith a healthy an d those w ith an unhealthy kind o f self-esteem. T h e healthier kind seemed to centre on a fairly w ell-fou n ded sense o f confidence, w ith a reaso n ab ly accurate view

o f on e’ s strengths in different situations and an ability to recognize

on e’ s w eaknesses. T h e other seemed to be p rim arily defensive and in volved a denial o f w eakn esses, a kind o f internal attem pt to talk on eself up and m aintain a positive sense o f on eself in the face o f threats to self-esteem. It w a s (and is) therefore fragile, like w histling in the d a rk , and reacts b ad ly to criticism . People w ith insecure high

self-esteem tend to be insensitive to others and to sh o w an excessive preoccu pation w ith them selves, w ith success, and w ith their image and appearance in the eyes o f others. T h is unhealthy high self-esteem is often called ‘threatened egotism ’ , ‘ insecure high self-esteem ’ , or ‘n arcissism ’ . D urin g the com p aratively short time over w h ich data are available to com pare trends in narcissism w ith o u t getting it m ixed up w ith real self-esteem, T w en ge has show n a rising trend. She foun d that by z o o 6, tw o-th irds o f A m erican college students scored abo ve w h a t had been the average narcissism score in 1 9 8 2 . The recognition that w h a t w e have seen is the rise o f an insecure narcissism - p articu larly am ong yo u n g people - rather than a rise in genuine self-esteem n o w seems w id ely accepted.

T H R E A T S T O T H E S O C I A L S E L F

So the picture o f self-esteem rising alon g w ith an xiety levels isn’t true. It is n o w fairly clear that the rises in a n xiety have been acco m ­ panied by rising narcissism and that the tw o have com m on roots. Both are caused by an increase in w h a t has been called ‘social evalu ative th reat’ . T h ere are n o w go o d pointers to the m ain sources o f stress in m odern societies. A s living w ith high levels o f stress is n o w recognized as h arm fu l to health, researchers have spent a lot o f

time tryin g to understand both h o w the body responds to stress and

w h a t the m ost im p ortan t sources o f stress are in society at large. M u ch o f the research has been focused on a central stress horm one called cortisol w hich can be easily m easured in saliva or blood. Its release is triggered by the brain and it serves to prepare us p h ysio ­ lo gically fo r dealing w ith potential threats and em ergencies. There

have n o w been num erous experim ents in w h ich volunteers have been invited to com e into a lab o ra to ry to have their saliv a ry cortisol levels m easured w h ile being expo sed to some situation o r task designed to be stressful. D ifferen t experim ents have used different

stressors: som e have tried askin g volunteers to do a series o f arith ­

metic problem s - som etim es pu b licly com p arin g results w ith those o f

others - some have exp o sed them to loud noises or asked them to w rite abo u t an u n pleasant experience, or filmed them w h ile doing a task. B ecause so m any different kinds o f stressor have been used in these experim ents, S ally D ickerson and M a rg a re t K em en y, both

psychologists at the U niversity o f C alifo rn ia , L o s A ngeles, realized

that they cou ld use the results o f all these experim ents to see w h a t kinds o f stressors m ost reliab ly caused p eop le’ s co rtiso l levels to rise .16

T h ey collected findings from 20 8 published reports o f experim ents

in w h ich people’ s cortisol levels w ere m easured w h ile they w ere

exp o sed to an experim ental stressor. T h e y classified all the different kinds o f stressor used in experim ents and found that: ‘task s that included a social-evaluative threat (such as threats to self-esteem or social status), in w h ich others could n egatively judge perform ance, p articu larly w hen the outcom e o f the perform an ce w a s u n con trol­

lable, p ro vo k ed larger an d m ore reliable co rtiso l changes than

stressors w ith o u t these particu lar th reats’ (p. 3 7 7 ) . Indeed, they sug­ gested that ‘H u m an beings are driven to preserve the social self and are vigilan t to threats that m ay jeop ard ize their social esteem or statu s’ (p. 3 5 7 ) . So cial evalu ative threats w ere those w hich created the p ossib ility fo r loss o f esteem. T h ey typ ically in volved the

presence o f an evalu ative audience in the experim ent, a potential fo r

negative social com parison such as scorin g w o rse than som eone else, o r h avin g y o u r p erform an ce videoed or recorded, so creating the potential fo r later evalu ation . T h e highest cortisol responses cam e

w h en a social evalu ative threat w a s com bined w ith a task in w hich

participan ts could not avo id failu re - fo r instance because the task w a s designed to be im possible, o r because there w a s too little time, or they w ere sim ply told they w ere failin g h ow ever they perform ed.

T h e finding that social evalu ative threats are the stressors w hich get to us m ost p o w e rfu lly fits w ell w ith the evidence o f rising an xiety accom panied by a narcissistic defence o f an insecure self-im age. A s

D ickerson and K em en y say, the ‘ social self’ w hich w e try to defend ‘reflects on e’s esteem and status, and is largely based on others’ perception o f on e’s w o rth ’ (p. 3 5 7 ).

A quite separate strand o f health research co rro b o rates and fills

out this picture. O ne o f the m ost im p ortant recent developm ents in

our understanding o f the factors exerting a m ajo r influence on health in rich countries has been the recognition o f the im portance o f psych o lo gical stress. W e w ill outline in C h apter 6 h o w frequent and/or prolonged stress affects the body, influencing m an y ph ysio ­

logical system s, including the im m une and card io vascu lar systems.

B ut w h a t m atters to us in this chapter is that the m ost po w erful sources o f stress affecting health seem to fall into three intensely social categories: lo w social status, lack o f friends, and stress in early life. A ll have been show n , in m any w ell-con trolled studies, to be seriously detrim ental to health and longevity.

M u ch the m ost plausible interpretation o f w h y these keep

crop pin g up as m arkers fo r stress in m odern societies is that they all affect - or reflect - the extent to w hich w e do or do not feel at ease and confident w ith each other. Insecurities w h ich can com e from a stressful early life have some sim ilarities w ith the insecurities w hich can com e from lo w social status, and each can exacerbate the effects

o f the other. Friendship has a protective effect because w e feel more secure and at ease w ith friends. Friends m ake yo u feel appreciated, they find you good co m p an y, en joy y o u r con versation - they like y o u . If, in con trast, w e lack friends and feel avoided by others, then

fe w o f us are thick-skinned enough not to fall prey to self-doubts, to w o rries that people find us unattractive and boring, that they think w e are stupid or so cially inept.

P R I D E , S H A M E A N D S T A T U S

The p sych o an alyst A lfre d A dler said ‘T o be hum an m eans to feel in ferio r.’ Perhaps he should have said ‘T o be hum an m eans being h ighly sensitive abo u t being regarded as in fe rio r.’ O ur sensitivity to such feelings m akes it easy to understand the contrasting effects o f high and lo w social status on confidence. H o w people see you m atters. W hile it is o f course possible to be upper-class and still feel totally inadequate, or to be low er-class and full o f confidence, in general the further up the social ladder yo u are, the m ore help the w o rld seems to give yo u in keeping the self-doubts at bay. If the social h ierarchy is seen - as it often is - as if it w ere a ranking o f the hum an race by ability, then the o u tw ard signs o f success or failure (the better jo b s, higher incom es, education, housing, ca r and clothes) all m ake a difference.

It’ s h ard to disregard social status because it com es so close to defining ou r w o rth and h ow m uch w e are valued. T o do w ell fo r y o u rse lf or to be successful is alm ost syn on ym ous w ith m ovin g up the social ladder. H igher status alm ost a lw a y s carries connotations

o f being better, superior, m ore successful and m ore able. I f you

d o n ’t w a n t to feel sm all, incapable, looked d o w n on or in ferior, it is not quite essential to a vo id lo w social status, but the further up the social ladder yo u are, the easier it becom es to feel a sense o f pride, dignity and self-confidence. Social com p arison s in creasin gly sh o w

you in a positive light - w hether they are com parison s o f w ealth ,

education, job status, w here you live, h olid ays, or an y other m arkers o f success.

N o t on ly do advertisers p lay on ou r sensitivity to social co m ­ parison s, k n o w in g w e w ill tend to buy things w h ich enhance h o w we

are seen, but, as w e shall see in C h ap ter 1 0 , one o f the m ost com m on

causes o f violence, and one w hich p lays a large part in explain in g w h y violence is m ore com m on in m ore unequal societies, is that it is often triggered by loss o f face and h um iliation when people feel looked dow n on and disrespected. By p layin g on ou r fears o f being seen as o f less w o rth , advertisers m ay even contribute to the level o f

violence in a society.

It w a s T h o m as Scheff, em eritus p ro fesso r o f so cio lo gy at the U n iversity o f C a lifo rn ia , Santa B arb a ra, w h o said that sham e w as

the social em o tio n .17 H e m eant alm ost exactly w h a t D ickerson and K em en y w ere referring to w hen they found that the m ost lik ely kinds o f stressors to raise levels o f stress horm ones w ere ‘ social evaluative th reats’ . B y ‘sham e’ he m eant the range o f em otions to do w ith feel­ ing fo olish , stupid, rid icu lou s, inadequate, defective, incom petent, a w k w a rd , expo sed , vulnerable and insecure. Sham e and its opposite,

pride, are rooted in the processes through w h ich w e internalize h ow w e im agine others see us. Sch eff called sham e the social em otion because pride and sham e provid e the social evalu ative feed back as w e experience ourselves as if through others’ eyes. Pride is the

pleasure and sham e the pain through w h ich w e are socialized, so that

w e learn, from early ch ildh ood o n w ard s, to behave in so cially accept­ able w a y s. N o r o f course does it stop in childhood: our sensitivity to sham e continues to provid e the basis fo r co n fo rm ity th roughout adult life. People often find even the sm allest infringem ent o f social

norm s in the presence o f others causes so m uch em barrassm ent that

they are left w ish in g they could just disappear, or that the ground w o u ld s w a llo w them up.

A lth o u gh the D ickerson and K em eny study found that it w as exp o su re to social evalu ative threats w h ich m ost reliab ly raised levels o f stress h orm ones, that does not tell us h o w frequently people

su ffer from such an xieties. A re they a very com m on part o f everyday life, or only occasion al? A n an sw er to that question com es from the health research sho w in g that lo w social status, lack o f friends, and a difficult early ch ildh ood are the m ost im portant m arkers o f p sych o so cial stress in m odern societies. I f ou r interpretation o f these three factors is right, it suggests that these kinds o f social an xiety

and insecurity are the m ost com m on sources o f stress in m odern societies. H elen L e w is, a p sych o an alyst w h o drew p eop le’ s attention to sham e em otions, thought she sa w very frequent beh aviou ral indications o f sham e or em barrassm ent - perhaps not m uch more than w e w o u ld call a m om entary feeling o f aw k w a rd n ess o r self­ consciousness - w h en her patients gave an em barrassed laugh or hesitated at particu lar points w h ile speaking in a w a y suggesting slight n ervo usn ess.18

F R O M C O M M U N I T Y T O M A S S S O C I E T Y

W h y have these social an xieties increased so dram atically over the last h a lf century - as T w e n g e ’ s studies show in g rising levels o f an xiety an d fragile, narcissistic egos suggest they have? W h y does the social ev alu ative threat seem so great? A plausible exp lan atio n is the break-u p o f the settled com m unities o f the past. People used to g ro w up k n o w in g , an d being k n o w n b y, m any o f the sam e people

all their lives. A lth ou gh geographical m o b ility had been increasing

fo r several generations, the last h alf century has seen a particu larly rap id rise. A t the beginning o f this period it w a s still com m on for people - in rural and urb an areas alike - never to have travelled m uch beyond the boundaries o f their im m ediate city or village

com m unity. M arried brothers and sisters, parents and grandparents,

tended to rem ain living nearb y and the com m unity consisted o f people w h o had often k n o w n each other fo r m uch o f their lives. But n o w that so m an y peop le m ove fro m w here they g re w up, k n o w ­ ledge o f neighbours tends to be superficial or non-existent. P eo ple’s

sense o f identity used to be em bedded in the com m unity to w hich

they belonged, in p eo p le’s real kn o w led ge o f each other, but n o w it

is cast ad rift in the an on ym ity o f m ass society. F am iliar faces have been replaced by a con stan t flu x o f strangers. A s a result, w h o w e are, identity itself, is endlessly open to question.

T h e problem is show n even in the difficulty w e have in distinguish­

ing betw een the concept o f the ‘esteem ’ in w h ich w e m ay or m ay

not be held by others, and ou r o w n self-esteem . The evidence o f our sensitivity to ‘social evalu ative th reat’ , coupled w ith T w e n g e ’ s evidence o f long-term rises in a n xiety and n arcissism , suggests that w e m ay - by the standards o f any previous society - have becom e h ighly self-conscious, obsessed w ith h o w w e appear to others,

w o rried that w e m ight com e across as u n attractive, borin g, stupid or w h atever, and con stan tly trying to m anage the im pressions w e m ake. A n d at the core o f our interactions w ith strangers is our concern at the social judgem ents an d evaluation s they m ight m ake: h o w do they rate us, did w e give a go o d account o f ourselves? T h is vu ln erability

is p art o f the m odern psych o lo gical condition and feeds directly into consum erism .

It is w ell k n o w n that these problem s are p articu larly difficult fo r adolescents. W hile their sense o f them selves is most uncertain, they have to cope in schools o f a thousand o r m ore o f their peers. It is h ard ly surprising that peer pressure becom es such a p o w erfu l force in their lives, that so m any are dissatisfied w ith w h a t they lo o k like, or succum b to depression and self-harm .

I N E Q U A L I T Y I N C R E A S E S E V A L U A T I O N A N X I E T I E S

A lth o u gh the rises in a n xiety that seem to centre on social evaluation

pre-date the rise in in equality, it is not difficult to see h o w rising inequality and social status differences m ay im pact on them. R ather than being entirely separate spheres, h o w m uch status and w ealth people achieve - fro m unskilled lo w -p aid w o r k to success, m oney and pre-em inence - affects not o n ly their sense o f them selves, but also h o w positively they are seen even by friends and fam ily. O ur need to feel valu ed and capable hum an beings m eans w e crave positive feed back and often react w ith anger even to im plied criticism . Social status carries the strongest m essages o f superiority and in feriority, and social m obility is w id ely seen as a process by w hich people are sorted by ability. Indeed, in jo b ap plication s and pro m o tio n s, w h ere discrim ination by age, sex, race or religion is proh ibited , it is the task o f the in terview panel to discrim inate between individuals exclu sively by ability - just as long as they do n ’t m ake inferences from gender or skin co lo u r, etc.

G reater inequality seems to heighten p eop le’ s social evaluation anxieties by increasing the im portance o f social status. Instead o f accepting each other as equals on the basis o f ou r com m on hum anity as w e m ight in m ore equal settings, getting the m easure o f each other becom es m ore im portant as status differences w iden. W e com e to see social p osition as a m ore im portan t feature o f a person ’ s identity. B etw een strangers it m ay often be the dom inant feature. A s R alp h

W a ld o Em erson , the nineteenth-century A m erican ph ilosoph er, said,

‘ T i s v e ry certain that each m an carries in his eye the ex a ct indi­ cation o f his ran k in the imm ense scale o f men, and w e are a lw ay s learning to read it.’ 19 Indeed, p sych ological experim ents suggest that w e m ake judgem ents o f each other’ s social status w ithin the first few seconds o f m eeting.20 N o w o n d er first im pressions count, an d no

w o n d er w e feel social evaluation anxieties! If inequalities are bigger, so that som e people seem to count for

alm ost everything and others fo r practically nothing, w h ere each one o f us is placed becom es m ore im portant. G reater inequality is likely to be accom panied by increased status com petition and

increased status an xiety. It is not sim ply that w h ere the stakes are

higher each o f us w o rries m ore abo u t w h ere he or she com es. It is also that w e are likely to p ay m ore attention to social status in h o w w e assess each other. Surveys h ave found that w hen choosing prospective m arriage partners, people in m ore unequal countries put

less em phasis on rom antic considerations and m ore on criteria such

as financial prospects, status and am bition, than do people in less unequal societies.21

S E L F - P R O M O T I O N R E P L A C E S S E L F - D E P R E C A T I O N A N D M O D E S T Y

C o m p arin g Ja p a n w ith the U S A , that is, the m ost equal w ith alm ost the m ost unequal o f the rich m ark et dem ocracies (see Figure 2 . 1 ) , research has revealed a stark con trast betw een the w a y people see and present them selves to others in the tw o countries. In Ja p a n ,

people ch oose a m uch m ore self-deprecating and self-critical w a y

o f presenting them selves, w hich contrasts sh arp ly w ith the much m ore self-enhancing style in the U S A . W hile A m ericans are m ore likely to attribute in dividual successes to their o w n abilities and their failures to extern al facto rs, the Ja p an e se tend to do just the o ppo site.22 M o re than tw en ty studies in Ja p a n have failed to find any evidence o f the m ore self-serving pattern o f attributions com m on in the U S A . In Ja p a n people tended to pass their successes o f f as if they w ere m ore a reflection o f luck than o f judgem ent, w h ile suggest­ ing their failures are p ro b ab ly attributable to their o w n lack o f

ability. T h is Ja p an e se pattern w a s also foun d in T a iw a n and China. R ath er than getting too caught up in psych o lo gical term in ology,

w e w o u ld do w ell to see these patterns as differences in h o w far people value personal m odesty, preferring to m aintain social bonds by not using their successes to build themselves up as m ore able than others. A s greater inequality increases status com petition an d social evalu ative threat, egos have to be propped up by self-prom oting

and self-enhancing strategies. M o d e sty easily becom es a casu alty o f inequality: w e becom e o u tw a rd ly tougher and hard er in the face o f greater exposu re to social ev alu ation an xieties, but in w ard ly - as the literature on narcissism suggests - p ro b ab ly m ore vulnerable, less able to take criticism , less go o d at personal relationships an d less able to recognize our ow n faults.

L I B E R T Y , E Q U A L I T Y A N D F R A T E R N I T Y

Th e dem and fo r ‘liberty, eq u ality and frate rn ity ’ during the French R evo lu tio n show s just h ow long the issues w e have been discussing here have been recognized. Th e slogan focused attention on the dim ensions o f social relations w h ich m atter m ost if w e are to create a better society and m ake a difference to the real q u ality o f our lives. ‘L ib e rty ’ m eant not being subservient or beholden to the feudal n ob ility and landed aristo cracy. It w a s liberty from the feudal shackles o f in feriority. Sim ilarly, ‘ fratern ity’ reflects a desire for greater m utuality and reciprocity in social relations. W e raise the sam e issues when w e talk abo u t com m unity, social cohesion or so lidarity. T h eir im portance to hum an w ellbein g is dem onstrated

repeatedly in research w h ich show s h o w beneficial friendship and

involvem ent in com m unity life is to health. ‘E q u ality ’ com es into the picture as a precondition fo r getting the other tw o right. N o t only do large inequalities produce all the problem s associated w ith social differences and the divisive class prejudices w h ich go w ith them, but, as later chapters show , it also w eak en s com m unity life, reduces trust, and increases violence.

P A R T T W O

The Costs of Inequality

Community life and social relations

Among the new objects that attracted my attention during my stay in the United States, none struck me with greater force than the equality of conditions. I easily perceived the enormous influence that this primary fact exercises on the workings of the society.

Alexis de Tocqueville, D emocracy in America

In A ugu st 2 0 0 5 H u rrican e K a trin a hit the G u lf C o ast o f the southern United States, devastating cities in M ississip p i and L o u isia n a, o ver­ w h elm ing flood protection system s, and leavin g 80 per cent o f the city o f N e w O rleans under w ater. A m an d ato ry evacuation order w a s issued fo r the city the d ay before the storm hit, but b y that tim e m ost public tran spo rt had shut dow n and fuel and rental cars w ere u n available. T h e city governm ent set up ‘ refuges o f last reso rt’

fo r people w h o co u ld n ’t get out o f N e w O rleans, including the Superdom e, a vast sports arena, w h ich ended up sheltering around 2 6 ,0 0 0 people, despite sections o f its r o o f being ripped o f f by the storm . A t least 1 , 8 3 6 people w ere killed by the hurricane, and another 7 0 0 people w ere m issing and unaccounted for.

W h at captured the attention o f the w o r ld ’s m edia in the afterm ath o f the storm as m uch as the p h ysical d evastation - the flattened houses, flooded streets, co llap sed h igh w ays and battered oil rigs - w a s w h a t seemed like the com plete b reak d o w n o f civilization in the city. There w ere num erous arrests and shoot-outs throughout the w eek fo llo w in g the hurricane. T elevision new s screens show ed desperate residents begging fo r help, fo r b ab y fo o d , fo r m edicine,

and then sw itched to im ages o f tro o p s, cruising the flooded streets in

boats - not evacuatin g people, not bringing them supplies, but, fully arm ed w ith autom atic w eap o n s, lo o k in g fo r looters.

Th is response to the chaos in N e w O rleans led to w idespread criticism and condem nation w ithin the U S . M a n y alleged that the lack o f trust between la w enforcem ent and m ilitary forces on the

one hand and the m ostly p o o r, b lack citizens o f N e w O rleans on the other, reflected deeper issues o f race and class. D urin g a w idely televised benefit concert fo r victim s o f the hurricane, m usician K an ye W est, burst out: ‘ I hate the w a y they p o rtra y us in the m edia. Y o u see a w h ite fam ily , it says, “ T h e y ’re lo o kin g fo r fo o d .” Y o u see

a b lack fam ily, it says, “ T h e y ’re lo o tin g .’” A s tro o ps w ere m obilized

to go into the city, L o u isian a G o v e rn o r K athleen B lan co said: ‘ T h ey have M i 6 s and are locked and loaded. These troops k n o w h o w to shoot and kill and I expect they w ill.’

The lack o f trust on disp lay during the rescue effo rts in N e w

O rleans w a s also ro u n d ly condem ned intern ation ally. C ountries aroun d the w o rld offered aid and assistance, w h ile their new s co v e r­ age w a s filled w ith criticism . W e can con trast the w a y in w hich troops in N e w O rleans seemed to be used prim arily to control the po p u latio n , w ith the rap id deploym ent o f unarm ed soldiers in rescue and relief m issions in C hin a after the devastating earth ­

q uake o f z o o 8 , a response w id ely applauded by the international com m unity.

T H E E Q U A L I T Y O F C O N D I T I O N S

A very different visio n o f A m erica is offered by one o f its earliest observers. A le xis de T o cq u eville travelled th roughout the United States in 1 8 3 1 . 23 H e met presidents and ex-presidents, m ayo rs, senators and judges, as w ell as o rd in ary citizens, and everyw here he w en t he w a s im pressed by the ‘ equality o f con d ition s’ (p. x i ) , ‘ the blending o f social ran ks and the abolition o f p rivileges’ - the w a y that society w as ‘ one single m ass’ (p. 7 2 5 ) (at least fo r whites). He w rote that ‘A m ericans o f all ages, conditions, and all dispositions con stan tly unite to gether’ (p. 59 6 ), that ‘strangers read ily congregate

in the sam e places and find neither danger nor advan tage in telling each other freely w h at they th in k ’ , their m anner being ‘n atu ral, open and unreserved’ (p. 6 5 6). A n d de T o cq ueville points out the w a y s in w h ich A m ericans su pport one another in times o f trouble:

Should some unforeseen accident occur on the public highway, people

run from all sides to help the victim; should some family fall foul of an unexpected disaster, a thousand strangers willingly open their purses . . .

( p . 6 6 1 )

D e T o cq ueville believed that the equality o f conditions he observed

had helped to develop and m aintain trust am ong A m ericans.

W H A T ’ S T R U S T G O T T O D O W I T H I T ?

But does in equality corrode trust and divide people - governm ent

fro m citizens, rich fro m p o o r, m in ority fro m m ajo rity? T h is chapter show s that the q u ality o f social relations deteriorates in less equal societies.

In equ ality, not su rp risin gly, is a p o w erfu l social divider, perhaps

because w e all tend to use differences in livin g standards as m arkers

o f status differences. W e tend to choose ou r friends from am on g our near equals and have little to do w ith those m uch richer or much po o rer. A n d w hen w e have less to do w ith other kinds o f people, it’ s harder fo r us to trust them. O u r position in the social hierarchy affects w h o w e see as part o f the in-group and w h o as out-group - us

and them - so affecting ou r ability to identify w ith and em pathize

w ith other people. Later in the b o ok , w e ’ll sh o w that inequality not on ly has an im pact on h o w m uch w e lo o k do w n on others because they have less than w e do, but also affects other kinds o f dis­ crim in ation , such as racism and sexism , w ith attitudes sometim es

justified b y statements like ‘they just do n ’t live like us’ . De Tocq ueville un derstood this point. A lifelong opponent o f

slavery, he w ro te abo u t the exclu sion o f both A frican -A m erican s and N a tiv e A m ericans from the liberty and eq u ality enjoyed b y other A m erican s.23 S lavery, he thought, could o n ly be m aintained because A frican -A m erican s w ere view ed as ‘ o th er’ , so m uch so that ‘the

E uropean is to other races w h a t m an him self is to the an im a ls’

(p. 3 7 1 ) . E m p ath y is only felt fo r those w e vie w as equ als, ‘ the same

feeling fo r one another does not ex ist between the different classes’ (p. 6 50 ). Prejudice, thought de T o cq u eville, w a s ‘ an im agin ary in eq u ality’ w hich fo llo w ed the ‘real inequality produced by w ealth and the la w ’ (p. 40 0).

E a rly socialists and others believed that m aterial in equality w as an obstacle to a w id er hum an harm on y, to a universal hum an brotherhood, sisterhood o r com radeship. The data w e present in this chapter suggest that this intuition w a s sound: inequality is divisive, and even sm all differences seem to m ake an im portan t difference.

I N C O M E I N E Q U A L I T Y A N D T R U S T

Figures 4 . 1 and 4 .2 sh o w that levels o f trust betw een m embers o f the public are lo w er in countries and states w here incom e differences

Income inequality

Figure 4 .1 The percentage o f people agreeing that ‘most people can be trusted’ is higher in more equal countries.

Income inequality

Figure 4.2 In m ore equal states m ore p eo p le agree that ‘most p eo p le can be trusted’ . (Data available fo r only forty-on e US states.)

are larg er. T h ese re la tio n sh ip s are stro n g en o u gh th a t w e can be co n fid en t th at th ey are n o t due to ch an ce. T h e in te rn atio n a l d a ta

o n tru st in F ig u re 4 .1 co m e fro m the E u ro p e a n an d W o rld V a lu e s

S u r v e y , a stu d y design ed to a llo w in te rn atio n a l co m p a riso n s o f

v a lu e s an d n o rm s. In each c o u n try , ra n d o m sam p les o f the p o p u ­

la tio n w e re a sk e d w h eth er o r n o t th ey a greed w ith the statem ent:

‘ M o s t p e o p le ca n be tru ste d .’ T h e d ifferen ces b etw een co u n tries are

larg e. P eo p le tru st each o th er m o st in the S ca n d in a v ia n co u n trie s an d

the N e th e rla n d s; S w ed en h as the h igh est levels o f tru st, w ith 66 per

cen t o f p eo p le feelin g th at th ey ca n tru st o th ers. T h e lo w e s t level o f

tru st is seen in P o rtu g a l, w h e re o n ly 1 0 p e r cen t o f the p o p u la tio n

b elieve th at o th ers can be tru sted . So ju st w ith in these rich , m a rk e t

d e m o crac ies, there are m o re th an s ix fo ld d iffere n ce s in levels o f

tru st, a n d , as the g ra p h sh o w s, h igh levels o f tru st a re lin k ed to lo w

levels o f in e q u a lity .

T h e d a ta o n tru st w ith in the U S A , sh o w n in F ig u re 4.2, are ta k en

fro m the fed era l g o v e rn m e n t’ s G e n e r a l S o c ia l S u r v e y , w h ic h h as

been m on itorin g social change in A m erica fo r m ore than a quarter o f a cen tu ry.24 In this su rvey, just as in the international surveys, people are asked w hether or not they agree that m ost people can be trusted. W ithin the U S A , there are fo u rfo ld differences in trust between states. N o rth D ak o ta has a level o f trust sim ilar to that o f Sw eden -

67 per cent feel they can trust other people - w h ereas in M ississip pi only 1 7 per cent o f the p op u latio n believe that people can be trusted. Ju s t as w ith the international d ata, lo w levels o f trust am ong the United States are related to high inequality.

The im p ortant m essage in these graphs o f trust and inequality is

that they indicate h ow different life m ust feel to people livin g in

these different societies. Im agine living som ew here w here 90 per cent o f the po pu latio n mistrusts one another and w h at that must m ean fo r the q u ality o f ev eryd ay life - the interactions between people at w o rk , on the street, in shops, in schools. In N o r w a y it is not unusual

to see cafes w ith tables and chairs on the pavem ent and blankets left

out fo r people to use if they feel ch illy w h ile h avin g a coffee. N o b o d y w o rries abou t custom ers or passers-by stealing the blankets. M a n y people feel nostalgic fo r tim e past, w h en they could leave their doors unlocked, and trusted that a lost w a lle t w o u ld be handed in. O f all

large U S cities, N e w O rleans is one o f the m ost unequal. T h is w as

the b ackgrou n d to the tensions and m istrust in the scenes o f chaos after H u rrican e K a trin a that w e described above.

C H I C K E N O R E G G ?

In the U S A , trust has fallen from a high o f 60 per cent in i9 6 0 , to a lo w o f less than 40 per cent by Z004.24 But does in equality create lo w levels o f trust, or does m istrust create inequality? W hich com es first? P olitical scientist R o b ert Putnam o f H a rv a rd U niversity, in his b o ok B o w lin g A lo n e , show s h o w in equality is related to ‘social

ca p ita l’ , by w hich he m eans the sum total o f peop le’ s involvem ent in com m unity life.25 H e says:

Community and equality are mutually reinforcing . . . Social capital and economic equality moved in tandem through most o f the twentieth

century. In terms of the distribution of wealth and income, America in the 19 5 0 s and 1960s was more egalitarian than it had been in more than a

century . . . those same decades were also the high point of social connect­ edness and civic engagement. Record highs in equality and social capital coincided . . . Conversely, the last third of the twentieth century was a time of growing inequality and eroding social capital . . . The timing of the two trends is striking: Sometime around 1 9 6 5 - 7 0 America reversed course and started becoming both less just economically and less well connected

socially and politically, (p. 359)

In another article, Putnam says:

the causal arrows are likely to run in both directions, with citizens in high social capital states likely to do more to reduce inequalities, and inequalities themselves likely to be socially divisive.26

T a k in g a m ore definite stance in his b o o k , T h e M o ra l F o u n d a tio n s o f T ru st, Eric U slan er, a political scientist at the U n iversity o f

M a ry la n d , believes that it is inequality that affects trust, not the other w a y ro u n d .27 I f w e live in societies w ith m ore social capital, then w e k n o w m ore people as friends and neighbours and that might increase our trust o f people w e k n o w , people w e feel are like us.

But U slaner points out that the kind o f trust that is being m easured

in surveys such as the E uropean and W o rld V alu es Survey is trust o f strangers, o f people w e do n ’t k n o w , people w h o are often not like us. U sing a w ealth o f data fro m different sources, he show s that people w h o trust others are optim ists, w ith a strong sense o f control o ver their lives. T h e kind o f parenting that people receive also affects

their trust o f other people. In a study w ith his colleague Bo R o th stein , U slaner sh o w s, using a

statistical test fo r cau sality, that inequality affects trust, but that there is ‘ no direct effect o f trust on in equality; rather, the causal direction starts w ith in eq uality’ .28’ p- 45 U slaner says that ‘ trust can ­

not thrive in an unequal w o r ld ’ and that incom e inequality is the ‘prim e m o ver’ o f trust, w ith a stronger im pact on trust than rates o f unem ploym ent, inflation or econom ic g ro w th .27 It is not average levels o f econom ic w ellb ein g that create trust, but econom ic equality. U slan er’ s graph sho w in g that trust has declined in the U S A during a

Gini index of inequality

Figure 4.3 As inequality increased, so trust declined.27̂ - 187

period in w hich inequality rap id ly increased, is show n in Figure 4 .3 .

The num bers on the graph sh o w fo r each y ea r ( 1 9 6 0 - 9 8 ) the relation between the level o f trust and inequality in that year.

C hanges in inequality and trust g o together o ver the years. W ith greater inequality, people are less carin g o f one another, there is less m utuality in relation ship s, peop le have to fend fo r themselves and get w h a t they can - so, in evitably, there is less trust. M istru st and inequality reinforce each other. A s de Tocq ueville pointed out, w e are less likely to em pathize w ith those not seen as equals; m aterial differences serve to divide us socially.

T R U S T M A T T E R S

Both Putnam and U slaner m ake the point that trust leads to co ­ o peration . U slaner show s that, in the U S A , people w h o trust others are m ore likely to donate time and m oney to helping other people. ‘ T ru sters’ also tend to believe in a com m on culture, that A m erica is held together by shared valu es, that ev eryb o d y should be treated w ith respect and tolerance. T h ey are also supportive o f the legal order.

T ru st affects the w ellbein g o f in d ividu als, as w ell as the w ellbeing

o f civic society. H igh levels o f trust m ean that people feel secure,

they have less to w o r r y about, they see others as co-operative rather

than com petitive. A num ber o f convincing studies in the USA have linked trust to health - people w ith high levels o f trust live lon ger.29 In fact, people w h o trust others benefit from living in com m unities w ith gen erally high levels o f trust, w h ereas people w h o are less trusting o f others fare w o rse in such n eigh b o u rh o o d s.30

T ru st, or lack o f it, m eant the difference betw een life and death

fo r som e people cau gh t up in the chaotic afterm ath o f H urrican e K a trin a . T ru st w a s also crucial fo r su rvival in the C h icago heatw ave o f 1 9 9 5 . Sociologist Eric K lin en berg, in his b o o k abo u t the heat­ w a v e ,31 show ed h o w p o o r A frican -A m erican s, living in areas w ith

lo w levels o f trust and high levels o f crim e, w ere too frightened to

open their w in d o w s or d o o rs, or leave their hom es to go to local co o lin g centres established by the city authorities. N eigh b o u rs didn ’t check on neighbours, and hundreds o f elderly an d vu ln erable people died. In eq u ally p o o r H isp an ic neighbou rhood s, characterized by high levels o f trust and active com m unity life, the risk o f death w as

m uch low er.

R A I D E R S A N D M A V E R I C K S

Perhaps another m ark er o f co rro ded social relations and lack o f

trust am ong people w a s the rap id rise in the p o p u larity o f the Sport U tility Vehicle (SU V ) through the 19 8 0 s and 19 9 0 s . These vehicles are k n o w n in the U K by the d ero gato ry term ‘ Chelsea tracto rs’ - Chelsea being a rich area o f L o n d o n , the nam e d raw s attention to the silliness o f driving rugged o ff-ro a d vehicles in busy urb an areas.

But the vehicles them selves have nam es that evoke im ages o f hunters and outdoorsm en - O utlan der, Pathfinder, C herokee, W ran gler, etc. O thers evoke an even tougher im age, o f soldiers and w a rrio r s, w ith nam es like T ro o p e r, D efen der, Shogun, R a id e r and C o m m ander. These are vehicles fo r the ‘ u rban ju n gle’ , not the real thing.

N o t only did the p o p u larity o f S U V s suggest a preoccu p ation with

lo o k in g tough, it also reflected gro w in g m istrust, and the need to feel safe fro m others. Jo s h Lau er, in his paper, ‘ D riven to extrem es’ ,

asked w h y m ilitary ruggedness became prized abo ve speed o r sleek­ ness, and w h a t the rise o f the S U V said abo u t A m erican society.32 H e concluded that the trend reflected A m erican attitudes to w ard s crim e and violence, an adm iration fo r rugged in dividualism and the im portance o f shutting on eself o f f from con tact w ith others - m is­

trust. These are not large vehicles born from a co-operative p u b lic­

spiritedness and a desire to give lifts to hitch-hikers - hitch-hiking started to decline just as inequality started to rise in the 19 7 0 s . A s one an thro p o lo gist has observed, people attem pt to shield them ­ selves fro m the threats o f a harsh and untrusting society ‘ b y ridin g in

S U V s , w h ich lo o k arm oured, and by trying to app ear as in tim idat­

ing as possible to potential attack ers’ .33 Pollster M ich ael A dam s, w ritin g ab o u t the con trastin g values o f the U S A and C an a d a , pointed out that m inivans outsell S U V s in C an a d a b y tw o to one - the ratio is reversed in A m erica (and C an a d a is o f course m ore equal

than A m erica).34 A ccom p an yin g the rise in S U V s w ere other signs o f

A m erican s’ increasing uneasiness and fear o f one another: gro w in g

num bers o f gated com m unities,35 and increasing sales o f hom e secu­ rity system s.32 In m ore recent years, due to the steeply rising cost o f filling their fuel tan ks, sales o f S U V s h ave declined, but people still w a n t that rugged im age - sales o f sm aller, to u gh -lo o k in g ‘cro ss-o ver’

vehicles continue to rise.

W O M E N ’ S S T A T U S

In several respects, m ore unequal societies seem m ore m asculine, at

least in terms o f the stereotypes. W hen w e put this to the test, w e found that just as levels o f trust and social relations are affected by in equality, so too is the status o f w om en.

In the U S A , the Institute fo r W om en ’ s P o licy R esearch produces

m easures o f the status o f w om en . U sing these m easures, researchers

at H a rv a rd U niversity show ed that w o m en ’s status w a s linked to state-level incom e in eq uality.36 T h ree o f the m easures are: w o m en ’s political participation , w o m en ’s em ploym ent and earnings, and w o m en ’s social and econom ic auton om y. W hen w e com bine these m easures fo r each U S state and relate them to state levels o f incom e

In de

x of

w

o m

en ’s

st at

u s

Incom e inequality

Figure 4.4 Women's status and inequality in US states.

in equality, w e also find that w o m en ’ s status is significantly w orse in m ore unequal states, although this is not a p articu larly strong relationship (Figure 4 .4). T h e fa irly w ide scatter o f points around the line on the graph show s that factors besides in equality affect w o m en ’ s status. N everth eless, there is a tendency that can n ot be put d o w n to chance, fo r few er w o m en to vote or hold political office, fo r

w o m en to earn less, and few er w o m en to com plete college degrees in m ore unequal states.

In tern ation ally w e find the sam e thing, and w e sh o w this relation ­ ship in Figure 4 .5 . C o m b in in g m easures o f the percentage o f w om en in the legislature, the m ale-fem ale incom e g a p , and the percentage o f w o m en com pleting higher education to m ake an in d ex o f w o m en ’ s status, w e find that m ore equal countries do significantly better.

Ja p a n is conspicuous am ong the m ost equal countries in that w o m en ’ s status is lo w er than w e w o u ld expect given its level o f inequality; Italy also has w o rse w o m en ’ s status than expected, and Sw eden does better. A s w ith the scattering o f points on the A m erican grap h abo ve, this show s that other facto rs are also influencing

In de

x of

w

o m

en ’s

st at

u s

Incom e inequality

Figure 4.5 Women’s status and inequality in rich countries.

w o m en ’s status. In both Ja p a n and Italy w om en have trad ition ally had lo w er status than m en, w h ereas Sw eden has a long tradition o f w o m en ’s rights and em pow erm ent. But again , the link between incom e inequality and w o m en ’ s status can n ot be exp lain ed by

chance alone, and there is a tendency fo r w o m en ’s status to be better in m ore equal countries.

Epidem iologists have found that in U S states w here w o m en ’ s status is higher, both men and w om en have lo w er death ra tes,36 and w o m en ’s status seems to m atter fo r all w om en , w hether rich or p o o r.37

T R U S T B E Y O N D B O R D E R S

N o t su rprisin gly, just as in dividuals w h o trust other people are m ore likely to give to ch arity, m ore equal countries are also m ore generous to po o rer countries. Th e United N a tio n s ’ target fo r spending on foreign developm ent aid is 0 .7 per cent o f G ro ss N a tio n a l Income.

Income inequality

Figure 4 .6 Spending on foreign aid and inequality in rich countries.

O n ly N o r w a y , Sw eden, D en m ark and the N etherlands meet that target - indeed, they are generous beyond w h a t the United N atio n s expects - and, as w e sh o w in Figure 4.6 using data from the O E C D ,38 m ore unequal countries spend significantly lo w er percent­ ages o f their incom e on foreign aid. Ja p a n and the U K m ight be seen as outliers on this graph. Perhaps Ja p a n ’ s lo w er than expected spending on aid reflects its w ith d raw al fro m the international stage fo llo w in g the Second W orld W ar, and the U K ’s higher than e x ­ pected spending reflects historical, colo n ial ties to m any developing

countries.

W H A T W E H A V E L E A R N E D

In this chapter, w e have show n that levels o f social trust are

connected to incom e inequality, but o f course show in g a correlation is not the sam e thing as show in g causality.

T h ere are several reasons w h y w e believe that eq u ality is the

precondition fo r greater trust (although alm ost certain ly there is a feed back loop betw een the tw o). O ne facto r is the strength o f the relationship, w h ich is show n by the steepness o f the lines in Figures 4 . 1 and 4 .2 . People in Sw eden are m uch m ore likely to trust each other than people in P ortugal. A n y alternative exp lan atio n w o u ld need to be just as strong, and in our o w n statistical m odels w e find that neither p o verty nor average standards o f living can explain our findings. W e also see a consistent association am ong both the United States and the developed countries. E arlier w e described h ow U slaner and Rothstein used a statistical m odel to sh o w the ordering

o f inequality and trust: inequality affects trust, not the other w a y round. The relationships between inequality an d w o m en ’s status and between inequality and foreign aid also add coherence and plau sib ility to our belief that inequality increases the social distance between different grou p s o f people, m akin g us less w illin g to see them as ‘ us’ rather than ‘them ’ .

In sum m ary, w e can think o f trust as an im portan t m ark er o f the w a y s in w hich greater m aterial eq u ality can help to create a cohesive, co-operative com m unity, to the benefit o f all.

Mental health and drug use

It is no measure of health to be well adjusted to a profoundly sick society. Krishnamurti

M E N T A L I L L N E S S I N T H E U K A N D U S A

C h ild ren ’ s mental health n o w m akes the fro n t pages o f n ew spapers, B ritain ’ s D a ily M a il fo r exam p le, under banner headlines such as

‘T H E D I S T U R B E D G E N E R A T I O N ’ . A m illion B ritish children - one in ten betw een the ages o f 5 and 1 6 - are estim ated to be m entally ill.39 It has been suggested that in an y secon dary school w ith 1,0 0 0 students, 50 w ill be severely depressed, 1 0 0 w ill be distressed, 1 0 - 2 0 w ill be su fferin g from obsessive-com pulsive dis­ o rder and between 5 - 1 0 girls w ill have an eating disord er.40 T h is is

backed up by a 20 08 report from the G o o d C h ild h o o d In q u iry, an independent inquiry com m issioned by the C h ild ren ’ s Society.41 A fter surveying thousands o f children, they report that increasing numbers o f children have mental health problem s, over a quarter regu larly

feeling depressed, m ostly as a result o f fam ily b reak d o w n and peer

pressure. In the U S A , 6 per cent o f children have been diagnosed w ith

A ttention D eficit H yp eractivity D iso rd er, a b eh aviou ral syndrom e characterized by serious distractibility, im pulsivity and restlessness.42

In a n ation al su rvey, alm ost 1 0 per cent o f children aged 3 - 1 7 had m oderate or severe difficulties in ‘the areas o f em otions, concen­ tratio n , beh aviou r, o r being able to get alon g w ith other peop le’ .43

A n d h o w are adults doing in these sam e tw o societies? In the U K ,

"We're all on the same wavelength. Everyone's on an anti-depressant!" ww

w .C

a rt

o o

n S

to ck

.c o

m

in a nation al su rvey conducted in 2 0 0 0 , 23 per cent o f adults had

either a neurotic disorder, a psych otic disorder, or w ere addicted to alcohol or drugs, 4 per cent o f adults h avin g m ore than one d iso rd er.44 In 2 0 0 5 , doctors in England alone w ro te 29 m illion pre­ scriptions fo r anti-depressant drugs, costing o ver £ 4 0 0 m illion to the N a tio n a l H ealth Service.45 In the U S A , one in fo u r adults have been

m en tally ill in the past year and alm ost a q u arter o f these episodes w ere severe; over their lifetim e m ore than h a lf w ill suffer from a mental illness.46 In 2 0 0 3 , the U S A spent $ 1 0 0 billion on mental health treatm ents.47

M E N T A L W E L L B E I N G

B efore w e turn to com parison s o f mental illness in other societies, it’ s w o rth asking - w h a t is a healthy mind?

M I N D , the N a tio n a l A sso ciatio n fo r M en ta l H ealth in the

U K , publishes a pam phlet called ‘H o w to Im prove Y o u r M en tal W ell-b ein g’ . It begins w ith the prem ise that:

Good mental health isn’t something you have, but something you do. To

be mentally healthy you must value and accept yourself.48

It concludes that people w h o are m entally w ell are able to lo o k after them selves, see them selves as v alu ab le people and judge themselves by reason able, rather than unrealistic, standards. People w h o d o n ’ t valu e themselves becom e frightened o f rejection; they keep others at

a distance, and get trapped in a vicious circle o f loneliness. It is also im portant to note that although people w ith mental

illness som etim es have changes in the levels o f certain chem icals in their b rains, n o b o d y has show n that these are causes o f depression, rather than changes caused by depression. Sim ilarly, although genetic vu ln erability m a y underlie some mental illness, this can ’t by itself exp lain the huge rises in illness in recent decades - ou r genes c a n ’ t change that fast.

A P P L E S A N D O R A N G E S ?

C an w e really com pare levels o f mental illness in different countries? D o n ’t different cultures have different labels fo r m ental disorders, and different standards o f n o rm ality, or tolerance o f differences? A ren ’t people in different societies m ore o r less reluctant to adm it to em otional problem s, o r drug use, or an y stigm atized condition?

N o t su rprisin gly, it h asn ’t a lw a y s been easy to get co m parab le

m easures o f h o w m an y people are sufferin g from m ental illness in

different countries. But this began to get easier in the 19 8 0 s , w hen researchers developed diagnostic interview s - sets o f questions that cou ld be asked by n on-psychiatrists and non-psych ologists, allo w in g researchers to m easure on a large scale the num bers o f people

m eeting diagnostic criteria fo r different mental illnessess.

In 19 9 8 , the W o rld H ealth O rgan ization set up the W orld M en tal H ealth Survey C o n so rtium in an attem pt to estim ate the num bers o f peop le w ith m ental illness in different countries, the severity o f their illness and patterns o f treatm ent. A lth o u gh their m ethods d o n ’t

entirely overcom e w o rries abou t cu ltu ral differences in interpreting

and respon din g to such questions, at least the sam e questions are being asked, in the sam e w a y , in different places. A m o n g our set o f rich developed countries, W H O surveys have been com pleted in nine: Belgium , France, G erm an y, Italy, Ja p a n , N etherlan d s, N e w Z e a la n d , Spain and the U S A .49' 50 A lth ou gh not strictly co m p arab le,

very sim ilar nation al surveys give estim ates o f the p ro po rtion o f the adult p op u latio n w ith mental illness in another three countries - A u stra lia ,51 C a n a d a 52 and the U K .44

I N C O M E I N E Q U A L I T Y A N D M E N T A L I L L N E S S

In Figure 5 . 1 w e use these surveys to sh o w the association in rich countries betw een incom e in equality and the p ro po rtion o f adults

w h o had been m entally ill in the tw elve m onths p rio r to being inter­

view ed. T h is is a strong relationship: a m uch higher percentage o f

Figure 5 .1 M ore people suffer from mental illnesses in more unequal countries.

the p o p u latio n su ffer from m ental illness in m ore unequal countries.

Such a close relationship can n ot be due to chance, indeed the co u n ­

tries line up alm ost perfectly, w ith only Italy standing out as having lo w er levels o f m ental illness than w e m ight expect, based on its level

o f incom e inequality. Ju s t as w e sa w w ith levels o f trust in the previous chapter, there

are big differences in the pro p o rtion o f people w ith m ental illness

(from 8 per cent to 26 per cent) between countries. In G erm an y, Italy, Ja p a n and Spain, few er than 1 in 1 0 people had been m entally ill w ithin the previous yea r; in A u stra lia, C an a d a , N e w Z e a la n d and the U K the num bers are m ore than 1 in 5 people; and in the U S A , as w e described abo ve, m ore than 1 in 4. O verall, it lo o ks as if d iffer­

ences in inequality tally w ith m ore than th reefold differences in the percentage o f people w ith m ental illness in different countries.

Fo r ou r nine countries w ith data from the W H O surveys, w e can also lo o k at sub-types o f m ental illness, specifically a n xiety dis­ orders, m ood disorders, im pulse-control disorders and addictions, as

w ell as a m easure o f severe mental illness. A n x ie ty disorders, im pulse-control disorders and severe illness are all stron gly co rre­ lated w ith in equality; m ood disorders less so. W e sa w in C h ap ter 3

h o w a n xiety has been increasing in developed countries in recent decades. A n x ie ty disorders represent the largest sub-group o f mental illness in all our countries. Indeed, the percentage o f all mental illnesses that are a n xiety disorders is itself significantly higher in

m ore unequal countries. U n fo rtu n ately, there are no international sources o f co m parab le data on the mental health o f children and adolescents.

T u rn in g n o w to our other test-bed, the fifty states o f the U S A , w e discovered som ething rather surprising. A lo n e am on g the num erous health and social problem s w e exam in e in this b o ok , w e foun d no relationship between adult m ale mental illness and incom e inequality

am ong the U S states. State-specific estim ates o f mental illness are collected both by the United States B eh avio ral R isk F acto r Surveillance Study and by the N a tio n a l Survey on D ru g Use and H ealth , but the lack o f a relationship betw een incom e inequality and m ental illness am ong men w a s consistent in both sources.

H o w ev e r, incom e in equality is associated w ith m ental illness in

adult w om en . It is not a p a rticu la rly strong relation ship, but too strong to be dism issed as chance. T h ere is also a sim ilar relation ­ ship w ith the mental health o f children. T h e N a tio n a l S urvey o f C h ild ren ’ s H ealth p rovides estim ates o f the percentage o f children in each state w ith ‘m oderate or severe difficulties in the area o f

em otions, con cen tration , beh avior, or getting alon g w ith others’ .43 A lth o u gh , as fo r adult w om en , the relationship w ith state inequality is not p a rticu la rly strong, children’ s m ental health is significantly correlated w ith state levels o f incom e inequality.

Th ere are several plausible exp lan atio n s fo r the lack o f an a sso cia ­ tion between the a v ailab le m easures o f adult mental health am ong men and inequality. In general, problem s related to inequality have steep social gradients (becom ing m ore com m on lo w er do w n the social lad d er).8 Some indicators suggest that m ental health in the U S A does not sh o w a consistent social gradient. W hether the e x p la n ­

ation fo r this lies in m ethods o f data collection, gender differences in

reporting m ental illness, the apparen t resilience o f ethnic m in ority

1 2

% 8

4

o W hite B lack A m erican A sian H ispanic

Indian

Figure 5.2 US adults reporting frequent mental distress, 1 9 9 3 - 2 0 0 1 .53

popu lation s to m ental illness (see Figure 5 .2 ), or a delay in being able to observe the effects o f gro w in g in equality, it is im portan t to rem em ber that, from an international perspective, levels o f mental

illness in the U S A as a w h ole are ex actly w h a t w e w o u ld expect,

given its high overall level o f inequality.

C L I N G I N G T O T H E L A D D E R

So w h y do m ore people tend to have mental health problem s in more unequal places? Psychologist and jou rn alist O liver Ja m e s uses an a n alo gy w ith infectious disease to exp lain the link. Th e ‘affluen za’

virus, accordin g to Ja m e s, is a ‘set o f valu es w h ich increase our

vu ln erability to em otional distress’ , w h ich he believes is m ore com m on in affluent societies.54 It entails placin g a high value on acqu irin g m oney and possessions, looking go o d in the eyes o f others and w an tin g to be fam ous. These kinds o f valu es place us at greater risk o f depression, a n xiety, substance abuse and p erson ality disorder, and are closely related to those w e discussed in C h ap ter 3. In another recent b o o k on the sam e subject, philosopher A la in de Botton describes ‘status a n x ie ty ’ as ‘ a w o rr y so pernicious as to be capable o f ruining extended stretches o f our liv es’ . W hen w e fail to m aintain our position in the social hierarchy w e are ‘ condem ned to consider the successful w ith bitterness and ourselves w ith sham e’ .55

Econ om ist R o b ert F ran k observes the sam e phenom enon and calls it ‘lu xu ry fever’.56 A s inequality gro w s and the super-rich at the top

spend m ore and m ore on lu xu ry go o d s, the desire fo r such things cascades do w n the incom e scale and the rest o f us struggle to co m ­ pete and keep up. A dvertisers p lay on this, m akin g us dissatisfied w ith w h a t w e have, and encouraging invidious social com parison s.

A n oth er econom ist, R ich ard L a y a rd , describes ou r ‘ addiction to incom e’ - the m ore w e have, the m ore w e feel w e need and the m ore time w e spend on striving fo r m aterial w ealth and p ossessions, at the expense o f ou r fam ily life, relation ships, and q u ality o f life.3

G iven the im portance o f social relationships fo r m ental health, it

is not surprising that societies w ith lo w levels o f trust and w eak er com m unity life are also those w ith w o rse m ental health.

I N E Q U A L I T Y A N D I L L E G A L D R U G S

L o w position in the social status h ierarch y is p ain fu l to m ost people, so it com es as no surprise to find out that the use o f illegal drugs, such as cocaine, m ariju an a and heroin, is m ore com m on in m ore unequal societies.

In tern ation ally, the United N a tio n s O ffice on D rugs and C rim e publishes a W o rld D r u g R e p o r t ,57 w h ich contains separate data on the use o f opiates (such as heroin), cocaine, can n abis, ecstasy and

am phetam ines. W e com bined these data to form a single index, givin g each drug catego ry the sam e w eight so that the figures w ere not dom inated by the use o f an y one drug. W e use this in d ex in Figure 5 .3 , w hich show s a strong tendency fo r drug use to be m ore com m on in m ore unequal countries.

W ithin the United States, there is also a tendency fo r addiction to

illegal drugs and deaths from drug overdose to be higher in m ore unequal states.58

In de

x of

dr

ug

u se

Incom e inequality

Figure 5.3 The use o f illegal drugs is more common in more unequal countries.

M O N K E Y B U S I N E S S

T h e im portance o f social status to ou r m ental w ellbein g is reflected in the chem ical b eh aviou r o f ou r brains. Serotonin and dopam ine are am ong the chem icals that p lay im portant roles in the regulation o f m ood: in hum ans, lo w levels o f dopam ine and serotonin have been linked to depression and other mental disorders. A lth ou gh w e must be cautious in extrap o latin g to hum ans, studies in anim als sh o w that lo w social status affects levels o f, and responses to, different chem i­ cals in the brain.

In a clever experim ent, researchers at W ak e Forest School o f M ed icine in N o rth C aro lin a to o k tw enty m acaque m onkeys and housed them fo r a w h ile in in dividual cages.59 T h ey n ext housed the anim als in groups o f fo u r and observed the social hierarchies w hich developed in each grou p, noting w hich anim als w ere dom inant and w h ich subordinate. T h ey scanned the m o n k ey’ s brains before and

after they w ere put into grou p s. N e x t, they taught the m onkeys that they cou ld adm inister cocaine to them selves by pressing a lever - they cou ld take as m uch o r as little as they liked.

T h e results o f this experim ent w ere rem arkab le. M o n k ey s that had becom e dom inant had m ore dopam ine activity in their brains than they had exhibited before becom ing dom inant, w h ile m onkeys that becam e subordinate when housed in grou p s show ed no changes in their brain chem istry. Th e dom inant m on keys to o k m uch less cocaine than the subordinate m onkeys. In effect, the subordinate m onkeys w ere m edicating themselves again st the im pact o f their lo w social status. T h is kind o f experim ental evidence in m onkeys adds plau sib ility to ou r inference that in equality is cau sally related to m ental illness.

A t the beginning o f this chapter w e m entioned the huge num ber o f

prescriptions w ritten fo r m ood-altering drugs in the U K and U S A ;

add these to the self-m edicating users o f illegal drugs and w e see the pain w ro u g h t by inequality on a very large scale.

Physical health and life expectancy

A sad soul can kill you quicker than a germ. John Steinbeck, Travels with Charley

M A T E R I A L A N D P S Y C H O S O C I A L D E T E R M I N A N T S O F H E A L T H

A s societies have becom e richer and ou r circum stances have changed, so the diseases w e su ffer from and the m ost im portant causes o f health and illness have changed.

T h e h istory o f public health is one o f shifting ideas abo u t the causes o f disease.60-61 In the nineteenth century, reform ers began to collect statistics w h ich show ed the burden o f ill-health and p re­ m ature death suffered by the p o o r living in city slum s. T h is led to the great reform s o f the San itary M ovem en t. D rain age and sew age system s, rubbish collection, public baths and decent housing, safer

w o rk in g conditions and im provem ents in fo o d hygiene - all brought m ajo r im provem ents in p op u latio n health, and life expectancy lengthened as p eop le’ s m aterial standards o f living advanced.

A s w e described in C h ap ter i , w h en infectious diseases lost their hold as the m ajo r causes o f death, the industrialized w o rld under­ w ent a shift, k n o w n as the ‘ep idem iological tran sitio n ’ , an d chronic diseases, such as heart disease and cancer, replaced infections as the m ajo r causes o f death and p o o r health. D urin g the greater part o f the twentieth century, the predom inant appro ach to im proving the health o f p o pu lation s w a s through ‘ lifestyle ch oices’ and ‘risk

“What do you mean, I have an ulcer? I give ulcers, I don’t get them!"

fac to rs’ to prevent these chronic conditions. Sm okin g, high -fat diets,

exercise and alcohol w ere the focus o f attention. But in the latter p art o f the twentieth century, researchers began to

m ake some surprising discoveries abou t the determ inants o f health. T h ey had started to believe that stress w as a cause o f chronic disease, p articu larly heart disease. H e a rt disease w a s then thought o f as the execu tive’s disease, caused by the excess stress experienced by

businessm en in responsible positions. Th e W hitehall I Study, a lo n g­ term fo llo w -u p study o f m ale civil servants, w a s set up in 1 9 6 7 to investigate the causes o f heart disease and other chronic illnesses. Research ers expected to find the highest risk o f heart disease am ong men in the highest status jo b s; instead, they foun d a strong inverse

asso ciatio n between p osition in the civil service h ierarchy and death rates. M en in the lo w est grade (m essengers, doorkeepers, etc.) had a death rate three times higher than that o f men in the highest grade (adm in istrators ).62~3

Further studies in W hitehall I, and a later study o f civil servants,

W hitehall II, w hich included w o m en , have show n that lo w jo b status is n ot on ly related to a higher risk o f heart disease: it is also related to som e cancers, chronic lung disease, gastrointestinal disease, de­ pression, suicide, sickness absence from w o r k , back pain and self­ reported health.64~6 So w a s it lo w status itself that w a s causing w o rse

health, or cou ld these relationship s be explain ed by differences in

lifestyle betw een civil servants in different grades? T h o se in lo w er grades w ere indeed m ore lik ely to be obese, to

sm oke, to have higher blood pressure and to be less p h ysically active, but these risk factors explain ed on ly one-third o f their increased

risk o f deaths from heart disease.67 A n d o f course factors such as

absolute poverty and unem ploym ent cannot exp lain the findings, because ev eryb o d y in these studies w a s in paid em ploym ent. O f all the facto rs that the W hitehall researchers h ave studied o ver the years, jo b stress and p e o p le’s sense o f control o ver their w o rk seem to m ake the m ost difference. T h ere are n o w num erous studies that

sh o w the sam e thing, in different societies and fo r m ost kinds o f ill-health - lo w social status has a clear im pact on physical health, and not ju st fo r peop le at the v e ry bottom o f the social h ierarchy. A s w ell as highlighting the im portance o f social status, this is the other

im portan t m essage from the W hitehall studies. There is a social gradient in health running right across society, and w here w e are placed in relation to other peop le m atters; those ab o ve us have better

health, those belo w us have w o rse health, fro m the very bottom

to the very to p .68 U nderstanding these health gradients m eans understanding w h y senior adm in istrators live longer than those in pro fessio n al and executive grad es, as w ell as understanding the w o rse health profiles o f the poor.

Besides o u r sense o f con trol o ver ou r lives, other factors w hich m ake a difference to ou r ph ysical health include ou r happiness, whether w e ’re optim istic or pessim istic, and w hether w e feel hostile or aggressive to w ard s other people. O u r psych o lo gical w ellb ein g has a direct im p act on ou r health, and w e ’re less likely to feel in con trol, h ap p y, optim istic, etc. if ou r social status is lo w .

It’s not ju st ou r social status and psych o lo gical w ellbein g that

affects ou r health. T h e relationship s w e h ave w ith other people

m atter too. T h is idea goes back as far as the w o r k on suicide by E m ile D urkheim , one o f the founding fathers o f so cio lo gy, in the late nineteenth cen tu ry.69 D urkheim show ed that the suicide rates o f different countries and p o p u lation s w ere related to h o w w ell

people w ere integrated into society and w h eth er or not societies

w ere undergoing rapid change and turm oil. But it w a s n ’t until the 19 7 0 s that epidem iologists began to investigate system atically h ow peop le’ s social n etw o rk s relate to health, sho w in g that people w ith few er friends w ere at higher risk o f death. H a v in g friends, being m arried, belonging to a religious gro u p or other association and

h aving peop le w h o w ill provid e su pport, are all protective o f h ealth .70-71

Social su pport and so cial n etw orks have also been linked both to the incidence o f ca rd io v ascu la r disease and to recovery from

heart attacks. In a strikin g experim ent, researchers have also show n

that people w ith friends are less likely to catch a cold w h en given the sam e m easured exp o su re to the co ld viru s - in fact the m ore friends they h ad, the m ore resistant they w e re .72 E xp erim en ts have also show n that ph ysical w o u n d s heal faster if people h ave good relationships w ith their intim ate p artn ers.73

Social status and social integration are now well established as

im portan t determ inants o f po p u latio n health and, increasingly,

researchers are also recognizing that stress in ea rly life, in the w om b

as w ell as in in fan cy and early ch ildh ood, has an im p ortan t influence on p eop le’ s health th roughout their lives.74-5 Stress in early life affects physical g ro w th , em otion al, social and cognitive develop­ ment, as w ell as later health and health b eh aviours. A n d the so cio ­

econom ic status o f the fam ilies in w h ich children live also determines

their lifelong trajectories o f health and developm ent.76 T a k en together, social status, social n etw o rk s and stress in early

ch ildh ood are w h a t researchers label ‘p sych osocial fac to rs’ , and these are o f increasing im portance in the rich, developed countries

w here m aterial living stan dards, as w e described in C h ap ter i , are

n o w high enough to have ceased to be im p ortan t direct determ inants o f po pu latio n health.

L I F E I S S H O R T W H E R E L I F E I S B R U T A L

E vo lu tio n a ry psych ologists M a r g o W ilson and M artin D a ly were interested in w hether adoptin g m ore im pulsive and risk y strategies w a s an evolved response to m ore stressful circum stances in w hich life is likely to be shorter. In m ore threatening circum stances, then, m ore reckless strategies are perhaps necessary to gain status, m axim ize sexu al opportun ities, and en joy at least some short-term gratification s. Perhaps on ly in m ore relaxed conditions, in w h ich a longer life is assured, can people a ffo rd to plan fo r a long-term fu tu re.77 T o test this h ypothesis, they collected data on the m urder rates fo r the seventy-seven com m unity areas o f C h icag o , and then they collected data on death rates fo r those sam e areas, su btract­ ing all o f the deaths caused by hom icide. W hen they put the tw o together, they show ed a rem ark ab ly close relationship , seen in Figure 6. i - n eighbourhoods w ith high hom icide rates w ere also

n eighbourhoods w h ere people w ere dying you nger fro m other causes as w ell. Som ething abo u t these n eighbourhoods seem ed to be affectin g both health and violence.

In C hapter 4 w e show ed h o w different developed countries and

H om

ic id

e ra

te

1 60

1 4 0

120

I O O

80

60

4 0

20

O 5 0 5 5 6 0 6 5 7 0 7 5 8 0

M a le life exp ectancy at birth (years)

Figure 6 .1 Homicide rates are related to male life expectancy in seventy - seven neighbourhoods in Chicago. (Calculation o f life expectancy included deaths from all causes except homicide.)77

U S states v a ry in the levels o f social trust that people feel. T h ere are

six fo ld differences in levels o f trust betw een developed countries and fo u rfo ld differences am on g U S states. W e m entioned that levels o f trust have been linked to pop u latio n health and, in fact, research on social cohesion an d so cial cap ital has m ushroom ed over the past

ten years o r so. M o re than fo rty papers on the links betw een health

and so cial capital have n o w been pu b lish ed .78 In the United States, epidem iologist Ichiro K a w a ch i and his

colleagues at the H a rv a rd School o f Public H ealth looked at death rates in thirty-nine states in w h ich the G eneral Social Surveys had

been conducted in the late 19 8 0 s .79 These surveys allo w ed them to count h o w m an y people in each state w ere m em bers o f vo lu n tary

•

• •

•

• •

• • • • • • •

• •

• •• • • •

• • •

• • •

• 1

• •

• •

• •

• •

organ ization s, such as church gro u p s an d unions. T h is m easure o f gro u p m em bership turned out to be a strong predictor o f deaths

fro m all causes com bined, as w ell as deaths from co ro n a ry heart dis­

ease, cancers, and infant deaths. Th e higher the grou p m em bership, the lo w er the death rate.

R o b e rt Putnam looked at social capital in relation to an index o f health and health care fo r the U S states.25 T h is in d ex included

in fo rm atio n on such things as the percentage o f babies born w ith

lo w birthw eight, the percentage o f m others receiving antenatal care, m an y different death rates, expenditure on health care, the num ber o f people w ith A I D S and cancer, im m unization rates, use o f car safety belts, and num bers o f hospital beds, am ong other factors. Th e health index w a s clo sely linked to social cap ital; states such as

M in n eso ta and V erm o n t had high levels o f social capital and scored

high on the health in d ex, states such as L o u isian a and N e v a d a scored b ad ly on both. C le arly , it’ s not just ou r in dividual social status that m atters fo r health, the social connections betw een us m atter too.

H E A L T H A N D W E A L T H

L e t’ s consider the health o f tw o babies born into tw o different societies.

B ab y A is born in one o f the richest countries in the w o rld ,

the U S A , hom e to m ore than h a lf o f the w o rld ’ s billionaires. It is a co u n try that spends som ew here between 4 0 - 5 0 per cent o f the w o rld ’ s total spending on health care, although it contains less than 5 per cent o f the w o r ld ’s p opu lation . Spending on drug treatments and high-tech scanning equipm ent is p a rticu la rly high. D o cto rs in

this co u n try earn alm ost tw ice as m uch as doctors elsew here and m edical care is often described as the best in the w o rld .

B ab y B is born in one o f the po o rer o f the w estern dem ocracies, G reece, w here average incom e is not m uch m ore than h a lf that o f the U S A . W hereas A m erica spends abo u t $ 6 ,0 0 0 per person per y ea r on health care, Greece spends less than $ 3 ,0 0 0 . T h is is in real term s, after takin g into account the different costs o f m edical care.

A n d G reece has six times few er high-tech scanners per person than

the U S A . Surely B ab y B ’ s chances o f a long and h ealthy life are w o rse than

B ab y A ’ s? In fact, B a b y A , born in the U S A , has a life expectan cy o f 1 . 2 years

less than B ab y B, born in Greece. A n d B ab y A has a 40 per cent

higher risk o f dyin g in the first year after birth than B ab y B. A m o n g developed countries, there are even bigger contrasts than the co m ­ p arison w e ’ve used here: babies born in the U S A are tw ice as likely to die in their first year than babies in Ja p a n , and the difference in

average life exp ectan cy betw een the U S A and Sw eden is three years,

between P ortu gal and Ja p a n it is over five years. Som e com parison s are even m ore shocking: in 19 9 0 , C o lin M c C o rd and H a ro ld Freem an in the D epartm ent o f Surgery at C o lu m b ia U n iversity calculated that b lack men in H arlem w ere less likely to reach the age

o f 65 than men in B an glad esh .80

A m o n g other things, our co m p ariso n between B ab y A an d B ab y B

8 2 -

Q.X aj

Ji 78 J

76

•Ja p a n

•Sweden

Spain»# israe| Aus„ a|ia. 'C a n a d a • France •Switzerland

Italy. Belgium •

•A ustria New Zealand* "G reece # Netherlands* Germany

•Singapore Fin|and. UK

Ireland • U S A *

• Denmark

• Portugal

3 0 0 0 4 0 0 0 5 0 0 0

H ealth expenditure per person ($)

6 0 0 0

Figure 6.2 L ife expectancy is unrelated to spending on health care in rich countries (currencies converted to reflect purchasing power).

show s that spending on health care and the a v ailab ility o f high-tech m edical care are not related to pop u latio n health. Figure 6 .2 show s that, in rich countries, there is no relation ship betw een the am ount o f health spending per person and life expectan cy.

T H E ‘ B I G I D E A ’

If average levels o f incom e d o n ’t m atter, and spending on high-tech

health care doesn ’t m atter, w h a t does? T here are n o w a large

num ber o f studies o f incom e in equality and health that com pare countries, A m erican states, o r other large regions, and the m ajority o f these studies sh o w that m ore egalitarian societies tend to be h ealth ier.10 T h is vast literature w a s given im petus b y a study b y one o f us, on inequality and death rates, published in the B ritish M e d ic a l

J o u r n a l in 1 9 9 2 .81 In 19 9 6 , the editors o f that jo u rn al, com m enting on further studies confirm ing the link between incom e inequality and health, w rote:

The big idea is that what matters in determining mortality and health in a society is less the overall wealth of that society and more how evenly

wealth is distributed. The more equally wealth is distributed the better the health of that society.82

Inequ ality is associated w ith lo w er life exp ectan cy, higher rates o f infant m ortality, shorter height, p o o r self-reported health, lo w

birthw eight, A I D S and depression. Figures 6 .3 - 6 . 6 sh o w incom e in equality in relation to life expectan cy fo r men and w o m en , and to in fan t m o rtality - first fo r the rich countries, and then fo r the U S states.

O f course, p o pu latio n averages hide the differences in health

w ith in an y po pu latio n , and these can be even m ore dram atic than

the differences b e tw e en countries. In the U K , health disparities have been a m ajo r item on the public health agenda fo r o ver twenty-five years, and the current N a tio n a l H ealth S e rv ic e P la n states that ‘ N o injustice is greater than the inequalities in health w h ich scar our

n a tio n .’ 83 In the late 19 9 0 s the difference in life exp ectan cy between

the low est and highest social class grou p s w a s 7 .3 years fo r men and

In fa

n t

de at

h s

pe r

io oo

liv

e bi

rt hs

Li

fe

ex pe

ct an

cy

in ye

ar s

L o w H igh

Incom e inequality

Figure 6.3 L ife expectancy is related to inequality in rich countries.

L o w H igh

Incom e inequality

Figure 6.4 Infant mortality is related to inequality in rich countries.

In fa

nt

de at

hs

pe r

io oo

liv

e bi

rt hs

Li

fe

ex pe

ct an

cy

in ye

ar s

L o w H igh

Incom e inequality

Figure 6.5 L ife expectancy is related to inequality in US states.

L o w H igh

Incom e inequality

Figure 6.6 Infant mortality is related to inequality in US states.

7 years fo r w o m en .84 Studies in the U S A often report even larger d if­ ferences, such as a 28 -y ea r difference in life expectan cy at age 1 6

betw een b lacks and whites living in some o f the p oorest and some

o f the richest a reas.85-7 T o have m an y y ea rs’ less life because y o u ’re w o rk in g -class rather than p ro fessio n al - no one can argue abou t the serious injustice that these num bers represent. N o te that, as the W hitehall study sh o w ed , these gaps cannot be explain ed a w a y by w o rse health beh aviou rs am o n g those lo w er do w n the social scale.88-90 W h at, then, if the cost o f that injustice is a three- or four- y ear shortening o f a v era g e life expectan cy if w e live in a m ore unequal society?

W e exam in ed several different causes o f death to see w h ich had the biggest class differences in health. W e found that deaths am ong w o rk in g -ag e adults, deaths from heart disease, and deaths from hom icide had the biggest class differences. In con trast, death rates

from prostate cancer had sm all class differences and breast cancer death rates w ere com pletely unrelated to social class. Th en w e looked at h o w those different death rates w ere affected by incom e inequality, and foun d that those w ith big class differences w ere m uch m ore sensitive to in eq u ality.8 W e also foun d that living in a

m ore equal place benefited ev eryb o d y, not just the p o o r. It’s w o rth repeating that health disparities are not sim ply a contrast between the ill-health o f the p o o r and the better health o f ev eryb o d y else. Instead, they run right across society so that even the reaso n ab ly w e ll-o ff have shorter lives than the very rich. L ik ew ise , the benefits

o f greater equality spread right across society, im p rovin g health fo r

everyone - not just those at the bottom . In other w o rd s, at alm ost an y level o f incom e, it’ s better to live in a m ore equal place.

A dram atic exam p le o f h o w reductions in in equality can lead to rapid im provem ents in health is the experience o f B ritain during the

tw o w o rld w a r s .91 Increases in life expectan cy fo r civilian s during

the w a r decades w ere tw ice those seen th roughout the rest o f the twentieth century. In the decades w h ich con tain the w o rld w a rs, life expectan cy increased betw een 6 and 7 years fo r men and w om en , w h ereas in the decades before, betw een and after, life expectan cy increased b y between 1 and 4 years. A lth ou gh the n atio n ’ s

nutritional status im p roved w ith ratio n in g in the Second W o rld W ar,

this w a s not true fo r the F irst W o rld W a r, and m aterial living stan ­

dards declined during both w ars. H o w ev e r, both w artim es were

characterized by full em ploym ent and co n sid erab ly n a rro w e r income differences - the result o f deliberate governm ent policies to prom ote co-o peratio n w ith the w a r effo rt. D urin g the Second W o rld W a r, for exam p le, w o rk in g-class incom es rose by 9 per cent, w h ile incom es o f the middle class fell by 7 per cent; rates o f relative po verty were

halved. T h e resulting sense o f cam araderie an d social cohesion not

on ly led to better health - crim e rates also fell.

U N D E R O U R S K I N

So h ow d o the stresses o f adverse experiences in ea rly life, o f lo w social status and lack o f social support m ake us u n w ell?92 The belief that the m ind affects the b o d y has been aroun d since ancient times, and m odern research has enhanced ou r understanding o f the

w a y s in w hich stress increases the risk o f ill-health, and pleasure and

happiness prom ote w ellbeing. T h e psyche affects the neural system and in turn the im m une system - when w e ’ re stressed or depressed or feeling hostile, w e are far m ore likely to develop a host o f bodily ills, including heart disease, infections and m ore rap id ageing.93

Stress disrupts our b o d y ’ s balan ce, interferes w ith w h a t biologists

call ‘h om eo stasis’ - the state w e ’re in w hen everything is running sm oothly and all our ph ysio lo gical processes are norm al.

W hen w e experience som e kind o f acute stress and experience som ething traum atic, ou r bodies go into the fight-or-flight resp on se.93 E n ergy stores are released, our b lo o d vessels constrict, clotting facto rs are released into the blood stream , anticipating in ju ry, and

the heart and lungs w o rk harder. O ur senses and m em ory are enhanced and our im m une system perks up. W e are prim ed and read y to fight or run a w a y fro m w h atever has caused the stress. I f the em ergency is over in a few m inutes, this am azing response is healthy

and protective, but w h en w e go on w o rryin g fo r w eeks or m onths

and stress becom es chronic, then our bodies are in a constant state o f an ticipation o f some challenge or threat, and all those fight-or-flight responses becom e dam aging.

The human body is superb at responding to the acute stress of a physical challenge, such as chasing down prey or escaping a predator. The circulatory, nervous and immune systems are mobilized

while the digestive and reproductive processes are suppressed. If the stress becomes chronic, though, the continual repetition of theses responses can cause major damage.

EFFECTS OF ACUTE STRESS

B r a i n ------------------------- Increased alertn ess and less percep tion o f pain

A d ren a l g la n d s ------------- Secrete h o rm o n es that m obilize en ergy supplies

R ep ro d u ctive o rg a n s — - R ep ro d u ctive fu n ctio n s 2 te m p o rarily suppressed

T h y m u s gland and o th er im m un e tissues Im m une system readied f o r p o ssible in jury

C ir c u la to r y sy ste m -------- H e a rt beats faster, and b lo o d vessels con strict to brin g m o re o x y g e n to m uscles

EFFECTS OF CHRONIC STRESS

B ra in Im paired m em ory and in creased risk o f depression

T h y m u s g lan d and o th er im m un e tissues D eterio rated im m une response

C ir c u la to r y system b lood pressure

h igher risk o f d isease

Figure 6.7 The biology o f stress.91

C hro n ic m obilization o f energy in the fo rm o f glucose into the blood stream can lead us to put on w eigh t in the w ro n g places

(central obesity) and even to diabetes; chronic constriction o f blood

vessels and raised levels o f b lood-clotting factors can lead to h yp er­ tension and heart disease. W hile acute m om en tary stress perks up our im m une system , chronic continuing stress suppresses im m unity and can lead to g ro w th failure in children, o vu latio n failu re in w o m en , erectile dysfunction in men and digestive problem s fo r all o f

us. N eu ro n s in som e areas o f the brain are dam aged and cognitive

function declines. W e have trouble sleeping. C hro n ic stress w ears us

do w n and w ears us out. In this chapter w e ’ve show n that there is a strong relationship

betw een inequality and m any different health outcom es, w ith a consistent picture in the U S A and developed countries. O ur belief that this is a cau sal relationship is enhanced by the coherent picture

that emerges from research on the psych o so cial determ inants o f health, and the social gradients in health in developed countries. Position in society m atters, fo r health and alternative explan atio n s, such as higher rates o f sm oking am ong the p o o r, do n ’ t account for these gradients. T here are n o w a num ber o f studies sh o w in g that incom e inequality affects health, even after adjusting fo r peop le’s in dividual incom es.94 T h e dram atic changes in incom e differences in Britain during the tw o w o rld w a rs w ere fo llo w e d b y rap id im p ro ve­ ments in life expectan cy. Sim ilarly, in Ja p a n , the influence o f the post-Second W o rld W a r A llied occupation on dem ilitarization, dem ocracy and redistribution o f w ealth and p o w e r led to an

egalitarian econom y and u n rivalled im provem ents in po pu latio n h ealth .95 In con trast, R u ssia has experienced dram atic decreases in life expectan cy since the ea rly 19 9 0 s , as it m oved from a centrally planned to a m arket econ om y, accom pan ied b y a rap id rise in incom e in eq u ality.96 Th e b io lo g y o f chronic stress is a plau sible path ­ w a y w h ich helps us to understand w h y unequal societies are alm ost a lw a y s unhealthy societies.

mmm la a i w u p o f p a t FRgnowfliweM 1

Obesity: wider income gaps, wider waists

Food is the most primitive form of comfort

Sheila Graham

O besity is increasing ra p id ly th roughout the developed w o rld . In som e countries rates have doubled in just a fe w years. O besity is

m easured by B o d y M a ss In dex ( B M I) * to take height into account and av o id labelling people as o verw eight just because they are tall. T h e W o rld H ealth O rgan ization has set standards fo r using B M I to cla ssify people as underw eight (B M I < 1 8 .5 ) , n orm al weight (B M I 1 8 . 5 - 2 4 . 9 ) , o verw eight (B M I Z 5 - 2 9 .9 ) and obese ( B M I >

30). In the U S A , in the late 19 7 0 s , close to h a lf the pop u latio n were o verw eigh t and 1 5 per cent w ere obese; n o w three-quarters o f the p o pu latio n are overw eight, and close to a third are obese. In the U K in 19 8 0 , abo u t 40 per cent o f the po pu latio n w ere o verw eigh t and less than 1 0 per cent w ere obese; n o w tw o-th ird s o f adults are o v er­ w eigh t and m ore than a fifth are obese. 97- 100 T h is is a m a jo r health crisis, because obesity is so b ad fo r health - it increases the risk o f

hypertension, type II diabetes, card io vascu lar disease, gallb lad der disease and some cancers. T ren ds in ch ildh ood obesity are n o w so serious that they are w id ely expected to lead to shorter life exp ec­ tancies fo r to d a y ’ s children. T h a t w o u ld be the first reversal in life expectan cy in m any developed countries since governm ents started keeping track in the nineteenth cen tu ry.101

A p a r t fro m the health consequences, obesity reduces em otional

* B M I = w eigh t in kg/(height in m)2

and social w ellbeing: overw eight and obese adults and children su ffer terribly. A 17 -y e a r -o ld fro m Illin ois, w eigh in g 409 lb (29 stone)

described her physical pain: ‘m y heart aches in m y chest and I have

arm pains and stu ff and it gets s c a ry ’ .102 But just as hurtful are the m em ories she has o f other children callin g her nam es at sch ool, her restricted so cial life and her feeling that her body is ‘ alm ost a prison to m e’ .

B ritain ’ s tabloid new sp aper, the S u n , featured three obese children in the spring o f 2 0 0 7 .103-105 Th e youngest, a b o y aged 8, w eighed 2 1 8 lb ( 1 5 . 5 stone) and w a s being bullied at school - w hen he

attended. H is w eight w a s so great that he often m issed school due to his difficulties in w a lk in g there an d b ack, and w a s exem pt fro m w earin g school u n iform because none w a s a v ailab le to fit him. H is elder sister, aged 9, w eigh ed 1 9 6 lb ( 1 4 stone) and w a s also being bullied and teased, b y both children and adults. She said she fo u n d it ‘h ard to breathe som etim es’ , and did not like ‘h avin g to w e a r ugly cloth es’ and being unable to fit on the rides at am usem ent parks. H eaviest w a s the oldest boy w h o , at the age o f 1 2 , w eighed 28 0 lbs (20 stone). H e w a s desperately u n h appy - expelled from tw o schools and suspended from a third, fo r lashing out at children w h o called

him nam es.

T H E ‘ O B E S O G E N I C ’ E N V I R O N M E N T

Many people believe that obesity is genetically determined, and genes do undoubtedly play a role in how susceptible different individuals are to becoming overweight. But the sudden rapid increase in obesity in many societies cannot be explained by genetic factors. The obesity epidemic is caused by changes in how we live. People often point to the changes in cost, ease of preparation and availability of energy-dense foods, to the spread of fast-food restaur­ ants, the development of the microwave, and the decline in cooking skills. Others point to the decline in physical activity, both at work and in leisure time, increasing car use and the reduction in physical education programmes in schools. Modern life, it seems, conspires to

m ake us fat. If there w a s no m ore to it than that, then w e might

expect to see m ore obesity am on g richer people, w h o are able to buy

m ore fo o d , m ore cars, etc., and high levels o f obesity in all w ealth y

societies. But this is not w h a t happens. D urin g the ep idem iological tran ­

sition, w h ich w e discussed in C hapters i and 6, in w h ich chronic diseases replaced infectious diseases as the leading causes o f death,

obesity changed its social distribution. In the p ast the rich w ere fat

and the p o o r w ere thin, but in developed countries these patterns are n o w reversed.106 T h e W o rld H ealth O rgan izatio n set up a study in the 19 8 0 s to m on itor trends in ca rd io v ascu la r diseases, and the risk factors fo r these diseases, including obesity, in tw en ty-six countries. It foun d that, as rates o f obesity have increased, their social gradient

has steepened.107 By the early 19 9 0 s obesity w a s m ore com m on

am on g po o rer w o m en , com pared to richer w o m en , in all tw en ty-six coun tries, and am o n g p o o rer men in all except five. A s jou rn alist P o lly T oyn b ee declared in a new spap er article in 2 0 0 4 : ‘F at is a class issu e.’ 108 Pointing to the high rates o f obesity in the U S A and the lo w rates am on g the S can din avian coun tries, w h ich p ro ve that w e d o n ’t find high levels o f obesity in all m odern, rich societies, she suggested that incom e in equality m ight contribute to the obesity epidem ic.

I N C O M E I N E Q U A L I T Y A N D O B E S I T Y

Figure 7 . 1 show s that levels o f obesity tend to be lo w er in countries w h ere incom e differences are sm aller. T h e data on obesity come fro m the International O besity T a s k Force an d sh o w the p roportion o f the adult p o p u latio n , both men and w o m en , w h o are obese - a

B o d y M a ss In dex (B M I) o f m ore than 3 0 .109 T h e differences between countries are large. In the U S A , just over

3 0 per cent o f adults are obese; a level m ore than tw elve times higher than Ja p a n , w h ere on ly 2.4 per cent o f adults are obese. Because these figures are fo r B M I , not just w eight, th ey’re not due to d iffer­

ences in average height.

Incom e inequality

Figure 7 . 1 M ore adults are obese in more unequal countries.

The sam e pattern can be seen in tern ation ally fo r children (Figure 7 .2 ). O u r figures on the percentage o f yo u n g people aged 1 3 and 1 5

w h o are overw eight, reported in the 2 0 0 7 U N I C E F report on child

w ellbein g, cam e o rig in ally from the W o rld H ealth O rgan izatio n ’ s H ealth B eh avio u r in School-age C hildren su rv ey .110 T h ere are no data fo r A u stra lia, N e w Z e a la n d or Ja p a n from this su rvey, but the relationship w ith in equality is still stron g enough to be sure it is not due to chance. The differences betw een countries are sm aller fo r o verw eight children than fo r adult obesity. In the cou n try w ith the

low est level, the N etherlan d s, 7 .6 per cent o f children aged 1 3 and 1 5 are overw eight, w h ich is one-third the rate in the U S A , w here 2 5 . 1 per cent are overw eight. (As these figures are based on children reporting their height and w eight, rather than being m easured, the true prevalence o f overw eigh t is p ro b a b ly higher in all countries, but that sh o u ldn ’t m ake m uch difference to h o w they are related to inequality.)

W ithin the U S A , there are no states w ith levels o f adult obesity lo w er than 20 per cent. C o lo ra d o has the low est obesity prevalence,

Incom e inequality

Figure 7.2 M ore children are overweight in more unequal countries.

at 2 1 . 5 per cent, com pared to 34 per cent in T e x a s , w h ich has the h igh est/' But the relationship w ith in equality is still strong enough fo r us to be confident it isn ’t due to chance. O ther researchers have found sim ilar relationships. One study found that higher state-level incom e inequality w a s associated w ith abd o m in al w eight gain in

m en ,111 others have found that incom e in equality increases the risk o f inactive lifestyles.112 O verw eigh t am ong the p o o r seems to be p articu larly strongly associated w ith incom e inequality.

F o r children in the U S A , w e obtained data fro m the N a tio n a l Survey o f C h ild ren ’ s H ealth (Figure 7 .4). Ju s t as fo r the international figures fo r children, these data are fo r o verw eight (rather than obese) children, aged 1 0 - 1 7 years. (The ch ild’ s height and w eight are reported by the parent, or the adult w h o k n o w s the child best.) The

"T h e data on adult o besity w ith in the U S A w ere m ade availab le to us b y Professor M a jid Ezzati fro m H a rv a rd U niversity Sch ool o f Public H ealth. Professor Ezzati bases his calculations o f the prevalence o f o besity in each state on actual m easures o f height an d weight.

Pe r

ce nt

o

ve rw

ei gh

t io

- i

7- ye

ar -o

ld s

Pe r

ce nt

o

b es

e

L o w H igh

Income inequality

Figure 7.3 M ore adults are obese in more unequal US states.113

L o w H igh

Incom e inequality

Figure 7.4 M ore children are overweight in more unequal US states.

relationship w ith inequality is even stronger fo r children than fo r adults.

E A T I N G F O R C O M F O R T . . .

Th e p a th w ays lin kin g incom e inequality to obesity are likely to include calorie intake and ph ysical activity. Indeed, ou r ow n research has show n that per cap ita calorie intake is higher in more unequal countries. T h is explain ed p art o f the relation ship between inequality and obesity, but less fo r w om en than fo r m en .114 O ther

researchers have show n that incom e in equality in U S states is related to p h ysical in a ctiv ity.112 It seems that people in m ore unequal societies are eating m ore and exercisin g less. B ut in studies in A u stra lia, the U K and Sw eden the am ount that people eat, and the

am ount o f exercise they do, fails to fu lly account fo r so cial class

differences in w eigh t gain and o b esity .115-18

C alo rie intake and exercise are only part o f the story. People w ith a lo n g history o f stress seem to respond to fo o d in different w a y s from people w h o are not stressed. T h eir bodies respond by depositing fat p articu larly rou n d the m iddle, in the abd om en , rather

than lo w er do w n on hips and th igh s.119-20 A s w e sa w in C h apter 6, chronic stress affects the action o f the h orm one co rtiso l, and researchers have found differences in co rtiso l and psych o lo gical vu ln erability to stress tests am on g men and w o m en w ith high levels o f abd om in al fat. People w h o accum ulate fat aroun d the middle are

at p articu larly high risk o f ob esity-associated illnesses.

T h e b o d y ’ s stress reaction causes another problem . N o t only does it m ake us put on w eigh t in the w o rst places, it can also increase ou r fo o d intake and change our fo o d choices, a pattern k n o w n as stress-eating or eating fo r com fort. In experim ents w ith rats, w h en the anim als are stressed they eat m ore su gar and fat. People w h o are ch ro n ically stressed tend either to o ver-eat and gain w eigh t, or under-eat and lose weight. In a study in Fin lan d, people w h ose eating w a s driven by stress ate sausages, h am burgers, pizza and ch ocolate, and dran k m ore alcohol than other p eop le.121

Scientists are starting to understand h o w co m fo rt eating m ay be a

w a y w e cope w ith particu lar changes in ou r p h ysio lo gy w hen w e are ch ro n ically stressed, changes that go w ith feelings o f a n x ie ty .122

Th e three obese children featured in the Sun new spaper, w h om w e described earlier, all seemed to have turned to co m fo rt eating to deal w ith fam ily break-ups. The nine-year-old girl said, ‘ C h o colate is the on ly thing I ’ m interested in. It’ s the only thing I live fo r . . . when

I ’m sad and w o rried I ju st e a t.’ H er older brother gained 2 1 0 pounds ( 1 5 stone) in five years, after his parents divorced.

A num ber o f years ago , the W all S treet j o u r n a l ran a series, ‘D ead ly D iet’, on the nutrition problem s o f A m erica ’ s inner cities.123 A m o n g the o verw eight people they interview ed w a s a 1 3 -year-old girl livin g in a violent h ousing project (estate), w h o said that food

and T V w ere a w a y o f calm ing her nerves. A n unem ployed w o m an w h o k n ew that her diet and drinking w ere dam agin g her liver and

arteries, still figured she ‘m ight as w ell live high on the h o g ’ w h ile she could. A gran dm other bringing up her gran dchildren because o f her dau gh ter’ s addiction to crack cocaine, said:

Before I was so upset that my daughter was on this crack I couldn’t eat. I turned to Pepsi - it was like a drug for me. I couldn’t function without it. I used to wake up with a Pepsi in my hand. A three-liter bottle would just see me through the day.

Recent research suggests that fo o d stim ulates the brains o f chronic over-eaters in just the sam e w a y s that drugs stim ulate the brains o f

a d d icts.124-6 Studies using brain scans have show n that obese people

respond both to fo od and to feeling full differently fro m thin p e o p le .127

. . . E A T I N G ( O R N O T ) F O R S T A T U S

But fo o d choices and diets aren ’t ju st dictated by the w a y w e feel - th ey’re also patterned b y social facto rs. W e m ake fo o d choices fo r com plicated cu ltu ral reasons - som etim es w e like fo od s w e grew up eating, w h ich represent hom e to us, som etim es w e w a n t food s

that represent a lifestyle w e ’re tryin g to achieve. W e o ffe r fo o d to other peop le to sh o w that w e love them , or to sh o w that w e ’ re

sophisticated, or that w e can a ffo rd to be generous. F o o d has p ro b ­

a b ly a lw ay s played this role; it’ s the n ecessary com ponent o f the feast, w ith all o f its social m eanings. But n o w , w ith the easy a v a il­ ability o f cheap, energy-dense fo o d s, w h atever social benefits m ight com e from frequent feasting, they are, so to speak, outw eighed by the d raw b ack s.

In the W all Street J o u r n a l’ s ‘ D ead ly D iet’ series, a recent

im m igrant from Puerto R ic o describes h ow her fam ily used to live on an unchanging diet o f rice, beans, vegetables, p o rk and dried fish. Since m ovin g to C h icag o , they h ave enjoyed fizzy drin ks, pizza, h am ­ burgers, sugared b reak fast cereals, hotdogs and ice-cream . ‘I can ’t a ffo rd to buy the children expen sive shoes or dresses . . . but fo o d is

easier so I let them eat w h atever they w a n t.’ M o st o f all, the fam ily en joy goin g to fast-fo o d restaurants and eat out tw ice a m onth, alth ou gh the children w o u ld like to go m ore often. ‘W e feel good w h en w e go to those places . . . w e feel like w e ’ re A m erican s, that

w e ’re here and w e belong here.’

A 17 -y e a r -o ld in N e w Je rse y described h o w being able to b u y fast fo o d pro ves y o u r financial status, show s that yo u have m on ey in y o u r po cket and are not h avin g to w a it fo r the w elfare cheque at the end o f the month.

A 37 -y ea r-o ld m an said he spent h alf his w ages on fast fo o d . On the day he w a s interview ed he had been to M cD o n ald s three times

and w a s planning to go to K en tu ck y Fried C hicken and a Chinese take-ou t shop before the d ay w a s out. But the fast-fo o d restaurants had a m eaning fo r him that w en t w ell beyond the cheap fo o d . Despite w o rk in g , he w a s hom eless and they had becom e his sanctuary:

He has no home of his own and shuttles between his aunt’s place in Brooklyn and a friend’s apartment in a Harlem housing project [estate]. ‘The atmosphere makes me feel comfortable and relaxed and you don’t have to rush,’ he says as he admires the hamburger restaurant’s shiny floors and the picture of George Washington Carver [a famous nineteenth- century black American] on a wall. Lulled by the soft piped-in music, he nods off for a moment and then adds: ‘ ain’t no hip-hop, ain’t no profanity. The picture, the plants, the w ay people keep things neat here, it makes you feel like you’re in civilization.’

A m em ber o f a H isp an ic street gan g eats all his m eals at fast-fo o d restaurants, boastin g that he h asn ’t eaten a m eal at hom e since he

w a s 1 6:

Kids here don’t want to eat their mother’ s food . . . everyone is tired of their mother’s food - rice and beans over and over. I wanted to live the life o f a man. Fast food gets you status and respect.

F A T I S A F E M I N I N E I S S U E ?

O ur o w n w o r k , like the studies o f other researchers, show s that the relation betw een incom e in equality and obesity is stronger fo r w o m en than fo r men. In the W o rld H ealth O rgan izatio n ’s study

in tw en ty-six countries the social gradient in o b esity is seen m ore

consistently, and tends to be steeper, fo r w om en than fo r men. In the

2 0 0 3 H ealth Survey fo r E n glan d , the positive asso ciatio n between lo w socio-econom ic status and o b esity is very clear fo r w o m en but am on g men there is no asso c iatio n .128

It m ight be that these patterns result fro m obesity h avin g a

stronger negative effect on social m o b ility fo r w o m en , than fo r men.

M a y b e obese yo u n g w om en su ffer m ore discrim ination in lab o u r m arkets and the m arriage m arket, than obese yo u n g men. O r m aybe lo w social status is m ore o f a risk facto r fo r obesity in w o m en than in men. T w o studies w ithin B ritish birth ‘co h o rts’ o ffer som e clues.

These studies are surveys o f large sam ples o f people born at the sam e tim e, and fo llo w ed fro m birth. A study o f people born in 1 9 4 6 found that u p w a rd ly m obile men and w o m en w ere less likely to be obese than those w h ose so cial class did n ’t change betw een childhood and ad u lth o o d .129 In the 1 9 7 0 coh ort obese w om en , but not m en, w ere

m ore likely never to have had gain fu l em ploym ent and not to have a p a rtn er.130

In the U S A and in B ritain , fem ale obesity in adolescence has been linked to lo w er earnings in a d u lth o o d .131-2 A lth o u gh not limited to w o m en , a recent su rvey o f m ore than 2 ,0 0 0 H u m an Resou rce pro fessio n als foun d that 93 per cent w o u ld fav o u r a norm al-w eigh t

job ap plican t over an eq u ally qualified o verw eight candidate. N e a r ly

50 per cent o f these p rofession als felt that o verw eigh t peop le w ere

less p ro d uctive; alm ost 3 3 per cent felt that obesity w a s a valid

reason not to hire som eb ody; and 4 0 per cent felt that overw eight people lacked self-disciplin e.133

A lth o u gh being o verw eight clearly ham pers social m o b ility, our o w n an alysis o f trends w ithin w o m en born in B ritain in 1 9 7 0 sug­ gests that this doesn ’ t exp lain the social gradient in obesity am ong

w o m en and, even in m iddle age, lo w social class is linked to weight

g a in .117

Y O U C A N N E V E R B E T O O R I C H O R T O O T H I N

Social class differences in the im portance o f b o d y size and in the b o d y im age to w ard s w h ich w o m en aspire also seem to contribute to

the social gradient in obesity. In the past, w o m en w ith vo lu ptuo u s bodies w ere m uch adm ired, but in m an y m odern, richer cultures, being thin signals high social class and attractiveness. B ritish w om en in higher social classes are m ore likely to m on itor their w eigh t and to be dieting than w om en in lo w er social class gro u ps, an d are also m ore dissatisfied w ith their b o d ies.134 W om en w h o m ove do w n the

social scale seem to place less em phasis on thinness and are m ore

satisfied w ith their bodies. C han ges in m arital status also p la y a role: in a U S study, w o m en w h o m arried gained m ore w eigh t than w om en w h o rem ained single or w o m en w h o divorced o r sep arated .135 A nd not all w om en w a n t to be thin - fo r exam ple in inner-city A frican - A m erican com m unities, thinness can be associated w ith an im age o f p o verty, hunger and being on w e lfa re, as w ell as A I D S and drug

addiction. A s one 19 -y e a r-o ld w o m an put it:

I’ve been a voluptuous female all my life. If I start losing a lot of weight, people will think I’m on drugs . . . in the ghetto, you just can’t afford to

look too thin.

H er w o rd s are a rem inder o f the w a y s in w h ich social class is related to being overw eight in the developin g w o rld , w h ere o n ly the affluent can a ffo rd to be fat. In w e alth y countries, it lo o ks as if w om en in

higher social classes are m ore likely to have asp iration s to thinness and be m ore able to achieve them.

But w h ile w o m en ’s body w eigh t m ay be m ost affected by social facto rs, men are certain ly not im m une. A recent 12 - y e a r study o f w o rk in g -ag e men in the U S A foun d that if they becam e unem ployed, they gained w e ig h t.136 W hen their annual incom e dropped they gained, on average, 5 .5 lbs.

T H E T H R I F T Y P H E N O T Y P E

One add itio n al idea that suggests a cau sal link between higher levels o f incom e in equality in a society and higher body w eights is k n o w n as the ‘ th rifty p h en o type’ hypothesis. Put sim ply, this theory suggests that w hen a pregnant w o m an is stressed, the developm ent o f her u n born child is m odified to prepare it fo r life in a stressful

environm ent. It isn ’t yet clear w h eth er stress h orm ones themselves do the dam age, or w hether stressed foetuses are less w ell nourished, or both things happen, but these ‘ th rifty p h en o type’ babies have a lo w er birthw eight and a lo w er m etabolic rate. In other w o rd s, they are adapted fo r an environm ent w here fo od is scarce - they

are sm all and need less food . In conditions o f scarcity during our ev o lu tio n ary past this ad aptation w o u ld have been beneficial, but in our m odern w o rld , w h ere stress during pregn an cy is unlikely to be due to fo o d shortages and babies are born into a w o rld o f plenty, it’s m alad aptive. Babies w ith a th irfty phenotype in a w o rld

w here fo o d is plentiful are m ore pron e to obesity, to diabetes and to card io vascu lar disease. A s this b o o k sh o w s, societies w ith higher levels o f incom e in equality have higher levels o f m istrust, illness, status insecurity, violence and other stressors, so the th rifty p h en o­ type m ay w ell be contributing to the high p revalence o f obesity in them.

T H E E Q U A L I T Y D I E T

It is clear that obesity and o verw eigh t are not problem s confined to the p o o r. In the U S A , abou t i z per cent o f the p o pu latio n are p o o r,

but m ore than 7 5 per cent are overw eight. In the U K , social class differences in w o m en ’s obesity can be seen all the w a y up the social ladder. W hile obesity affects on ly 1 6 per cent o f ‘higher m anagerial and p ro fessio n al’ w o m en , just b elo w them , 20 per cent o f low er m an agerial and pro fessio n al w o m en are obese. It’s h ard to argue

in the face o f these facts that the obesity epidem ic is due to p o o r

nutritional k n ow led ge am on g the uneducated. In a study o f middle- aged B ritish w o m e n ,137 84 per cent k n ew they should be eating five fruits and vegetables each day, and another study show ed that obese people are better than thinner people at guessing the calo rie content

o f snack fo o d s .138 A n oth er piece o f evidence that it’ s relative, not absolu te, levels o f

incom e that m atter fo r obesity com es from studies in w h ich people are asked to describe subjectively their place in the social hierarchy. Research ers sh o w subjects a d iagram o f a ladder and tell them that at the top are people w ith the highest status, and at the bottom people w ith the lo w est status, and then ask them to place an ‘X ’ on

the ladder to m ark their o w n standing. It has been show n that this m easure o f subjective social status is linked to an unhealthy pattern o f fat distribu tio n 139 and to o b esity140 - in other w o rd s, obesity w as m ore stron gly related to p e o p le’s subjective sense o f their status than to their actual education or incom e.

If w e can observe that changes in societal incom e inequality are fo llo w e d b y changes in obesity, this w o u ld also be supportive evidence fo r a causal association . A n exam ple o f a society that has experienced a rapid increase in inequality is post-reunification G erm an y. A fter the fall o f the Berlin W all, inequality increased in the

form er E ast G e rm a n y ,141 and there is evidence from studies fo llo w ­ ing people over time that this social disruption led to increases in the body m ass index o f children, yo u n g adults and m o th ers.142

H ealth and social policies fo r obesity treatm ent and prevention tend to focus on the in d ivid u al; these policies try to educate people

abo u t the risks associated w ith being o verw eigh t, and try to coach them into better habits. But these appro ach es o v erlo o k the

reason s w h y people continue to live a sedentary lifestyle and eat an unhealthy diet, h o w these b eh aviours give co m fo rt o r status, w h y there is a social gradient in obesity, h o w depression and stress in pregn an cy p la y a role. B ecause b eh aviou r changes are easier fo r people w h o feel in con trol and in a go o d em otion al state, lessening the burdens o f in equality could m ake an im p ortan t con tribution to w a rd s resolvin g the epidem ic o f obesity.

Educational performance

Our progress as a nation can be no swifter than our pro­

gress in education. The human mind is our fundamental

resource. John F. Kennedy, Special message to the Congress

on Education, 20 February 19 6 1

A cro ss the developed w o rld , and across the p olitical spectrum , ev eryb o d y agrees abo u t the im portance o f education. It’ s g o o d for society, w hich needs the contributions and econom ic prod uctivity - n ot to mention the t a x - o f a skilled w o rk fo rc e , and it’s good fo r in dividuals. People w ith m ore education earn m ore, are more satisfied w ith their w o rk an d leisure tim e, are less likely to be un­ em ployed, m ore likely to be h ealthy, less likely to be crim in als, m ore likely to volunteer their time and vote in election s.143 In 2 0 0 6 , acco rd­ ing to the U S D epartm ent o f L a b o r, if yo u had been to high school but didn ’ t graduate w ith a dip lo m a, you earned an average o f $ 4 1 9

per w eek. T h a t sum rose to $ 5 9 5 if you had the dip lo m a, up to $ 1 , 0 3 9 if y o u ’ d gone on to college and got a b ach elo r’ s degree, and rose to o ver $ 1 , 2 0 0 fo r an advan ced degree.144

T H E H O M E A D V A N T A G E

A lth ou gh g o o d schools m ake a difference, the biggest influence on edu cation al attainm ent, h o w w ell a child perform s in school and later in higher ed ucation, is fam ily back gro u n d. In a report on the

/ .C a r t o o n S t o c k .c o m

n/ r\

future o f education in B ritain , M elissa Benn and Fio n a M illa r

describe h ow :

One of the biggest problems facing British schools is the gap between rich and poor, and the enormous disparity in children’ s home backgrounds and

the social and cultural capital they bring to the educational table.145, p-23

C hild ren do better if their parents have higher incom es and more education them selves, and they do better if they com e fro m hom es w h ere they have a place to study, w here there are reference books an d n ew spapers, and w h ere education is v a lu e d .146 Parental in vo lve­ ment in ch ildren ’ s education is even m ore im portant.

So w h y , w hen all developed societies are com m itted to education and eq u ality o f o pp o rtu n ity (at least in th eory), do disadvan taged children do less w ell at school and miss out on the m yriad benefits o f ed ucation, h ow ever go o d the school system? A s w e shall see, some

societies com e a lot closer to ach ievin g eq u ality o f o pp o rtu n ity than

others.

U N E Q U A L A T T A I N M E N T

Figure 8 .1 show s that international edu cation al scores are closely related to incom e inequality and Figure 8 .2 show s the sam e relation ­ ship fo r the U S A . M o re unequal countries and m ore unequal states have w o rse education al attainm ent - and these relationship s are strong enough fo r us to be sure that they are not due to chance.

C o m p a rab le international data on edu cation al achievem ent come from the Program m e fo r In tern ation al Student Assessm ent (P IS A ), w h ich w a s set up to adm inister standardized tests to 1 5 -year-olds in schools in different countries. T h e program m e began in 43 countries in z o o o , and assesses children every three y ears, typ ically testing betw een 4 ,5 0 0 and 1 0 ,0 0 0 children in each cou n try each time; sch ools are ran dom ly selected. P I S A tests 15 -y e a r-o ld s because they are com ing to the end o f co m p u lso ry education in m ost countries. E ach su rvey gives tests in readin g, m athem atical and scientific literacy. T h e go al is to test h o w w ell children can ap p ly know ledge

and skills.

Incom e inequality

Figure 8 .1 Maths and literacy scores o f 15-year-olds are lower in more unequal countries.148-9

Incom e inequality

Figure 8.z Maths and literacy scores o f eighth-graders are lower in more unequal US states.

F o r consistency w ith the data availab le fo r the U S , w e com bine

n atio n al average scores fo r readin g and m aths on ly and plo t them

against incom e in equality (Figure 8 .1 ) . H o w ev e r, if scientific literacy scores are added in it m akes little difference to the results. N o data w ere a vailab le fo r the U K fro m P I S A Z003, as too few schools agreed to take part in the su rvey to meet the P I S A standards. The

sam e strong international relationship w ith incom e in equality has

been show n fo r adult literacy scores as w ell, using data fro m the In tern ation al A d u lt L iteracy S u rv e y .147

T o exam in e the sam e relationship am ong the fifty states o f the U S A , w e com bined m aths and reading perform an ce scores for

eighth-graders (aged aroun d 1 4 years old) from the U S D epartm ent

o f E du catio n , N a tio n a l Center fo r E du catio n Statistics fo r 20 03 (Figure 8.2). T h e scores are significantly lo w er in states w ith w ider incom e differences.

A s a further test, w e lo o ked at the p ro p o rtion o f children droppin g out o f high school in the U S A . A s Figure 8.3 show s, children are

m uch m ore likely to drop out o f school in m ore unequal states. The

Incom e inequality

Figure 8.3 M ore children drop out o f high school in more unequal US states.

states w ith the lo w est drop-out rates are A la s k a , W yo m in g, U tah, M in n eso ta and N e w H am psh ire, w ith drop-out rates aro un d i z per cent. In three states, M ississip p i, L o u isian a and K en tu cky, m ore than a q u arter o f children drop out o f high school w ith no educational

qualifications. Y o u m ight think that this striking association is due to absolute

p o verty - th at kids drop out o f high school m ore frequently in p o o r states, so that they can start earning soon er and contribute to the fam ily budget. A n d it is true that high school drop-out rates are

higher in p o o r states, but po verty and in equality have independent

effects. P o verty does not exp lain the in equality effect. N o state has a p o verty rate higher than 1 7 per cent but d ro p-o ut rates are above 2.0 per cent in sixteen states and dropp in g out is not confined to the p o o r.

S T A N D A R D S O F P E R F O R M A N C E

It is often assum ed that the desire to raise n ation al standards o f perform ance in fields such as education is quite separate fro m the

desire to reduce edu cation al inequalities w ithin a society. But the

truth m ay be alm ost the opposite o f this. It lo o ks as if the achieve­ ment o f higher n atio n al standards o f edu cation al perform an ce m ay actu ally depend on reducing the social gradient in educational achievem ent in each country. D o u g las W illm s, p ro fesso r o f ed u ca­ tion at the U niversity o f N e w B ru n sw ick , C an a d a , has provid ed

striking illustrations o f th is.150 In Figure 8.4 w e sh o w the relation

between adult literacy scores from the International A d u lt Literacy Survey and their p aren ts’ level o f education - in F inland, Belgium , the U K and the U S A .

This figure suggests that even if y o u r parents are w ell educated -

and so presum ably o f high social status - the cou n try yo u live in

m akes som e difference to y o u r educational success. But fo r those lo w er d o w n the social scale w ith less w ell-educated parents, it m akes a very m uch larger difference. A n im portan t poin t to note, lo o kin g at these fo u r countries, is the steepness o f the social gradient - steepest

in the U S A and the U K w h ere in equality is high, flatter in F inland

L it

er ac

y sc

o re

school school higher

P aren ts’ education

Figure 8.4 Literacy scores in relation to parents’ education in four coun­ tries (data source: International Adult Literacy Survey).

and Belgium , w h ich are m ore equal. It is also clear that an im portant influence on the average literacy scores - on nation al levels o f achievem ent - in each o f these countries is the steepness o f the social

gradient. Th e U S A and U K w ill have lo w average scores, pulled d o w n across the social gradient.

W illm s has dem onstrated that the pattern w e ’ve show n in Figure 8.4 holds m ore w id ely - in tern ation ally am on g tw elve developed countries, as w ell as am ong C an ad ian provin ces and the states o f the U S A .151 A s w ell as the tendency to w a rd s divergence -

larger differences at the bottom o f the so cial gradient than at the top - he says ‘ there is a strong inverse relationship betw een average proficiency levels and the slope o f the socioecon om ic grad ien ts’ .

E p idem iologist A rju m an d Siddiqi and colleagues have also looked at social gradients in reading literacy in 1 5 -year-olds, using data from

P I S A 2 0 0 0 .152 T h ey foun d that countries w ith a long h istory o f w e l­ fare state pro visio n did better and, like W illm s, report that countries w ith higher average scores have sm aller social differences in reading

literacy. Finland and Sw eden h ave high average reading scores and lo w levels o f inequality in readin g scores; G reece and P ortugal have lo w average scores and a high degree o f inequality in reading literacy. Siddiqi and colleagues do, h ow ever, note som e exceptions to this general pattern. N e w Z e a la n d and the U K have high average

reading scores, but a high degree o f social in equality in readin g liter­ acy. O n the other hand N o r w a y com bines a rather m ediocre average score w ith ve ry little socio-econom ic inequality in reading literacy. One ex p lan atio n offered by these researchers is that N e w Z e a la n d and the U K have a greater p ro p o rtion o f children w h o should sit the

tests, but do not, because they have dropped out, or are truants.

E D U C A T I O N A L W E L F A R E

Siddiqi and colleagues em phasize that high reading scores and lo w

social inequalities in readin g literacy are found in nations ‘ m arked by stronger w elfare p ro v isio n s’ . T h is is a poin t w e w ill return to in C h ap ter i z , w h en w e lo o k at public spending on education in relation to incom e inequality. But h o w else m ight incom e inequality

affect education al outcom es? O ne im p ortant connection is likely to be through the im pact

o f in equality on the q u ality o f fam ily life and relationships. Social inequalities in early childhood developm ent are entrenched long before the start o f fo rm al education. W e k n o w a lot n o w abou t the im p ortance o f the early years fo r later developm ent - learning begins at birth an d the first fe w years o f life are a critical period fo r brain

developm ent. T h is early learning can be enhanced or inhibited by the environm ent in w h ich a child g ro w s up. A n ation w id e study in the U K foun d that, by the age o f three years, children from d isad v an ­ taged b ackgro u n ds w ere already ed u catio n ally up to a y ea r behind children from m ore privileged h om es.153

Essential fo r early learning is a stim ulating social environm ent.

B abies an d yo u n g children need to be in carin g, responsive en viro n ­ ments. T h e y need to be talked to, loved and interacted w ith. T h ey need opportunities to p lay , talk an d exp lo re their w o rld , an d they need to be encouraged w ithin safe lim its, rather than restricted in

their activities or punished. A ll o f these things are harder fo r parents and other care-givers to p ro vid e w h en they are p o o r, or stressed, or unsupported.

In C hapter 4 w e described h o w the general q u ality o f social relationship s is lo w er in m ore unequal societies, and in C hapters 5

and 6 w e show ed h o w inequality is linked to p o o r ph ysical and

m ental health and m ore substance abuse. It’ s not a great leap then to think h o w life in a m ore hierarch ical, m istrustful society m ight affect intim ate, dom estic relationship s and fam ily life. D om estic conflict and violence, paren tal mental illness, p o verty o f time and resources

w ill all com bine to affect child developm ent. The results o f these

stresses can perhaps be seen in an an alysis by econom ists R o b ert F ran k and A d am Levin e, o f C o rn ell U niversity. T h ey show ed that in the United States, counties that had the largest increases in incom e inequality w ere the sam e counties that experienced the largest rises

in divorce ra tes.154 Children living in low -incom e fam ilies experience

m ore fam ily conflict and disruption and are m ore likely to w itness or experience violence, as w ell as to be living in m ore crow d ed , noisy an d substan dard h ou sin g155 - the q u ality o f the hom e environm ent is directly related to in com e.156 T h e w a y parents behave in response to relative p o verty m ediates its im pact on children - there is evidence

that som e fam ilies are resilient to such p roblem s, w h ile others react

w ith m ore punitive an d unresponsive parenting, even to the extent o f becom ing neglectful or a b u siv e .157-8 It is im portan t, once again, to note that difficulties in fam ily relationship s and paren tin g are not confined to the p o o r. So cio lo gist A nnette L areau describes h ow paren tin g differs betw een m iddle-class, w o rk in g -class and poor fam ilies in A m erica: there are k ey differences in the organ izatio n o f daily life, the use o f lan guage, and the degree to w h ich fam ilies are so cially con n ected.159 W e h ave foun d that w ithin the U K M illen n ium C o h o rt Study, a large survey o f children born in 2 0 0 0 and 2 0 0 1 ,

even m others in the second fro m the top social class gro u p are m ore likely to report feeling incom petent as a paren t or h avin g a p o o r relationship w ith their children, com pared to those in the topm ost group.

Societies can do a lot to am eliorate the stresses on fam ilies and to su pp o rt early ch ildh ood developm ent. F ro m the very start o f life,

some societies do m ore than others to prom ote a secure attachm ent

between m other an d in fan t through the pro visio n o f p aid m aternity leave fo r m others w h o w o rk . U sing data on the duration o f paid

m aternity leave, provid ed by the C learin gh ou se on International D evelopm ents in C h ild , Y o u th and F am ily Policies at C olu m b ia U n iversity, w e foun d that m ore equal countries pro vid ed longer periods o f p aid m aternity leave.

Sweden provid es paren tal leave (which can be divided between m others and fathers) w ith 80 per cent w age replacem ent until the child is 1 8 m onths old; a further three m onths can be taken at a flat rate o f p ay, an d then another three m onths o f unpaid leave on top o f that. N o r w a y gives parents (again either m other or father) a year o f

leave at 80 per cent w a g e replacem ent, or fo rty -tw o w eek s at 10 0 per cent. In con trast, the U S A and A u stralia p ro vid e no statutory entitlement to paid leave - in A u stra lia parents can have a year o f unpaid leave, in the U S A , tw elve w eeks.

A s w ell as allo w in g paren tal leave, societies can im p rove the qu ality o f early ch ildh ood through the pro visio n o f fam ily a llo w ­ ances and t a x benefits, social h ousing, health care, program m es to prom ote w ork/life balan ce, en forcin g child su pport paym ents and,

perhaps m ost im p ortan tly, through the pro visio n o f h igh -quality early ch ildh ood education. E a rly ch ildh ood education program m es can foster ph ysical an d cognitive developm ent, as well as social and em otion al developm en t.160' 62 T h ey can alter the long-term

trajectories o f ch ildren ’s lives, and cost-benefit analyses sh o w that

they are high-yield investm ents. In experim ents, disadvan taged children w h o have received h igh -quality early ch ildh ood education are less likely to need rem edial education, less likely to becom e in volved in crim e, and they earn m ore as a d u lts.160 A ll o f this

adds up to a substantial return on governm ent investm ents in such

program m es.

U N E Q U A L L E A R N I N G O P P O R T U N I T I E S

So far w e have described w a y s in w h ich greater inequality m ay affect ch ildren ’s developm ent through its im pact on fam ily life and relationships. But there is also evidence o f m ore direct effects o f in eq uality on ch ildren ’ s cognitive abilities and learning.

In 2 0 0 4 , W orld B an k econom ists K a rla H o ff and P riyan ka Pandey reported the results o f a rem arkab le exp erim en t.163 T h ey to o k 3 2 1 high-caste and 3 2 1 low -caste r i to 12 -y e a r -o ld boys fro m scattered ru ral villages in In dia, and set them the task o f solving mazes. First, the b o ys did the puzzles w ith o u t being a w a re o f each o ther’s caste. U nder this condition the low -caste boys did just as w ell w ith the

m azes as the high-caste boys, indeed slightly better. Th en , the experim ent w a s repeated, but this time each b o y w a s

asked to confirm an announcem ent o f his nam e, village, fath er’ s and g ran d fath er’ s nam es, and caste. A fte r this public announcem ent o f caste, the boys did m ore m azes, and this tim e there w a s a large caste

gap in h o w w ell they did - the perform an ce o f the low -caste boys d ropped significantly (Figure 8.5).

T h is is striking evidence that perform an ce and b eh aviou r in an education al task can be p ro fo u n d ly affected b y the w a y w e feel w e are seen and judged b y others. W hen w e expect to be view ed as in ferio r, our abilities seem to be diminished.

T h e sam e phenom enon has been dem onstrated in experim ents w ith w h ite and b lack high-school students in A m erica, m ost co n vin c­ in gly by social psych ologists C lau de Steele at Stan fo rd U niversity, and Jo s h u a A ro n so n at N e w Y o r k U n iversity.164 In one study they adm inistered a standardized test used fo r college students’ adm ission

to grad uate program m es. In one condition, the students w ere told

that the test w a s a m easure o f ability; in a second condition, the students w ere told that the test w a s n o t a m easure o f ability. The w h ite students perform ed eq u ally under both conditions, but the black students perform ed m uch w o rse w hen they thought their ab il­

ity w a s being judged. Steele and A ro n so n labelled this effect ‘stereo­ type th reat’ and it’ s n o w been show n that it is a general effect, w hich applies to sex differences as w ell as racial and ethnic differen ces.165

N u

m b

er

of

m az

es

so lv

ed

6

Figure 8.5 The effect o f caste identity on perform ance in Indian school boys. 163

D espite the w o rk w e m entioned on so cial an xie ty and the effects

o f being judged n egatively w h ich w e discussed in C h ap ter 3 , it is

perhaps surprising h o w easily stereotypes and stereotype threats are

established, even in artificial conditions. Ja n e E llio tt, an A m erican schoolteacher, conducted an experim ent w ith her students in 19 6 8 , in an effo rt to teach them abo u t racial in equality and in ju stice.166 She told them th at scientists had show n that people w ith blue eyes

w ere m ore intelligent an d m ore lik ely to succeed than people w ith b ro w n eyes, w h o w ere lazy and stupid. She divided her class into blue-eyed and brow n -eyed gro u p s, and gave the blue-eyed grou p extra privileges, praise and attention. T h e blue-eyed grou p q u ickly asserted its su p eriority over the b row n -eyed children, treating them contem ptuously, and their school perform an ce im p roved. The b row n -eyed grou p just as q u ick ly adopted a subm issive tim idity, and their m arks declined. A fte r a fe w d ays, E llio tt told the children she had go t the in form ation m ixed up and that actu ally it w a s b row n eyes that indicated superiority. T h e classroo m situation rap id ly reversed.

HI H igh caste ] L o w caste

■■I I

|___ | C aste C aste

unannounced announced

N e w developm ents in n eu ro lo gy provid e b iological explan ation s fo r h o w our learning is affected b y our feelin gs.167 W e learn best in

stim ulating environm ents w h en w e feel sure w e can succeed. W hen w e feel h app y or confident ou r brains benefit from the release o f dopam in e, the re w a rd chem ical, w h ich also helps w ith m em ory, attention and prob lem solvin g. W e also benefit fro m serotonin w hich im proves m ood, and fro m adrenaline w h ich helps us to perform at ou r best. W hen w e feel threatened, helpless and stressed, our bodies are flooded by the h orm one co rtiso l w h ich inhibits ou r think­

ing and m em ory. So inequalities o f the kind w e have been describing in this chapter, in society and in ou r schools, have a direct and dem onstrable effect on our brains, on ou r learning and educational achievem ent.

D I F F E R E N T S T R O K E S F O R D I F F E R E N T F O L K S

A n oth er w a y in w h ich in equality directly affects educational achievem ent is through its im pact on the asp iration s, norm s and

valu es o f people w h o find them selves lo w er d o w n the social hier­ archy. W hile education is view ed by the m iddle class and by teachers and p o licy m akers as the w a y u p w ard s and o u tw ard s fo r the poor and w o rk in g class, these values are not a lw ay s subscribed to by the p o o r and w o rk in g class them selves.

In her 1 0 0 6 b o ok E d u c a tio n a l F a ilu re a n d W o rk in g C lass W hite C h ild re n in B rita in , an th ro po lo gist G illian E van s describes the w o rk in g -class culture o f B erm on dsey, in east L o n d o n .168 She show s h o w the kinds o f activities expected o f children in schools fit w ith the w a y m iddle-class parents expect their children to p la y and inter­ act at hom e, but clash w ith the w a y in w h ich w o rk in g -class fam ilies care fo r, and interact w ith, their children. T o a degree, w o rk in g-class people resist the im position o f education and m iddle-class values, because becom ing educated w o u ld require them to give up w a y s o f being that they valu e. O ne w o m an tells E van s that being ‘co m m o n ’ m eans ‘ k n o w in ’ ’ o w to ’ ave a go o d laugh ’cos y o u ’ re not stuck u p ’ . T h e things that the w o m en she describes like to talk abo u t are their

fam ilies, their health, w o r k and w a y s to get m oney, h o u sew o rk , relation ship s, shopping, sex and gossip. T a lk in g abou t abstract ideas, bo ok s and culture, is seen as posh and pretentious. T h e ch il­ dren o f these w o rk in g -class m others are con strain ed b y m inim al rules in their hom es. E van s describes children w h o are allo w ed to eat

and drink w h a t they like, w hen they like; to sm oke at hom e; to do

h om ew ork or not, as they please. ‘ I f they w an t to learn, they w ill, if they d o n ’ t, they w o n ’t and th at’ s th a t.’ O f course these fam ilies w an t the best fo r their children, but that ‘ best’ isn ’ t a lw ay s ‘ education, education, ed u catio n ’ .

T h at p o o r and w o rk in g -cla ss children resist fo rm al education and

m iddle-class values does not, o f course, m ean that they have no asp iration s or am bitions. In fact, w hen w e first looked at data on ch ildren ’s asp iration s fro m a U N I C E F report on ch ildh ood w e ll­ bein g,110 w e w ere surprised at its relation ship to incom e inequality (Figure 8.6). M o re children reported lo w asp iration s in m ore equal

countries; in unequal countries children w ere m ore likely to have high asp iration s. Som e o f this m ay be accounted fo r by the fact that in m ore equal societies, less-skilled w o r k m ay be less stigm atized,

Incom e inequality

Figure 8.6 Aspirations o f 1 5 -year-olds and inequality in rich countries.

in com parison to m ore unequal societies w h ere career choices are dom inated by rather star-struck ideas o f financial success and im ages o f glam o u r and celebrity.

In m ore unequal countries, w e found a larger gap between asp iration s and actu al opportunities and expectation s. If w e com ­ p are Figure 8 .1 on maths and reading scores in different countries to Figure 8.6, it is clear that aspiration s are higher in countries w h ere educational achievem ent is lo w er. M o re children m ight be asp irin g to higher-status jo b s, but few er o f them w ill be qualified to get them. If in equality leads to unrealistic hopes it m ust also lead

to disappointm ent. G illian E van s quotes a teacher at an inner-city p rim ary school,

w h o summed up the corrosive effect o f inequality on children:

These kids don’t know they’re working class; they w on’t know that until

they leave school and realize that the dreams they’ve nurtured through childhood can’t come true.168

In the next tw o chapters w e ’ ll sh o w h o w yo u n g w o m en and you ng men in m ore unequal societies respond to their lo w social status, and in C h ap ter 1 1 w e ’ ll return to the theme o f education and life chances

w h en w e exam ine the im pact o f inequality on social m obility.

"Is this birds and bees chat going to take long? I'm late fo r my pre-natal class. "

Teenage births: recycling deprivation

Just saying ‘N o ’ prevents teenage pregnancy the w ay ‘Have a nice day’ cures chronic depression.

Faye Wattleton, Conference speech, Seattle, 1988

In the sum m er o f z o o 5, three sisters hit the headlines o f B ritain ’ s tabloid n ew spapers - all three w ere teenage m others. T h e youngest w a s the first o f the girls to becom e pregnant and had her b ab y at the age o f 1 2 . ‘W e w ere in bed at m y m um ’ s house m essing around and sex ju st sort o f h appen ed ,’ she said; ‘ I did n ’t tell anyone because I w a s too scared and didn ’ t k n o w w h a t to do . . . I w ish it had h appened to som eone else.’ 169 Soon after, the n ext older sister had a b ab y at age 1 4 . ‘It w a s just one o f those things. I thought it w ou ld never happen to m e,’ she said. ‘ A t first I w an ted an abo rtion because I d id n ’t w a n t to be like [my sister], but I co u ld n ’t go through w ith it.’ T h e oldest sister, the last o f the girls to find out she w a s pregnant, gave birth aged 1 6 ; unlike her sisters she seemed to w elcom e m other­ h ood. ‘I left school . . . as I w a s n ’t really interested,’ she adm itted, ‘all m y friends w ere h avin g babies and I w an ted to be a m um , t o o ’ . A t the time their stories becam e new s, the girls w ere all living at hom e w ith their m other, sharing their b edroom s w ith their babies, the youngest tw o struggling w ith sch ool, and all three tryin g to get by on social security benefits. W ith no qualification s and no support fro m the fathers o f their babies, their futures w ere bleak. M ed ia

com m entators and m em bers o f the public w ere qu ick to condem n the sisters and their m other, p o rtra y in g them as feckless scroungers. ‘ M eet the kid sisters . . . benefit b o n an za’ . . . ‘ G irls’ babies are the

real vic tim s,’ exclaim ed the n ew sp ap e rs.170-71 T h eir m other blam ed the lack o f sex education in school.

W H Y I T M A T T E R S

Th e press furore brings society’s fears and concerns aroun d teenage m otherhood into sharp focus. O ften described as ‘ babies having bab ies’ , teenage m oth erh ood is seen as b ad fo r the m other, b ad fo r

the b ab y and bad fo r society.

Th ere is no doubt that babies born to teenage m others are m ore

likely to h ave lo w birthw eight, to be born prem aturely, to be at higher risk o f dying in in fan cy and, as they g ro w up, to be at greater risk o f educational failu re, juvenile crim e an d becom ing teenage parents them selves.172-3 G irls w h o give birth as teenagers are m ore

likely to be p o o r and uneducated. But are all the b ad things a sso ci­

ated w ith teenage birth really caused b y the age o f the m other? O r are they sim ply a result o f the cultural w o rld in w h ich teenage mothers give birth?

T h is issue is h otly debated. O n the one h and, som e argue that teenage m otherhood is n o t a health prob lem because yo u n g age is

not in itself a cause o f w o rse o u tco m es.174 In fact, am on g p o o r A frican -A m erican s, cum ulative exp o su re to p o verty and stress across their lifetim es com prom ises their health to such an extent that their babies do better if these w om en have their children at a yo u n g

a g e.175-6 T h is idea is k n o w n as ‘w eath erin g’ an d suggests that, fo r

p o o r and disadvan taged w o m en , p ostpon in g pregn an cy until later ages doesn ’ t actu ally m ean that they have healthier babies. O thers have sh o w n that the children o f teenage m others are m ore likely to end up excluded fro m m ainstream society, w ith w o rse ph ysical

and em otion al health and m ore deprivation . T h is is true even after

takin g accou n t o f other ch ildh ood circum stances such as social class, education, w h eth er the parents w ere m arried or not, the p aren ts’ person alities, and so o n .177 But although w e can som etim es separate out the influences o f m aternal age and econom ic circum stances in research studies, in real life they often seem in extricab ly intertwined

and teenage m otherhood is associated w ith an inter-generational cycle o f d ep rivatio n .178

B ut h o w ex actly are yo u n g w o m en ’s in d ividu al experiences and choices - their personal choices abo u t sleeping w ith their boyfriends, choices around con traception and abo rtion , choices abou t qu alifica­ tions and careers, shaped by the society they live in? Lik e the issues discussed in earlier chapters, the teenage birth rate is stron gly related to relative deprivation and to inequality.

B O R N U N E Q U A L

There are social class differences in both teenage conceptions and births but the differences are smaller for conceptions than for births, because middle-class young women are more likely to have abortions. Teenage birth rates are higher in communities that also have high divorce rates, low levels of trust and low social cohesion, high unemployment, poverty, and high crime rates. 173 It has been suggested by others that teenage motherhood is a choice that women make when they feel they have no other prospects for achieving the social credentials of adulthood, such as a stable intimate relationship or rewarding employment. 179 Sociologist Kristin Luker claims that it is ‘the discouraged among the disadvantaged’ who become teenage mothers. 180

But it is im p ortant to rem em ber that it isn ’ t on ly p o o r you ng

w o m en w h o becom e teenage m others: like all the problem s w e have

looked at, inequality in teenage birth rates runs right across society. In Figure 9 . 1 , w e sh o w the percentage o f yo u n g B ritish w om en w h o becom e teenage m others in relation to household incom e. E ach year alm ost 5 per cent o f teenagers livin g in the p oorest q u arter o f homes have a first b ab y, fo u r times the rate in the richest quarter. But even

in the second richest qu arter o f households the rate is double that o f the richest q u arter (2.4 per cent and 1 . 2 per cent). Sim ilar patterns are seen in the United States. A lth ou gh m ost o f these births are to o lder teenagers, aged 1 8 - 1 9 y ears, the pattern is evident, an d even stronger, fo r the 1 5 - 1 7 - y e a r - o ld s .

_G

I 4

c u H 2

4 .8 %

''

■ i ®

2 .9 %

z .4 %

i.z%

________1 Poorest quarter

H ou seh old incom e

R ichest quarter

Figure 9 .1 There is a gradient in teenage birth rates by household income, from poorest to richest.m

•h— -—----- ---------------------------------------------------------------------- L o w H igh

Incom e inequality

Figure 9.2 Teenage birth rates are higher in more unequal countries.185

Figure 9 .2 sh o w s that the international teenage birth rates pro vid ed by U N I C E F 182 are related to incom e inequality and Figure

9 .3 show s the sam e relationship fo r the fifty states o f the U S A , using teen pregn an cy rates fro m the U S N a tio n a l V ita l Statistics System 183 and the A lan G u ttm ach er Institute.184 T h ere is a strong tendency for m ore unequal countries and m ore unequal states to have higher teenage birth rates - m uch too strong to be attributable to chance.

T h e U N I C E F repo rt on teenage births show ed that at least one

and a quarter m illion teenagers becom e pregnant each y ea r in the rich O E C D countries and abo u t three-quarters o f a m illion go on to becom e teenage m o th ers.182 T h e differences in teen birth rates betw een countries are striking. T h e U S A and U K top the charts.

A t the top o f the league in our usual grou p o f rich countries, the

U S A has a teenage birth rate o f 5 2 . 1 (per 1,0 0 0 w om en aged 1 5 - 1 9 ) , m ore than fo u r times the E U average and m ore than ten times higher than that o f Ja p a n , w h ich has a rate o f 4.6.

Rachel Gold and colleagues have studied income inequality and teenage births in the USA, and shown that teen birth rates are

Incom e inequality

Fgure 9.3 Teenage pregnancy rates are higher in more unequal US states.

highest in the m ost u n equ al, as w ell as the m ost relatively deprived counties. She also reported that the effect o f in equality w a s strongest fo r the youngest m others, those aged 1 5 - 1 7 y e a r s .186 F o r the US states, w e sh o w data fo r live births an d abo rtion s com bined. T h ere

are substantial differences in pregn an cy rates betw een U S states.

M ississipp i has a rate close to tw ice that o f Utah. W e m ight expect patterns o f conceptions, abo rtion s and births to

be influenced by factors such as religion and ethnicity. W e’ d expect predom in an tly C ath o lic countries to have high rates o f teenage births, because o f lo w rates o f abo rtion . But, w h ile predom in an tly

C ath o lic Portugal and Ireland have high rates that w o u ld indeed

fit this alternative exp lan atio n , Italy and Spain have unexpectedly lo w rates, although they are also predom in an tly C ath o lic. W ithin countries, different ethnic grou ps can have different cultures and values aroun d sex u ality , con traception , a b o rtion , ea rly m arriage

and w o m en ’ s roles in society. In the U S A , fo r exam p le, H isp an ic

and A frican -A m erican girls are alm ost tw ice as lik ely to be teenage m others as w hite girls, and in the U K sim ilarly, co m p aratively high rates are seen in the B angladeshi and C arib b ean com m u n ities.182 But, because these com m unities are m in o rity p o p u latio n s, these d iffer­

ences d o n ’t actu ally have much im pact on the ran k in g o f countries

and states by teenage pregnancy or birth rates, and so d o n ’t affect our interpretation o f the link w ith inequality.

But hidden w ithin the sim ple relationships revealed in Figures 9 .2 and 9 .3 are the real-life com plexities o f w h a t it m eans to be a teenage m other in an y particu lar cou n try. F o r exam p le, in Ja p a n ,

Greece and Italy, m ore than h a lf o f the teenagers giving birth are

m arried - in fact in Ja p a n , 86 per cent o f teen m others are m arried, w hereas in the U S A , the U K and N e w Z e a la n d , less than a quarter o f these m others are m a rrie d .182 So not only do these latter countries have higher o verall rates o f teen births, but those births are m ore

likely to be associated w ith the b ro ad range o f health and social

problem s that w e think o f as typical consequences o f early m oth er­ h ood - problem s that affect both the m other and the child. W ithin the U S A , H isp an ic teenage m others are m ore likely to be m arried than those fro m other ethnic gro u p s, but they are also m ore likely to

be p o o r;187-8 the sam e is true fo r B angladeshis in the U K .

So w h a t do w e k n o w abo u t w h o becom es a teenage m other that can help us understand this p articu lar effect o f inequality?

T H E F A S T L A N E T O A D U L T H O O D

Interestingly, there is not m uch o f a connection between teenage

birth rates and birth rates fo r w o m en o f a ll ages in rich countries.

T h e m ost unequal countries, the U S , U K , N e w Z e a la n d and

P o rtu gal, have much higher teenage birth rates relative to older w o m en ’ s birth rates than the m ore equal countries, such as Ja p a n , Sw eden, N o r w a y and F inland, w hich have teenage birth rates

that are low er relative to the rates o f birth o f older w o m e n .182 So

w h atever drives teenage birth rates up in m ore unequal countries is unconnected w ith the facto rs driving o verall fertility. U nequal societies affect teenage ch ildbearin g in particu lar.

A report from the R o w n tree Fo un d atio n called Y o u n g P e o p le ’s C h a n g in g R o u te s to In d e p e n d e n c e , w h ich com pares h o w children born in 1 9 5 8 and 1 9 7 0 g re w up, describes a ‘w iden ing gap between

those on the fast and the slo w lanes to a d u lth o o d ’ .189 In the slow lane, yo u n g people born into fam ilies in the higher socio-econom ic classes spend a long time in education and career train in g, putting o ff m arriage and childbearing until they are established as successful adults. F o r yo u n g people on the fast track , truncated education often

leads them into a disjointed pattern o f unem ploym ent, lo w -p aid w o rk and training schem es, rather than an ordered, u p w ard career trajectory.

A s sociologists H ila ry G ra h a m and Elizabeth M cD erm o tt point out, teenage m oth erh ood is a p a th w a y through w h ich w om en

becom e excluded from the activities and connections o f the w id er society, and a w a y in w hich generations becom e trapped by in­ e q u a lity .190 But as w ell as the constraints that relative p overty im poses on life chances fo r yo u n g people, there seem to be additional reasons w h y teenage m oth erh ood is sensitive to degrees o f inequality in society.

E A R L Y M A T U R I T Y A N D A B S E N T F A T H E R S

T h e first o f these add itio n al reasons w a s touched on in C h ap ter 8,

w h ere w e discussed the im pact o f inequality on fam ily relationships

and stress in early life. Exp erien ces in early ch ildh ood m ay be ju st as relevant to teenage m otherhood as the education al and econom ic opportunities availab le to adolescents. In 1 9 9 1 , p sych ologist J a y B elsk y at the U niversity o f Lo n d o n and his colleagues p ro po sed a theory, based on evolu tio n ary p sy ch o lo g y, in w h ich experiences in

early childhood w o u ld lead in d ividu als to w ard s either a q u a n tity o r a q u a lity reproductive strategy, depending on h o w stressful their early experiences had b een .191 T h ey suggested that peop le w h o learned, w h ile gro w in g up, ‘ to perceive others as u n tru stw o rth y, relationships as o pportu n istic and self-serving, and resources as

scarce and/or u n predictable’ w o u ld reach b io lo g ical m atu rity earlier,

be sex u ally active earlier, be m ore lik ely to fo rm short-term relatio n ­ ships and m ake less investm ent in parenting. In con trast, people w h o g ro w up learning ‘ to perceive others as tru stw o rth y, relationships as enduring and m utually rew ard in g and resources m ore or less

constantly a v a ila b le ’ w o u ld m ature later, defer sexu al activity, be

better at form in g long-term relationships and invest m ore h eavily in

their children’ s developm ent. In the w o rld in w h ich hum ans evolved, these different strategies

m ake sense. I f you ca n ’t rely on y o u r m ate or other people, and you

ca n ’ t rely on resources, then it m ay once have m ade sense to get

started early and have lots o f children - at least som e w ill survive.

But if yo u can trust y o u r partner and fam ily to be com m itted to you and to provid e fo r y o u , it m akes sense to have few er children and to devote m ore attention and resources to each one.

R ach el G o ld and colleagues foun d that the relationship between

in equality and teenage birth rates in the U S A m ight be acting

through the im pact o f inequality on social cap ital, w h ich we discussed in C h ap ter 4 .192 A m o n g U S states, that is, those w ith low er levels o f so cial cohesion, civic engagem ent an d m utual trust - exactly

the conditions w h ich m ight fa v o u r a quan tity strategy - teenage birth rates are higher.

Several studies have also sh o w n that ea rly conflict and the absence o f a fath er d o predict earlier m atu ration - girls in such a situation becom e p h ysically m ature an d start their periods earlier than girls w h o g ro w up w ith o u t those sources o f stress.193-4 A n d reaching

pu b erty earlier increases the likelih ood o f girls becom ing sexu ally

active at an early age and o f teenage m o th erh o o d .195 Father absence m ay be p articu larly im p ortan t fo r teenage preg­

n ancy. In a study o f tw o large sam ples in the U S A and N e w Z e a la n d , p sych ologist Bruce Ellis and his colleagues fo llo w ed girls

from ea rly childhood through to a d u lth o o d .196 In both countries,

the longer a fath er w a s absent fro m the fam ily , the m ore likely it

w a s that his daughter w o u ld have sex at a yo u n g age an d becom e a teenage m other - and this strong effect could not be exp lain ed a w a y b y b eh aviou ral problem s o f the girls, by fam ily stress, parenting style, socio-econom ic status, or by differences in the neighbourhoods

in w h ich the girls grew up. So there m ay be deep-seated adaptive processes w hich lead from m ore stressful and unequal societies - perhaps p a rticu la rly from lo w social status - to higher teenage birth rates. U n fo rtu n ately, w h ile w e can obtain international data on single-parent households, being a single parent m eans very different

things in different countries, and there are no international data that

tell us h o w m any fathers are absent from their ch ildren ’s lives.

W H A T A B O U T T H E D A D S ?

T h ro u gh o u t this chapter, w e ’ve been discussing the prob lem o f teenage parenting exclu sively in term s o f teenage m others, but w h at abo u t the fathers? L e t’s return to the story o f the three sisters. The father o f the 12 -y e a r -o ld g ir l’ s b ab y left her sh o rtly after his son w as born. T h e boy nam ed by the m iddle sister as the father o f her little girl denied h avin g sex w ith her and dem anded a patern ity test. And the 38-year-o ld father o f the oldest sister’s b ab y already had at least fo u r other children.

Sociologists G ra h a m and M cD erm o tt discuss w h at has been learned fro m studies w h ere researchers talk at length to yo u n g w om en abo u t their experiences. W h at they sh o w is that these sisters’

experiences w ith their bab ies’ fathers are ty p ic a l.190 M o th erh o o d is a w a y in w h ich yo u n g w om en in deprived circum stances join adult social n etw o rk s - n etw o rk s w h ich u su ally include their o w n mothers and other relatives, and these supportive n etw orks help them transcend the social stigm a o f being a teenage m other. A ccord in g to G ra h a m and M cD erm o tt, yo u n g w om en prioritize their relatio n ­ ships w ith the babies, o ver their often difficult relationships w ith the

bab ies’ fathers, because they feel this relationship is a ‘ m ore certain source o f intim acy than the heterosexu al relationships they had . . . exp erien ced’ .

Y o u n g men living in areas o f high unem ploym ent and lo w w ages often ca n ’ t o ffer m uch in the w a y o f stability o r support. In co m ­

munities w ith high levels o f teenage m oth erh ood, yo u n g men are

them selves tryin g to cope w ith the m an y difficulties that inequality inflicts on their lives, and yo u n g fath erh o o d adds to those stresses.

I O

Violence: gaining respect

Where justice is denied, where poverty is enforced, where ignorance prevails and where any one class is made to

feel that society is in an organized conspiracy to oppress, rob and degrade them, neither persons nor property will be safe.

Frederick Douglas, Speech on the 24th anniversary of emancipation, Washington, D C , 18 8 6

A s w e began to w rite this chapter, violence w a s in the headlines on both sides o f the A tlan tic. In the U S A , an 1 8-year-old m an w ith a shotgun entered a shopping m all in Salt L a k e C ity , U tah, killin g five

people and w o u n d in g fo u r others, apparen tly at ran dom , before

being shot dead by police. In the U K , there w a s a w a v e o f killings in South L o n d o n , including the m urder o f three teenage b o ys in less than a fortnight. But perhaps the story that best illustrates w h at this chapter is abo u t occurred in M a rc h 2 0 0 6 , in a quiet suburb o f C in cin n ati, O hio. C harles M a rtin , a 66 -year-o ld, telephoned the

em ergency services.197 ‘I just killed a k id ,’ he told the o p erato r, ‘ I shot him w ith a god dam n 4 1 0 shotgun tw ice .’ M r M artin had shot his 15 -y e a r-o ld neighbour. Th e b o y ’ s crime? H e had run across M r M a r tin ’ s law n . ‘ K id ’s just been givin g me a bunch o f shit, m akin g the other kids harass me and my p la c e .’

V io len ce is a real w o rr y in m an y p eop le’ s lives. In the m ost recent British C rim e Surveys, 35 per cent o f peop le said they w ere very w o rried or fairly w o rried abo u t being a victim o f m ugging, 3 3 per cent w o rried abo u t physical attack , 24 per cent w o rried abo u t rape,

" A n d f i n a l l y , w o u l d y o u s a y y o u r f e a r o f c r i m e h a d i n c r e a s e d ? "

and 1 3 per cent w o rried abo u t ra cia lly m otivated violence. M o re

than a quarter o f the people w h o responded said they w ere w o rried abo u t being insulted or pestered in p u b lic.198 Surveys in A m erica and A u stra lia repo rt sim ilar findings - in fact fear o f crim e an d violence m ay be as big a problem as the actual level o f violence. V e ry few

people are victim s o f violent crim e, but fear o f violence affects the

q u ality o f life o f m an y m ore. F ear o f violence d isp ro po rtio n ately affects the vulnerable - the p o o r, w om en and m in ority g ro u p s.199 In m any places, w o m en feel n ervous goin g out at night or com ing hom e late; old people dou ble-lock their d oors and w o n ’t open them to strangers. These are im p ortan t infringem ents o f basic hum an

freedom s. P eo p le’ s fears o f crim e, violence and an ti-social b eh aviou r d o n ’t

a lw a y s m atch up w ith rates an d trends in crim e and violence. A recent do w n -sw in g in hom icide rates in A m erica (which has n o w ended), w a s not m atched b y a reduction in peop le’s fear o f violence. W e w ill return to recent trends later. First, let’ s turn ou r attention

to differences in rates o f actu al violence betw een different societies an d lo o k at som e o f the sim ilarities and the differences between them.

In som e w a y s patterns o f violence are re m ark ab ly consistent acro ss time and space. In different places an d at different times, v io ­ lent acts are overw h elm in gly perpetrated by m en, and m ost o f those men are in their teens or early twenties. In her b o o k , T h e A n t a n d the P e a c o c k , ph iloso p h er an d evo lu tio n ary p sych o lo g ist H elena C ro n in show s h o w clo sely correlated the age an d sex characteristics

o f m urderers are in different places.200 W e reproduce her graph

sh o w in g m urder rates, co m parin g C h icag o w ith E n glan d and W ales (Figure 1 0 . 1 ) . T h e age o f the perp etrator is show n alon g the bottom ; up the side is the m urder rate, and there are separate lines fo r men and w om en . It is im m ediately apparen t that m urder rates p e ak in the late teens and early tw enties fo r men, an d that rates fo r w o m en are m uch lo w er at all ages. T h e age an d sex distribution is astonishingly

sim ilar both in C h icag o and in E n glan d and W ales. H o w ev e r, w h at is less o b viou s is that the scales on the left- and right-hand sides o f the graph are very different. O n the left-hand side o f the graph, the scale show s hom icide rates per m illion people in E n glan d and

England an d W ales C h icago

- 9 0 0 - 8 0 0 - 7 0 0 - 6 0 0

- 5 0 0 - 4 0 0 - 3 0 0 - 2 0 0 - 1 0 0 -o

,s°> y * I ? t f ' 1 ? & t? S1* b*' <o°> A°l tblx^ y ./ y ./v y .s~ y * .sk j >./> ./■ .a

A ge o f perpetrator (years)

Figure ro .r Homicides by age and sex o f perpetrator. England and Wales com pared with Chicago.100

W ales, goin g from zero to 3 0 . O n the right-hand side, the scale

show s hom icide rates in C h icag o , an d here the scale runs from zero to 9 0 0 m urders per m illion. D espite the strikin g sim ilarities in the patterns o f age and sex distribution , there is som ething fun d am en ­ tally different in these places; the city o f C h icag o had a m urder rate

3 0 times higher than the rate in E n glan d and W ales. O n top o f the

bio lo gical sim ilarities there are huge environm ental differences. V io len t crim es are alm ost u n k n ow n in som e societies. In the U S A ,

a child is killed by a gun every three hours. D espite h avin g a m uch lo w er rate than the U S A , the U K is a violent society, com pared to m an y other countries: o ver a m illion violent crim es w ere recorded in

2 0 0 5 - 2 0 0 6 . A n d w ithin an y society, w h ile it is gen erally yo u n g men

w h o are violen t, m ost yo u n g men are not. Ju s t as it is the discou r­ aged an d disadvan taged am o n g yo u n g w om en w h o becom e teenage m others, it is p o o r yo u n g men fro m disadvan taged n eighbourhoods w h o are m ost likely to be both victim s and perpetrators o f violence. W hy?

‘ i f y o u a i n ’ t g o t p r i d e , Y O U G O T N O T H I N G . ,201’ p 29

Ja m e s G illigan is a psych iatrist at H a rv a rd M ed ical S ch ool, where

he directs the Center fo r the Study o f V iolen ce, and has w o rk e d on

violence prevention fo r m ore than thirty years. H e w a s in charge o f

m ental health services fo r the M assach usetts prison system fo r m any years, and for m ost o f his years as a clinical psych iatrist he w o rk e d w ith the m ost violent o f offenders in prison s and prison m ental h ospitals. In his b o ok s, V io le n c e 202 and P re v e n tin g V io le n c e ,201 he

argues that acts o f violence are ‘attem pts to w a rd o ff or elim inate the feeling o f sham e and h um iliation - a feeling that is p ain fu l, and can even be intolerable and overw h elm in g - and replace it w ith its opposite, the feeling o f p rid e’ . T im e after time, w h en talk in g to men w h o had com m itted violen t offences, he discovered that the triggers

to violence had in volved threats - or perceived threats - to pride,

acts that instigated feelings o f h um iliation o r sham e. Som etim es the incidents that led to violence seemed in credibly trivial, but they all evoked sham e. A yo u n g neighbour w a lk in g disrespectfully across y o u r im m aculate law n . . . the p o p u lar kids in the school harassing

yo u and calling yo u a faggo t . . . being fired fro m y o u r jo b . . . your

w o m an leaving yo u fo r another m an . . . som eone lo o kin g at you ‘fu n n y’ . . .

G illigan goes so far as to say that he has ‘yet to see a serious act o f violence that w a s not p ro vo k ed by the experience o f feeling sham ed and hum iliated . . . and that did not represent the attem pt to . . .

undo this “ loss o f fa c e ’” .202’ p- 110 A n d w e can all recognize these feelings, even if w e w o u ld never go so far as to act on them . W e recognize the stom ach-clenching feelings o f sham e and em b arrass­ m ent, the m ortification that w e feel burning us up w h en w e m ake ourselves lo o k fo olish in the eyes o f others. W e k n o w h o w im portant it is to feel liked, respected, and v a lu e d .203 But if all o f us feel these things, w h y is it predom in an tly am on g yo u n g m en that those feelings escalate into violen t acts?

H ere the w o rk o f ev olu tio n ary p sych ologists M a r g o W ilson and M a rtin D a ly helps to m ake sense o f these patterns o f violence. In

their 19 8 8 b o ok H o m ic id e 204 and a w ealth o f b o ok s, chapters and articles since, they use statistical, an th ro po lo gical and historical

data to sh o w h o w yo u n g men have strong incentives to achieve and

m aintain as high a social status as they can - because their success in sex u al com petition depends on statu s.77’ 205-8 W hile lo o ks and ph ysical attractiveness are m ore im portan t fo r w o m en , it is status that m atters m ost fo r sex u al success am ong men. P sychologist D avid Buss foun d that w o m en valu e the financial status o f prospective partners ro u gh ly tw ice as m uch as men d o .209 So w h ile w om en try to enhance their sexu al attractiveness w ith clothes and m ake-up, men com pete fo r status. T h is exp lain s not on ly w h y feeling put d o w n , disrespected and hum iliated are the m ost com m on trigger fo r violence; it also ex p lain s w h y m ost violence is betw een men - men have m ore to w in or lose fro m h avin g (or failin g to gain) status. R eckless, even violen t b eh aviou r com es fro m yo u n g men at the

bottom o f society, deprived o f all the m arkers o f status, w h o m ust struggle to m aintain face and w h a t little status they have, often reacting ex p lo siv ely w h en it is threatened.

But w h ile it seems clear that the p ropen sity fo r violence am on g yo u n g men lies p a rtially in evolved psych o lo gical adaptation s related

to sex u al com petition, m ost men are n ot violent. So w h a t factors e xp lain w h y some societies seem better than others at preventing or con trollin g these im pulses to violence?

I N E Q U A L I T Y I S ‘ S T R U C T U R A L ’ V I O L E N C E

Th e sim ple an sw er is that increased in equality ups the stakes in the com petition fo r status: status m atters even m ore. T h e im pact o f

in equality on violence is even better established and accepted

than the other effects o f in equality th at w e discuss in this b o o k .203 In this ch apter w e sh o w relationships betw een violence and inequality fo r the sam e countries and the sam e time period as w e use in other chapters. M a n y sim ilar graph s have been published by other

researchers, fo r other time periods or sets o f countries, including

one co verin g m ore than fifty countries betw een 1 9 7 0 and 1 9 9 4 from

researchers at the W o rld B an k .207-210 A large b o d y o f evidence show s a clear relationship between greater inequality and higher hom icide rates. A s early as 1 9 9 3 , crim inologists H sieh and Pugh w ro te a review w hich included thirty-five analyses o f incom e inequality and violen t crim e.211 A ll but one found a positive link betw een the tw o -

as in equality increased so did violent crim e. H om icid es and assaults

w ere m ost closely associated w ith incom e in equality, and robbery and rape less so. W e have found the sam e relationship s w h en lo o k ­ ing at m ore recently published studies.10 H om icid es are m ore com ­ m on in the m ore unequal areas in cities ran gin g from M an h a ttan to

R io de Ja n e iro , and in the m ore unequal A m erican states and cities

and C an ad ian provinces. Figure 10.2. show s that in tern ation al hom icide rates from the

United N a tio n s S u rv eys on C rim e T ren d s a n d the O p e ra tio n s o f C rim in a l Ju s tic e S ystem s212 are related to incom e in equality, and Figure 1 0 .3 show s the sam e relationship fo r the U S A , using hom icide

rates from the Federal Bureau o f In vestigation .213 Th e differences

betw een some countries in the first grap h are very large. T h e U S A is

Incom e inequality

Figure 10 .2 Homicides are more common in m ore unequal countries.

Incom e inequality

Figure 10 .3 H omicides are more common in more unequal US states.

once again at the top o f the league table o f the rich countries. Its

m urder rate is 64 per m illion , m ore than fo u r times higher than the U K ( 1 5 per m illion) and m ore than tw elve times higher than Ja p a n , w hich has a rate o f on ly 5 .2 per m illion. T w o countries take rather unusual p ositions in this graph , com pared to w h ere they sit in m any

o f ou r other chapters: Singapore has a much lo w er hom icide rate

than w e m ight expect, and Finland has a higher rate. Interestingly, although international relationships between gun ow n ership and violent crim e are com plicated (for instance, gun ow n ership is linked to m urders in vo lvin g fem ale victim s but not m ale victim s),214 in the United N a tio n s In tern ation al Study on Firearm R egu latio n , F inland

had the highest p ro po rtion o f h ouseholds w ith guns, and Singapore

had the lo w est rate o f gun o w n ersh ip .215 D espite these exception s, the trend fo r m ore unequal countries to have higher hom icide rates is w ell established.

In the U S A , although no data w ere availab le fo r W yom in g, the relationship betw een in equality and hom icides is still significant and

the differences between states are alm ost as great as the differences

betw een countries. L o u isian a has a m urder rate o f 1 0 7 per m illion,

m ore than seven times higher than that o f N e w H am p sh ire and

Io w a , w hich are bottom o f the league table w ith m urder rates o f 1 5 per m illion. T h e hom icide rate in A la sk a is much higher than w e w o u ld expect, given its relatively lo w in equality, and rates in N e w Y o r k , C onnecticut and M assach usetts are low er. In the United States, tw o out o f every three m urders are com m itted w ith guns, and

hom icide rates are higher in states w here m ore people o w n gu n s.216 A m o n g the states on ou r grap h , A la sk a has the highest rate o f gun ow nership o f all, and N e w Y o r k , C onnecticut and M assach usetts are am on g the lo w est.217 If w e a llo w fo r gun o w n ership, w e find a slightly stronger relationship betw een inequality and hom icides.

H A V E N S I N A H E A R T L E S S W O R L D

W e have already seen som e features o f m ore unequal societies that help to tie violence to inequality - fam ily life counts, schools and n eighbourhoods are im portan t, and status com petition m atters.

In C h ap ter 8 w e m entioned a study w h ich foun d that divorce rates are higher in m ore unequal A m erican counties. In his b o o k , L ife W ithout F ath er, sociologist D a v id Popenoe describes h o w 60 per cent o f A m erica ’ s rapists, 7 2 per cent o f juvenile m urderers and

7 0 per cent o f long-term prisoners grew up in fatherless hom es.218 Th e effect o f fatherlessness on delinquency and violence is only p artly explain ed by these fam ilies being poorer. W h y do fathers m atter so much?

O ne researcher has described the b eh aviou r o f boys and young men w h o g ro w up w ith o u t fathers as ‘ h yp erm ascu lin e’ , w ith boys

engaging in ‘rigidly o verco m p en sato ry m asculine b eh aviors’ 219, pp- 1-2 - crim es against property and peop le, aggression and exp lo itatio n and short-term sexu al conquests. T h is could be seen as the male version o f the quantity versus q u ality strategy in hum an relatio n ­ ships that w e described in relation to teenage m others in C h apter 9.

T h e absence o f a fath er m ay predispose som e boys to a different reproductive strategy: shifting the balance a w a y fro m long-term relationships an d putting m ore em phasis on status com petition.

F athers can , o f cou rse, act as positive role m odels fo r their sons. F athers can teach b o ys, just by being present in the fam ily, the positive aspects o f m an hood - h o w to relate to the opposite sex , h o w to be a responsible adult, h o w to be independent and assertive, yet

included w ith , and connected to, other people. P articu larly im p ort­

ant is the w a y in w h ich fathers can provid e au th o rity and discipline fo r teenage b o ys; w ith o u t that security, yo u n g men are m ore in­ fluenced by their peers and m ore likely to engage in the kinds o f an ti-social b eh aviou r so often seen w hen gro u p s o f yo u n g men get together. B ut fathers can also be negative role m odels. O ne study

foun d that, although children had m ore b eh aviou ral problem s the

less time they had lived w ith their fathers, this w a s not true when the fathers them selves had b eh aviou ral pro b lem s.220 If the fathers engaged in anti-social beh aviou r, then their children w ere at higher risk w hen they spent m o re time living w ith them.

Perhaps m ost im p ortan tly, fathers love their children in a w a y that

studies sh o w step-parents do not. T h is is not, o f course, to say that m ost step-fathers and other men d o n ’t lo vin gly raise other m en’s children, but on average children livin g w ith their b iological fathers are less likely to be abused, less lik ely to be delinquent, less likely

to drop out o f school, less likely to be em o tion ally neglected. P sy­

chiatrist G illigan says o f the violen t men he w o rk e d w ith 201’ ?-36:

They had been subjected to a degree of child abuse that was o ff the scale of anything I had previously thought o f describing with that term. M any had been beaten nearly to death, raped repeatedly or prostituted, or neglected to a life-threatening degree by parents too disabled to care for

their child. And of those who had not experienced these extremes of physical abuse or neglect, my colleagues and I found that they had experi­ enced a degree of emotional abuse that had been just as damaging . . . in which they served as the scapegoat for whatever feelings of shame

and humiliation their parents had suffered and then attempted to rid themselves o f by transferring them onto their child, by subjecting him to systematic and chronic shaming and humiliation, taunting and ridicule.

The increased fam ily b reak d o w n and fam ily stress in unequal societies leads to inter-generational cycles o f violence, just as much as inter-generational cycles o f teenage m otherhood.

O f course it isn ’ t just the fam ily environm ent that can breed sham e, h um iliation and violence. C hildren experience things in their schools and in their n eighbou rhood s that influence the pro b ab ility

that they w ill turn to violence w hen their status is threatened. The A m erican high-school m assacres have show n us the significance o f bullying as a trigger to violen ce.221-2

In U N I C E F ’ s 2 0 0 7 report on child w ellb ein g in rich countries, there are m easures o f h o w often yo u n g people in different countries w ere involved in p h ysical fighting, had been the victim o f bullying,

or foun d their peers w ere not ‘kind and h elp fu l’ . 110 W e com bined these three m easures into an in d ex o f ch ildren ’s experiences o f conflict and foun d that it w a s significantly correlated w ith income in equality, as show n in Figure 1 0 .4 . In m ore unequal societies children experience m ore b ullyin g, fights and conflict. A n d there is

no better predictor o f later violence than ch ildh ood violence.

En viron m en tal influences on rates o f violence have been

Incom e inequality

Figure 10 .4 There is more conflict between children in m ore unequal countries (based on percentages reporting fighting, bullying and finding peers not kind and helpful).

recognized fo r a long time. In the 19 4 0 s , sociologists o f the C h icago

School described h o w som e n eighbourhoods had persistent repu ta­

tions fo r violence o ver the years - different po pu latio n s m oved in and out but the sam e p o o r n eighbourhoods rem ained dangerous, w h o ever w a s livin g in them .223 In C h icag o , n eighbou rhood s are often identified w ith a p a rticu la r ethnic grou p. So a n eighb ou rhood w hich

m ight once have been an enclave o f Irish im m igrants an d their

descendants later becom es a Polish com m unity, and later still a Latin o n eighbourhood. W h at the C h icag o school sociologists drew attention to w a s the persistent effect o f deprivatio n and p o verty in p o o r n eighbourhoods - on w h o ever lived there. In n eighbourhoods

w here people ca n ’t trust one anoth er, w here there are high levels

o f fear and grou ps o f youths hanging aroun d on street corners, neighbours w o n ’ t intervene fo r the com m on go o d - they feel helpless in the face o f public disturbance, drug dealing, prostitution , graffiti and litter. So cio lo gist R o b ert Sam pson and colleagues at H a rv a rd U n iversity have show n that violen t crim e rates are lo w er in cohesive

n eighbourhoods w h ere residents have close ties w ith one another and are w illin g to act fo r the com m on go o d , even takin g into account factors such as po verty, p rio r violence, the concentration o f im m igrants and residential stab ility.224 In the U S A p o o r neigh­ bourhoods have becom e ghettos, ring-fenced and neglected by the

better-off w h o m ove o u t.225 A lth ough neighbours in areas w ith lo w levels o f trust (see C h a p ­

ter 4) m ay feel less inclined to intervene fo r the com m on go o d , they seem to be m ore pugn acious. In B o w lin g A lo n e , sociologist R o b ert Putnam linked a m easure o f aggression to levels o f social capital in U S states. In a su rvey, people w ere asked to say w h eth er they agreed

or disagreed w ith the sentence: ‘I ’ d do better than average in a fist figh t.’ Putnam says citizens in states w ith lo w social capital are ‘read ier fo r a fight (perhaps because they need to be), an d they are predisposed to m ayhem ’ .25’ p- 310 W hen w e an alyse this m easure o f pugn acity in relation to in equality w ithin states, w e find ju st as strong a relation as Putnam show ed w ith social cap ital (Figure 1 0 .5 ) .

So violence is m ost often a response to disrespect, h um iliation and loss o f face, and is u su ally a m ale response to these triggers. Even w ithin the m ost violen t o f societies, m ost people d o n ’t react violen tly

Incom e inequality

Figure 10 .5 In less equal states more people think they w o u ld do better than average in a fist fight.

to these triggers because they have w a y s o f achieving and m ain tain ­

ing their self-respect and sense o f status in other w a y s. T h ey m ight

have m ore o f the trap pin gs o f status - a go o d education, nice houses and cars, good jo b s, new clothes. T h ey m ay h ave fam ily , friends and colleagues w h o esteem them , o r qualification s they are prou d o f, or skills that are valu ed and valu ab le, or education that gives them status and hope fo r the future. A s a result, although ev eryb o d y experiences disrespect and h um iliation at tim es, they d o n ’t all

becom e violen t; w e all experience loss o f face but w e d o n ’t turn rou n d and shoot som ebody. In m ore unequal societies m ore people lack these protections and buffers. Sham e and h um iliation becom e m ore sensitive issues in m ore h ierarch ical societies: status becom es m ore im portant, status com petition increases and m ore people are

deprived o f access to m arkers o f status and social success. A n d if y o u r source o f pride is y o u r im m aculate law n , y o u ’re go in g to be m ore than a bit an noyed w h en that pride gets tram pled on.

P E A K S A N D T R O U G H S

H om icide rates in A m erica, after rising fo r decades, peaked in the early 19 9 0 s , then fell to their low est level in the early z o o o s. In Z005, they started to rise a ga in .226 Sim ilarly, after peakin g in the early 19 9 0 s , teenage pregn an cy and birth rates began to fall in A m erica, and the decline w a s p articu larly steep fo r A frican -A m erican s.227 But in z o o 6 , the teenage birth rate also started to rise again , and the biggest reversal w a s fo r A frican -A m erican w o m en .228

Som e people h ave tried to exp lain the decline in violence by pointing to changes in po licin g o r drug use or access to guns, or even the ‘m issin g’ coh ort o f yo u n g men w h o w ere n o t born because o f increased access to abortion . E xp la n atio n s fo r the fall in teenage birth rates focused on changes in the num ber o f teenagers w h o are

sexu ally active and increasing contraceptive use. But w h a t influences

w hether or not yo u n g people use drugs, buy guns, have sex or use contraception? W h y are hom icides and teenage births n o w rising again? A n d h o w do these trends m atch up w ith changes in in eq ual­ ity? W h y have hom icides and teenage births m oved in parallel?

T o exam in e this in m ore detail, w e need data on recent short-term

fluctuations in o verall incom e in equality in the U S A . The best data com e from a co llab o rative team o f researchers fro m the U S A , China and the U K , w h o have produced a series o f annual estim ates.229 These sh o w in equality rising through the 19 8 0 s to a peak in the

early 19 9 0 s . T h e fo llo w in g decade sa w an o verall decline in in­

eq u ality, w ith an upturn since z o o o . So there is a reason ab le m atch betw een recent trends in hom icides, teenage births and inequality - rising through the early 19 9 0 s an d declining fo r a decade or so, w ith a very recent upturn.

A lth ou gh violence and teenage births are co m p lex issues and rates

in each can respond to lots o f other influences, the d o w n w a rd trends through the 19 9 0 s w ere consistent w ith im provem ents in the relative incom es o f people at the very bottom o f the incom e distribution. The distribution o f incom e can be m ore stretched out o ver some parts o f its range than others. A society m ay get m ore unequal because the p o o r are getting left further behind the m iddle, or because the rich

are pulling further ahead. A n d w h o suffers fro m lo w social status m ay also v a ry fro m one society to another. A m o n g societies w ith the

sam e overall level o f in equality, in one it m ay be the elderly w h o

are m ost deprived relative to the rest o f society, in another it m ay be ethnic m inority grou ps.

F ro m the early 19 9 0 s in A m erica there w a s a p a rticu la rly dram atic decline in relative p o verty and unem ploym ent fo r yo u n g people at

the bottom o f the social h ierarchy. A lth o u gh the rich continued to

pull further a w a y from the bulk o f the p o p u latio n , fro m the early 19 9 0 s the rela tive position o f the very poorest A m erican s began to im p ro ve.230-31 A s violence and teenage births are so clo sely co n ­ nected to relative deprivation and concentrated in the p oorest areas,

it is w h a t happens at the very bottom that m atters m ost - hence the

trends in violence and teenage b irth s.232 T h ese trends, during the 19 9 0 s , con trast w ith w h a t had been

happening p revio u sly. T h e decades leading up to the 19 9 0 s saw a lo n g sustained deterioration in opportunities and status fo r young people at the bottom o f both A m erican and B ritish society. In

the U S A , from abo u t 1 9 7 0 through the early 19 9 0 s , the earning position o f yo u n g men declined, and em ploym ent prospects for yo u n g people w h o dropped out o f high school or w h o com pleted high school but d id n ’t go on to college w o rsen ed ,233 and violence and teenage births increased. In a recent study, dem ograph er C ynthia

C o len and her colleagues sh o w ed that fallin g levels o f u n em p loy­ m ent during the 19 9 0 s exp lain ed 85 per cent o f the decline in rates o f first births to 1 8 - 1 9 - y e a r - o ld A frican -A m erican s.234 T h is w a s the gro u p experien cin g the biggest drop in teen births. W elfare reform and changes in the availa b ility o f a b o rtion , in contrast, appeared to

h ave had little im pact. In the U K , the im pact o f the econom ic recession and w idening

incom e differences during the 19 8 0 s can also be traced in the hom icide rate. A s health geo grap h er D an n y D o rlin g pointed out, w ith respect to these trends:235, pp- 36-7

There is no natural level of murder . . . For murder rates to rise in par­ ticular places . . . people have to be made to feel more worthless. Then there are more fights, more brawls, more scuffles, more bottles and more

knifes and more young men die . . . These are the same young men who saw many of their counterparts, brought up in better circumstances and in different parts of Britain, gain good work, or university education, or both, and become richer than any similarly sized cohort of such young ages in British history.

In sum m ary, w e can see that the asso ciatio n between inequality and violence is stron g an d consistent; it’ s been dem onstrated in m any different time periods and settings. R ecent evidence o f the close co rrelation betw een ups and do w n s in in equality and violence sh o w that if inequality is lessened, levels o f violence also decline.

A n d the ev olu tio n ary im portance o f sham e and hum iliation provides

a plausible ex p lan atio n o f w h y m ore unequal societies su ffer m ore violence.

Imprisonment and punishment

The degree of civilization in a society can be judged by entering its prisons.

F y o d o r D o s t o e v s k y , The House o f the D ead

In the U S A , prison p o pu lation s have been increasing steadily since the early 19 7 0 s . In 1 9 7 8 there w ere over 450 ,0 0 0 people in ja il, by 2005 there w ere o ver z m illion: the num bers had q u adrupled. In the U K , the num bers have doubled since 1 9 9 0 , clim bing fro m around 46,000 to 80,000 in Z007. In fact, in F eb ru ary Z007, the U K ’ s jails w ere so full that the H om e Secretary w ro te to judges, askin g them to

send on ly the m ost serious crim inals to prison. Th is contrasts sh arply w ith w h a t has been happening in some

other rich countries. T h ro u gh the 19 9 0 s , the priso n p o pu latio n w as stable in Sw eden and declined in Fin lan d; it rose by on ly 8 per cent in D en m ark , 9 per cent in Ja p a n .236 M o re recently, rates have been fallin g in Ireland, A u stria, France and G e rm a n y.237

C R I M E O R P U N I S H M E N T ?

T h e num ber o f people locked up in prison is influenced by three

things: the rate at w h ich crim es are actu ally com m itted, the tendency to send convicted crim inals to priso n fo r p articu lar crim es, and the lengths o f prison sentences. C hanges in an y o f these three can lead to changes in the p ro p o rtion o f the po p u latio n in prison at a n y point in

'.C a

rt o

o n

S to

ck .c

o m

time. W e’ve already described the tendency fo r violen t crim es to be

m ore com m on in m ore unequal societies in C h ap ter 1 0 . W h at has

been happening to crim e rates in the U S A and U K as rates o f im pris­ onm ent have skyrocketed?

C rim in ologists A lfre d Blum stein and A llen Beck have exam ined the gro w th in the U S prison p o p u latio n .238 O n ly i z per cent o f the

gro w th in state p risoners betw een 1 9 8 0 an d 1 9 9 6 cou ld be put dow n

to increases in crim inal offending (dom inated by a rise in drug- related crime). T h e other 88 per cent o f increased im prisonm ent w a s due to the increasing likelih oo d that convicted crim inals were sent to prison rather than being given non-custodial sentences, and

to the increased length o f prison sentences. In federal p rison s, longer

prison sentences are the m ain reason fo r the rise in the num ber o f prisoners. ‘T h ree-strikes’ law s, m inim um m an d ato ry sentences and ‘truth-in-sentencing’ law s (i.e., no rem ission) m ean that som e co n ­ victed crim inals are receiving long sentences fo r m inor crim es. In

C a lifo rn ia in 2 0 0 4 , there w ere 3 6 0 people serving life sentences fo r

sho pliftin g.239 In the U K , prison num bers have also gro w n because o f longer

sentences and the increased use o f custodial sentences fo r offences that a few years ago w o u ld have been punished w ith a fine o r com ­ m unity sentence.240 A bo u t fo rty prison sentences fo r shopliftin g are

handed out every d ay in the U K . C rim e rates in the U K w ere falling

as in e xo ra b ly as im prisonm ent rates w ere rising. Th e prison system in the N etherlan d s has been described by

crim inologist D a v id D o w n es, pro fesso r em eritus o f social adm inis­ tration at the Lo n d o n School o f E co n o m ics.241 H e describes h ow tw o-th ird s o f the difference betw een the lo w rate o f im prisonm ent

in the N etherlan d s and the m uch higher rate in the U K is due to the different use o f cu stodial sentences and the length o f those sentences, rather than differences in rates o f crime.

C o m p arin g different countries, M a r c M a u e r o f the Sentencing

Project242 show s that in the U S A , people are sent to priso n more often, and fo r lon ger, fo r p ro p erty and drug crim es than they are in C a n a d a , W est G erm an y and E n glan d and W ales. F o r exam p le, in the U S A burglars received average sentences o f sixteen m onths, w hereas in C an a d a the average sentence w a s five m onths. A n d variatio n s in

crim e rates d id n ’t exp lain m ore than a sm all am ount o f the variatio n in rates o f im prisonm ent w hen researchers lo o ked at A u stra lia, N e w Z e a la n d and a num ber o f E u ro p ean countries. If crim e rates

ca n ’ t exp la in different rates o f im prisonm ent, can inequality do better?

I M P R I S O N M E N T A N D I N E Q U A L I T Y

W e used statistics on the p ro p o rtion o f the p o pu latio n im ­

prison ed in different countries fro m the United N a tio n s S u rv e y on C rim e T ren d s a n d the O p era tio n s o f C rim in a l Ju s tic e S ystem s.212 Figure i i . i show s (on a lo g scale) that m ore unequal countries have higher rates o f im prisonm ent than m ore equal countries.

In the U S A there are 5 7 6 people in priso n per 10 0 ,0 0 0 , w h ich is

m ore than fo u r and a h a lf tim es higher than the U K , at 1 2 4 per 10 0 ,0 0 0 , and m ore than fourteen times higher than Ja p a n , w hich has a rate o f 40 per 10 0 ,0 0 0 . Even if the U S A and Singapore are

Incom e inequality

Figure 1 1 . 1 M ore people are imprisoned in more unequal countries.1*9

Incom e inequality

Figure 1 1 . 2 M ore people are imprisoned in more unequal US states.1*9

exclu d ed as outliers, the relationship is ro b u st am on g the rem aining countries.

F o r the fifty states o f the U S A , figures fo r im prisonm ent in 1 9 9 7 - 8

com e fro m the U S D epartm ent o f Ju stice , B ureau o f Ju stice

Statistics.243 A s Figure 1 1 . 2 sh o w s, there is again a strong relationship

betw een im prisonm ent and in equality, and big differences between states - L o u isian a im prisons people at m ore than six times the rate o f M in n eso ta.

Th e other thing to notice on this graph is that states are show n

using tw o different sym bols. T h e circles represent states that have

abolished the death pen alty; diam onds are states w hich have retained it.

A s w e pointed out in C h ap ter 2 , these relationships w ith in ­ eq u ality occur fo r problem s w hich have steep social gradients w ithin

societies. There is a strong social gradient in im prisonm ent, w ith

people o f lo w er class, incom e and education m uch m ore likely to be sent to prison than people higher up the social scale. Th e rarity o f m iddle-class peop le being im prisoned is highlighted by the fact that

tw o sociologists at C alifo rn ia State Polytechnic thought it w o rth ­ w hile to publish a research paper describing a m iddle-class inm ate’s adap tation to prison life.244

R a c ia l and ethnic disparities in rates o f im prisonm ent are one w a y

o f show in g the inequalities in risk o f being im prisoned. In A m erica,

the ra cia l g ap can be m easured as the ratio betw een im prisonm ent rates fo r w h ites and b lack s.245 H a w a ii is the on ly state w here the risk o f being im prisoned doesn ’t seem to d iffer m uch by race. Th ere, the risk o f being im prisoned if you are b lack is 1 . 3 4 times as high as if you are w h ite. In every other state o f the union ratios are greater

than 2. T h e ratio is 6 .04 fo r the U S A as a w h o le and rises to 1 3 . 1 5

fo r N e w Je rsey . T h ere is a sim ilar picture in the U K , w h ere mem bers o f ethnic m inorities are m uch m ore likely to end up in p riso n .246 A re these ethnic inequalities a result o f ethnic disparities in rates o f crim es com m itted? R esearch on yo u n g A m erican s suggests n o t.247

T w en ty-five per cent o f w hite youths in A m erica have com m itted

one violen t offence by age 1 7 , com pared to 3 6 per cent o f A frican - A m erican s, ethnic rates o f p ro perty crim e are the sam e, and A frican - A m erican youth com m it few er drug crim es. But A frican -A m erican youth are overw h elm in gly m ore likely to be arrested, to be detained,

to be ch arged, to be charged as if an adult and to be im prisoned.

The sam e pattern is true fo r A frican -A m erican and H isp an ic adults, w h o are treated m ore h arsh ly than w hites at every stage o f ju dicial proceedings.248 Facin g the sam e charges, w hite defendants are far m ore likely to have the charges again st them reduced, or to be offered ‘diversio n ’ - a deferm ent or suspension o f prosecution if the

offen der agrees to certain conditions, such as com pleting a drug

rehabilitation program m e.

D E G R E E S O F C I V I L I Z A T I O N

Prison data sh o w us that m ore unequal societies are m ore punitive. There are other indicators o f this in the w a y s that offen ders are treated in different penal system s. First, as Figure 1 1 . 2 show s, m ore unequal U S states are m ore lik ely to retain the death penalty.

Second, h o w prisoners are treated seems to differ.

D iscussing the N etherlan d s, D av id D o w n es describes h o w a group o f crim inal law yers, crim inologists and psych iatrists cam e together

to influence the prison system . T h e y believed that:

the offender must be treated as a thinking and feeling fellow human being, capable of responding to insights offered in the course of a dialogue . . . with therapeutic agents.241’ p-147

T h is ph ilosoph y h as, he says, resulted in a prison system that

em phasizes treatm ent and rehabilitation . It a llo w s hom e leave and interruptions to sentences, as w ell as extensive use o f p aro le and pardon s. Prisoners are housed in single cells, relations am ong prisoners and betw een prisoners and sta ff are go o d , and p ro ­ gram m es fo r ed ucation, train in g and recreation are considered a

m odel o f best practice. A lth o u gh the system has toughened up som ew hat since the 19 8 0 s in response to rising crim e (m ostly a co n ­ sequence o f rising rates o f drug traffick in g and the use o f the N etherlan d s as a base fo r international organized crim e), it remains

ch aracteristically hum ane and decent. Ja p a n is another cou n try w ith a very lo w rate o f im prisonm ent.

P rison environm ents there have been described as ‘ havens o f tran q u illity’ .249 T h e Ja p an e se ju dicial system exercises rem arkable fle xib ility in prosecution and crim inal proceedings. O ffenders w h o confess to their crim es and express regret and a desire to reform are gen erally trusted to do so, by police, judges and the public at large.

One crim inologist w rites that:

the vast majority [of those prosecuted] . . . confess, display repentance, negotiate for their victims’ pardon and submit to the mercy of the authorities. In return they are treated with extraordinary leniency.250’ p-495

M a n y custodial sentences are suspended, even fo r serious crimes

that in other countries w o u ld lead to long m an dato ry sentences. A p p aren tly, m ost prison inm ates agree that their sentences are app ro priate. Prisoners are housed in sleeping room s holding up to eight people, and m eals are taken in these sm all grou p settings. Prisoners w o rk a fo rty-h o u r w eek and have access to train in g and

recreational activities. D iscipline is strict, w ith ex a ct rules o f con­ duct, but this seems to serve to m aintain a calm atm osphere rather

than p ro v o k e an aggressive reaction. Prison sta ff are expected to act as m o ral educators and lay counsellors as w ell as guards.

The picture is fa r starker in the priso n system s o f the U S A . Th e harshness o f the U S prison system s at fed eral, state and county levels has led to repeated condem nations by such bodies as A m nesty

In tern ation al,251-2 H u m an R ights W atch 253-4 and the United N a tio n s

Com m ittee against T o rtu re .255 T h eir concerns relate to such p ra c­ tices as the incarceration o f children in adult p rison s, the treatm ent o f the m entally ill and learning disabled, the prevalence o f sexu al assaults w ithin p rison s, the shackling o f w o m en inm ates during

childbirth, the use o f electro-shock devices to con trol prisoners,

the use o f prolon ged so litary confinem ent and the b rutality and ill- treatm ent som etim es perpetrated by police and prison gu ards, p articu larly against ethnic m inorities, m igrants and h om osexu als.

Em inent A m erican crim in ologist Jo h n Irw in has spent time

studying high-security prison s, cou n ty jails and S olan o State Prison

in C a lifo rn ia , a m edium -security facility housing aroun d 6 ,0 0 0 prisoners, w here prisoners are crow d ed together, w ith very limited access to recreation facilities o r education, train in g or substance abuse program m es.256 H e describes serious psych o lo gical h arm done to prison ers, and their difficulties in coping w ith the w o rld outside

w hen released, across all security levels and types o f institutions.

In som e prisons, inm ates are denied recreation al activities, in­ cluding television and sp ort activities. In others, prisoners h ave to p ay fo r health care, as w ell as ro o m and b oard . Som e have brought back ‘priso n strip e’ u niform s and chain gangs. ‘A m erica ’s toughest

s h e r iff, Jo e A rp a io , has becom e fam o u s fo r his ‘tent city ’ cou n ty jail

in the A rizon a desert, w h ere p risoners live under can vas, despite tem peratures that can rise to I 3 0 ° F , and are fed on m eals costing less than i o p (20 cents) per h ead.257-*

A m erica ’s developm ent o f the ‘su p erm ax ’ priso n ,201 facilities designed to create a perm anent state o f social iso latio n , has been co n ­ demned b y the United N a tio n s C om m ittee on T o rtu re .255 Som etim es free-standing, but som etim es constructed as ‘priso n s-w ith in -p rison s’ , these are facilities w here prisoners are kept in so litary confinem ent fo r twenty-three hours out o f ev ery day. Inm ates leave their cells only fo r so litary exercise or show ers. M ed ical an thro p o lo gist L o rn a

R h o d es, w h o has w o rk e d in a su perm ax, describes p riso n ers’ lives as characterized by ‘ lack o f m ovem ent, stim ulation and social

co n tact’ .259 Prisoners kept in such conditions often are (or become)

m entally ill and are unprepared fo r eventual release: they have no

m eaningful w o rk , get no train in g or education. Estim ates v a ry , but as m an y as 4 0 ,0 0 0 people m ay be im prisoned under these conditions, and new su p erm ax prisons continue to be built.

T here is, o f course, considerable v a riatio n in prison regim es w ithin

the U S A . A recent report by the C om m ittee on Safety and A buse

in A m erica ’ s prisons gives a com prehensive picture o f the problem s o f the system , an d describes som e o f the m ore hum ane systems and practices.260 A health care initiative in M assach usetts provid es con tin uity o f care fo r prisoners w ithin prison and in the com m unity after their release. M a ry la n d has an ex em p lary program m e for

screening inm ates fo r m ental illness. V erm o n t ensures that prisoners

have access to lo w -cost telephone calls to m aintain their contacts w ith the outside w o rld . A n d in M in n eso ta there is a high-security prison that em phasizes hum an con tact, n atural light and sensory stim ulation, regu lar exercise and the need to treat inm ates with

dignity and respect. If you lo o k b ack at Figure 1 1 . 2 , yo u can see that

m ost o f these exam ples com e from am on g the m ore equal U S states. N o t on ly do the higher rates o f im prisonm ent in m ore unequal

societies seem to reflect m ore punitive sentencing rather than crime rates, but both the harshness o f the prison system s and use o f capital

punishm ent point in the sam e direction.

D O E S P R I S O N W O R K ?

Perhaps a high rate o f im prisonm ent, and a h arsh system fo r dealing w ith crim inals w o u ld seem w o rth w h ile if prison w o rk e d to deter

crim e and protect the p u b li c / Instead, the consensus am on g experts w o rld w id e seems to be that it doesn ’t w o rk very w e ll.261-4 Prison

* Jo h n Irw in writes that while im prisonm ent is g en erally believed to have four ‘ o ffic ia l’ purposes - retribution fo r crim es com m itted, deterrence, incapacitation o f dangerous crim inals and the rehabilitation o f crim inals, in fact three other purposes have shaped A m erica’ s rates and conditions o f im prisonm ent. T hese ‘ u n official’

psych iatrist Ja m e s G illigan says that the ‘m o st effe c tiv e w a y to

turn a n o n -v io le n t p e rso n in to a v io le n t o n e is to s e n d h im to p r is o n ’ ,201> p- 117 In fact, im prisonm ent doesn’t seem to w o rk as w ell n o w as it used to in the U S : paro le vio latio n and repeat o ffen din g are an increasing facto r in the gro w th o f im prisonm ent rates. Between

1 9 8 0 and 19 9 6 , prison adm issions fo r p aro le vio latio n s rose from

1 8 per cent to 3 5 per cent.238 L o n g sentences seem to be less o f a deterrent than higher conviction rates, and the longer som eone is incarcerated, the hard er it is fo r them to ad ap t to life outside. G illigan says that:

the criminal justice and penal systems have been operating under a huge

mistake, namely, the belief that punishment will deter, prevent or inhibit

violence, when in fact it is the most powerful stimulant of violence that we have yet discovered.201’ p- 116

Some effo rts to use punishm ent system s to deter crim e are not just

ineffective, they actu ally increase crim e. In the U K , the introduction

o f A n ti-So cial B eh avio u r O rders (A S B O s ) fo r delinquent youths has

been co n tro versial, p a rtly because they can crim inalize b eh aviour that is otherw ise la w fu l, but also because the acquisition o f an A S B O has com e to be seen as a rite o f p assage and badge o f h on ou r am on g som e yo u n g p eop le.265-6

A lthough there seems to be a gro w in g consensus am on g experts

that prison doesn ’t w o r k , it is difficult to find g o o d , co m p arab le data on re-offen din g rates in different countries. I f a co u n try im prisons a sm aller p ro p o rtion o f its citizens, these are m ore likely to be hardened crim inals than those im prisoned under a harsher regime.

So w e m ight expect countries w ith lo w er o verall rates o f im p rison ­

ment to have higher rates o f re-offen din g. In fact, there appears to be a trend to w ard s higher rates o f re-offen din g in m ore punitive system s (in the U S A and U K , re-offen din g rates are gen erally reported to be betw een 60 and 65 per cent) and lo w er rates in

purposes are class control - the need to protect honest m iddle-class citizens from the dangerous crim inal und erclass; scapegoating - diverting attention a w a y from m ore serious social problem s (and here he singles o ut g ro w in g inequalities in w ealth and incom e); and using the th reat o f the dangerous class fo r political gain. 256

less harsh environm ents (Sweden an d Ja p a n are reported to have

recidivism rates betw een 35 and 4 0 per cent).

H A R D E N I N G A T T I T U D E S

W e’ve seen that im prisonm ent rates are not determ ined by crime rates so much as by differences in o fficial attitudes to w ard s punish­ ment versus rehabilitation and reform . In societies w ith greater in equality, w here the social distances betw een people are greater,

w h ere attitudes o f ‘us and them ’ are m ore entrenched and w h ere lack

o f trust and fear o f crim e are rife, public and p o licy m akers alike are m ore w illin g to im prison peop le and ad o pt punitive attitudes to w a rd s the ‘crim inal elem ents’ o f society. M o re unequal societies are h arsher, tougher places. A n d as prison is not p a rticu la rly effec­ tive fo r either deterrence or rehabilitation , then a society m ust only

be w illin g to m aintain a high rate (and high cost) o f im prisonm ent

fo r reasons unrelated to effectiveness. Societies that im prison m ore peop le also spend less o f their w ealth

on w e lfa re fo r their citizens. T h is is true o f the U S states an d also o f O E C D coun tries.267-8 C rim in ologists D avid D o w n es and K irstine

H ansen report that this phenom enon o f ‘ penal exp an sio n and

w elfare co n tractio n ’ has becom e m ore pronounced o ver the past couple o f decades. In his b o o k C rim e a n d P u n ish m en t in A m erica , published in 1 9 9 8 , sociologist E llio tt C u rrie points out that, since 19 8 4 , the state o f C a lifo rn ia built on ly one n ew college but twenty-

one new p riso n s.264 In m ore unequal societies, m oney is diverted

a w a y from positive spending on w e lfa re, education, etc., into the crim inal and judicial system s. A m o n g our grou p o f rich countries, there is a significant co rrelation betw een incom e inequality and the num ber o f police and internal security officers per 10 0 ,0 0 0 people.212 Sw eden em ploys 1 8 1 police per 10 0 ,0 0 0 p eople, w h ile Portugal

has 4 5 0 . O ur im pression is that, in m ore equal countries and societies,

legal and ju dicial system s, prosecution procedures and sentencing, as w ell as penal system s, are developed in consultation w ith experts - crim in ologists, law y ers, prison p sychiatrists and psych o lo gists, etc.,

and so reflect both theoretical and evidence-based considerations o f w h a t w o rk s to deter crim e and rehabilitate offenders. In con trast,

m ore unequal countries and states seem to have developed legal

fram ew o rk s and penal system s in response to m edia and political pressure, a desire to get tough on crim e and be seen to be doing so, rather than on a considered reflection on w h a t w o rk s and w h a t doesn’ t. Jo h n Silverm an, w ritin g fo r the U K ’s E co n o m ic and Social

R esearch C o u n cil, says that prisons are effective on ly ‘ as a means o f an sw ering a sustained m edia battering w ith an apparen t sh o w o f fo rce ’ .269 In conclusion, D o w n es and H an so n deserve to be quoted in fu ll:268, pp- 4-5

A growing fear of crime and loss o f confidence in the criminal justice system among the population, . . . made the general public more favourable towards harsh criminal justice policies. Thus, in certain countries, in particular the United States and to a lesser extent the United Kingdom - public demand for tougher and longer sentences has been met by public policy and election campaigns which have been fought and won on the grounds of the punitiveness o f penal policy. In other countries, such as Sweden and Finland, where the government provides greater ‘insulation against emotions generated by moral panic and long-term cycles of tolerance and intolerance’ (Tonry, 1 999),270 citizens have been less likely to call for, and to support, harsher penal policies and the government has resisted the urge to implement such plans.

Social mobility: unequal opportunities

A l l th e p e o p l e lik e u s a r e W e , a n d e v e r y o n e e ls e is

T h e y . R u d y a r d K i p l i n g , We and They

In som e h istorical and m odern societies, social m o b ility has been virtu a lly im possible. W here social status is determ ined by religious or legal system s, such as the H in du caste system , the feu dal systems

o f m edieval E uro pe, or slavery, there is little or no opportun ity

fo r people to m ove up or d o w n the social ladder. But in m odern m arket dem ocracies, people can m ove up or dow n w ithin their lifetim e (intra -genera tio n al m obility) or o ffsp rin g can m ove up and do w n relative to their parents (in ter-gen eratio n al m obility). Th e p ossib ility o f social m o b ility is w h a t w e m ean w h en w e talk abo u t equality o f opportun ity: the idea that an yb o d y , by their ow n m erits and h ard w o rk , can achieve a better social or econom ic position fo r them selves and their fam ily. U nlike greater equality itself, eq u ality o f o ppo rtu n ity is valu ed across the political spectrum , at least in theory. Even if they do nothing to actively prom ote social m obility, v e ry few politician s w o u ld take a public stance again st equal op portu n ity. So h o w m obile are ou r rich m arket

dem ocracies? It’s not easy to m easure social m obility in societies. D o in g so

requires longitudinal data - studies that track people o ver tim e to see w h ere they started from and w h ere they end up. O ne convenient

w a y is to take in c o m e m o b ility as a m easure o f social m obility: to see h o w m uch peop le’ s incom es change over their lifetim es, or h o w m uch they earn in com parison to their parents. T o m easure

w w w .C a r to o n S t o c k .<

inter-generational m obility these longitudinal studies need to cover periods o f as m uch as thirty y ears, in order fo r the o ffsp rin g to estab­

lish their position in the incom e hierarchy. W hen w e have incom e data fo r parents and o ffsprin g, social m obility can be m easured as the correlation betw een the tw o . If the co rrelation between paren t’ s incom e and ch ild’ s incom e is high, that m eans that rich parents tend to have children w h o are also rich, and p o o r parents tend to have children w h o stay p o o r. W hen the correlation is lo w , children’ s incom e is less influenced b y w h eth er their parents w ere rich or poor. (These com parison s are not affected by the fact that average incom es

are n o w higher than they used to be.)

L I K E F A T H E R , L I K E S O N ?

C o m p arab le international data on inter-generational social m obility

are availab le fo r on ly a few o f our rich countries. W e take our figures fro m a study b y econom ist J o B landen and colleagues at the Lon d on S ch ool o f E co n o m ics.271 U sing large, representative longitudinal studies fo r eight countries, these researchers w ere able to calculate so cial m obility as the correlation between fath ers’ incom es when

their sons w ere born and so n s’ incom es at age thirty. D espite h av ­ ing data fo r o n ly eight countries, the relationship betw een inter- gen eration al social m obility and incom e in equality is v e ry strong. Figure i z . i show s that countries w ith bigger incom e differences tend to have much lo w er social m obility. In fact, far from enabling the

id eo lo gy o f the A m erican D ream , the U S A has the low est m obility

rate am on g these eight countries. The U K also has lo w social m ob ility, W est G erm an y com es in the m iddle, and C an a d a and the S can din avian countries have m uch higher m obility.

W ith data for so fe w countries w e need to be cau tiou s, p articu larly

as there are no data o f this sort that a llo w us to estim ate social m o b ility fo r each state and test the relationship w ith in equality inde­ pendently in the U S A . But other o b servation s, lo o k in g at changes in social m obility o ver tim e, public spending on education, changes in geo graph ical segregation, the w o rk o f sociologists on m atters o f

taste and p sych ologists on displaced aggression, and so-called group

Incom e inequality

Figure i z . i Social mobility is lower in more unequal countries.149

density effects on health, lend plau sib ility to the picture w e see in Figure i z . i .

Th e first o f these observation s is that, after slo w ly increasing from 1 9 5 0 to 1 9 8 0 , social m obility in the U S A declined ra p id ly, as incom e differences w idened dram atically in the later p art o f the century.

F igure i z . z uses data from T h e State o f W o rk in g A m e ric a 10 0 6 / 7 report. Th e height o f each colum n show s the p o w e r o f fath ers’ incom e to determ ine the incom e o f their sons, so s h o rte r bars indicate m ore social m obility: fath ers’ incom es are less predictive

o f so n s’ incom es. H igher bars indicate less m obility: rich fathers are

m ore likely to have rich sons and p o o r fathers to have p o o r sons. D ata fro m the 19 8 0 s and 19 9 0 s sh o w that abo u t 3 6 per cent

o f children w h ose parents w ere in the bottom fifth o f the w ealth distribution end up in that sam e b ottom fifth them selves as adults, and am on g children w h o se parents w ere in the top fifth fo r w ealth ,

3 6 per cent o f them can be found in the sam e top fifth.272 T h o se at the top can m aintain their w ealth and status, those at the bottom

■o QJc rt

■q, 4°X

1 9 5 0 i 9 6 0 1 9 7 0 1 9 8 0 1 9 9 0 2 0 0 0

Figure 12..2 Social mobility in the USA increased to 19 8 0 and then decreased.272

find it difficult to clim b up the incom e ladder, but there is m ore flexib ility in the m iddle. Inter-generational social m o b ility has also

been fallin g in B ritain over the time period that incom e differences

have w iden ed.271 A second ob servation that supports ou r belief that greater incom e

in equality reduces social m obility com es from data on spending on education. E du cation is gen erally thought o f as the m ain engine o f social m obility in m odern dem ocracies - people w ith m ore education

earn m ore and have higher social status. W e sa w in C h ap ter 8 h ow inequality affects education al achievem ents and asp iration s, but it’ s w o rth noting that, am ong the eight countries fo r w h ich w e have in fo rm atio n abo u t so cial m obility, public expenditure on education

(elem entary/prim ary and high/secondary schools) is stron gly linked to the degree o f incom e equality. In N o r w a y , the m ost equal o f the eight, alm ost all (9 7.8 per cent) spending on school education is public expen ditu re.273 In con trast, in the U S A , the least equal o f this group o f countries, o n ly abou t tw o-th ird s (6 8 .2 per cent) o f the spending on school education is public m oney. T h is is likely to have a substantial im pact on social differences in access to higher education.

M O V I N G U P W A R D S , M O V I N G O U T

A third type o f evidence that m ay confirm the co rrelation between incom e inequality and social m o b ility is the w a y in w hich greater

social distances becom e translated into greater geo grap h ical segrega­ tion betw een rich and p o o r in m ore unequal societies.

A s in equality has increased since the 19 7 0 s in the U S A , so to o has the geograp hical segregation o f rich and p o o r.274 P olitical econom ist Paul Ja r g o w s k y has an alysed data fro m the 1 9 7 0 , 1 9 8 0 and 19 9 0

U S Census and show n that the residential co n cen tratio n o f p o v e r ty

increased o ver that p erio d .275-6 N e igh b o u rh o o d concentration o f po verty is a m easure that tells us w h a t p ro po rtion o f p o o r people in a city live in high -poverty areas. Ja r g o w s k y estim ates that in 1 9 7 0 abou t one in fo u r p o o r b lacks lived in high -p overty neighb ou r­

h oods, but b y 19 9 0 that pro p o rtion had risen to one in three.

A m o n g w h ites, p o verty concentration doubled during the tw o decades, w h ile incom e differences w ere w idening. W hen po verty concentration is high, p o o r people are not on ly coping w ith their o w n p o verty but also the consequences o f the p o verty o f their neigh­ bours. Between the 1 9 9 0 and the 20 0 0 census, Ja rg o w s k y reports a

decline in po verty con cen tration , p a rticu la rly fo r black A m erican s in the inner cities, w h ich goes alon g w ith the im provem ents in the re la ­ tive position o f the very poorest A m erican s w h ich w e described at the end o f C h ap ter i o . 277 Even as p o verty concentration has declined in the inner city, th ough, it has g ro w n in the inner ring o f suburbs

and, w ith the recent econom ic dow n tu rn in A m erica, Ja r g o w s k y w arn s that the gains o f the 19 9 0 s m ay have alread y been reversed.

A sim ilar pattern o f segregation by p o verty and w ealth during a period o f increasing incom e differences has been takin g place in the U K .278 T h e rich are w illin g to p a y to live sep arately from the p o o r,279

and residential segregation alon g econom ic lines increased th ro ugh ­

out the 19 8 0 s and 19 9 0 s .280 The im age o f the ‘sink estate’ p ro vo k es ju st as clear a picture o f a deprived underclass as does the im age o f the ghetto an d the b arrio in the U S A .

Research ers on both sides o f the A tlan tic are clear that increased incom e in equality is responsible fo r increasing the segregation o f rich

and p o o r.281-3 T h e concentration o f p o o r people in p o o r areas in­ creases all kinds o f stress, deprivation and difficulty - fro m increased com m uting times fo r those w h o have to leave deprived com m unities to find w o rk elsew here, to increased risk o f traffic accidents, w orse sch ools, p o o r levels o f services, exp o su re to gan g violence, pollution and so on. S o ciologist W illiam Ju liu s W ilson, in his classic study o f inner-city poverty, refers to p o o r people in p o o r n eighbourhoods as the ‘tru ly disad v an taged ’ .225 T w o studies fro m the U S A have show n that residential econom ic segregation increases peop le’ s risk o f dyin g, and one show ed that m ore unequal cities w ere also m ore

econ om ically segregated.284' 5 T h ese processes w ill o f course feed

back into further reductions in social m obility.

M A T T E R S O F T A S T E - A N D C U L T U R E

So social m obility is lo w er and geograp hical segregation greater in

m ore unequal societies. It is as if greater in equality m akes the social structure o f society m ore rigid an d m ovem ent up and d o w n the

social ladder m ore difficult. Th e w o rk o f French sociologist Pierre B ourdieu also helps us

to understand h o w social m obility becom es m ore lim ited w ithin m ore h ierarch ical societies.286 H e describes h o w m aterial differences betw een people, the am ount o f m oney and resources they have, becom e o verlaid w ith cultural m arkers o f social difference, w hich becom e m atters o f sn obb ery and prejudice. W e all use m atters o f taste as m arks o f distinction and social class - w e judge people by their accent, clothing, language, choice o f readin g m atter, the tele­ visio n program m es they w a tch , the fo o d they eat, the sports they play , the m usic they prefer, an d their appreciation - or lack o f it -

o f art. M idd le-class and upper-class people have the right accents, k n o w

h o w to behave in ‘ polite so ciety’ , k n o w that education can enhance

their advan tages. T h e y pass all o f this on to their children, so that they in turn w ill succeed in school and w o r k , m ake go o d m arriages, find h igh -payin g jo b s, etc. T h is is h o w elites becom e established and

m aintain their elite status.

People can use m arkers o f distinction and class, their ‘go o d taste’ , to m ain tain their positio n , but th roughout the social h ierarchy

people also use discrim ination and d o w n w a rd prejudice to prevent

those b elo w them from im p rovin g their status. D espite the m odern id eo lo gy o f eq u ality o f opportu n ity, these m atters o f taste and class still keep people in their place - stopping them from believing they

can better their position and sap pin g their confidence if they try. The

experim ents on stereotype threat described in C h ap ter 8 sh o w h o w strong the effects on p erform an ce can be. B ourdieu calls the actions by w h ich the elite m aintain their distinction s y m b o lic v io le n c e ; w e m ight ju st as easily call them discrim ination an d snobbery. A lth ou gh racial prejudice is w id ely condem ned, class prejudice is, despite the sim ilarities, rarely m entioned.

These so cial system s o f taste, w h ich define w h a t is h igh b ro w and cultured, and w h at is lo w b ro w or p o p u lar, con stan tly shift in co n ­ tent but are a lw ay s w ith us. T h e exam ples that B ourdieu collected in the 19 6 0 s seem very dated n o w , but illustrate the point. H e found that different social class grou ps preferred different types o f m usic; the lo w er social class grou ps preferred the catch y tune o f the ‘ Blue D an u b e’, w h ile the upper classes expressed a preference fo r the

m ore ‘difficu lt’ ‘W ell-Tem pered C la v ie r’ . T h e upper classes preferred abstract art and experim en tal novels, w h ile the lo w er classes liked representational pictures and a go o d plot. But if ev eryb o d y starts to en jo y Bach and Picasso an d Ja m e s Jo y c e , then upper-class taste w ill shift to appreciate som ething n ew - elitism is m aintained by shifting

the b oundaries. W h at B ourdieu is describing is an ‘ econ om y o f cultural g o o d s’ , and inequalities in that econ om y affect people alm ost as p ro fo u n d ly as inequalities in incom e.

In her b o o k , W atching the E n g lis h , an th ro po lo gist K ate F o x describes the social class m arkers o f the English — in con versation ,

hom es, cars, clothes, fo o d and m o re.287 Jo se p h Epstein does the same fo r the U S A in S n o b b e r y : T h e A m eric a n V ersio n .m Both bo ok s are am using, as w ell as erudite, and it’ s difficult not to laugh at ou r ow n pretensions and the p o o r taste o f others.

In the U K , fo r exam p le, yo u can tell if som eone is w o rk in g class,

m iddle class or upper class by w h eth er they call their evening meal tea , din ner’ or ‘ su p per’ . B y w hether they call their m other ‘m am ’ ,

‘m u m ’ o r ‘ m um m y’ , by w h eth er they go out to a ‘d o ’ , a ‘ fun ction ’ or

a ‘p a rty ’ , an d so on. Sn obbery, says E pstein, is ‘sitting in y o u r B M W 7 4 o i and feeling

quietly, assuredly better than the p o o r v u lgarian . . . w h o pulls up n ext to yo u at the stoplight in his garish C ad illa c. It is the calm pleasure w ith w h ich yo u greet the new s that the son o f the w o m an y o u have just been introduced to is m ajo rin g in ph o tojou rn alism at

A rizon a State U n iversity w h ile y o u r o w n daughter is studying art

h istory at H a rv a rd . . . ’ But snobbishness and taste turn out to be a zero-sum gam e. Epstein goes on to point out that another d a y , at another stoplight, a Bentley w ill pull up next to y o u r pathetic B M W , and yo u m ay be introduced to a w o m an w h ose son is studying

classics at O xfo rd . T h e w a y s in w h ich class and taste and sn obb ery w o rk to constrain

peop le’ s opportunities and w ellb ein g are, in reality, pain fu l and per­ vasive. T h e y are form s o f discrim ination and social exclu sion . In their 1 9 7 2 b o ok , T h e H id d e n In ju ries o f C lass, sociologists R ich ard Sennett an d Jo n a th a n C o b b described the psych o lo gical dam age

done to w o rk in g -class men in B osto n , w h o had com e to vie w their failu res to get on in the w o rld as a result o f their o w n inadequacies, resulting in feelings o f hostility, resentm ent and sham e.289 M o re recently, sociologist Sim on C h a rlesw o rth , in an in terview w ith a w o rk in g-class m an in R o th erh am , in the English M id la n d s, is told

h o w asham ed the m an feels encountering a m iddle-class w o m a n .290

Even w ith o u t anything being said betw een them , he is im m ediately filled w ith a sense o f his in ferio rity, becom es self-conscious and even tu ally hostile and angry:

I went in to the social [Social Security Office] the other day . . . there were chairs and a space next to this stuck-up cow, you know, slim, attractive,

middle class, and I didn’t want to sit with her, you feel you shouldn’t . . . I became all conscious, of my weight, I felt overweight, I start sweating, I start bungling, shuffling, I just thought ‘no, I’m not going to sit there, I don’ t want to put her out’, I don’t want to feel that she’s put out, you

don’t want to bother them . . . you know you insult them . . . the w ay they look at you like they’ re disgusted . . . they look at you like you’re invading their area . . . you know, straight away . . . you feel ‘I shouldn’t be there’

. . . it makes you not want to go out. What it is, it’ s a form o f violence . . . right, it’s like a barrier saying ‘listen low-life, don’t even [voice rises with pain and anger] come near me! . . . ‘What the fuck are you doing in my space . . . We pay to get away from scum like you . . . It fucking stresses you, you get exhausted . . . It’s everywhere . . . I mean, I clocked her

[looked at her] like they clock us, right, . . . and I thought ‘fuck me, I ain’t even sitting there’ . She would be uncomfortable, and it’ll embarrass me, you know, [voice rises in anger/pain.] . . . Just sitting there, you know what I’m trying to say? . . . It’s like a common understanding, you know how they feel, you feel it, I’m telling you . . . They are fuck all, they’ve got

nothing, but it’s that air about them you know, they’ve got the right body,

the clothes and everything, the confidence, the attitude, know what I m e an .. . . We [sadly, voice drops] ain’t got it, we can’t have it. We walk in like w e’ve been beaten, dragging our feet when w e’re walking in . . . you feel like you want to hide . . .

T H E B I C Y C L I N G R E A C T I O N

Bigger differences in m aterial w ealth m ake status differences m ore im p ortant, and in m ore unequal societies the w eigh t o f d o w n w a rd

prejudice is bound to be h eavier; there is m ore social distance

between the ‘h av e s’ at the very top and the ‘ h ave-n ots’ at the bottom . In effect, greater in equality increases d o w n w a rd social prejudices. W e m ain tain social status by sho w in g su periority to those b elow . T h o se deprived o f status try to regain it by takin g it out on m ore vulnerable peop le b elo w them. T w o lines o f doggerel ca p ­ ture these processes. T h e English say ‘Th e captain k icks the cabin

boy and the cabin b o y k icks the ca t’ , describing the d o w n w a rd flo w ° f aggression and resentm ent, w h ile a line fro m an A m erican rhym e fam o u sly describes B oston as the place, ‘w h ere the L o w ells ta lk only to C ab o ts, and the C ab o ts talk on ly to G o d ’ , in vo k in g the snobbery

and social clim bing o f people lo o kin g up to those abo ve them.

W hen people react to a p ro v o catio n fro m som eone w ith higher status by redirecting their aggression on to som eone o f lo w er status, psych ologists label it d is p la c e d a g g ressio n .291 E xam p les include: the m an w h o is berated by his boss and com es hom e and shouts at his

w ife and children; the higher degree o f aggression in w o rk p laces w here supervisors treat w o rk e rs u n fa irly ;292 the w a y s in w hich

people in deprived com m unities react to an in flu x o f foreign im m i­

gran ts;293-4 and the w a y s in w h ich prisoners w h o are bullied turn on others b elow them - p articu larly sex offen ders - in the prison

h ierarch y.295 In his b o ok , T h e H o t H o u s e , w h ich describes life inside a high-

security prison in the U S , Pete E arle y tells a story abo u t a m an in

prison w ith a life sentence fo r m u rd er.296’ pp- 74-5 B ow les had been

incarcerated fo r the first tim e at the age o f 1 5 w h en he w a s sent to a juvenile refo rm ato ry. T h e d ay he arrived, an o lder, bigger b o y came

up to him:

‘Hey, what size shoes do you w ear?’ the boy asked.

‘Don’t know’ said Bowles ‘Let me see one of ’em will ya?’ the boy asked politely. Bowles sat down on the floor and removed a shoe. The older boy took

off one of his own shoes and put on Bowles’s. ‘H ow ’bout letting me see the other one?’ ‘I took o ff my other shoe and handed it to him,’ Bowles remembered,

‘ and he puts it on and ties it and then walks over to this table and every

boy in the place starts laughing at me. That’s when I realized I am the butt of the joke.’

B ow les grabbed a p o o l cue and attacked the b o y, fo r w h ich he received a w eek o f h ard lab o u r. W hen a n ew b o y arrived at the

re fo rm ato ry the fo llo w in g w eek , ‘he too w a s con fron ted by a boy w h o dem anded his shoes. O n ly this time it w a s B ow les w h o w as tak in g advan tage o f the new kid. “ It w a s m y turn to dish it o u t,” he

recalled. “ I had earned that rig h t.’” In the sam e b o o k , E arley tells alm ost ex a ctly the sam e sto ry again,

o n ly this time he describes a m an ’ s reaction to being sexu ally

assaulted and sodom ized on his first night in a cou n ty ja il at the age o f 1 6. S ix years later, arrested in another to w n , he is put in a jail cell w ith a ‘ kid, p ro b ab ly seventeen or so, and yo u k n o w w h a t I did? I fucked h im .’ 296’ PP-430-31

D isplaced aggression am on g non-hum an prim ates has been

labelled ‘the bicycling reactio n ’ . P rim atologist V o lk e r Sum m er

explain s that the im age being con ju red up is o f som eone on a racing bicycle, bo w in g to their superiors, w h ile k ick in g do w n on those

beneath. H e w a s describing h o w anim als livin g in strict social hierarchies appease dom inant anim als and attack in ferior ones. Psychologists Jim Sidanius and Felicia Pratto have suggested that hum an gro u p conflict and o ppression, such as racism and sexism ,

stem from the w a y in w hich inequality gives rise to in dividual and institutional discrim ination and the degree to w hich people are com plicit or resistant to som e so cial grou ps being dom inant over o thers.297 In m ore unequal societies, m ore people are oriented to w ard s dom inance; in m ore egalitarian societies, m ore people are oriented to w ard s inclusiveness and em pathy.

O u r final piece o f evidence that incom e in eq uality causes lo w er social m o b ility com es from research w h ich helps to exp la in w h y stigm atized grou ps o f people livin g in m ore unequal societies can feel m ore co m fo rtab le w hen separated fro m the people w h o lo o k dow n on them. In a p o w erfu l illustration o f h o w discrim ination and

prejudice dam age peop le’s w ellbein g, research show s that the health o f ethnic m in ority gro u p s w h o live in areas w ith m ore peop le like themselves is som etim es better than that o f their m ore affluent counterparts w h o live in areas w ith m ore o f the dom inant ethnic g ro u p .298 T h is is called a ‘grou p den sity’ effect, and w a s first show n

in relation to m ental illness. Studies in L o n d o n , fo r exam p le, have

show n a higher incidence o f schizophrenia am on g ethnic m inorities living in n eighbourhoods w ith few er people like them selves,299 and the sam e has been show n fo r suicide300 and self-h arm .301 M o re recently, studies in the United States have dem onstrated the sam e effects fo r heart disease302- 3 and lo w birth w eight.304" 8 G en erally,

living in a p o o rer area is associated w ith w o rse health. M em bers o f ethnic m inorities w h o live in areas w h ere there are few like them ­ selves tend to be m ore affluent, and to live in better n eighbourhoods, than those w h o live in areas w ith a higher concentration. So to find that these m ore ethnically isolated in dividuals are som etim es less h ealthy is surprising. The p ro b ab le exp lan atio n is that, through the eyes o f the m ajo rity com m unity, they becom e m ore aw are o f be­ longing to a low -status m in ority gro u p and perhaps encounter m ore frequent prejudice and discrim ination and have less support. T h at

the psych o lo gical effects o f stigm a are som etim es strong enough

to override the health benefits o f m aterial advan tage tells us a lot abo u t the po w er o f inequality and brings us back to the im portance o f social status, so cial su pport and friendship, and the influence o f social a n xiety and stigm a discussed in C hap ter 3.

Bigger incom e differences seem to so lid ify the social structure and

decrease the chances o f u p w ard m obility. W here there are greater inequalities o f outcom e, equal o ppo rtu n ity is a significantly more distant prospect.

P A R T T H R E E

A Better Society

w w w .C a r t o o n S t o c k .

*3

Dysfunctional societies

N o m a n is a n I s l a n d , e n t ir e o f it s e lf ; e v e r y m a n is a p ie c e o f

th e c o n t in e n t , a p a r t o f th e m a in .

John Donne, M editation X V I I

T h e last nine chapters have sh o w n , am on g the rich developed countries and am on g the fifty states o f the United States, that m ost o f the im portant health and social problem s o f the rich w o rld are m ore com m on in m ore unequal societies. In both settings the relationships are to o strong to be dism issed as chance findings. The im p ortance o f these relationships can scarcely be overestim ated. First, the differences betw een m ore and less equal societies are large - problem s are anything from three times to ten times as com m on in the m ore unequal societies. Second, these differences are not d iffer­ ences between high- and lo w -risk groups w ithin p o p u lation s w hich m ight ap p ly only to a sm all p ro p o rtion o f the p o p u latio n , o r just to the p o o r. R a th e r, they are differences betw een the prevalence o f different problem s w h ich ap p ly to w h o le population s.

D Y S F U N C T I O N A L S O C I E T I E S

O ne o f the points w h ich em erge from C hapters 4 - 1 2 . is a tendency fo r som e countries to do w ell on just abo u t everything and others to do bad ly. Y < fl®can predict a co u n try ’ s perform an ce on one outcom e fro m a k n ow led ge o f others. If - fo r instance - a cou n try does badly on health, you can predict w ith som e confidence that it w ill also

im p rison a larger pro p o rtion o f its p o p u latio n , have m ore teenage

pregnancies, lo w er literacy scores, m ore obesity, w o rse mental health, and so on. Inequ ality seems to m ake countries so cially dysfu nction al across a w ide range o f outcom es.

In tern ation ally, at the healthy end o f the distribution w e a lw ay s

seem to find the Scan din avian countries and Ja p a n . A t the opposite

end, su fferin g high rates o f m ost o f the health and social problem s, are u sually the U S A , P ortugal and the U K . T h e sam e is true am ong the fifty states o f the U S A . A m o n g those that tend to p erfo rm w ell across the b o ard are N e w H am p sh ire, M in n eso ta, N o rth D ak o ta

and V erm o n t, and am on g those w h ich do least w ell are M ississip p i,

L o u isian a and A lab am a.

Figure 1 3 . 1 sum m arizes our findings. It is an ex a ct co p y o f Figure 2 .2 . It show s again the relation ship betw een inequality and our com bined In dex o f H ealth and So cial Problem s. T h is grap h also show s that the relationship is not dependent on an y p articu lar group

o f countries - fo r instance those at either end o f the distribution.

Income inequality

F ig u r e 1 3 . 1 Health and social problem s are more common in more unequal countries.

Instead it is ro b u st across the range o f in equality foun d in the

developed m arket dem ocracies. Even though w e som etim es find less stron g relationships am on g ou r analyses o f the fifty U S states, in the in tern ation al an alyses the U S A as a w h o le is just w here its inequality w o u ld lead us to expect.

T h o u g h som e cou n tries’ figures are p resum ab ly m ore accurate

than others, it is clearly im portan t that w e do not ch erry-pick the data. T h a t is w h y w e have used the sam e set o f inequality data, published by the United N a tio n s, throughout. In the an alyses o f the A m erican states w e have used the U S census data as published. H o w ev e r, even if som eone had a strong objection to the figures for

one or other society, it w o u ld clearly not change the overall picture presented in Figure 1 3 . 1 . T h e sam e applies to the figures w e use for all the health and social problem s. E ach set is as p rovid ed at source - w e take them as published w ith no ifs or buts.

The on ly social problem w e h ave encountered w h ich tends to be

m ore com m on in m ore equal countries (but not significantly am ong

m ore equal states in the U S A ) is, perhaps su rprisin gly, suicide. The reasons fo r this are tw o fo ld . First, in som e countries suicide is not m ore com m on lo w er do w n the social scale. In B ritain a well-defined social gradient has o n ly em erged in recent decades. Second, suicide is

often inversely related to hom icide. T here seems to be som ething in the p sych ological cliche that an ger som etim es goes in and sometimes goes out: do yo u blam e yo u rse lf or others fo r things that go w ron g? In C h ap ter 3 w e noted the rise in the tendency to blam e the outside w o rld - defensive n arcissism - and the contrasts between the U S and Ja p a n . It is notable that in a paper on health in H arlem in N e w

Y o r k , suicide w a s the only cause o f death w h ich w a s less com m on there than in the rest o f the U S A .80

E V E R Y O N E B E N E F I T S

A com m on response to research findings in the social sciences is for people to say they are o b vio u s, an d then perhaps to add a little scorn ­ fully, that there w a s no need to do all that expen sive w o rk to tell us w h a t w e already kn ew . V e r y often, h ow ever, that sense o f k n o w in g

only seeps in w ith the benefit o f hindsight, after research results have been m ade k n o w n . T r y ask in g people to predict the results

in advan ce and it is clear that all sorts o f different things can seem

perfectly plausible. H a v in g lo o ked at the evidence in the preceding chapters o f h o w inequality is related to the prevalence o f so m any problem s, w e hope that m ost readers w ill feel the picture m akes im m ediate intuitive sense. Indeed, it m ay seem ob vio u s that problem s associated w ith relative deprivation should be m ore com m on in m ore u nequal societies. H o w ev e r, if yo u a sk people w h y greater equality reduces these problem s, m uch the m ost com m on guess is that it m ust

be because m ore equal societies have few er p o o r people. T h e assu m p ­ tion is that greater eq u ality helps those at the bottom . A s w ell as being on ly a m inor p art o f the pro per ex p la n atio n , it is an assu m p­ tion w h ich reflects ou r failure to recognize very im p ortan t processes

affecting ou r lives and the societies w e are part of. T h e truth is that

the vast m a jo rity o f the po pu latio n is harm ed by greater inequality. One o f the clues, and one w hich w e in itially found surprising,

is just h o w big the differences betw een societies are in the rates o f the va rio u s problem s discussed in C hapters 4 - 1 2 . A cro ss w h o le

po pu latio n s, rates o f m ental illness are five times higher in the m ost

unequal com pared to the least unequal societies. Sim ilarly, in m ore unequal societies people are five times as likely to be im prisoned, six times as likely to be clin ically obese, and m urder rates m ay be m any times higher. The reason w h y these differences are so big is, quite

sim ply, because the effects o f in equality are not confined just to the

least w e ll-o ff: instead they affect the vast m ajo rity o f the p opu lation . T o take an exam p le, the reason w h y life exp ectan cy is 4 .5 years shorter fo r the average A m erican than it is fo r the average Ja p an e se, is not p rim arily because the poorest 1 0 per cent o f A m ericans suffer

a life expectan cy deficit ten times as large (i.e., forty-five years) w hile

the rest o f the pop u latio n does as w ell as the Ja p an e se. A s epi­ dem iologist M ich ael M a rm o t frequ en tly points out, you cou ld take a w a y all the health problem s o f the p o o r and still leave m ost o f the problem o f health inequalities untouched. O r, to lo o k at it another

w a y , even if yo u take the death rates just o f w h ite A m erican s, they still do w o rse - as w e shall see in a m om ent - than the po pu latio n s o f m ost other developed countries.

C o m p ariso n s o f health in different groups o f the pop u latio n in

m ore and less equal societies sh o w that the benefits o f greater eq u al­

ity are very w idespread. M o st recently, a study in the Jo u r n a l o f the A m e ric a n M e d ic a l A sso c ia tio n com p ared health am on g m iddle-aged men in the U S A and E n glan d (not the w h ole U K ) .315 T o increase co m p arab ility the study w as confined to the n o n -H ispan ic white p o pu lation s in both countries. People w ere divided into both income

and educational categories. In Figure 1 3 . 2 rates o f diabetes, h yp er­

tension, cancer, lung disease and heart disease are show n in each o f three education al categories - high, m edium an d lo w . Th e A m erican rates are the d ark er bars in the b ackgrou n d and those fo r England are the lighter ones in fron t. T here is a consistent tendency fo r rates

o f these conditions to be higher in the U S than in E n glan d , not just am on g the less w ell-educated, but across all educational levels. The sam e w a s also true o f death rates and v a rio u s b iological m arkers such as blood pressure, cholesterol and stress m easures.

T h o u g h this is only just apparen t, the authors o f the study say

that the social class differences in health tend to be steeper in the

U S A than in E n glan d regardless o f w hether people are classified by incom e o r ed u catio n .316

L o w H ig h L o w H ig h L o w H ig h L o w H ig h L o w H ig h

D iabetes H ypertension C an cer L u n g disease H eart disease

Figure 1 3 .2 Rates o f illness are low er at both lo w and high educational levels in England com pared to the U S A .315

D ea

th s

pe r

10 0

,0 0

0

In that co m pariso n , E n glan d w a s the m ore equal and the healtheir o f the tw o countries. But there have also been sim ilar co m p ariso n s o f death rates in Sweden w ith those in E n glan d an d W ales. T o allo w accurate co m p ariso n s, Sw edish researchers classified a large num ber o f Sw edish deaths accordin g to the British o ccu patio n al class classifi­

cation . T h e classification runs from unskilled m an u al occupation s in class V at the bottom , to pro fessio n al occupation s in class I at the top. Figure 1 3 . 3 show s the differences they foun d in death rates fo r w o rk in g -ag e m en.317 Sw eden, as the m ore equal o f the tw o countries, had lo w er death rates in all occu pation al classes; so much so that

their highest death rates - in the lo w est classes - are lo w er than the

highest class in E n glan d and W ales. A n oth er sim ilar study com pared in fan t m o rtality rates in Sweden

w ith E n glan d and W ales.318 Infant deaths w ere classified by fath er’ s occu pation and o ccu pation s w ere again coded the sam e w a y in each

cou n try. T h e results are show n in Figure 1 3 . 4 . D eaths o f babies born

to single parents, w h ich can n ot be coded b y fath er’s occu pation , are show n separately. O nce again , the Sw edish death rates are lo w er right across the society. (N ote that as both these studies w ere published some time ago , the actual death rates they sh o w are

co n sid erab ly higher than the current ones.)

800

600

4 00

□ England and Wales 3] Sweden

Social class

Figure 13 .3 Death rates among working-age men are low er in all occupational classes in Sweden com pared to England and Wales.317

in ta

nt

de at

hs

pe r

1, 0

0 0

Kngland and Wales

m Sweden

1 0 - —

'w High;

mothers Single Low

v Father’s social class

F ig u r e 1 3 . 4 Infant mortality rates are low er in all occupational classes in Sweden than in England and Wales.3ls

C o m p ariso n s have also been m ade betw een the m ore and less equal o f the fifty states o f the U S A . H ere to o the benefits o f sm aller incom e differences in the m ore equal states seem to spread across all incom e grou p s. O ne study concluded that ‘ incom e inequality exerts a co m parab le effect across all po pu latio n su bgro u p s’ , w hether

people are classified by education, race or incom e - so m uch so that

the authors suggested that inequality acted like a pollu tan t spread th roughou t so ciety.319 In a study o f ou r o w n , w e lo o ked at the relationship betw een m edian cou n ty incom e and death rates in all

counties o f the U S A .8 W e com pared the relationship betw een county m edian incom e and county death rates accordin g to w h eth er the counties w ere in the tw enty-five m ore equal states or the twenty-five less equal states. A s Figure 1 3 - 5 show s, in both the m ore and less equal states, p o o rer counties tended - as expected - to have higher death rates. H o w e v e r at all levels o f incom e, death rates w ere low er in the tw enty-five m ore equal states than in the tw enty-five less equal

states. C o m p a rin g counties at each level o f incom e show ed that the benefits o f greater eq u ality w ere largest in the po o rer counties, but still existed even in the richest counties. In its essentials the picture is m uch like that show n in Figures L3.3 and 1 3 . 4 co m parin g Sweden

St an

da rd

iz ed

m

or ta

lit y

ra te

s

County median household income ($)

F ig u r e 1 3 . 5 The relation between county median income and county death rates according to whether the counties are in the twenty-five more equal states or the twenty-five less equal states.

w ith E n glan d and W ales. Ju s t as am on g U S counties, w here the benefits o f greater state eq u ality extended to all incom e gro u ps, so the benefits o f S w eden ’s greater eq u ality extended across all classes, but w ere biggest in the lo w est classes.

Figure 8.4 in C hapter 8, w hich com pared yo u n g p eo p le’ s literacy scores across different countries accordin g to their p aren ts’ level o f education (and so indirectly accordin g to the social status o f their fam ily o f upbringing) also show ed that the benefits o f greater equality extend th roughout society. In m ore equal F inland and

B elgium the benefits o f greater eq u ality w ere, once again, bigger at the bottom o f the social ladder than in less equal U K and U S A . But even the children o f parents w ith the very highest levels o f education did better in Finland and Belgium than they did in the m ore unequal U K or U S A .

A question w hich is often asked is w h eth er even the rich benefit

from greater equality. Perhaps, as Jo h n D onne said, ‘N o m an is an

Islan d ’ even fro m the effects o f inequality. T h e evidence w e have

been discussing typ ically divides the p o pu latio n into three or four incom e or education al gro u ps, or o ccasio n ally (as in Figure 1 3 .4 ) into s ix occu pation al classes. In those an alyses it lo o ks as if even the richest groups do benefit. But if, w hen w e talk o f ‘the rich ’ , we

m ean m illion aires, celebrities, people in the m edia, running large

businesses or m ak in g the n ew s, w e can on ly guess h ow they m ight be affected. W e m ight feel w e live in a w o rld peopled by faces and nam es w hich keep crop p in g up in the m edia, but such people actu ally m ake up on ly a tiny fractio n o f r per cent o f the popu lation

and they are just to o sm all a p ro p o rtion o f the p o pu latio n to

lo o k at separately. W ith out data on such a sm all m in ority w e can o n ly guess w hether or not they are likely to escape the increased violence, drugs or m ental illness o f m ore unequal societies. Th e lives and deaths o f celebrities such as B ritney Spears, Jo h n Len n o n , K u rt C o b a in , M a rily n M o n ro e , the assassinated K en n edy brothers,

Princess D ian a or Princess M a rg a re t, suggest they m ight not. W hat the studies do m ake clear, h ow ever, is that greater eq u ality brings substantial gains even in the top o ccu patio n al class and am on g the richest or best-educated q u arter or third o f the po p u latio n , w hich include the sm all m in ority o f the seriously rich. In short, w h eth er w e lo o k at states or countries, the benefits o f greater eq u ality seem to be shared across the vast m a jo rity o f the p o pu lation . O n ly because the benefits o f greater equality are so w id ely shared can the differences in the rates o f problem s betw een societies be as large as it is.

A s the research findings have com e in over the years, the w id e­ spread nature o f the benefits o f greater equality seem ed at first so p a ra d o x ica l that they called everything into question. Several attem pts by international co llab o rative grou p s to com pare health inequalities in different countries suggested that health inequalities did not d iffer very m uch fro m one cou n try to another. T h is seemed inconsistent w ith the evidence that health w a s better in m ore equal societies. H o w could greater equality im p rove health unless it did so by n arro w in g the health differences betw een rich and po o r? A t the time this seemed a m ajo r stum bling block. N o w , h ow ever, w e can see h o w the tw o sets o f findings are consistent. Sm aller income

differences im p rove health fo r everyone, but m ake a bigger difference to the health o f the p o o r than the rich. I f sm aller incom e differences lead to ro u gh ly the sam e percentage reduction in death rates across

the w h o le society then, w hen m easured in relative term s, the d iffer­

ences in death rates betw een rich and p o o r w ill rem ain unchanged. Suppose death rates are 60 per 10 0 ,0 0 0 people in the bottom class and on ly 20 per 10 0 ,0 0 0 in the top one. I f yo u then k n o ck 50 per cent o ff death rates in all gro u ps, you w ill have reduced the death rate by 3 0 in the bottom gro u p and by 1 0 at the top. B ut although the p o o r have had m uch the biggest absolute decline in death rates, there is

still a th reefold relative class difference in death rates. W h atever

the percentage reduction in death rates, as long as it applies right across society, it w ill m ake m ost difference to the p o o r but still leave relative m easures o f the difference unchanged.

W e can n o w see that the studies w hich once lo o ked p a ra d o x ica l

w ere in fac t telling us som ething im portan t abo u t the effects o f

greater equality. By suggesting that m ore and less equal societies contained sim ilar relative health differentials w ithin them, they w ere telling us that everyone receives ro u gh ly pro p o rtion al benefits from greater eq u ality. T h ere are n o w several studies o f this issue using data

fo r U S states,8* 319>320 and at least five in tern ation al ones, w h ich p ro ­

vide consistent evidence that, rather than being confined to the p o o r, the benefits o f greater eq u ality are w id ely sp re ad .1S2' 315’ 317>318’ 321

O T H E R E X P L A N A T I O N S ?

It is clear that there is som ething w h ich affects h o w w ell o r b ad ly societies do across a w id e range o f social p roblem s, but h o w sure can w e be that it is inequality? B efore discussing w h eth er in equality plays a cau sal role, let us first see w h eth er there m ight be an y quite different explan atio n s.

A lth ou gh people h ave o ccasio n ally suggested that it is the E n glish ­ speakin g countries w h ich do b ad ly, that doesn ’t exp lain m uch o f the evidence. F o r exam p le, take mental health, w h ere the w o rst perform ers am on g the countries fo r w h ich there is co m p arab le data are En glish-speakin g. In C h ap ter 5 w e show ed that the highest rates

D Y S F U N C T I O N A L S O C I E T I E S

are in the U S A , fo llo w ed in turn by A u stra lia, U K , N e w Z e alan d and C an a d a . But even am on g those countries there is a very strong correlation betw een the prevalence o f m ental illness and inequality.

So inequality exp lain s w h y En glish-speakin g countries do b ad ly, a n d it exp lain s w h ich ones do better or w o rse than others.

N o r is it just the U S A and B ritain , tw o countries w h ich do have a lot in com m on , w h ich do b ad ly on m ost outcom es. P o rtu gal also does bad ly. Its p o o r perform an ce is consistent w ith its high levels o f inequality, but P ortu gal and the U S A cou ld h ard ly be less alike in other respects. H o w ev e r, the p ro o f that these relationships are not sim ply a reflection o f som ething w ro n g w ith E n glish-sp eakin g cu l­ tures is that even if yo u delete them from Figure 1 3 . 1 (p. 1 7 4 ) there is still a close relationship between inequality and the In dex o f H ealth and Social Problem s am on g the rem aining countries. The

sam e applies to the dom inance o f the N o rd ic countries at the other end o f the distribution. T h ey clearly share som e im portan t cultural characteristics. But, like the English speaking countries, if yo u delete them fro m Figure 1 3 . 1 , a strong relationship rem ains betw een in­ equality and the In dex am ong the rem aining countries.

A lth o u gh that puts paid to the only o b viou s cultural explan atio n s,

it’ s w o rth pointing out some interesting contrasts between countries. F o r exam p le, although P ortugal does b ad ly, Spain fares at least as w ell as the average - despite the fact that they share a border, they lived under dictators until the mid 19 7 0 s , and have m an y other cul­ tu ral sim ilarities. Y e t all that seems to be trum ped by the differences in inequality. Th e cou n try w h ich does best o f all is Ja p a n , but Ja p a n is, in other respects, as different as it cou ld be fro m Sw eden, w h ich is the n ext best perform er. T h in k o f the con trastin g fam ily structures and the position o f w om en in Ja p a n and Sw eden. In both cases these tw o countries com e at opposite ends o f the spectrum . Sw eden has

a very high pro p o rtion o f births outside m arriage and w om en are

alm ost eq u ally represented in politics. In Ja p a n the opposite is true. T h ere is a sim ilar stark con trast between the pro p o rtion o f w o m en in paid em ploym ent in the tw o countries. Even h o w they get their greater equality is quite different. Sw eden does it through redistribu­ tive taxes and benefits and a large w e lfa re state. A s a p ro p o rtion o f n ation al incom e, public social expenditure in Ja p a n is, in contrast to

Sw eden, am o n g the lo w est o f the m a jo r developed countries. Ja p a n gets its high degree o f eq u ality not so much fro m redistribution as

fro m a greater eq u ality o f m arket incom es, o f earnings b e fo re taxes and benefits. Y e t despite the differences, both countries do w ell - as their n a rro w incom e differences, but alm ost nothing else, w o u ld lead us to expect.

T h is leads us to another im p ortan t point: greater eq u ality can be gained either b y using taxes and benefits to redistribute very unequal incom es or by greater eq u ality in gross incom es before taxes and benefits, w h ich leaves less need fo r redistribution. So big governm ent m ay not a lw a y s be n ecessary to gain the advan tages o f a m ore equal

society. T h e sam e applies to other areas o f governm ent expenditure.

F o r countries in ou r international an alysis, w e collected O E C D figures on public so cial expenditure as a pro p o rtion o f G ro ss D om estic Product and found it entirely unrelated to our In dex o f Flealth and Social Problem s. Perhaps rather counter-intuitively, it

also m ade no difference to the association between in equality and

the In d ex. P art o f the reason fo r this is that governm ents m a y spend either to prevent social problem s o r, w h ere incom e differences have w idened, to deal w ith the consequences.

E xam p les o f these contrasting routes to greater eq u ality w hich w e h ave seen in the international data can also be foun d am on g the

fifty states o f the U S A . A lth ou gh the states w h ich perfo rm w ell are

dom inated by ones w h ich have m ore generous w e lfa re p ro visio n s, the state w h ich perform s best is N e w H am p sh ire, w hich has am ong the lo w est public social expenditure o f an y state. Lik e Ja p a n , it appears to get its high degree o f eq u ality through an unusual eq u ality o f m arket incom es. R esearch using data fo r U S states w hich tried to see w hether better w e lfa re services explain ed the better perform an ce o f m ore equal states fo u n d that although - in the U S setting - services appear to m ake a difference, they do not account fu lly fo r w h y m ore equal states do so m uch better.309 T h e really im portan t im plication is that h o w a society becom es m ore equal is less im portan t than w h eth er or not it actu ally does so.

E T H N I C I T Y A N D I N E Q U A L I T Y

People som etim es w o n d er w h eth er ethnic division s in societies accou n t fo r the relationship betw een inequality and the higher

frequ en cy o f health and social problem s. T h ere are tw o reasons for

thinking that there m ight be a link. First is the idea that som e ethnic grou ps are inherently less cap able and m ore lik ely to have problem s. Th is m ust be rejected because it is sim ply an expression o f racial prejudice. Th e other, m ore serious, p ossib ility is that m inorities often do w o rse because they are excluded from the education al and job

opportunities needed to do w ell. In this v ie w , prejudice against m inorities m ight cause ethnic divisions to be associated w ith bigger incom e differences and, flo w in g from this, also w ith w o rse health and m ore frequent social problem s. T h is w o u ld , h ow ever, produce a relation between incom e inequality and w o rse scores on ou r index

through very m uch the sam e processes as are responsible fo r the relation ship w h erever it occurs. Ethnic division s m ay increase social exclu sion and discrim ination, but ill-health and social problem s becom e m ore com m on the greater the relative deprivation people experience - w h atever their ethnicity.

People nearer the bottom o f society alm ost a lw ay s face d o w n w a rd discrim ination and prejudice. T h ere are o f course im p ortant d iffer­ ences between w h a t is seen as class prejudice in societies w ithou t ethnic division s, and as racial prejudice w h ere there are. A lth ough the cultural m arks o f class are derived inherently fro m status

differentiation, they are less indelible than differences in skin colour. But w hen differences in ethnicity, religion or lan guage com e to be seen as m arkers o f lo w social status and attract vario u s d o w n w a rd prejudices, social divisions and discrim ination m ay increase.

In the U S A , state incom e in equality is closely related to the

pro p o rtion o f A frican -A m erican s in the state’ s po pu latio n . T h e states

w ith w id er incom e differences tend to be those w ith larger A frican - A m erican popu lation s. T h e sam e states also have w o rse outcom es - fo r instance fo r health - am on g both the b lack a n d the w h ite p o p u ­ lation . T h e ethnic divide increases prejudice and so w iden s incom e

differences. T h e result is that both com m unities suffer. R a th e r than

w hites en joyin g greater privileges resulting fro m a larger and less

w ell-p aid b lack com m unity, the consequence is that life exp ectan cy is shorter am on g both b lack and w h ite population s.

So the an sw er to the question as to w h eth er w h a t appear to be the effects o f in equality m ay actu ally be the result o f ethnic divisions is

that the tw o in volve m ost o f the sam e processes and should not be seen as alternative explan atio n s. Th e prejudice w h ich often attaches to ethnic divisions m ay increase inequality and its effects. W here ethnic differences h ave becom e stron gly associated w ith social status division s, ethnic divisions m ay p ro vid e alm ost as go o d an in dicator o f the scale o f social status differentiation as incom e inequality. In this situation it has been claim ed that incom e differences are trum ped, statistically speakin g, by ethnic differences in the U S A .310

H o w ev e r, other papers exam in in g this claim have rejected it.311-13 Th e U S A , w ith its ethnic d ivision s, is o n ly one o f a great m an y co n ­ texts in w h ich the im p act o f incom e in equality has been tested. W e review ed 1 6 8 published reports o f research exam in in g the effect o f inequality on health, and there are n o w aroun d 2 0 0 in a ll.10 In m any o f these (for exam ple Portugal) there is no p ossib ility that effects

could be attributed to ethnic divisions. A n international study w hich included a m easure o f each co u n try ’ s ethnic m ix , foun d that it did not accou n t fo r the tendency fo r m ore unequal societies to be less h ealth y.314

S I N G L E P A R E N T S

A s w e noted near the beginning o f this chapter, it is u sually the sam e countries that do w ell and the sam e ones w h ich do b ad ly w h atever

health or so cial problem s w e lo o k at. T h e fact that so m an y quite different problem s share the sam e in tern ation al pattern im plies that they have a com m on underlying cause. T h e question is w hether that com m on cause is inequality. A n oth er altern ative po ssib ility is that these problem s m ight all be rooted in the b reak d o w n o f the tw o-

parent fam ily as the unit in w h ich children are brought up. T h ere is a tendency to blam e a w ide range o f social problem s on bad parenting

- p articu larly resulting from the increased prevalence o f single

parents.

D ata com parin g children b rought up in single parent fam ilies w ith

those b rought up by tw o parent fam ilies alm ost a lw a y s show s that the children o f single parents do less w ell. M o re co n tro versial is the question o f h ow much this reflects differences in m oth ers’ education and m aternal depression ,397 h o w much is due to the tendency for

single paren t fam ilies to be p o o rer, and h o w much results from less

go o d parent-child relationship s. U su ally all these facto rs are found to m ake substantial contributions.

Th e p ro p o rtion o f parents w h o are single varies d ram atically from one nation to another. In countries like G reece, o n ly abo u t 4 per cent

o f fam ilies w ith children are single p arent fam ilies, but in others, like

the U S A , B ritain and N e w Z e a la n d , it rises to alm ost 3 0 per cent. C o u ld this exp lain w h y children in some countries do less w ell than others? R ath er than in equality, is the real issue the problem s o f single paren th ood? T o find out, w e lo o ked to see if the U N I C E F in d ex o f child w ellbein g w a s related to the pro p o rtion o f parents w h o were

single parents in each cou n try. T h e surprisin g results are show n in

Figure 1 3 . 6 . T h ere is no connection between the pro p o rtion o f single parents an d nation al standards o f child w ellbeing. T h is contrasts sharp ly w ith the strong relationship between child w ellb ein g and incom e inequality show n in Figure 2 .6 (see p. 23).

T h a t there is so little connection at the international level between

child w ellbein g and the p ro p o rtion o f single parents is pro b ab ly p a rtly a reflection o f the extent to w h ich w elfare system s in some countries protect single paren t fam ilies fro m poverty. R ecent O E C D figures suggest that on ly 6 per cent o f Sw edish single parents with job s, and 1 8 per cent o f those w ith o u t, w ere in relative po verty, as again st 3 6 and 9 2 per cent fo r the U S A .399 The figures fo r the U K are 7 per cent fo r single parents w ith jobs and 39 per cent fo r those w ithou t. T h e pro visio n o f childcare w h ich enables single parents to w o rk m ust also be im portant.

G iven the po litical controversies around the p ro visio n o f state su pport to single parents, tw o points are w o rth noting. First, that it seems to be possible to safegu ard children against m ost o f the adverse effects o f being b rought up by lone p arents, and second, that

Better -

£ TD

ULtUJ U

W orse -

Norway • • Spain

Italy • • Belgium • Canada

Ireland* Portugal. ‘ Austria

• France

• Germany

O 1 0 2 0 3 0

Lone parents as % o f all households w ith dependent children

Figure 13 .6 Child wellbeing is not related to the proportion o f single parents.39*

denying state su pport does not seem to reduce the p ro po rtion o f single parents.

D I F F E R E N T H I S T O R I E S

A n oth er ex p lan atio n som etim es suggested fo r w h y incom e inequality is related to health and social problem s is that w h a t m atters is not

the inequality itself, but the historical factors w h ich led societies to

becom e m ore or less equal in the first place - as if in equality stood, alm ost as a statistical m onum ent, to a h istory o f division. T h is is m ost often suggested in relation to the U S A w hen people notice that the m ore unequal states are u su ally (but not alw ays) the south ­

ern states o f the C o n fed eracy w ith their histories o f plantation econom ies dependent on slave lab o u r. H o w ev e r, the degree o f eq u ality or in equality in every setting has its o w n p articu lar history. If w e lo o k to see h o w Sweden becam e m ore equ al, or h o w B ritain and a num ber o f other countries h ave recently becom e m uch less so,

o r h o w the regions o f R u ssia or C h in a developed va ry in g am ounts

o f equality or in equality, w e get different stories in every case. A nd

o f course these different backgrou n ds are im portant: there is no doubt that there are, in each case, specific historical exp lan atio n s o f w h y some countries, states o r regions are n o w m ore or less unequal than others. But the prevalence o f ill-health and o f social problem s in

those societies is not sim ply a patternless reflection o f so m any

unique histories. It is instead patterned accordin g to the am ount o f inequality w hich has resulted from those unique histories. W hat seems to m atter therefore is not h o w societies got to w here they are n o w , but w h e re - in terms o f their level o f in equality - it is that they

h ave n o w got to. T h at does not m ean that these relations w ith in equality are set in

stone fo r all time. W h at does change things is the stage o f econom ic developm ent a society has reached. In this b o o k ou r focus is e x ­ clu sively on the rich developed societies. But it is clear that a num ber o f outcom es, including health and violence, are also related to

in equality in less developed countries. W h at happens during the course o f econom ic developm ent is that som e problem s reverse their social gradients an d this changes their association s w ith inequality. In po o rer societies both obesity and heart disease are m ore com m on

am on g the rich, but as societies get richer they tend to reverse their

social distribution and becom e m ore com m on am ong the p o o r. A s a result, w e find that am on g p o o rer countries it is the m ore unequal ones w h ich have m ore underw eight people - the opposite o f the pattern am on g the rich countries show n in C h ap ter 7. Th e age o f m enarche also changes its so cial distribution during the course o f econom ic developm ent. W hen m ore o f the p o o r w ere u n dern our­

ished they reached sexu al m atu rity later than girls in richer fam ilies. W ith the rise in living standards that pattern too has reversed - per­ h aps contributing to the gradient in teenage pregnancies described in C h ap ter 9. A ll in all, it lo o ks as if econom ic gro w th and social status differences are the m ost p o w e rfu l determ inants o f m any aspects o f

ou r lives.

C A U S A L I T Y

The relationships betw een in equality and p o o r health and social problem s are to o stron g to be attributable to chance; they occur independently in both ou r test-beds; and those between inequality and both violence and health have been dem onstrated a large num ber o f times in quite different settings, using data from different sources. But association on its o w n does not p ro ve cau sality and,

even if there is a cau sal relation ship, it do esn ’t tell us w h a t is cause

and w h a t is effect.

T h e graph s we have show n have all been cross-section al - that is, they have show n relationships at a p articu lar point in time rather than as they change in each cou n try o ver time. H o w ev e r these cro ss­ sectional relationships cou ld only keep crop pin g up if so m eh o w they

changed together. I f health and in equality w en t their separate w a y s and passed b y only coin ciden tally, like ships in the night, w e w o u ld not keep catching repeated glim pses o f them in close fo rm atio n . T h ere is u sually not enough in tern ation ally co m p arab le data to track relationships o ver tim e, but it has been possible to lo o k at changes in

health and inequality. O ne study foun d that changes between L975 and 1 9 8 5 in the p ro p o rtion o f the po pu latio n livin g on less than h alf the nation al average incom e am on g w h a t w ere then the tw elve mem bers o f the E u ro p ean U nion w ere significantly related to changes in life exp e cta n cy .81 Sim ilarly, the decrease in life expectan cy

in E astern E uropean countries in the six years fo llo w in g the collapse

o f com m unism ( 1 9 8 9 - 9 5 ) w a s show n to be greatest in the countries w hich sa w the m ost rapid w idening o f incom e differences. A longer- term and p articu larly striking exam p le o f h o w incom e distribution and health change o ver time is the w a y in w h ich the U S A and Ja p a n

sw ap ped places in the international league table o f life expectan cy

in developed countries. In the 1 9 5 0 s , health in the U S A w a s only surpassed b y a fe w countries. Ja p a n on the other h and did bad ly. But b y the 19 8 0 s Ja p a n had the highest life exp ectan cy o f all developed countries and the U S A had slipped do w n the league and w a s w ell

on the w a y to its current po sitio n as num ber 30 in the developed

w o rld . C ru cia lly , Ja p a n e se incom e differences n arro w ed during the

fo rty years after the Second W o rld W ar. Its health im p roved rap id ly,

o vertak in g other countries, and its crim e rate (alm ost alone am ong

developed countries) decreased. M ea n w h ile, U S incom e differences w idened from abo u t 1 9 7 0 o n w ards.

C h ap ter 3 provided a general exp lan atio n o f w h y w e are so sensi­ tive to inequality, and in each o f C hapters 4 - 1 2 w e have suggested

cau sal links specific to each health and social problem . E arlier in this

chapter w e sa w w h y cultural facto rs can n ot be regarded as rival exp lan atio n s o f the association s w ith inequality. W h at other e x p la ­ nation m ight there be if one w an ted to reject the idea o f a causal relationship? C o u ld in equality and each o f the social problem s be caused by som e other un kn ow n factor?

W eak relationship s m ay som etim es turn out to be a mere m irage

reflecting the influence o f som e underlying facto r, but that is much less plausible as an exp lan atio n o f relationship s as close as these. The fact that ou r In d ex is not significantly related to average incom es in either ou r international test-bed or am on g the U S states alm ost certain ly rules out any underlying facto r directly related to m aterial

living standards. O ur an alysis earlier in this chapter also rules out governm ent social expenditure as a possible alternative explan atio n . A s fo r other possible hidden facto rs, it seems u nlikely that such an im p ortan t cau sal fac to r w ill suddenly com e to light w hich not only determ ines inequality but w h ich also causes everything fro m poor health to obesity and high prison population s.

T h a t leaves the question o f w h ich w a y cau sality goes. O ccasio n ­ ally w hen w e describe ou r findings people suggest that instead o f inequality causing everything else, perhaps it all w o rk s the other w a y rou n d and health and social problem s cause bigger incom e d iffer­ ences. O f course, in the real w o rld these things do not happen in clearly defined steps w hich w o u ld a llo w us to see w h ich com es first. T h e lim ited evidence fro m studies o f changes o ver time tells us only that they tend to change together. C o u ld it be that people w h o succum b to health or social problem s suffer a loss o f incom e and that tends to increase inequality? Perhaps people w h o are sick or very o verw eight are less likely to have jobs or to be given prom otion . C o u ld this exp lain w h y countries w ith w o rse health and social problem s are m ore unequal?

T h e short an sw er is no - or at least, not m uch. First, it do esn ’ t explain w h y societies that do b ad ly on any p articu lar health or social problem tend to do b ad ly on all o f them. If they are not all caused at least p a rtly by the sam e thing, then there w o u ld be no reason w h y

countries w h ich, fo r instance, have high obesity rates should also have high prison po pu latio n s. Second, som e o f the health and social problem s are u n likely to lead to serious loss o f incom e. U sing the U N I C E F in d ex w e show ed that m any ch ildh ood outcom es w ere w o rse in m ore unequal countries. But lo w child w ellb ein g w ill not have a m ajo r influence on incom e in equality am ong adults.

N o r could higher hom icide rates be considered as a m a jo r cause o f inequality even if the num bers w ere m uch higher. N o r fo r that m atter could exp an d in g prison p opu lation s lead to w id er incom e differences - rather the reverse, because m easures o f in equality are u su ally based on m easures o f h ousehold incom e w h ich leave

out institutionalized p opu lation s. A lth o u gh it could be argued that

teenage parents m ight increase in equality because they are often single and p o o r, w e have seen that even w hen m ore equal countries have a high p ro po rtion o f single parents th at does not explain nation al differences in child w ellb ein g. T h is is p a rtly because gener­ ous w e lfa re systems ensure that v e ry m uch sm aller pro po rtion s o f

them are in po verty than in m ore unequal countries. H o w ev e r, there is a m ore fun d am en tal objection to the idea that

cau sality m ight go fro m social problem s to in equality. E arlier in this chapter w e show ed that it w a s people at alm ost all incom e levels, not just the p o o r, w h o do w o rse in m ore unequal societies. Even w hen you com pare grou p s o f people w ith the sam e incom e, you find that those in m ore unequal societies do w o rse than those on the sam e incom e in m ore equal societies. T h o u g h some m ore unequal societies have m ore p o o r people, m ost o f the relationship w ith in eq uality is, as w e pointed out earlier, not explain ed by the p o o r: the effects are

m uch m ore w idesp read. So even if there is som e loss o f incom e

am on g those w h o are sick or affected by som e social prob lem , this does not begin to exp lain w h y people w h o rem ain on perfectly go o d incom es still do w o rse in m ore unequal societies.

A n oth er alternative ap p ro ach is to suggest that the real cause is not incom e distribution but som ething m ore like changes in

id eo lo gy, a shift perhaps to a m ore in dividualistic econom ic p h ilo ­ sophy or vie w o f society, such as the so-called ‘ n eo-lib eral’ thinking. D ifferen t ideologies w ill o f course affect not on ly governm ent policies but also decisions taken in econom ic institutions throughout society. T h ey are one o f very m any different facto rs w h ich can affect

the scale o f incom e differences. But to say that a change in id eology can affect incom e distribution is not at all the sam e as sayin g that it can also affect all the health and social problem s w e have discussed - regardless o f w h a t happens to incom e distribution. A lth o u gh it does lo o k as if neo-liberal policies w idened incom e differences (see

C h ap ter 16 ) there w a s no governm ent intention to lo w er social

cohesion or to increase violence, teenage births, obesity, drug abuse and everything else. So w h ile changes in governm ent id eo lo gy m ay som etim es be am on g the causes o f changes in incom e distribution, this is not part o f a p ack age o f policies intended to increase the

prevalence o f social problem s. T h eir increase is, instead, an un­

intended consequence o f the changes in incom e distribution. R ather than challenging the cau sal role o f in equality in increasing health and social problem s, if governm ents u nderstood the consequences o f w iden ing incom e differences they w o u ld be keener to prevent them.

Econom ists have never suggested that p o o r health and social problem s w ere the real determ inants o f incom e inequality. Instead they have concentrated on the contributions o f things like ta xe s and benefits, international com petition, changing techn ology and the m ix o f skills needed b y industry. N o n e o f these is o b v io u sly connected to the frequency o f health and social problem s. In C h ap ter 1 6 w e shall

touch on the facto rs responsible fo r m ajo r changes in in equality in

different countries. A difficulty in p ro vin g cau sality is that w e cannot experim en tally

reduce the inequalities in h a lf ou r sam ple o f countries and not in the others an d then w a it to see w h a t happens. But p u rely o b servation al

research can still produce p o w erfu l science - as astron o m y show s. T h ere are, h ow ever, som e experim en tal studies w h ich do support ca u sality w o rk in g in the w a y our argum ent suggests. Som e o f them have alread y been m entioned in earlier chapters. In C h ap ter 8 on education w e described experim ents w h ich sh o w h o w m uch p eop le’ s

perform an ce is affected by being categorized as so cially inferior.

Indian children fro m lo w er castes solved mazes just as w ell as those fro m higher castes - until their lo w caste w a s m ade kn o w n . E x p e ri­

ments in the United States have show n that A frican -A m erican students (but not w h ite students) do less w ell w h en they are told a test is a test o f ability than they do on the sam e test w hen they are told it is not a test o f ability. W e also described the fam ou s ‘ blue-

eyes’ experim ents w ith school children w h ich show ed the same processes at w o rk .

Som etim es asso ciatio n s w hich are on ly observed am ong hum an beings can be sh o w n to be cau sal in an im al experim ents. Fo r

instance, studies o f civil servants sh o w ca rd io v ascu la r health declines w ith declining social status. But h o w can w e tell w hether the dam age is caused by lo w social status rather than b y p o o rer m aterial co n ­ ditions? E xp erim en ts w ith m acaque m onkeys m ake the an sw er clear. M ac a q u e s fo rm status hierarchies but w ith captive colonies it is possible to ensure all anim als live in the sam e m aterial conditions: they are given the sam e diet and live in the sam e com pounds. In addition , it is possible to m anipulate social status by m ovin g anim als between grou p s. If yo u take lo w -status anim als fro m different

grou ps and house them together, some have to becom e high-status. Sim ilarly, if you put high-status an im als together som e w ill becom e low -status. A n im als w h ich m ove do w n in these conditions have been found to h ave a rap id build-up o f atherosclerosis in their arteries.322 Sim ilar experim ents also suggest a cau sal relation ship between lo w

social status and the accum ulation o f abd om in al fa t.323 In C h apter 5 w e m entioned other anim al experim ents w h ich show ed that when cocaine w a s m ade a v ailab le to m onkeys in these conditions, it w a s taken m ore b y lo w social status an im als - as if to offset their lo w er dopam ine activity .59 L a stly , the p rim ary im p ortance o f inequality

has been confirm ed by researchers using statistical m ethods designed to check the causal p a th w ay s through w h ich in eq uality affects levels o f trust o r bullying in sch o o ls.27’ 400’ 402

A lth ou gh w e k n o w o f no experim ents confirm ing the cau sality o f the relation betw een inequality and violence, w e invite anyone to go

into a p o o r p art o f to w n and try ran d o m ly insulting a fe w people.

W e have discussed the reasons fo r thinking that these links are cau sal fro m a num ber o f different perspectives. But as philosophers

o f science, such as Sir K a rl P opp er, have em phasized, an essential

elem ent in judgin g the success o f an y th eory is w hether it m akes successful predictions. A successful th eory is one w h ich predicts the existence o f p revio u sly u n kn ow n phenom ena or relationships w hich can then be verified. The theory that m ore equal societies were healthier arose from one set o f international data. T h ere have n o w been a very large num ber o f tests (about 200) o f that th eory in d iffer­

ent settings. W ith the exception o f studies w h ich looked at inequality in sm all local areas, an overw h elm in g m ajo rity o f these tests confirm ed the theory. Second, if the link is cau sal it im plies that

there m ust be a mechanism . T h e search fo r a m echanism led to the

disco very that social relationships (as m easured by social cohesion,

trust, involvem ent in com m unity life and lo w levels o f violence) are better in m ore equal societies. T h is happened at a tim e w h en the im portance o f social relation ship s to health w a s beginning to be m ore w id ely recognized. T h ird , the th eory that p o o r health m ight be

one o f a range o f problem s w ith social gradients related to inequality has been tested (initially on cause-specific death rates as described earlier in this chapter) and has since been am ply confirm ed in tw o different settings as w e have described in C h apters 4 - 1 2 . Fourth, at a time w hen there w a s no reason to think that in equality had

p sych osocial effects, the relation betw een health and eq u ality seemed

to im p ly that in equality m ust be affecting health through psych o ­ social processes related to social differentiation. T h a t inequality does have p o w erfu l psych o so cial effects is n o w confirm ed by its links (show n in earlier chapters) w ith the qu ality o f social relations and

num erous b eh aviou ral outcom es. It is very difficult to see h o w the enorm ous variatio n s w h ich exist

fro m one society to another in the level o f problem s associated w ith lo w social status can be exp lain ed w ith o u t accepting that inequality is the com m on denom inator, and a hugely dam agin g force.

A ccepting this does not in vo lve a huge theoretical leap. T w o points should be kept in m ind. First, the evidence m erely confirm s the com m on intuition that inequality is divisive and so cially co rro ­ sive. Second, everyone k n o w s that w ithin our societies ill health and social problem s are related to social status and are m ost com m on in the m ost deprived neighbou rhood s. T h o u g h yo u could once have

been forgiven fo r thinking that this m erely reflected a tendency fo r the vu ln erable to end up at the bottom o f society, it is n o w o b viou s that this fails to exp lain w h y these problem s are so m uch m ore

com m on in m ore unequal societies. T h is b o ok sim ply points out that if you increase the incom e and status differences related to these problem s, then - u n surp risin gly - the problem s all becom e m ore com m on.

1 4

Our social inheritance

Gifts make friends and friends make gifts. M arshall Sahlins, Stone A ge Economics

L O O K I N G B E F O R E L E A P I N G

A lth o u gh attitudes to in equality have a lw a y s been central to the disagreem ent betw een the political right and left, few w o u ld not prefer a friendlier society, w ith less violence, better m ental health, m ore involvem ent in com m un ity life - and so on. N o w that w e have

show n that reducing in equality leads to a very m uch better society, the m ain sticking point is w h eth er people believe greater equality is attainable. O u r an alysis has not o f course com pared existing societies w ith im p ossibly egalitarian im agin ary ones: it is not about

u topias o r the exten t o f hum an perfectibility. E veryth in g w e have seen com es from co m pariso n s o f existin g societies, and those societies have not been p articu larly unusual or odd ones. Instead, we have lo o ked exclu sively at differences betw een the w o r ld ’ s richest and m ost successful econom ies, all o f w h ich en joy dem ocratic institutions and freedom o f speech. T here can be no dou b t w h a tso ­

ever that hum an beings are capable o f livin g w ell in societies w ith

inequalities as sm all - fo r instance - as Ja p a n an d the N o rd ic co u n ­ tries. F a r fro m being im p ractical, the im plications o f ou r findings are p ro b ab ly m ore consistent w ith the institutional structures o f m arket dem ocracy than som e people - at either end o f the po litical spectrum

- w o u ld like to believe. Som e m ay still feel hesitant to take the evidence at face value.

I can therefore conclude that the primates are indeed social animals,”

From the van tage point o f m ore unequal countries, it m ay seem genuinely perplexin g and difficult to understand h o w som e, a p p ar­ ently sim ilar, countries can function w ith so m uch less inequality. Evidence that m aterial self-interest is the govern in g principle of hum an life seems to be everyw h ere. Th e efficiency o f the m arket econ om y seems to prove that greed and avarice are, as econom ic th eory assum es, the overridin g hum an m otivation s. Even the burden o f crim e appears to spring fro m the difficulty o f stopping people breakin g the rules to satisfy selfish desires. Signs o f a carin g, sharing,

hum an nature seem thin on the ground. Som e o f this scepticism m ight be allayed by a m ore fundam ental

u nderstanding o f h ow w e, as hum an beings, are dam aged by in­ equality and have the cap acity fo r som ething else. W e need to under­ stand h o w , w ithou t genetically re-engineering ourselves, greater

equality allo w s a m ore sociable hum an nature to emerge.

T W O S I D E S O F T H E C O I N

In ou r research fo r this b o o k , social status and friendship have kept crop pin g up together, linked in extricab ly as a p air o f opposites.

First, they are linked as determ inants o f the health o f each in d ivid ­ ual. A s w e sa w in C h ap ter 6, friendship and involvem ent in social life are highly protective o f go o d health, w h ile lo w social status, or bigger status differences and m ore inequality, are harm ful. Second, the tw o are again linked as they v a ry in societies. W e sa w in C h a p ­ ter 4 that as in equality increases, so ciab ility as m easured by the strength o f com m unity life, h o w m uch people trust each other, and the frequency o f violence, declines. T h ey crop up together fo r a third tim e in people’s tendency to choose friends fro m am ong their near equals: larger differences in status or w ealth create a social gu lf

betw een people. W h at binds social status and friendship together in these different

w a y s? T h e exp lan atio n is sim ple. T h ey represent the tw o opposite w a y s in w hich hum an beings can com e together. Social status stratification , like ran k in g systems or pecking orders am on g anim als, are fun dam en tally orderings based on p o w e r and coercion , on

privileged access to resources, regardless o f o th ers’ needs. In its most naked and anim al fo rm , m ight is right and the w eak est eat last.

Friendship is alm ost ex a ctly the opposite kind o f relationship. It is abo u t recipro city, m u tu ality, sharing, social o b ligation s, co ­ o peration and recognition o f each o ther’ s needs. G ifts are sym bols o f friendship because they dem onstrate that the giver and receiver do

not com pete fo r access to necessities, but instead recognize and

respond to each o ther’ s needs. In the w ell-chosen w o rd s o f M a rsh a ll Sahlins, a social an th ro po lo gist, ‘gifts m ake friends and friends m ake g ifts’ .324 Food -sh arin g and eating together c a rry the sam e sym bolic m essage, and they do so p articu larly p o w e rfu lly because fo o d is the

m ost fundam ental o f all m aterial necessities. In tim es o f scarcity,

com petition fo r fo o d has the poten tial to be extra o rd in a rily socially destructive.

F R I E N D O R F O E

Social status and friendship are so im portan t to us because they

reflect different w a y s o f dealing w ith w h at is perhaps the m ost fu n ­ dam ental problem o f social organ izatio n and p olitical life am ong anim als and hum ans alike. Because m em bers o f the sam e species have the sam e needs as each other, they have the potential to be each

o ther’s w o rst rivals, com peting fo r alm ost everything - fo r food ,

shelter, sexu al partn ers, a co m fo rtab le place to sit in the shade, a good nesting site - indeed fo r all scarce co m fo rts an d necessities. A s a result, am ong very m any species the m ost frequent conflicts take place not so m uch betw een m em bers o f different species, despite the danger o f predators, but between m em bers o f the sam e species.

A lo w -status b ab o o n has to spend m uch m ore time keeping out o f the w a y o f a dom inant b ab oon than in a vo id in g lions. M o st o f the bite m arks and scars w h ich subordinate anim als bear com e from m ore dom inant m em bers o f their o w n species. Y o u can see signs o f riva lry w ithin species all aroun d us - you have o n ly to w atch birds at a garden feeder, or dogs fighting, o r think o f the banned sport o f co ck fighting: in each case the conflicts are w ithin the species.

H u m an beings h ave to deal w ith the sam e problem . W riting in the

seventeenth century, T h o m as H o b b es m ade the danger o f conflict,

caused by riva lry fo r scarce resources, the basis o f his political p h ilo so p h y .325 A s w e all have the sam e needs, com petition fo r scarce necessities w o u ld lead to a continuous conflict o f ‘ every m an against every m an ’ . H o b b es believed that, because o f this danger, the m ost im p ortan t task o f governm ent w a s sim ply to keep the peace. He assum ed that, w ith o u t the firm hand o f governm ent, life ‘ in a state o f n ature’ w o u ld be ‘ so litary, p o o r, nasty, brutish, and sh o rt’ .

But perhaps H o b b es m issed an im portan t p art o f the story. A s w ell as the potential fo r conflict, hum an beings have a unique poten tial to be each other’ s best source o f co-operation , learning,

love and assistance o f every kind. W hile there’ s not m uch that

ostriches or otters can do fo r an injured m em ber o f their ow n species, am ong hum ans there is. But it’ s not just that w e are able to give each other care and protection. Because m ost o f our abilities are learned, w e depend on others fo r the acquisition o f ou r life skills.

S im ilarly, our unique cap acity fo r specialization and division o f lab o u r m eans that hum an beings have an u n rivalled potential to benefit fro m co-operation . So as w ell as the potential to be each o ther’s w o rst riv a ls, w e also have the potential to be each o th er’s greatest source o f co m fo rt and security.

W e have becom e attentive to friendship and social status because

the q u ality o f social relationships has a lw a y s been crucial to w e ll­ being, determ ining w h eth er other people are feared rivals or vital sources o f security, co-o p eratio n and support. So im portant are these dim ensions o f social life that lack o f friends and lo w social status are

am on g the m ost im p ortan t sources o f chronic stress affecting the

health o f p o pu lation s in rich countries today. A lth o u gh H o b b es w a s right abo u t the u n derlyin g prob lem o f the

dangers o f com petition betw een m em bers o f the sam e species, his v ie w o f h o w societies m anaged before the developm ent o f go vern ­ ments w ith the p o w e r to keep the peace w a s very w ide o f the m ark.

N o w that w e h ave m uch m ore k n ow led ge o f hunting and gathering societies it is clear that ou r ancestors did not live in a state o f co n ­ tinuous conflict. Instead, as Sahlins pointed out, they had other w a y s o f keeping the peace.324 T o av o id the ‘w a rre o f each again st a ll’ , social and econom ic life w a s based on system s o f gift exchan ge, food

sharing, and on a very high degree o f equality. These served to

minim ize an im osity and keep relation s sweet. F orm s o f exchange in vo lvin g direct expression s o f self-interest, such as buying and sell­

ing or b arter, w ere u su ally regarded as so cially unacceptable and outlaw ed.

These patterns dem onstrate the fundam ental truth: system s o f m aterial or econom ic relations are system s o f social relations.

E C O N O M I C E X P E R I M E N T S

E con om ic theory has trad ition ally w o rk e d on the assum ption that hum an b eh aviou r cou ld be explain ed largely in terms o f an inherent tendency to m axim ize m aterial self-interest. B ut a series o f ex p e ri­

ments using econom ic gam es h ave n o w show n h o w fa r fro m the truth this is.

In the ‘ ultim atum g am e’ , volunteers are ran d o m ly paired but rem ain an on ym o u s to each other an d do not meet. A k n o w n sum o f m oney is given to the ‘p ro p o ser’ w h o then divides it as he or she pleases w ith the ‘resp on d er’ . A ll the responders do is m erely accept

or reject the offer. I f rejected, neither partner gets an yth in g, but if it is accepted, they each keep the shares o f m oney offered.

T h ey p lay this gam e on ly once, so there is no p oin t in rejecting a sm all o ffer to try to force the p ro p o ser to be m ore generous n ext time - they k n o w there isn ’t goin g to be a n ext time. In this situation,

self-interested responders should accept an y o ffe r, h ow ever derisory, and self-interested proposers should o ffer the sm allest positive am ount, ju st enough to ensure that a responder accepts it.

A lth o u gh experim ents sh o w that this is ex a ctly h o w chim panzees beh ave,326 it is not w h a t happens am on g hum an beings. In practice,

the average o ffer m ade by people in developed societies is u sually betw een 4 3 and 48 per cent, w ith 50 per cent as the m ost com m on o ffe r.327 A t direct cost to ourselves, w e com e close to sharing equ ally even w ith people w e never meet an d w ill never interact w ith again.

R espon ders tend to reject offers b elo w abo u t 20 per cent. R ejected

o ffers are m oney w h ich the respon d er chooses to lose in o rder to

punish the pro p o ser an d prevent them benefiting from m ak in g a

m ean offer. T h e hum an desire to punish even at som e personal cost has been called ‘ altruistic punishm ent’ , and it plays an im p ortant

role in rein forcin g co-operative b eh aviou r and preventing people freeloading.

A lth ou gh the studies o f h o w people played the ultim atum game w ere not concerned w ith the levels o f in equality in each society, they are, nevertheless, abo u t h o w eq u ally or un equ ally people choose to divide m oney between them selves and som eone else. T h ey are con ­

cerned w ith w h a t peop le feel is a prop er w a y to treat others (even w hen there is no direct con tact betw een them an d they bear the cost o f an y generosity). T h e egalitarian preferences people reveal in the ultim atum gam e seem to fly in the face o f the actual inequalities in our societies.

C H I M P S A N D B O N O B O S

Som e non-hum an prim ates are m uch m ore h ierarch ical than others. L o o k in g at their different social system s, it often seems as if the am ount o f conflict, the qu ality o f social relations and the relation ­ ship betw een the sexes are functions o f h o w hierarchical they are. H u m an beings are not o f course bound to an y one social system. O ur adap tab ility has enabled us to live in v e ry different social structures, both very egalitarian and very h ierarch ical. But som e o f

the sam e effects o f h ierarch y on other aspects o f our social systems still seem to be visible - even though the b eh aviou ral patterns are driven by culture rather than by instinct. Less h ierarch ical societies are less m ale-dom inated so, as w e sa w in C h ap ter 4 , the position o f w o m en is better. Sim ilarly, the q u ality o f social relations in m ore

equal societies is less hostile. People trust each other m ore and co m ­

m unity life is stronger (Chapter 4), there is less violence (Chapter 10 ) and punishm ent is less harsh (C h apter 1 1 ) .

A ro u n d six or seven m illion years ago the branch o f the ev olu tio n ­ a ry tree from w hich w e have em erged split from that w h ich led to tw o different species o f ape: chim panzees and b o n ob o s. G enetically

w e are eq u ally clo sely related to both o f them , yet there are striking differences in their social b eh aviou r and they illustrate sharply

con trastin g w a y s o f solving the H o b b esian prob lem o f the potential fo r conflict o ver scarce resources.

Bands o f chim panzees are headed by a dom inant m ale w h o gains his position largely on the basis o f su perior size, strength, an d an

ability to form alliances - often including su pp o rt from fem ales.

D om inance hierarchies in any species are orderings o f access to scarce resources, including - as fa r as m ales are concerned - re p ro ­ ductive access to fem ales. R an k in gs w ithin the dom inance hierarchy are established and m aintained through frequent contests, displays and assessm ents o f strength. In the w o rd s o f prim atologists F ran s de

W aal and F ran s Lan tin g:

Chimpanzees go through elaborate rituals in which one individual communicates its status to the other. Particularly between adult males, one male will literally grovel in the dust, uttering panting grunts, while the other stands bipedally performing a mild intimidation display to make

clear who ranks above whom.328- p- 30

B on o b o s, on the other hand, behave very differently. N o t o n ly is there m uch less conflict between n eighbouring grou p s o f bonobos than betw een neighbouring groups o f chim ps, but bon obos - again unlike chim ps - have a high degree o f sex eq uality. Fem ales are

at least as im portan t as m ales, and dom inance hierarchies are m uch

less pronounced. A lth ou gh m ales are slightly larger than fem ales, fem ales are u sually allo w ed to eat first. O ften dubbed the ‘ carin g, sh arin g’ apes, they engage in sexu al activity - including m utual m asturbation - frequently and in an y com bin ation o f sexes and ages.

Sex has evolved not only to serve reproductive functions, but also to relieve tensions in situations w h ich, in other species, m ight cause conflict. A s de W aal says, ‘ sex is the glue o f b o n ob o so ciety’ .329’ p- 99 It eases conflict, signals friendliness, and calm s stressful situations. B on o b o s use sex to solve the prob lem o f h o w to av o id conflict o ver access to scarce resources. Feeding time is app aren tly the peak o f

sexu al activity. Even before fo o d is th ro w n into their enclosure, male

bon obos get erections and m ales and fem ales invite both opposite and sam e-sex partners fo r sex. Possible conflict o ver non -food resources is dealt w ith in the sam e w a y .

A lth ou gh sexu al activity is not a prelim in ary to feeding am ong

hum ans, eating is a peak o f so ciality - w h eth er in the form o f shared

fam ily m eals, m eals w ith friends, feasts and banquets, or even in the religious sym bolism o f sharing bread and w in e at com m union.

Sum m ing up the b eh aviou ral difference between chim ps and b o n ob o s, de W aal and Lan tin g said: ‘If, o f the tw in concepts o f sex and p o w er, the chim panzee has an appetite fo r the second, the bo n ob o clearly has one fo r the first. The chim panzee resolves sexu al

issues (disputes) w ith p o w er; the b on obo resolves p o w e r issues w ith se x .’ 328, p- 32 Perhaps as a result o f these differences, bo n ob o s are, as research has show n , better at co-operative tasks than chim ps.

So w h a t m akes the difference? Interestingly, a section o f D N A ,

k n o w n to be im p ortant in the regulation o f so cial, sexu al and

parenting beh aviou r, has been foun d to d iffer betw een chim ps and b o n o b o s.329 It is perhaps co m fo rtin g to k n o w that, at least in this section o f D N A , hum ans h ave the b on obo rather than the chimp pattern, suggesting that ou r com m on ancestor m ay h ave had a

preference fo r m akin g love rather than w a r.

T H E S O C I A L B R A I N

T h e fact that w e can sim ultaneously agree w ith Sartre that ‘hell is

other p e o p le’ and also recognize that other peop le can be heaven, sh o w s h o w deeply enm eshed in social life w e are. R esearch looking fo r the m ost potent sources o f stress affectin g the card io vascu lar system concluded that ‘conflicts and tensions w ith other people are by far the m ost distressing events in d aily life in terms o f both initial

and enduring effects on em otional w ellb ein g’ - m ore so than the dem ands o f w o rk , m oney w o rries o r other difficulties.330 T h e quality o f ou r relations w ith other people has a lw a y s been so crucial not on ly to w ellb ein g, but also to su rvival an d to reproductive success, that social interaction has been one o f the m ost p o w erfu l influences on the evolution o f the hum an brain.

A rem arkab le indication o f this is the im pressively close relation ­ ship, first pointed out by the prim ato lo gist R o b in D u n b ar, between the n orm al gro u p size o f each species o f prim ate (whether they are so litary, go abou t in pairs, or in sm aller or larger troupes) and the

p ro po rtion o f the brain m ade up o f the n eo co rtex .331 T h e larger the group size, the m ore n eo co rtex w e seem to need to cope w ith social life. O u r P alaeolith ic ancestors u su ally lived in larger com m unities

than other prim ates, and the n eo co rtex m akes up a larger part o f our brains than it does o f prim ates’ brains. Because its gro w th w a s key to the enlargem ent o f the hum an brain, the relationship suggests that the reason w h y w e becam e clever m ay have been a response to the dem ands o f social life.

H u m an beings are - the w o rld over - preoccupied w ith social

interaction, w ith w h a t people h ave said, w h at they m ight h ave been thinking, w h eth er they w ere k in d, o ff-h an d , rude . . . , w h y they behaved as they did, w h a t their m otivation s w ere, and h o w w e should respond. A ll that social processing depends on the acquisition o f a basic set o f social skills such as the ability to recognize and distinguish between faces, to use lan guage, to infer each o ther’ s

thoughts and feelings fro m body lan guage, to recognize each o ther’ s peculiarities, to understand and heed w h a t are acceptable and un ­ acceptable w a y s o f behaving in ou r society, to recognize and m anage the im pressions others fo rm o f us, and o f course a basic ability to m ake friends and to handle conflict. B ut the reasons w h y our

brains have developed as social organ s to handle social interaction is

not just to provid e am usem ent, but because o f the p aram o un t im portance o f getting ou r social relationships right. T h at is w h y w e m ind abo u t them. Th e reason w h y other people can be heaven or hell is because they have the potential to be ou r w o rst rivals and

com petitors as w ell as to be our best source o f co -o peratio n , care

and security.

O U R D U A L I N H E R I T A N C E

D ifferen t form s o f so cial organ ization pro vid e different selective environm ents. C haracteristics w h ich are successful in one setting m ay not be so in another. A s a result, hum an beings have had to develop different m ental tool-kits w h ich equip them to operate both

in dom inance hierarchies and in egalitarian societies. D om inance

and affiliative strategies are part o f ou r deep psych o lo gical m ake-up.

T h ro u gh them w e k n o w h o w to m ake and keep friends, h o w to com pete fo r status, and w hen each o f these tw o con trastin g social

strategies is approp riate. D om inance strategies are alm ost certainly pre-hum an in origin.

T h e y w o u ld not have been app ro priate to life in the predom in an tly egalitarian societies o f Stone A ge hum an hunters and gatherers. In pre-hum an dom inance hierarchies w e not only developed ch aracter­

istics w h ich help attain and express high status, but also strategies

fo r m akin g the best o f lo w status if that turns out to be ou r lot. The dan ger, p articu larly fo r m ales in some species, is that lo w social status is an ev o lu tio n ary dead end. T o av o id that, a certain am ount o f risk -tak in g and opportunism m ay be desirable.

C om petin g effectively fo r status requires m uch m ore than a desire

fo r high social status and an aversion to lo w status. It requires a high

degree o f attentiveness to status differentials and the ability to m ake accurate social com parison s o f strength and status: it is im portan t to be able to distinguish accu rately betw een w in n ab le and u nw innable status conflicts. In m any species life and lim b often depend on

k n o w in g w hen to b ack o ff and w hen to challenge a dom inant anim al fo r ran k. M a x im iz in g status depends on being seen as superior. T his is fertile psych o lo gical ground fo r the developm ent and expression o f form s o f d o w n w a rd prejudice, discrim ination and snobbishness intended to express superiority. A n d the m ore w e feel devalued by those ab o ve us and the few er status resources w e have to fall back

on, the greater w ill be the desire to regain som e sense o f self-w orth by asserting su periority over an y m ore vu ln erable grou p s. T h is is likely to be the source o f the so-called ‘ bicycling reactio n ’ m entioned in C h ap ter 1 2 - so called because it is as if people b o w to their superiors w h ile k ick in g d o w n on inferiors.

A lth ou gh it is often thought that the pursuit o f status is a p a r­

ticu larly m asculine ch aracteristic, w e should not forget h o w much this is likely to be a response to the fem ale preference fo r high-status m ales. A s H en ry K issin ger said: ‘P o w er is the ultim ate a p h ro d isiac.’

D espite the m odern im pression o f the perm anence an d u n iversal­ ity o f inequality, in the tim e-scale o f hum an h istory and p rehistory, it

is the current h ighly unequal societies w h ich are exceptio n al. For o ver 9 0 per cent o f ou r existence as hum an beings w e lived, alm ost

exclu sively, in highly egalitarian societies. F o r perhaps as m uch as

the last tw o m illion y ears, coverin g the vast m ajo rity o f the time w e have been ‘ an ato m ically m o d ern ’ (that is to say , lo o kin g m uch as w e do n o w ), hum an beings lived in re m ark ab ly egalitarian hunting and gatherin g - or fo rag in g - g ro u p s.332-5 M o d e rn inequality arose and spread w ith the developm ent o f agriculture. Th e characteristics w h ich w o u ld have been selected as successful in m ore egalitarian societies w o u ld h ave been very different from those selected in dom inance hierarchies.

R ath er than reflecting an ev olu tio n ary o u tbreak o f selflessness,

studies o f m odern and recent hunter-gatherer societies suggest that

they m aintained eq u ality not on ly through the institutions o f fo od sharing an d recip rocal gift exchan ge, but also through w h a t have been called ‘counter-dom inance strategies’ .331 Sharin g w a s w h a t has been described as ‘vigilan t sh arin g’ , w ith people w atch in g to see that they go t their fair share. T h e counter-dom inance strategies through

w hich these societies m aintained their equality functioned alm ost

as alliances o f everyone against an yon e w h ose b eh aviou r threatened peop le’s sense o f their o w n auton om y and equality. Th e suggestion is that these strategies m ay have developed as a generalized form o f the kind o f alliances w hich p rim atologists often describe being

form ed betw een tw o o r three an im als to enable them to gan g up on

and depose the dom inant m ale. O bservatio n al studies o f m odern and recent fo rag in g societies suggest that counter-dom inance strategies n o rm ally in volve anything fro m teasing and ridicule to ostracism and violence, w hich are turned again st anyone w h o tries to dom inate others. A n im portant point abo u t these societies is that they sh o w that the selfish desires o f in dividuals fo r greater w ealth an d pre­

eminence can be contained or diverted to less socially dam aging form s o f expression .

A num ber o f psych o lo gical characteristics w o u ld have been selected to help us m anage in egalitarian societies. These are lik ely to include ou r strong conception and valu atio n o f fairness, w hich m akes it easier fo r people to reach agreem ent w ith o u t conflict when sharing scarce resources. V isib le even in yo u n g children, ou r concern fo r fairness som etim es seems so strong that w e m ight w o n d er h o w it is that social systems w ith great inequality are tolerated. Sim ilarly,

the sense o f indebtedness (now recognized as universal in hum an societies) w hich w e experience after h aving received a gift, serves to prom p t recip ro city and prevent freeloadin g, so sustaining friendship. A s the experim ental econom ic gam es w hich w e discussed show ed, there is also evidence that w e can feel sufficiently infuriated by u nfairness that w e are w illin g to punish, even at som e person al cost to ourselves.

A n oth er characteristic w h ich is perhaps im portan t is our tendency to feel a com m on sense o f identity and interdependence w ith those

w ith w h o m w e share fo o d and other resources as equals. T h ey form the in -group, the ‘ u s’ , w ith w h o m w e em pathize and share a sense o f identity. In vario u s religious institutions and political organizations sharin g has been used to create a sense o f brotherhood or sisterhood, and w h eth er w e say a society has an ‘exten ded’ o r ‘n u clear’ fam ily system is a m atter o f the extent o f the sharing gro u p - w h eth er m ore

distant relations have a call on each o th er’ s resources. W ritin g in the m iddle o f the nineteenth century, de T o cq u eville believed that substantial differences in m aterial living standards betw een people w a s a fo rm id ab le barrier to em path y.23 A s w e sa w in C h apter 4, he thought the differences in m aterial conditions prevented the French

n ob ility from em pathizing w ith the sufferings o f the p easan try, and also exp lain ed w h y A m erican slave ow n ers w ere so unaffected by the su fferin g o f their slaves. H e also thought the strong com m unity life he sa w am ong whites on his visit to the U S A in 1 8 3 0 w a s a reflection o f w h a t he called ‘the equality o f co n d itio n s’ .

A very im portant source o f the close social integration in an

egalitarian com m unity is the sense o f self-realization w e can get w hen w e successfully meet others’ needs. T h is is often seen as a m ysterious q u ality, alm ost as if it w ere above exp lan atio n . It com es o f course from ou r need to feel valu ed by others. W e gain a sense

o f being valued w hen w e do things w h ich others appreciate. The best w a y o f ensuring that w e rem ained included in the co-operative hunting and gathering group and reducing the risk o f being cast out, ostracized, and preyed upon, w a s to do things w h ich people appreciated. N o w a d a y s , w hether it is co o k in g a nice m eal, telling jokes o r pro vid in g fo r peop le’s needs in other w a y s, it can give rise

to a sense o f self-w orth . It is this cap acity - n o w m ost visible in

parenting - w h ich, long before the developm ent o f m arket m ech­

anism s an d w age lab o u r, enabled hum ans - alm ost uniquely - to

gain the benefits o f a division o f lab o u r and specialization w ithin co-operative groups o f interdependent individuals.

W e have then social strategies to deal w ith very different kinds o f social organ ization . A t one extrem e, dom inance hierarchies are

abou t self-advancem ent and status com petition. Individuals have

to be self-reliant and other people are encountered m ain ly as rivals fo r fo o d and mates. A t the other extrem e is m utual interdependence and co-operation , in w h ich each person ’ s security depends on the

q u ality o f their relationships w ith others, and a sense o f self-w orth com es less from status than from the con tribution m ade to the w ellbein g o f others. R a th e r than the overt pu rsu it o f m aterial self­ interest, affiliative strategies depend on m utuality, reciprocity and the cap acity fo r em path y and em otion al bonding.

In practice, o f cou rse, go d and m am m on co exist in every society and the territory o f each varies depending on the sphere o f life, the econom ic system an d on in d ividu al differences.

E A R L Y E X P E R I E N C E

So different are the kinds o f society w h ich hum ans have had to cope w ith that the processes w hich ad ap t us to deal w ith an y given social system start very ea rly in life. G ro w in g up in a society w h ere yo u m ust be prepared to treat others with suspicion, w atch y o u r back and fight fo r w h a t yo u can get, requires very different skills from those needed in a society w here yo u depend on em pathy, reciprocity

and co-operation . P sychologists and others have alw ay s told us

that the nature o f a ch ild ’ s early life affects the developm ent o f their person ality and the kind o f people they g ro w up to be in adult life. E xam p les o f a special cap acity in ea rly life to ad ap t to local en viron ­ m ental circum stances exist th roughout anim al and even plan t life. In

h um ans, stress responses and processes shaping o u r em otional and m ental characteristics go through a kind o f tuning, or program m in g, process w h ich starts in the w o m b and continues through early childhood. T h e levels o f stress w h ich w om en experience in pregnancy

are passed on to affect the developm ent o f babies before birth. Stress h orm ones cross the placental barrier and affect the b a b y ’ s horm one

levels and gro w th in the w om b.

A lso im portant in influencing children’ s developm ent is the stress they experience them selves in infancy. Th e q u ality o f care and nurture, the qu ality o f attachm ent and h o w m uch conflict there is, all affect stress horm ones and the ch ild’ s em otional and cognitive

developm ent. A lth o u gh not yet identified in hum ans, sensitive

periods in early life m ay som etim es in volve ‘epigenetic’ processes by w h ich ea rly expo su res and experience m ay sw itch particu lar genes on o r o ff to pattern developm ent in the longer term . D ifferences in nursing beh aviou r in m other rats have been show n to affect gene expression in their offsp rin g , so p ro vid in g w a y s o f adaptin g to the

environm ent in the light o f ea rly exp erien ce.336

In the past, there w a s a strong tendency sim ply to regard children w h o had had a very stressful ea rly life as ‘ d am aged’ . B ut it looks increasingly as if w h at is happening is that ea rly experience is being used to adapt the child to deal w ith con trastin g kinds o f social reality. T h e em otional m ake-up w h ich prepares yo u to live in a soci­

ety in w h ich you have to fend fo r yo u rself, w atch y o u r back and fight fo r every bit y o u can get, is very different from w h a t is needed if yo u g ro w up in a society in w h ich (to take the opposite extreme) you depend on em pathy, reciprocity and co-operation , and in w hich y o u r security depends on m aintaining go o d relations w ith others. C hildren w h o experience m ore stress in early life m ay be m ore

aggressive, less em pathetic, and p ro b ab ly better at dealing w ith conflict. In effect, ea rly life serves to provid e a taster o f the q u ality o f social relations yo u are likely to have to cope w ith in adulthood.

So im portant are these processes that w e need to see parenting as

part o f a system fo r passing on the adu lt’s experience o f adversity

to the child. W hen people talk o f p o o r parenting, or say people lack paren tin g skills, the truth is often that the w a y parents treat their children actu ally serves to pass on their experience o f adversity to the child. A lthough this is u su ally an unconscious process, in w hich

the paren t sim ply feels short-tem pered, depressed or at their w it’ s

end, it is som etim es also conscious. In a recent court case, three w om en w ere found to have encouraged their toddlers to fight -

go ad in g them to hit each other in the face and to kick a sibling w h o had fallen to the gro u n d .337 Th e ch ildren ’s grandm other show ed no rem orse, insisting that it ‘w o u ld harden them u p ’ . G iven their experience o f life, that w a s clearly w h a t they thought w a s needed. M a n y studies have show n that fo rm s o f b eh aviou r experienced in

ch ildh ood tend to be m irrored in adu lthood. C hildren w h o have, fo r

exam p le, experienced violence or abuse are m ore likely to becom e abusing and violent w hen they reach adulthood.

The effects o f ea rly experience are long-lasting. C hildren stressed in early life, or w h ose mothers w ere stressed during pregnancy,

are m ore likely to su ffer in m iddle and old age fro m a num ber o f

stress-related diseases - including heart disease, diabetes and stroke.

The result is that som e o f the effects o f w idening incom e differences in a society m ay not be short-lived. Increased inequality means that m ore fam ilies su ffer the strains o f livin g on relatively lo w incom es, and num erous studies have show n the dam agin g effects on child

developm ent. W hen parents experience m ore adversity, fam ily life

suffers, and the children g ro w up less em pathetic but read y to deal w ith m ore antagonistic relationships.

M a n y o f the problem s w h ich w e have seen to be related to in­ equality in volve adult responses to status com petition. But w e have

also found that a num ber o f problem s affectin g children are related

to in equality. These include juvenile conflict, p o o r peer relationships and educational perform ance at school, ch ildh o o d obesity, infant m o rtality and teenage pregnancy. Problem s such as these are likely to reflect the w a y the stresses o f a m ore unequal society - o f lo w social status - have penetrated fam ily life and relationships.

Inequ ality is associated w ith less go o d outcom es o f m an y kinds

because it leads to a deterioration in the q u ality o f relationships. A n im portant p art o f the reason w h y countries such as Sw eden, Finland and N o r w a y score w ell on the U N I C E F in d ex o f child w ellb ein g is that their w elfare systems have kept rates o f relative p o verty lo w

am ong fam ilies.

M I R R O R N E U R O N S A N D E M P A T H Y

T o vie w the pu rsu it o f greater equality as a process o f shoe-horning societies into an u n co m fo rtab ly tight-fitting shoe reflects a failure to recognize our hum an social potential. If w e understood ou r social needs and susceptibilities w e w o u ld see that a less unequal society causes dram atically lo w er rates o f ill-health and social problem s because it provid es us w ith a better-fitting shoe.

M ir ro r neurons are a strikin g exam ple o f h o w ou r b iology

establishes us as deeply social beings. W hen w e w atch someone doin g som ething, m irro r neurons in ou r brains fire as if to produce the sam e actio n s.338 The system is likely to have developed to serve learning by im itation. W atching a person doing a particu lar

sequence o f actions - one research paper uses the exam ple o f a

curtsey - as an extern al ob server, does not tell yo u h o w to do it y o u rse lf n early as w ell as if y o u r brain w a s acting as i f yo u were m ak in g the sam e m ovem ents in sym pathy. T o do the sam e thing you need to experience it from inside.

U su ally, o f course, there is no visible sign o f the internal processes

o f identification that enable us to put ourselves inside each o ther’s actions. H o w ev e r, the electrical activity triggered by these special­ ized neurons is detectable in the m uscles. It has been suggested that sim ilar processes m ight be behind our ability to em pathize w ith each other and even behind the w a y people som etim es flinch w hile

w atch in g a film if they see pain inflicted on som eone else. W e react as if it w a s happening to us.

T h o ugh equipped w ith the potential to em pathize v e ry closely w ith others, h o w m uch w e develop and use this potential is again affected by early childhood.

O X Y T O C I N A N D T R U S T

A n oth er exam ple o f h o w our b io lo g y dovetails w ith the nature o f social relations in volves a h orm one called o x yto c in and its effects on

ou r w illingness to trust each other. In C hap ter 4 w e sa w that people

in m ore unequal societies w ere m uch less likely to trust each other.

T ru st is o f course an im portan t ingredient in an y society, but it

becom es essential in m odern developed societies w ith a high degree o f interdependence.

In m an y different species, o xyto cin affects so cial attachm ent and bonding, both bonding between m other and child, and pair-

bonding betw een sexu al partners. Its prod uction is stim ulated by

ph ysical contact during sexu al intercourse, in childbirth and in breastfeeding w here it controls m ilk let-dow n. H o w ev e r, in a num ­ ber o f m am m alian species, including hum ans, it also has a role in social interaction m ore gen erally, affectin g a p p ro ach and avoidan ce

behaviour. T h e effects o f o x yto c in on p eop le’ s w illingness to trust each other

w a s tested in an experim ent in vo lvin g a trust gam e.339 T h e results show ed that those given o xyto cin w ere m uch m ore likely to trust their partner. In sim ilar experim ents it w a s foun d that these effects w o rk e d both w a y s round: not o n ly does receiving o x yto cin m ake people m ore likely to trust, but being trusted also leads to increases in o xyto cin . These effects w ere foun d even w h en the only evidence

o f trust or m istrust betw een people w a s the num erical decisions com m unicated through com puter term in als.340

C O - O P E R A T I V E P L E A S U R E A N D P A I N F U L E X C L U S I O N

O ther experim ents have show n h o w the sense o f co-operation stim ulates the rew ard centres in the brain. The experience o f m utual co-operation , even in the absence o f face-to-face con tact or real

com m unication, leads reliab ly to stim ulation o f the rew ard centres. T h e researchers suggested that the neural rew ard n etw o rk s serve to encourage reciprocity and m utuality w h ile resisting the tem ptation to act selfishly.341

In con trast to the rew ard s o f co -o peratio n , experim ents using brain scans have show n that the pain o f social exclu sio n in volves the sam e areas o f the brain as are stim ulated w hen som eone experiences ph ysical pain. N a o m i Eisenberger, a psych o lo gist at U C L A , got

volunteers to p lay a com puter bat-an d-ball gam e w ith , as it seemed on the screen, tw o other p articip an ts.342 T h e p ro g ram w a s arranged

so that after a w h ile the other tw o virtu al particip an ts w o u ld start to pass the b all just betw een each other, so exclu d in g the experim ental subject. B rain scans show ed that the areas o f the brain activated by this experience o f exclu sion w ere the sam e areas as are activated by ph ysical pain. In vario u s species o f m onkeys these sam e brain areas

have been foun d to p lay a role in o ffsp rin g callin g fo r, and mothers p ro vid in g, m aternal protection.

Th ese connections have a lw a y s been understood intuitively. W hen w e talk abo u t ‘ hurt feelin gs’ or a ‘ broken h eart’ w e recognize the connection betw een p h ysical pain and the social pain caused by the b reakin g o f close social bonds, by exclu sio n and ostracism .

E v o lu tio n a ry p sych ologists have show n that the tendency to ostracize people w h o do not co-operate, and to exclude them from the shared proceeds o f co-operation , is a p o w erfu l w a y o f m aintaining high stan d ards o f co -o p eratio n .343 A n d , just as the ultim atum game sho w ed that people w ere w illin g to punish a m ean a llo cato r by reject­ ing - at some cost to them selves - allocation s that seemed u n fair, so

w e ap p ear to have a desire to exclude people w h o do not co-operate. So cial pain is o f course central to rejection and is the opposite o f

the pleasures - discussed earlier - o f being valu ed or o f the sense o f self-realization w hich can com e fro m others’ appreciation o f w h at w e have done fo r them. T h e p o w e rs o f inclusion and exclusion indicate ou r fundam ental need fo r social integration and are, no dou b t, p art o f the exp lan atio n o f w h y friendship and social in vo lve­ ment are so protective o f health (Chapter 6).

So cial class and status differences alm ost certainly cause sim ilar form s o f social p ain. U n fairn ess, inequality and the rejection o f co-operation are all form s o f exclu sion . The experim ents w hich dem onstrated the p erform an ce effects o f being classified as inferior (which w e saw in C h ap ter 8 am on g Indian children in different castes, in experim ents w ith school children, and am ong A frican - A m erican students told they w ere doing tests o f ability) indicated the

social p ain related to exclu sion . Part o f the sam e picture is the social

pain w h ich som etim es triggers violence (C h apter 10 ) w h en people feel they are put do w n , hum iliated or suffer loss o f face.

F o r a species w hich thrives on friendship and en joys co-operation and trust, w h ich has a stron g sense o f fairness, w h ich is equipped w ith m irro r neurons allo w in g us to learn ou r w a y o f life through a process o f identification, it is clear that social structures w h ich create relationships based on in equality, in ferio rity and social exclu sion

m ust inflict a great deal o f social pain . In this light w e can perhaps begin not o n ly to see w h y m ore unequal societies are so so cially d ys­ functional but, through that, perhaps also to feel m ore confident that a m ore hum ane society m a y be a great deal m ore practical than the

highly unequal ones in w h ich so m an y o f us live n o w .

1 5

Equality and sustainability

T h e o n e w h o d ie s w i t h m o s t t o y s w in s .

US bumper sticker

O ver the n ext generation o r so, politics seem likely to be dom inated either by effo rts to prevent ru n aw a y glo b al w arm in g o r, if they fail, by attem pts to deal w ith its consequences. C a rb o n em issions per head in rich countries are betw een tw o and five times higher than the

w o rld average. But cutting their em issions by a h a lf or four-fifths w ill not be enough: w o rld to tals are alread y to o high and allow an ces m ust be m ade fo r econom ic gro w th in p o o rer countries.

H o w m ight greater equality and policies to reduce carbon em issions go together? G iven w h a t in equality does to a society, and

p a rticu la rly h o w it heightens com petitive consum ption, it lo o k s not on ly as if the tw o are com plem entary, but also that governm ents m ay be unable to m ake big enough cuts in carb o n em issions w ithou t also reducing inequality.

S U S T A I N A B I L I T Y A N D T H E Q U A L I T Y O F L I F E

E v er since the B ran d t R e p o rt in 1 9 8 0 people have suggested that social and environm ental su stain ability go together. It is fortunate

that ju st w hen the hum an species discovers that the environm ent can n ot abso rb further increases in em issions, w e also learn that fur­ ther econom ic g ro w th in the developed w o rld no longer im proves

w w w .C a r t o o n S t o c k .c o m

Li fe

ex

p ec

ta n

cy

(y ea

rs )

health, happiness or m easures o f w ellbeing. O n top o f that, w e have n o w seen that there are w a y s o f im p rovin g the q u ality o f life in rich countries w ith o u t further econom ic gro w th .

But if w e do not need to consum e m ore, w h a t w o u ld be the co n ­

sequences o f consum ing less? W o u ld m akin g the necessary cuts in carb o n em issions m ean reducing present m aterial living standards b elo w w h a t people in the rich w o rld could accept as an adequate q u ality o f life? Is su stain ability com patible w ith retaining ou r qu ality o f life?

A starting point fo r an sw ering this question is Figure 1 5 . 1 . It

show s life expectan cy in relation to C 0 2 em issions per head o f p o p ­ ulation am ong rich an d p o o r countries. Because carb on em issions tend to go up as societies get richer, it lo o ks very like the relationship between life exp ectan cy and N a tio n a l Incom e per person show n in

Figure 1 . 1 . H o w ev e r, w h a t w e can n o w see is that som e countries

achieve life expectancies close to 80 years at a fractio n o f the C O 2 em issions com m on in the richest countries. It should therefore be

8 0 -

7 0 -

60 -

5° -

4 0 -

Figure 1 5 . 1 L o w infant mortality can be achieved without high carbon emissions,344

( Costa Rica \* Cuba* \ Chit

v • • • fGeorgia..^^ •

•Sri Lanka China'"*'. Colombia* • •

. a s .fid V .■> ... . *Peru N“ * * . M r .,I .

•Jamaica * [V u8al •Slovenia

^ Croatia^, ̂ • Poland

Belgium f Israel , Norway iermany** • Finland Denmark Îreland

• • Pakistan • Bangladesh •Nepal

•islfir ••sen* ► V •Mauriti

O u

•Cote d'lv

Burkina Faso

V\ , • Zambia

■ Ruanda

>OJ 2

I 5 10 15 C O 2 em issions (metric tons per capita)

possible to m ake dram atic reductions in em issions in m ost rich co u n ­ tries w ith o u t an y loss o f health an d w ellbein g - even on the basis

o f current inefficient techn ology based m ain ly on non-renew able sources o f energy.

The circle in the top left hand corner o f Figure 1 5 . 1 show s (again on the basis o f current technology) the area in w h ich societies seem to be able to gain go o d health at the minim um environm ental cost. A s

the vertical line through the centre o f the circle is a rough estim ate o f w o rld average C 0 2 em issions, the graph suggests that all countries o f the w o rld have the poten tial to achieve high life expectancies w ithou t exceeding current w o rld C O 2 em issions.

But because current g lo b al em issions are alread y causing such rap id g lo b al w arm in g, w e need to reduce w o rld em issions far b elow current levels. T h at can only be achieved b y m ore energy efficient w a y s o f living and the developm ent o f sustainable sources o f energy. Such changes w o u ld shift the circle (m arking the low est en viro n ­ mental costs at w h ich high levels o f health and w ellbein g can be

achieved) leftw ards and p ro b ab ly upw ard s. A n oth er an sw er to the question w hether su stain ability is co m ­

patible w ith retaining our high q u ality o f life com es fro m the W orld W ild life Fund (W W F ). It an alysed data relating the q u ality o f life in each co u n try to the size o f the ecological footp rin t per head o f p o p u latio n .345 T o m easure the q u ality o f life they used the U N

H u m an D evelopm ent In d ex (H D I) w h ich com bines life expectan cy, education and G ro ss D om estic Product per capita. Figure 1 5 . 2 uses W W F data to sh o w the relation betw een each co u n try ’s ecological fo otp rin t per head and its score on the U N H u m an D evelopm ent Index. Scarcely a single co u n try com bines a q u ality o f life (above the

W W F threshold o f 0.8 on the H D I) w ith an ecological footp rin t

which is g lo b ally sustainable. C u b a is the o n ly one w h ich does so. Despite its much lo w er incom e levels, its life expectan cy and infant m o rtality rates are alm ost identical to those in the United States.

Th e fact that at least one cou n try m anages to com bine acceptable living standards w ith a sustainable econ om y, pro ves that it can be

done. H o w ev e r, because the co m bin ation is achieved w ith o u t access to the greenest and m ost fuel-efficient techn ology m eans it cou ld be done m ore easily in countries w ith access to m ore advan ced tech-

E co

lo gi

ca l

to o

tp ri

n t

pe r

ca p

it a

IO

8

6

4

2

O

n o lo g y than C u b a has. W ith the advan tages o f p o w e r generation from ren ew ables, en viron m en tally friendly n ew technologies and greater eq u ality, w e can be confident that it is possible to com bine su stain ability w ith a high qu ality o f life. B efore leaving Figure 1 5 .2 it is w o rth noting that much o f the reason w h y the highest scores on the H D I are achieved by countries w ith the largest ecological feet is m erely a reflection o f the fact that G ro ss D om estic P roduct per head is one o f the com ponents o f the H D I .

R E D U C I N G C A R B O N E M I S S I O N S F A I R L Y

Im p ro vin g the real q u ality o f ou r lives at lo w er levels o f co n ­ sum ption is only one o f the contributions equality can m ake to reducing carb o n em issions. T h ere are tw o others. F irst, if policies to cut em issions are to gain public acceptance, they m ust be seen to

1_0 5

-o

3 ■a

' 15 j5

"o

1 |5

Saudi Arabia •

•

• ^ Lebanon • <

w 1 i j u 1 u # South Africa • • f ft World biocapacity available per person a ^ • [

• United Arab Emirates

• USA

Finland • • Canada • Kuwait Australia

• Estonia ®0 Sweden N ew Zealand • ! § # UK

Spain • _ . • • f M Ireland Russia

' # • J * Portugal •

' • Slovenia

• / • • I • • Costa Rica ■ i

. . . _ • Gambia Bmswana • • • M 1 ^ • •

S, * . " *7 .. .. Mali iajikistan

• Cuba

0 .2 0 .4 0.6 0.8 1 U N H um an D evelopm ent Index, 200 3

Figure 1 5 .2 Human wellbeing and sustainability.3*5

be applied fairly . T h e richer yo u are and the m ore yo u spend, the m ore you are likely to contribute to g lo b al w arm in g. T h e carbon em issions caused by the consum ption o f a rich person m ay be ten times as high as the consum ption o f a po o rer person in the sam e

society. If the rich are the w o rst offen ders, then fair remedies must

surely affect them m ost. Policies that squeezed the p o o r w h ile a llo w ­ ing the rich to continue to produce much higher levels o f em issions w o u ld be unlikely to gain w idespread public support.

A system o f individual carb o n ration s has been p rop osed as one w a y o f reducing carb o n em issions fairly . T h e total perm issible

level o f em issions can be divided by the pop u latio n to give an equal

share, or q u o ta, o f a llo w a b le em issions per head. T here is an obvious parallel here w ith the egalitarian policies im plem ented in B ritain during the Second W o rld W ar: to gain public co-o p eratio n in the w a r effo rt, the burden had to be seen to be fa irly shared. Titm uss regarded this as the ration ale fo r the in troduction o f ration in g and

m ore progressive incom e ta xes, as w ell as fo r subsidizing necessities

and ta x in g lu xu ries.346 O ne suggestion n o w is that people should use an electronic card to co ver paym ents fo r fuel, p o w e r and air travel. T h ose using less than their ration w o u ld be able to sell their unused

allo catio n back to a carb o n b an k , from w here it cou ld then be

bought by richer people w an tin g to use m ore than their allo catio n o f

fuel and p o w er. Under such a system o f ‘tradeab le carb o n q u o tas’ high consum ers w o u ld be com pensating lo w consum ers, and incom e w o u ld be redistributed from rich to p o o r. In 2 0 0 6 the then M in ister fo r the Environ m en t in B ritain , D av id M ilib an d , pro p o sed such a system and a sm all trial w a s begun in M an chester in 2 0 0 7 . T o safe­

g u ard the p o o r it m ay be necessary to prevent people selling unused parts o f their ration till the end o f the period it co vers, so only allow an ces already saved could be traded.

N E W T E C H N O L O G Y I S N O T E N O U G H O N I T S O W N

W e m ight hope that new techn ology w ill save us from the rigours o f carb o n rationing. H o w ev e r, although green in n ovation s w hich

reduce fuel consum ption and carb o n em issions are an essential part

o f the change w e need to m ake, they can n ot solve the problem on

their o w n . Im agine that a n ew generation o f car engines is in tro­ duced w h ich halve fuel consum ption. D rivin g w o u ld then be cheaper and that w o u ld save us m oney, but it is m oney w h ich w e w o u ld alm ost certain ly spend on som ething else. W e m ight spend it on

driving m ore, or on buying a bigger car, or on m ore po w er-h un gry

electrical equipm ent - perhaps a bigger fridge-freezer. But h ow ever w e spend the m on ey put b ack in ou r pockets by m ore efficient car engines, ou r add ition al consum ption w ill p ro b a b ly add to carbon em issions elsew here and lose m uch o f the o rigin al environm ental

benefit. The sam e logic applies in alm ost all areas. M o re po w er-

efficient w ash in g m achines or better insulated houses w ill help the environm ent; but they also cut ou r bills, and that im m ediately means w e lose som e o f the environm ental gain by spending the saved m oney on som ething else. A s cars have becom e m ore fuel-efficient

w e have chosen to drive further. A s houses have becom e better insulated w e have raised stan d ards o f heating, and as w e put in en ergy-savin g light bulbs the chances are that w e start to think it

doesn ’t m atter so m uch leaving them on. Because en ergy-savin g in n ovation s m ean that w e can buy m ore,

they are like econom ic g ro w th . T h o ugh they give us higher m aterial living standards fo r an y level o f carb o n em issions, m uch o f the carb o n savings get sw a llo w ed up by higher livin g stan dards. The only question is h o w m uch o f the benefits o f greener techn ology get eaten up in higher consum ption. A s m an y countries have adopted sm aller, m ore fuel-efficient cars, n ation al em issions have usually continued to rise despite the increased efficiency.

A S T E A D Y - S T A T E E C O N O M Y

It is clear that w e have to m ove to som ething m ore like the steady- state econ om y first p roposed by econom ist H erm an D a ly .347 But h ow do w e do that w h en , as M u rra y B oo k ch in , the A m erican social ecologist and lib ertarian ph ilosoph er, said, ‘ C ap italism can no more be “ p ersu ad ed ” to lim it gro w th than a hum an being can be “ per­ su ad ed ” to stop breathin g’ ?348 W hen D aly developed the con cept o f a steady-state econom y people w ere m ore concerned abou t using up

the ea rth ’s finite m ineral and agricultu ral resources than they w ere

w ith g lo b al w arm in g. H e suggested that w e should have ph ysical quotas on the extractio n o f m inerals and that the use o f the w o r ld ’s resources should be prevented fro m gro w in g. Lim itin g w o rld oil and co al prod uction m ight turn out to be a ve ry effective w a y o f lim iting g lo b al w arm in g. In n ovation and change w o u ld then be con cen ­

trated on using finite resources m ore effectively fo r the benefit o f hum ankind.

T h in k o f m aterial living standards as given by the stock o f goods in use, rather than the rate o f flo w from consum ption to w aste. The

faster things w e a r o u t and need replacing, the m ore they contribute

to the flo w and to w aste. I f m aterial living stan d ards depend on the

good s w e have in use, then each thing that w ears out is a subtraction from that. R ath er than serving as consum ers, helping business to keep sales up, w e need incentives to build and m aintain longer- lasting good s o f every kind.

C le a rly an y system fo r tacklin g these problem s has to treat rich and

p o o r countries differently. In dia, producin g 1 . 6 tonnes o f carb o n per person an n u ally, can n ot be treated the sam e as the U S A , prod ucin g 2 4 .0 per person. A n y regu lato ry system has to include policies fo r ‘ con traction and con vergen ce’ or ‘cap and sh are’ . Both approaches

prop ose a year-o n -year con traction in perm itted em issions levels,

leading to an eventual convergence on equal per capita em issions across the planet.

It w o u ld be a m istake to think that a steady-state econ om y w o u ld mean stagn ation an d lack o f change. P a ra d o x ic a lly , the transition to a sustainable steady state econ om y w o u ld create huge dem ands

fo r in n ovation and change. T ry in g to get m ore fro m the limited resources availab le has a lw a y s been one o f the fundam ental drivers

o f in n ovation and technical ch an ge.349 F ixin g limits on resource con­

sum ption and em issions w o u ld require in n ovation as never before. A s w e shall see in the next chapter, continued rapid techn olog­ ical advan ces, such as digitization, electronic com m unications and virtu al system s, creating ‘w eigh tless’ sectors o f the econ om y, m ake it very m uch easier to com bine high living standards w ith lo w resource

consum ption and em issions. It is often suggested that invention and in n ovation go w ith

in equality and depend on the prom ise o f in d ividu al financial in­ centives. H o w ev e r, Figure 1 5 . 3 suggests the co n trary - that m ore equal societies tend to be m ore creative. It show s that there is a ten­

dency fo r m ore patents to be granted per head o f po p u latio n in m ore

equal societies than in less equal ones. W hether this is because talent goes undeveloped or w asted in m ore unequal societies, or w hether h ierarch y breeds co n fo rm ity, is an yo n e’ s guess. But it does suggest that greater equality w ill not m ake societies less adaptable.

4 0 -

• Finland

• N o rw ay

• Sweden

• Austria

Switzerland*

• Ireland • N e w Zealand

6 <D ( X

g 1 0 - ^ 3 Japan •

o -

H ig h

Th e second link betw een greater eq u ality and the prevention o f glo b al w arm in g in volves the consum erism w hich m akes it so much

harder to con tain econom ic activity w ithin sustainable levels. O ur

addiction to shopping and spending m akes m an y people think that w e have already lost the battle again st g lo b al w a rm in g. A s w ell as leading m ost o f us into an ostrich-like denial o f its im plications fo r ou r w a y o f life, the strength o f our consum erist tendencies

has reduced governm ents to a state o f p aralysis, to o nervous o f the

electorate to im plem ent an y p o licy capable o f m akin g a real d iffer­

ence. H o w are w e to tran sfo rm this culture and m ake it possible to reduce the threat to the planet?

G reater equality gives us the crucial key to reducing the cultural

pressure to consum e. In a period w h en people seem to h ave been

less gu arded, H en ry W allich, a fo rm er go vern o r o f the Federal

Reserve and pro fesso r o f econom ics at Y a le , said: ‘ G ro w th is a su b ­ stitute fo r equality o f incom e. So long as there is gro w th there is hope, and that m akes large incom e differentials to lera b le .’350 But this relation holds both w a y s round. It is not sim ply that g ro w th is

a substitute fo r eq u ality, it is that greater eq u ality m akes gro w th

m uch less necessary. It is a precondition fo r a steady-state econom y. A great deal o f w h a t drives consum ption is status com petition. Fo r

m ost o f us it p ro b ab ly feels less like being com petitive and m ore like a kind o f defensiveness: if w e d o n ’ t raise our standards, w e get left behind and everything starts to lo o k d o w d y , shab by and out o f date.

R o b ert F ran k , an econom ist at C o rn ell U n iversity, has described

h o w stan d ards are inherently relative and in volve com p arison s w ith others. In his b o o k , F a llin g B e h in d : F lo w risin g in e q u a lity harm s the m id d le class (20 0 7 ), he puts it like th is:351

N o one denies that a car experienced in 19 5 0 as having brisk acceleration would seem sluggish to most drivers today. Similarly, a house of given size

is more likely to be viewed as spacious the larger it is relative to other houses in the same local environment. And an effective interview suit is one that compares favorably with those worn by other applicants for the

same job. In short, evaluation depends always and everywhere on context,

(pp. viii-ix)

T h e problem is that second-class goods m ake us lo o k like second- class people. B y com parison w ith the rich and fam o u s, the rest o f us app ear second-rate and in ferio r, an d the bigger the differences, the m ore noticeable and im portant they becom e. A s inequality increases status com petition, w e have to struggle harder to keep up. W hile the rich m ay believe their w illingness to spend huge sums on a w a tch , a car or som e other lu xu ry item reflects their appreciation o f

the ‘attention to detail’ or ‘craftsm an sh ip’ , w h a t really m akes the difference is w h a t their purchases say abo u t them relative to the rest o f us. A s every advertiser k n o w s, it serves to set them a p art as people o f distinction - social distinction. O n ly the best people can have

nothing but the best. T h e other side o f this coin is that the consum ption o f the rich

reduces everyone else’ s satisfaction w ith w h a t they have, b y show in g it up as in ferior - as less than the best. In his b o o k , H a p p in e s s, R ic h a rd L a y a rd , foun der o f the Centre fo r E con om ic Perform ance at the Lo n d o n School o f Econ om ics, treated this dissatisfaction

as a cost w hich the rich im pose on the rest o f society.3 R a th e r as if it

w ere sm oke fro m a fac to ry chim ney, he estim ated the cost that the rich should p ay fo r it. H e w a s, h ow ever, u n aw are o f the effects o f inequality on the health and social problem s w hich w e have outlined. H e based his calculation s solely on the loss o f satisfaction ,

or happiness, am on g the rest o f the pop u latio n and concluded that a

60 per cent t a x rate on the better-off m ight cover that cost (presum ­ a b ly that should be o ver and abo ve the t a x rates other people pay).

T h e idea that inequality ratchets up the com petitive pressure to consum e is not just speculation. It has ob servable effects. W hile

in equality has been rising in the U S A and B ritain , there has been a long-term decline in savings and a rise in debt. R o b ert F ran k notes th at in 1 9 9 8 , even though the A m erican econ om y w a s boom in g as never before, one fam ily in sixty-eigh t filed fo r b an k ru p tcy - four times the rate in the early 19 8 0 s before the m ost dram atic rises in in eq u ality.351 B y Z 0 0 2 , unpaid credit card debt w a s $ 9 ,0 0 0 fo r the average card-holder. L o o k in g at changes over a ten-year period,

F ran k found that ban kru ptcy rates rose m ost in parts o f the U S A w here inequality had risen m o st.154- 351 T h e gro w th o f inequality m ade it hard er fo r people to m ain tain standards relative to others. Th e increased pressure to consum e led people to save less and b o rro w m ore to such an extent that the exp an sio n o f consum er dem and becam e one o f the m ain drivers o f the long econom ic boom

and financial speculation w hich ended in crisis. T h is fits w ell w ith the fact that spending on advertising also varies w ith inequality - in m ore unequal countries a higher p ro p o rtion o f G ro ss D om estic P roduct is spent on advertisin g, w ith the U S A and N e w Z e a la n d spending tw ice as m uch as N o r w a y and D enm ark.

A n oth er in dicator o f h o w in equality increases the pressure to consum e com es from the w a y w o rk in g h ours v a ry in different countries in relation to inequality. A study o f w o rk in g hours in O E C D countries by Sam B ow les, p ro fesso r em eritus o f econom ics at the U n iversity o f M assach usetts, sh o w ed not o n ly that m ore unequal countries tend to have longer w o rk in g hours, but also that d iffer­ ences in w o rk in g hours changed in line w ith changes in inequality o ver several decades.352 Th e relationship between greater inequality and longer w o rk in g hours is show n in Figure 1 5 . 3 . People in m ore unequal countries do the equivalent o f tw o o r three m on th s’ extra w o rk a year. A loss o f the equivalent o f an extra eight o r tw elve w e ek s’ h o lid ay is a high price to p a y fo r inequality.

A n oth er study, this time using data w ithin the U S A , found that m arried w om en w ere m ore likely to go out to w o rk if their sister’s h usband earned m ore than their o w n h usb an d .353 A sim ilar study suggested that the decisions m arried w o m en m ake abo u t taking paid w o rk are also affected by less personal inequalities: it looked

at w om en w h o w ere m arried to em ployed men and foun d that they

w ere m ore likely to take a jo b them selves if they lived in an area in w h ich m en’s incom es w ere m ore u n eq u al.354

Th e evidence w e have described fro m a num ber o f different sources on savin gs, debt, b an k ru p tcy rates, spending on advertising

and w o rk in g hours, all concurs w ith the v ie w that inequality does

indeed increase the pressure to consum e. I f an im portan t p art o f consum erism is driven b y em ulation, status com petition, or sim ply h avin g to run to keep up w ith everyone else, and is b asically abou t

2100

2000

(A 1900M3O 1800

O £ § 1700 G S

DO i ^ O On3Ul >

15 0 0

14 0 0

1 3 0 0

1 . 3 0 1.5 0 1 . 7 0 1 . 9 0 2 .1 0 Z . 3 0 2.50

Earnings inequality: ratio of rich to middle ($>0th:50th percentile)

F ig u r e 1 5 . 4 People w ork longer in more unequal societies.351

social appearances and positio n , this w o u ld exp lain w h y we continue to pursue econom ic gro w th despite its apparen t lack o f

benefits. If everyone w an ts m ore m oney because it im proves self­ im age and status in relation to others, then each p erso n ’ s desire to be richer does not add up to a societal desire fo r econom ic grow th. H o w m uch peop le’ s desire fo r m ore incom e is really a desire for higher status has been dem onstrated in a sim ple experim ent. People

w ere asked to say w h eth er th ey’ d prefer to be less w e ll-o ff than others in a rich society, or have a m uch lo w er incom e in a poorer society but be better-off than others. F ifty per cent o f the p a r­ ticipants thought they w o u ld trade as m uch as h a lf their real incom e if they cou ld live in a society in w h ich they w o u ld be better o ff than

others.355 This show s h o w m uch w e valu e status and exp lain s w h y (as w e saw in C h ap ter 2) the incom e differences w ithin rich societies m atter so much m ore than the incom e differences betw een them. O nce w e have enough o f the necessities o f life, it is the relativities w h ich m atter.

W hen B ow les and P ark first dem onstrated the relationship

Canada . •

U K O A

* * S

N orw ay O

O Germany

• •

O Netherlands

• / 0 Country average • Annual data

between inequality and w o rk in g hours (Figure 1 5 . 3 ) , they quoted

Th o rstein V eblen , w h o said: ‘T h e only practicable m eans o f im ­

pressing o n e’ s pecu n iary ability on the unsym pathetic observers o f on e’ s ev eryd ay life is an unrem itting dem onstration o f the ability to p a y .’ V eb len ’ s T h e o r y o f the L e is u re C lass, published in 18 9 9 , w as the first m a jo r w o r k on the relation ship betw een consum ption and social stratification. It w a s he w h o introduced the term ‘conspicuous

co n su m p tio n ’ and em phasized the im portance o f ‘pecu n iary em ula­ tio n ’ and ‘ invidious co m p ariso n s’ .356 B ecause the advertising industry p lays on insecurities abo u t h o w w e are seen, it has m ade us m ore a w a re o f the psych o lo g y o f consum ption. But V eblen w rote long before w e w ere so b om barded w ith advertisin g. So rather than blam ing these problem s entirely on advertisin g, w e should recognize

that it sim ply am plifies and m akes use o f vulnerabilities w h ich are there a n yw ay . Econom ists n o w use the term ‘V eblen effect’ to refer to the w a y goods are chosen fo r their social valu e rather than their usefulness. A n d research confirm s that the tendency to lo o k for goods w h ich con fer status and prestige is indeed stronger fo r things w h ich are m ore visible to others.

T o o often consum erism is regarded as if it reflected a fundam ental hum an m aterial self-interest and possessiveness. T h at, h ow ever, could h ard ly be further fro m the truth. O ur alm ost neurotic need to shop and consum e is instead a reflection o f h o w deeply social w e are.

L ivin g in unequal and individualistic societies, w e use possessions to

sh o w ourselves in a g o o d light, to m ake a positive im p ression, and to av o id appearin g incom petent or inadequate in the eyes o f others. C on sum erism show s h o w p o w e rfu lly w e are affected by each other. O nce w e h ave enough o f the basic necessities fo r com fort, possessions m atter less and less in them selves, an d are used m ore and

m ore fo r w h a t they say abo u t their ow n ers. Id eally, our im pressions

o f each other w o u ld depend on face-to-face interactions in the course o f com m unity life, rather than on o u tw a rd appearan ces in the absence o f real k n ow led ge o f each other. T h a t p oin t takes us back to the discussion in C hapter 4 o f the evidence that inequality

w eak en s com m unity life. Th e w eak en in g o f com m unity life and the gro w th o f consum erism are related.

If, to cut carb o n em issions, w e need to lim it econom ic gro w th

severely in the rich countries, then it is im portant to k n o w that

this does not m ean sacrificing im provem ents in the real q u ality o f life

- in the q u ality o f life as m easured by health, happiness, friendship and com m unity life, w hich really m atters. H o w ev e r, rather than sim ply h aving few er o f all the lu xu ries w hich substitute fo r and pre­ vent us recognizing ou r m ore fundam ental needs, inequality has to

be reduced sim ultaneously. W e need to create m ore equal societies able to m eet ou r real social needs. Instead o f policies to deal w ith

g lo b al w arm in g being experienced sim ply as im p osing lim its on the possibilities o f m aterial satisfactio n , they need to be coupled w ith egalitarian policies w h ich steer us to n ew and m ore fundam ental w a y s o f im p rovin g the q u ality o f ou r lives. T h e change is abou t a

historic shift in the sources o f hum an satisfaction from econom ic

gro w th to a m ore sociable society. In his speech accepting the 2 0 0 7 N o b el Peace Prize on b eh alf o f

the Intergovernm ental Panel on C lim ate C h an ge w h ich he chairs, R a jen d ra P achauri described h o w g lo b al w arm in g w o u ld reduce agricultu ral yields, fo o d and w a ter supplies fo r hundreds o f m illions

o f people and so lead to increasing conflict. (He spoke before the con tribution o f biofu el crops to rising w o rld fo o d prices had been clearly recognized.) T h e task o f responding adeq uately to the threat o f g lo b al w arm in g needs to be seen as bigger and m ore im portant than a n y o f us. But if everyone - in d ividu als, co rp o ratio n s, w hole nations - feels it is alm ost their duty to get round regu lation s, to

e x p lo it w h atever loopholes they can (as has long been taken as the norm w ith taxatio n ) then the task is lost. A s w e w rite, tankers o f biofuels are crossing the A tlan tic fro m E u ro p e to the U S A and back in o rder to pick up the U S governm ent subsidy paid w hen small quantities o f petroleum are added, w h ich cou ld just as w ell have

been added in E urop e w ith o u t every litre crossing the A tlan tic twice. R eversin g the intended effect o f regulations fo r private gain is an expression o f the dom inance o f attitudes w h ich m ake it m uch harder to resp ond adeq uately to the threat o f g lo b al w arm in g.

T a ck lin g clim ate change depends on w o rld co-o peratio n like never

before: w e cannot succeed if in practice everyone is tryin g to cir­ cum vent the regulations. C heatin g on regulation s and the pursuit o f short-term sectional or self-interest becom es not just anti-social

but anti-hum anity. Policies to reduce carb o n em issions depend on a w id er sense o f social respon sibility, o f co-o p eratio n and p u b lic­ spiritedness. H ere again the evidence suggests that m ore equal societies do better. W e have seen (C h apter 4) that they are m ore

so cially cohesive and have higher levels o f trust w h ich foster p u b lic­ spiritedness. W e have also seen h o w this carries over into inter­ n ation al relations: m ore equal societies give m ore in developm ent aid and score better on the G lo b a l Peace In dex. A n indication that a greater sense o f public resp onsibility in m ore eq u al countries m ight affect h o w societies respond to environm ental issues can be seen in

Figure 1 5 . 5 , w h ich show s that they tend to recycle a higher p r o ­

portion o f their w aste. Th e data com es fro m A u stra lia ’ s Planet A rk F oun d ation T ru st.357 W e sh o w each co u n try ’s ran kin g fo r the p ro po rtion o f w aste that they recycle. A n oth er indication o f a stronger sense o f pu b lic resp on sibility com es from an international

su rvey o f opinions o f business leaders. A s ou r colleagues, R o b erto

De V o g li and D av id G im en o, pointed out, business leaders in m ore

Incom e inequality

Figure 15 .5 M ore equal countries recycle a higher proportion o f their waste.

equal countries are m ore stron gly in fav o u r o f their governm ents

com plyin g w ith international environm ental agreem ents than busi­

ness leaders in m ore unequal coun tries.404,405 So rather than assum ing that w e are stuck w ith levels o f self­

interested consum erism , individualism and m aterialism w h ich must defeat an y attem pts to develop sustainable econom ic system s, we need to recognize th at these are not fixed expression s o f hum an nature. Instead they reflect the characteristics o f the societies in w hich

w e find ourselves and v a ry even fro m one rich m arket dem ocracy

to another. A t the m ost fundam ental level, w h a t reducing inequality is abou t is shifting the balance from the divisive, self-interested consum erism driven by status com petition, to w ard s a m ore so cially integrated and affiliative society. G reater eq u ality can help us develop the public ethos and com m itm ent to w o rk in g together w h ich w e need

if w e are goin g to solve the problem s w h ich threaten us all. As

w artim e leaders k n ew , if a society has to pull together, policies must be seen to be fair and incom e differences have to be reduced.

Building the future

Turning corporations loose and letting the profit motive run amok is not a prescription for a more liveable world.

Tom Scholz, Interview with the Sierra Club

B efore discussing w h a t should be done to m ake ou r societies more equ al, it is w o rth pointing out that focusin g attention on the in­

equalities w ithin them does not mean ign orin g the international

inequalities betw een rich and p o o r countries. T h e evidence strongly suggests that n a rro w in g incom e differences w ithin rich countries w ill m ake them m ore responsive to the needs o f po o rer countries. In C h ap ter 4 w e show ed (Figure 4.6) that m ore equal countries tend to

p ay a higher p ro po rtion o f their nation al incom e in foreign aid.

C o m p ared to the m ost unequal countries, som e o f the m ost equal devote fo u r times the p ro p o rtion o f their n ation al incom e to aid. M o re unequal countries also seem to be m ore belligerent inter­ n atio n ally. In eq u ality is related to w o rse scores on the G lo b a l Peace In d ex, w h ich com bines m easures o f m ilitarization w ith m easures o f

dom estic and in tern ation al conflict, and m easures o f security, hum an

rights and stability. (It is prod uced by V isio n s o f H u m an ity in conjunction w ith the E con om ist Intelligence U nit.)358

I f w e turn instead to the p a rt countries play in in tern ation al trade agreem ents or, fo r instance, in negotiations on reducing carbon em issions, w e find that m ore equal countries take position s on these

issues w h ich are likely to be m ore beneficial to developing countries. It lo o ks as if the inequalities w h ich affect the w a y people treat

each other w ithin their o w n societies also affect the norm s and

expectation s they bring to bear on international issues. G ro w in g up and livin g in a m ore unequal society affects p e o p le’ s assum ptions abou t hum an nature. W e have seen h o w in eq uality affects trust, com m unity life and violence, and h o w - through the q u ality o f

early life - it predisposes people to be m ore or less affiliative, em pa- thetic or aggressive. O b v io u sly these issues are clo sely related to the increased status com petition and consum erism w e discussed in the p revious chapter. It im plies that if w e put our o w n houses in order, w e m ay lo o k m ore sym pathetically on developing countries.

A T R A N S F O R M A T I O N

But h o w can w e m ake ou r societies m ore equal? T a lk abo u t greater eq u ality w o rries som e people. T ry in g to a lla y these fears at a

N a tio n a l P olicy A sso ciatio n conference on health inequalities in W ashin gton , one o f us pointed out that as all the data cam e from rich developed m arket dem ocracies and w e w ere on ly talk in g abou t the differences betw een them , it surely w o u ld n ’ t take a revolu tion

to put things right. But w hen I t D o e s n ’t T a k e a R e v o lu tio n appeared

as the title o f the N a tio n a l P olicy A sso c iatio n ’s b ooklet from the

conference, it w a s surprising to find a few people w h o thought it w o u ld .

A s B ill K e rry , one o f the foun ders o f the E q u ality T ru st, put it, if w e are go in g to achieve a m ajo r n a rro w in g o f incom e differences

w h ile resp onding effectively to g lo b al w arm in g, w h a t is required

am ounts to a tran sfo rm atio n o f ou r societies, a tran sform ation w h ich w ill not be furthered by a departure fro m peaceful methods but one w h ich is u n likely to be achieved by tinkering w ith m inor p o licy options. A social m ovem ent fo r greater equality needs a sus­

tained sense o f direction and a vie w o f h o w w e can achieve the

necessary econom ic and social changes. T h e k ey is to m ap out w a y s

in w hich the new society can begin to g ro w w ithin and alongside the institutions it m ay g ra d u a lly m arginalize and replace. T h a t is w h at m ak in g change is really abo u t. R a th e r than sim ply w aitin g fo r governm ent to do it fo r us, w e have to start m akin g it in ou r lives and in the institutions o f ou r society straight a w a y . W h at w e need is

not one big revo lu tio n but a continuous stream o f sm all changes in a consistent direction. A n d to give ourselves the best chance o f m ak in g the necessary tran sfo rm atio n o f society w e need to rem em ­ ber that the aim is to m ake a m ore sociable society, w h ich means a vo id in g the disruption and d islocation w h ich increase insecurity and fear and so often end in a disastrous b acklash. T h e aim is to increase peop le’ s sense o f security and to reduce fear; to m ake every­ one feel that a m ore equal society not on ly has ro o m fo r them but also that it offers a m ore fulfilling life than is possible in a society dom inated by h ierarch y and inequality.

In the past, w h en argum ents abou t inequality centred on the p riv atio n s o f the p o o r an d on w h a t is fair, reducing inequality depended on co a x in g or scaring the better-off into adoptin g a more altruistic attitude to the p o o r. But n o w w e k n o w that inequality affects so m any outcom es, across so m uch o f society, all that has

changed. The tran sfo rm atio n o f ou r society is a project in w h ich w e all have a shared interest. G reater eq u ality is the g a tew ay to a society cap able o f im p rovin g the q u ality o f life fo r all o f us and an essential step in the developm ent o f a sustainable econom ic system.

It is often said th at greater eq u ality is im possible because people are not equal. B ut that is a co n fu sion: equality does not m ean being the sam e. People did not becom e the sam e w h en the principle o f eq u ality before the la w w a s established. N o r - as is often claim ed - does reducing m aterial inequality m ean lo w erin g standards or levelling to a com m on m ediocrity. W ealth, p articu larly inherited

w ealth , is a p o o r in d icator o f genuine m erit - hence G eorge Bernard S h a w ’s assertion that: ‘ O n ly w h ere there is p ecu n iary eq u ality can the distinction o f m erit stand o u t.’359-p-71 Perhaps that m akes Sweden a p articu larly suitable hom e fo r the system o f N o b e l prizes.

W e see no indication that standards o f intellectual, artistic or

sportin g achievem ent are lo w er in the m ore equal societies in our

analyses. Indeed, m ak in g a large part o f the p o pu latio n feel devalued can surely only lo w er standards. A lth ou gh a baseball team is not a m icrocosm o f society, a w ell-con trolled study o f o ver 1,6 0 0 players in tw enty-nine team s o ver a nine-year p eriod foun d that m ajo r league baseball team s w ith sm aller incom e differences am on g players do significantly better than the m ore unequal ones.360 A n d w e s a w in

earlier chapters that m ore equal countries have higher o verall levels

o f attainm ent in m an y different fields.

T H E P O L I C Y F A I L U R E

Politics w a s once seen as a w a y o f im p rovin g peop le’s social and em otional w ellb ein g b y changing their econom ic circum stances. But o ver the last few decades the bigger picture has been lost. People are n o w m ore likely to see p sych o so cial w ellb ein g as dependent on w h at

can be done at the in d ividu al level, using cognitive b eh aviou ral th erapy - one person at a time - or on p ro vid in g su pport in early ch ildh ood, or on the reassertion o f religious or ‘fa m ily ’ values. H o w ev e r, it is n o w clear that incom e distribution provid es po licy m akers w ith a w a y o f im p rovin g the p sych osocial w ellb ein g o f w hole popu lation s. Politicians have an o pp o rtu n ity to do genuine good.

Attem pts to deal w ith health and social problem s through the pro visio n o f specialized services have proved expen sive and, at best, only p a rtially effective. E valu atio n s o f even some o f the m ost im portant services, such as police and m edical care, suggest that they are not am on g the m ost p o w e rfu l determ inants o f crim e levels or standards o f p o pu latio n health. O ther services, such as social w o rk or drug rehabilitation , ex ist to treat - or process - their vario u s client grou ps, rather than to dim inish the prevalence o f social problem s. O n the occasion s w hen governm ent agencies do announce policies ostensibly aim ed at prevention - at decreasing obesity, reducing health inequalities, or tryin g to cut rates o f drug abuse - it usually lo o k s m ore like a fo rm o f political w in d o w -d ressin g, a disp lay o f

go o d intentions, intended to give the im pression o f a governm ent

actively getting to grips w ith problem s. Som etim es, w hen policies w ill o b v io u sly fall ve ry far short o f their targets, yo u w o n d er whether even those w h o form u lated them , or w h o w rite the official docum ents, ever really believed their p ro p o sals w o u ld have any

m easurable im pact. T a k e health inequalities, fo r exam ple. F o r ten years B ritain has

had a governm ent com m itted to n a rro w in g the health gap between rich and p o o r. In an independent review o f p o licy in different

countries, a D utch expert said B ritain w a s ahead o f other countries in im plem enting policies to reduce health in equalities.361 H o w ever, health inequalities in B ritain h ave show n little or no tendency to decline. It is as if advisers and researchers o f all kinds kn ew , alm ost

u n con sciou sly, that realistic solutions can n ot be given serious consideration.

R a th e r than reducing in equality itself, the initiatives aim ed at tack lin g health or social problem s are n early a lw ay s attem pts to b reak the links betw een socio-econom ic d isadvan tage and the p ro b ­ lems it produces. T h e unstated hope is that people - p a rticu la rly the

p o o r - can carry on in the sam e circum stances, but w ill so m eh o w no longer succum b to m ental illness, teenage pregn an cy, educational failu re, obesity or drugs.

E very problem is seen as needing its o w n solution - unrelated to others. People are encouraged to take exercise, not to have un ­

protected sex, to say no to drugs, to try to re la x , to sort out their w o r k - life balance, and to give their children ‘q u a lity ’ time. T h e only thing that m any o f these policies do have in com m on is that they often seem to be based on the belief that the p o o r need to be taught to be m ore sensible. T h e glarin gly o b viou s fact that these problem s

h ave com m on roots in inequality and relative deprivation disappears fro m view .

T R E N D S I N I N E Q U A L I T Y

Inequ ality has risen in m an y, but not all, developed countries over the last fe w decades. Figures 1 6 . 1 and 1 6 . 2 sh o w the w iden ing gap betw een the incom es o f rich and p o o r in B ritain an d the United States o ver a thirty-year period. T h e figures sh o w the w idening

gap betw een the top and bottom 1 0 per cent in each cou n try. Both countries experienced very dram atic rises in inequality w h ich peaked in the early 19 9 0 s and have changed rather little since then. In both countries in equality rem ains at levels alm ost unprecedented since records began - certain ly higher than it has been fo r several generations. F ew other developed countries have show n quite such dram atic increases in inequality o ver this period, but on ly a very few

Ri si

ng

ga p

be tw

ee n

ric h

an d

po or

, re

la ti

ve

to 19

75

Ri si

ng

ga p

be tw

ee n

ric h

an d

po oi

; re

la tiv

e to

19 7

5

Figure 1 6 . 1 The widening gap between the incomes o f the richest and poorest 10 per cent in Britain 1 9 7 s (=1) to 2 0 0 5 -2 0 0 6 .

Figure 16 .2 The widening gap between the incomes o f the richest and poorest 10 per cent in the USA 19 7 5 (=1) to 2004.

- such as T h e N etherlan d s - seem to have avo ided them entirely. O thers, like Sw eden, w hich avo ided them initially, have had steep rises since the early 19 9 0 s .

Th e figures sh o w in g w iden ing incom e in equality in B ritain and the United States leave no room fo r doubt that incom e differences do change su bstan tially o ver time and that they are n o w not fa r short o f 4 0 per cent greater than they w ere in the m id -19 7 0 s .

If things can change so ra p id ly, then there are go o d reason s to feel confident that w e can create a society in w h ich the real q u ality o f life an d o f hum an relationships is fa r higher than it is n o w .

W henever governm ents have really w an ted to increase equality, policies to do so have not been lackin g. The historical evidence confirm s the prim acy o f political w ill. R a th e r than greater equality w aitin g till w ell-m ean in g governm ents think they can a ffo rd to m ake societies m ore equ al, governm ents have u su ally not pursued m ore egalitarian policies until they thought their su rvival depended on it.

In the early L990S a W orld B an k report pointed out that rap id eco­ nom ic gro w th in a num ber o f E ast A sian countries w a s underpinned by gro w in g eq u ality.366 In tryin g to explain w h y governm ents had adopted m ore egalitarian policies, the report said that it w a s because they faced crises o f legitim acy and needed to gain p o p u lar support. The governm ents in T a iw a n and H o n g K o n g faced rival claim s from the C om m unist Chinese governm ent. South K o re a faced N o rth K o re a , and the governm ents o f Sin gapore and the Philippines faced gu errilla forces. D escribing p o licy in these countries, Jo h n Page, w ritin g in a L994 W o rld B an k pu b lication , said:

Very explicit mechanisms were used to demonstrate the intent that all would have a share of future wealth. Korea and Taiwan carried out comprehensive land reform programs; Indonesia used rice and fertilizer price policies to raise rural incomes; M alaysia introduced explicit wealth sharing programs to improve the lot of ethnic M alays vis-a-vis the better off ethnic Chinese; Hong Kong and Singapore undertook massive public housing programs; in several economies, governments assisted workers’ cooperatives and established programs to encourage small and medium­ sized enterprises. Whatever the form, these programs demonstrated that the government intended for all to share in the benefits of growth.367

Ja p a n o w es its status as the m ost equal o f the developed countries

partly to the fact that the w h o le establishm ent had been hum iliated

by defeat in the Second W orld W a r, and partly to the su p po rt fo r political and econom ic reconstruction - including d raw in g up a new constitution - provid ed by disinterested, and re m ark ab ly far-sighted, A m erican advisers w o rk in g under G eneral M a c A rth u r.95

O ther exam p les o f increases in eq u ality have sim ilar origins.

B ism arck ’s early developm ent o f form s o f social insurance w ere part o f his attem pt to gain p o p u lar su pport fo r his project o f unifying the G erm an states. B ritain became su bstan tially m ore equal during both the F irst and Second W orld W ars as p art o f an attem pt to gain

su pport fo r the w a r effo rt by m ak in g people feel the burden o f w a r w a s eq u ally shared. A s R ic h a rd T itm uss put it: ‘ I f the co o peratio n o f the m asses w a s th ought to be essential [to the w a r effo rt], then inequalities had to be reduced an d the p yram id o f social stratifica­ tion had to be flatten ed.’368

Sw eden ’ s greater eq u ality originated in the Social D em ocratic P arty ’s electoral v ic to ry in 1 9 3 2 w h ich had been preceded by violent lab o u r disputes in w h ich troops had opened fire on saw m ill w o rk ers. A s prim e m inister alm ost co n tin u ou sly fro m 1 9 3 2 to 19 4 6 , Per A lb in H an sso n w a s able, during Sw edish rearm am ent and the w a r, to push through his aim o f m akin g Sw eden ‘a classless society’ and

‘the peop le’s h om e’ . T u rn in g n o w to exam ples o f w h ere incom e differences have

w idened rather than n a rro w e d , the central role o f politics is no less clear. In Figures 1 6 . 1 and 1 6 . 2 w e sa w the w iden ing o f incom e d if­ ferences in B ritain and the U S A w h ich took place p articu larly during the 19 8 0 s and early 19 9 0 s . Paul K ru gm an , the N o b e l Prize w in nin g econom ist, analyses the reasons fo r rising inequality in the U S A . He says that the con ven tion al ex p lan atio n is that it w a s driven by a rising dem and fo r skilled lab o u r, resulting m ain ly fro m the spread o f in form ation techn ology, and b y the im p ort o f cheap go o d s, such as textiles, rep lacin g less skilled lab o u r. H o w ev e r, he dism isses these

exp lan atio n s sayin g that econom etric research suggests that they are o n ly a sm all part o f the picture. H e also points out that factors such as these do not exp lain the ru n aw a y incom es at the top - fo r instance am on g C E O s - w hich w a s one o f the m ain features o f the g ro w th in

in equality, and adds that although these forces have been at w o rk in all rich countries, incom e differences w idened in on ly some o f them. C oun tries in w hich in equality did not increase during the 19 8 0 s and early ’90s include C a n a d a , F ran ce, Ja p a n , N etherlan d s, Spain and

S w itzerlan d.406,407 C on fin in g his attention largely to the U S A , K ru gm an argues that,

rather than m arket fo rces, rising inequality w a s driven by ‘ changes in institutions, norm s and political p o w e r’ . H e em phasizes the w e a k ­ ening o f trade unions, the abandonm ent o f pro d uctivity sharing agreem ents, the influence o f the po litical right, and governm ent changes in taxes and benefits. H e cou ld also have added the failure

to m aintain adequate m inim um w a g e legislation. D espite substantial differences betw een countries, the basic trends

in incom e distribution seen in the U S A th roughou t the twentieth century can be seen in m an y countries. A fte r inequalities rose to a

p e ak just before the G reat C rash o f 1 9 2 9 , they then n arro w ed so

d ram atically in the later 1 9 3 0 s and early ’40 s that the period is som etim es referred to as the ‘ G re at C o m p ressio n ’ . Incom e d iffer­ ences then rem ained n a rro w e r until the later 19 7 0 s or the mid 19 8 0 s . T h e y then started to w iden ra p id ly again until just before the

m ost recent financial crash , w h ere they reached levels o f inequality

not seen since just before the 1 9 2 9 crash. M o st research on changes in incom e distribution is concerned

w ith dividing up the com ponents o f the o verall trends: h o w much is due to w idening differences in earnings? H o w m uch to changes in taxes and benefits? H o w m uch is due to sim ultaneous gro w th in

w o rk less and tw o -earn er households? A n d then, at the n ext causal level d o w n , h o w m uch o f the w iden ing differences in earnings is due to w e ak e r trade unions and h o w m uch is due to a decline in dem and fo r unskilled lab o u r? But the truth is that the m a jo r changes in incom e distribution in an y cou n try are alm ost never attributable

sim ply to m arket forces influencing w a g e rates. W h at w e see instead is som ething m uch m ore like the changes in institutions, norm s and the use o f political p o w e r w h ich K ru gm an describes in the U S A . D ifferences in p re -tax earnings rise, ta x rates are m ade less progres­ sive, benefits are cut, the la w is changed to w eak en trade union p o w ers and so on. T ogether these are a fa irly clear sign o f a change

in norm s and p olitical o u tlook. If that w ere not so, and w idening

differences in earnings w ere p o litically unacceptable, governm ents

w ou ld have acted to reduce rather than increase the differences. In B ritain it w a s not until after the change o f governm ent in 1 9 9 7 that an y such attem pts w ere m ade.

T here can be even less doubt abo u t the po litical nature o f the com pression o f incom e differences before and during the Second W o rld W ar. A gain st a b ack d ro p o f the D epression , unprecedented levels o f unem ploym ent and gro w in g signs o f social unrest, coupled, p resum ab ly, w ith a fear o f the spread o f com m unism , governm ents to o k action. In the U S A President R o o sevelt in augurated the N e w

D eal in the early L930S and, w ith the com ing o f w a r, m an y go v e rn ­ ments reduced incom e differences even m ore dram atically.

If ‘m ark et fo rces’ w ere the real drivers o f in eq uality, it is unlikely that the p o st-w a r settlem ent w o u ld h ave rem ained intact fo r three or fo u r decades before incom e differences began to w iden again m ore ra p id ly in the 19 8 0 s . T h e ending o f that consensus w a s ve ry clearly related to a rig h tw ard shift in po litical opinion. Th e trium ph o f the n ew right exto llin g the benefits o f the free m arket an d the dom inance o f m onetarist econom ics w ere enshrined in the po litical leadership o f R e ag an in the U S A and T h atcher in B ritain. C om m unism had ceased to be a realistic threat and m any governm ents privatized w h a t had been state ow ned public utilities.

T o recognize h o w po litical attitudes sweep across the international

scene, w e have only to lo o k at the w a y revo lu tio n ary upheavals o f 18 4 8 sh o o k h alf a dozen different E uro p ean countries, or rem em ber the radicalism o f the 19 6 0 s , or h o w m any com m unist governm ents collapsed in 1 9 8 9 - 9 0 . A n in d icator that the w iden ing incom e d iffer­ ences o f the 19 8 0 s resulted from another such change in the p olitical

w in d is that, w ith the exception o f C an a d a , they w idened m ost

rap id ly in English speakin g countries - in B ritain , the U S A , N e w Z e a la n d and A u stralia - accom pan ied in each case by a free-m arket id eo lo gy and by policies designed to create a m ore ‘ flexib le’ lab o u r force. Stronger linguistic and id eo lo gical connections m eant that English speakin g countries caught the disease qu ick ly fro m each other and caught it badly.

A study w hich an alysed trends in in equality during the 19 8 0 s and

19 9 0 s in A u stra lia, C an a d a , G erm an y, Ja p a n , Sw eden, the United K in gd o m and the United States, found that the m ost im portant single facto r w a s trade union m em bership.370 A lth o u gh high levels o f un­

em ploym ent w eak en the bargainin g p o w e r o f lab o u r, in this study, declines in trade union m em bership w ere m ost clo sely associated w ith w idening incom e differences.

N o t on ly the extent o f u nionization but p ro visio n s fo r lab o u r representation in com panies are also likely to affect w a g e settle­ ments. Th e C om m ission o f the E u ro p ean Union requires m inim um

stan d ards o f representation and con sultation fo r all larger com ­ panies but it is not clear h o w far practice in different countries con form s to w h a t w a s intended. In Ja p a n , h ow ever, there is often a m uch closer relationship betw een m anagem ent and unions. Indeed, the Ja p a n e se Federation o f E m ployers A sso ciatio n found that 1 5 per

cent o f the directors o f large com panies w ere form er trade union

o ffic ials.371 In the countries o f the E uropean U nion the earnings o f som e 7 0 per cent o f em ployees are covered by collective agreem ents, com pared to only 1 5 per cent in the U S A . A t 3 5 per cent, the figure

fo r the U K is am on g the lo w est in the E U .

D I F F E R E N T R O U T E S T O G R E A T E R E Q U A L I T Y

R a th e r than suggesting a p articu lar route or set o f policies to n arro w

incom e differences, it is p ro b ab ly better to point out that there are m any different w a y s o f reaching the sam e destination. In C h ap ter 1 3 w e show ed that although the m ore equal countries often get their greater equality through redistributive ta xe s and benefits and through a large w e lfa re state, countries like Ja p a n m anage to achieve lo w levels o f inequality b e fo re ta xe s and benefits. Ja p a n e se d iffer­ ences in gross earnings (before taxes and benefits) are sm aller, so there is less need fo r large-scale redistribution. T h is is h o w Ja p a n m anages to be so m uch m ore equal than the U S , even though its social security tran sfers w ere a sm aller p ro p o rtion o f G D P than

so cial security transfers in the U S A .362 A lth o u gh , o f all the countries

included in ou r analyses, the U S A and Ja p a n are at opposite

extrem es in term s o f in equality, the pro p o rtion o f their G D P taken up by governm ent social expenditure is sm all in both cases: they com e second and third lo w est o f the countries in ou r analysis.

Sim ilar evidence that there are very different routes to greater equality can also be seen am on g the A m erican states.363 Th e total t a x burden in each state as a percentage o f incom e is com pletely

unrelated to inequality. Because V erm o n t and N e w H am psh ire are neighbouring N e w E n glan d states, the con trast betw een them is p a r­ ticularly striking. V erm o n t has the highest ta x burden o f an y state o f the union, w h ile N e w H am psh ire has the second lo w est - beaten only

by A la sk a . Y e t N e w H am psh ire has the best perform an ce o f a n y state

on ou r In d ex o f H ealth and Social Problem s and is clo sely fo llo w ed by V erm o n t w h ich is third best. T h ey both also do w ell on equality: despite their ra d ically different ta x a tio n , they are the fourth and sixth m ost equal states respectively. Th e need fo r redistribution depends on h o w unequal incom es are before ta xe s and benefits.

Both the international and U S state com p arison s send the sam e

m essage: there are quite different ro ad s to the greater eq u ality w hich im proves health and reduces social problem s. A s w e said in C h a p ­ ter 1 3 , w h a t m atters is the level o f in equality you finish up w ith , not h o w yo u get it. H o w ev e r, in the figures there is also a clear w arn in g fo r those w h o m ight w a n t to place lo w public expenditure and ta x a tio n at the top o f their list o f priorities. If you fail to av o id high

inequality, yo u w ill need m ore priso n s and m ore police. Y o u w ill have to deal w ith higher rates o f mental illness, drug abuse and every other kind o f problem . I f keeping ta xe s and benefits do w n leads to w ider incom e differences, the need to deal w ith the ensuing so cial ills m ay fo rce you to raise public expen diture to cope.

There m ay be a choice betw een using pu b lic expenditure to cope w ith social h arm w h ere in equality is high, or to p a y fo r real social benefits w here it is lo w . A n exam ple o f this balance shifting in the w ro n g direction can be seen in the U S A during the period since 19 8 0 , w h en incom e in equality increased p a rticu la rly rap id ly. D urin g

that period, public expenditure on prison s increased six times as fast

as public expenditure on ed ucation, and a num ber o f states have n o w reached a point w h ere they are spending as m uch public m oney on prisons as on higher ed u catio n .364

N o t only w o u ld it be preferable to live in societies w h ere m oney can be spent on education rather than on p rison s, but policies to

su pport fam ilies in early ch ildh ood w o u ld have m eant that m any o f those in prison w o u ld have been earning and payin g taxes instead o f being a burden on public funds. A s w e saw in C hapter 8, pre-school pro visio n s can be a profitable long-term investm ent: children w h o

receive these services are less likely to need special education and,

w hen they reach adu lth o o d, they are m ore likely to be earning and less likely to be dependent on w elfare o r to incur costs through crim e.36S

It is tem pting to say that there are tw o quite different paths to greater eq uality, one using taxes and benefits to redistribute incom e

fro m the rich to the p o o r, and the other ach ievin g n a rro w e r d iffer­ ences in gross m arket incom es before an y redistribution. B ut the tw o strategies are not m u tu ally exclu sive or inconsistent w ith each other. In the pursuit o f greater eq u ality w e should use both strategies: to rely on one w ith o u t the other w o u ld be to fight in equality w ith one

arm tied behind y o u r back. N everth eless, it is w o rth rem em bering that the argum ent fo r greater equality is not necessarily an argum ent fo r big governm ent. G iven that there are m any different w a y s o f dim inishing inequality, w h a t m atters is creating the necessary p o liti­

cal w ill to pursue an y o f them.

P O L I T I C A L W I L L

So if it all boils do w n to politics, h o w can w e create the necessary po litical w ill to n a rro w incom e differences? T h e strength o f the

evidence that a m ore equal society is a better society has a key role to p la y in changing public opin ion . M a n y people have a stron g per­ sonal belief in greater equality and fairness, but these valu es have rem ained private intuitions w h ich they fear others do not share. The ad van tage o f the gro w in g b o d y o f evidence o f the h arm inflicted by in eq uality is that it turns w h a t w ere p u rely p erson al intuitions into pu b licly dem onstrable facts. T h is w ill su bstan tially increase the co n ­ fidence o f those w h o have a lw a y s shared these values and encourage them to take action. In addition , some people w ill change their view s

in the light o f the n ew evidence. M a n y people are seriously w o rried abou t the m an y signs o f social failu re in ou r societies and search fo r exp lan atio n s.

P olitical differences are m ore a reflection o f different beliefs about the solution to problem s than o f disagreem ents abou t w h a t the problem s are. A lm o st everyon e, regardless o f their politics, w o u ld prefer to live in a safer and m ore frien d ly society. E veryon e w ill agree that a g o o d society w o u ld have few er o f all the health and social problem s w e have lo o ked at. Th e argum ent is therefore abou t solutions. A lth ou gh people h ave suggested m an y w a y s o f helping

in dividuals facin g p articu lar difficulties, the evidence presented in this b o o k suggests that greater eq u ality can address a w ide range o f problem s across w h o le societies. A n d if greater eq u ality is also an im portant com ponent o f policies to tackle g lo b al w arm in g, there is

m uch to recom m end it. R ecent research in B ritain using fo cu s groups

has sh o w n that an understanding o f the effects o f inequality can have a p o w erfu l influence on p e o p le’ s attitudes to it.408 P articipants draw n fro m across the social and political spectrum w ere show n the evidence pro vid ed in this b o ok on h o w in equality affects trust, child conflict, an d mental illness. A s w ell as finding the relationships

intuitively plau sible, they w ere also m oved b y them. M a n y o f those

p revio u sly opposed to greater eq u ality changed their m inds. Even people w h o rejected appeals fo r greater fairness w ere in fav o u r o f greater equality w hen it w as presented as p art o f a social vision aroun d im p rovin g the q u ality o f life fo r everyone. In terms o f creat­

ing the necessary po litical w ill, the evidence w a s regarded as one o f the m ost im portan t reasons fo r reducing inequality.

Fo r several decades progressive politics have been seriously w eaken ed by the loss o f a n y concept o f a better society. People have argued fo r piecem eal im provem ents in different areas o f life,

cam paigned against n ew environm ental threats or fo r better treat­

ment o f asylum seekers, and have dem onstrated again st m ilitary interventions. But n ow h ere is there a p o p u lar m ovem ent capable o f inspiring people w ith a vision o f h o w to m ake society a su bstan tially better place to live fo r the vast m ajo rity. W ith ou t that visio n politics

w ill ra rely p ro v o k e m ore than a y aw n .

Y e t m ost people do w a n t change. In the first chapter o f this b o ok

w e referred to a research report called Y ea rn in g fo r B a la n c e , w hich

sho w ed that three-quarters or m ore o f A m erican s felt that society

had lost touch w ith w h a t really m attered.1 C onsum erism and m ateri­ alism , they felt, w ere w in nin g out o ver m ore im portan t values to do w ith friends, fam ily and com m unity. A lth ou gh politician s recognize a deep-seated m alaise, and so cam paign fo r votes, sayin g that they stand fo r ‘ch an ge’ , they som etim es seem to have fe w ideas fo r change

w h ich go deeper than differences in the p ersonal im ages they project.

T h ere is no suggestion that they have an y v ie w o f h o w to begin changing daily life into som ething m ore jo y fu l and fulfilling.

Public opinion p o lls suggest that there is a substantial desire for n a rro w e r incom e differences. In B ritain o ver the last tw en ty years

polls have show n that the p ro p o rtion o f the po pu latio n w h o think

that incom e differences are to o big has averaged aroun d 80 per cent and has rarely dipped b elo w 75 per cent - even though m ost people underestim ate h o w big incom e differences actu ally are. In the U S A , the 2 0 0 5 M a x w e ll Poll on C ivic Engagem ent reported that over 80 per cent o f the po pu latio n thought the extent o f in equality w a s

a p roblem , and alm ost 60 per cent thought the governm ent should try to reduce it. G a llu p p olls betw een 19 8 4 and 20 0 3 w h ich asked A m erican s w hether incom e and w ealth w ere fa irly distributed or should be m ore evenly distributed, found that o ver 60 per cent o f the po pu latio n thought they should be m ore evenly distributed.369

C O R P O R A T E P O W E R - T H E E L E P H A N T I N T H E L I V I N G R O O M

Part o f the problem o f po litical w ill is the feeling that w e do not h ave the means to m ake an y difference. W e m ay all decry the vast w ealth o f the super-rich, but w h a t can w e do? U nions can , as the evidence suggests, m ake som e difference, but it is h ard to escape the conclusion that the high levels o f inequality in our societies reflect

the concentrations o f p o w e r in ou r econom ic institutions. The

institutions in w h ich w e are em ployed are, after all, the m ain source o f incom e inequality. It is there that valu e is created and divided between the v a rio u s grad atio n s o f em ployees. It is there that the

inequities w h ich necessitate redistribution are set up. A nd it is there that w e are m ost exp licitly placed in a ran k-ordered hierarch y, superiors and in feriors, bosses an d subordinates.

In 2 0 0 7 ch ief executives o f 3 6 5 o f the largest U S com panies

received w ell over 5 0 0 tim es the p a y o f their average em ployee, and these differences w ere getting bigger. In m an y o f the top com panies the ch ief executive is paid m ore in each day than the average w o rk e r is in a year. A m o n g the Fortune 500 com panies the p ay gap in 2 0 0 7 w as close to ten times as big as it w a s in 19 8 0 , w h en the long rise in incom e inequality w a s just beginning.

B ecause the ratio o f C E O p a y to average w o rk e r p a y varies

som uch between large an d sm all com panies and fro m one sector to another, it is difficult to com pare like w ith like w h en m ak in g inter­ nation al com parison s. H o w ev e r, an attem pt (from a respected source) to m ake such co m pariso n s, suggests th at ratios o f C E O

com pensation to the p ay o f pro d uctio n w o rk e rs in m an u factu rin g

m ight be 1 6 : 1 in Ja p a n , 2 l : l in Sw eden, 3 1 : 1 in the U K and 4 4 : 1 in the U S A .372

A ccordin g to the annual survey o f ch ief execu tives’ p a y carried out by the G u a r d ia n , b o ard ro o m p a y in the 1 0 0 com panies included in the Fin an cial T im es Stock E xch an ge in d ex in B ritain has risen in

successive years b y 1 6 per cent, 1 3 per cent, 28 per cent and m ost

recently ( 2 0 0 6 - 2 0 0 7 ) by 3 7 per cent at a period w h en inflation w a s rarely m ore than 2 per cen t.373 Th e average p a y (including bonuses) fo r the ch ief executives o f top com panies stood at just under £ 2 .9 m illion. A fte r review in g em pirical research, the International L a b o u r

O rgan izatio n concluded that there is little or no evidence o f a re la ­

tionship betw een executive p ay and co m p an y perform an ce and sug­ gested that these excessive salaries are likely to reflect the dom inant b argainin g p osition o f execu tives.374

T o p business p a y has fa r outstripped anything in the public sector.

In the U S A , the tw en ty highest-paid people w o rk in g in public

traded co rp o ratio n s received alm ost 4 0 times as m uch as the tw enty highest-paid peop le in the n on-profit sector, and 20 0 times m ore than the tw en ty highest-paid generals or cabinet secretaries in the Federal G o vern m en t.375

It seems likely that the den ation alization o f m a jo r industries and

the privatization o f large num bers o f frien d ly societies, m utuals, building societies, providen t societies and credit unions, w h ich had been controlled by their m em bers, m ay have m ade a substantial co n ­ tribution to the w iden ing incom e differences show n in Figures 1 6 . 1 and 1 6 . 2 . It w a s com m on practice fo r C E O s and other senior

m an agers to receive huge sala ry increases shortly after conversion to pro fit-m akin g co rp o ratio n s. T h is p ro b ab ly exp lain s som e o f the sharp rise in inequality w h ich Figure 1 6 . 1 show s to o k place in B ritain aroun d the m id -19 8 0 s . B ritish T elecom w a s privatized in 1 9 8 3 , B ritish G a s in 19 8 6 , fo llo w e d by a flood o f m ajo r com panies

in 1 9 8 7 . T h e international extent o f the w iden ing o f incom e in­

eq u ality is also consistent w ith a con tribution fro m privatization. N u m ero u s co rp o ratio n s are n o w bigger than m any nation states.

In the w o rd s o f the United N a tio n s C on feren ce on T ra d e and D evelopm ent ( U N C T A D ) :376

Twenty-nine of the w orld’s 10 0 largest economic entities are transnational corporations (T N C s), according to a new U N C T A D list that ranks both countries and T N C s on the basis of value added. O f the 200 T N C s with the highest assets abroad in 2000, E xxon is the biggest in terms of value

added ($63 billion). It ranks 45th on the new list, making it comparable in

economic size to the economies o f Chile or Pakistan. Nigeria comes in just between DaimlerChrysler and General Electric, while Philip M orris is on a par with Tunisia, Slovakia and Guatemala.

U sing different m easures, other estim ates suggest that h a lf o f the

w o r ld ’s largest econom ies are m u ltin ationals, and that G eneral M o to rs is bigger than D en m ark, that D aim lerC h rysler is bigger than P o lan d; R o y a l Dutch/Shell bigger than V en ezuela, and So ny bigger than P akistan. Lik e the aristo cratic ow n ership o f huge tracts o f land, w hich in 1 7 9 1 T o m Paine attacked in his T h e R ig h ts o f M a n ,377 these p roductive assets rem ain effectively in the hands o f a very

few , very rich peop le, an d m ake ou r claim s to real dem ocracy look pretty thin.

In T o m P ain e’ s lifetim e the capitalist system w a s in its infancy. A s an advocate o f eq u ality and dem ocracy, he focused his attack on the landed aristo cracy, the n o b ility, the m o n arch y, and on their ow n ership o f huge sw athes o f land. H e seems to have assum ed

that the m arket system , then in vo lvin g m ain ly sm all traders and craftsm en, w o u ld rem ain sm all-scale, fairly eg alitarian , and so com ­ patible w ith dem ocracy. H a d he foreseen h o w the developm ent o f huge m ultin ational co rp o ratio n s w o u ld surpass the concentrations o f w ealth and undem ocratic p o w e r o f his day, he w o u ld surely have included them in his sights. It is not possible to discuss w a y s o f reducing incom e differences w ith o u t discussing w h a t can be done abou t these bastions o f w ealth , p o w e r and privilege.

T h e failed experim ent w ith state ow n ership in the centrally planned econom ies o f the form er Soviet U nion and E astern E urope w as intended, am ong other things, to pro vid e a solution to the p ro b ­ lem o f the gro w in g concentration o f productive p o w e r in private hands. But concentrating that p o w e r into the hands o f the state w a s not only som etim es hugely inefficient, but invited co rrup tio n , led to the denial o f im p ortan t basic freedom s and harm ed public life. T h at failu re seems to have m ade us feel there are no w o rk a b le alternatives to the standard capitalist m odel and prevented us thinking creatively abo u t other m ore dem ocratic and egalitarian m ethods. W e blinker ourselves to the fact that there are lots o f alternatives, m any o f w hich are alread y p a rt o f ou r lives and flourishing all aroun d us.

A L T E R N A T I V E S

In his b o o k , A m e ric a B e y o n d C a p ita lism : R e c la im in g o u r W ealth, o u r L ib e r ty a n d o u r D e m o c ra c y , G a r A lp ero vitz, a pro fesso r o f political econ om y at the U n iversity o f M a r y la n d , sum m arizes the v a riety and scale o f the alternatives operatin g in the U S A .378 H e em phasizes the huge size o f the non-profit sector. In the tw en ty largest U S cities alm ost 40 per cent o f the 20 0 largest enterprises are

n on-profit organ ization s like universities and m edical institutions. H e mentions the 2,0 0 0 m un icipal electric utilities w h ich supply 40 m illion A m ericans w ith electricity. L a rg ely because they are not h avin g to m ake profits fo r shareholders they are often cheaper - an

average o f 1 r per cent cheaper, A lp ero vitz says - than profit-m akin g

com pan ies, and m an y p a y p articu lar attention to su stain ability and the developm ent o f ren ew able sources o f p o w er. A lso at the local

level, he discusses organ ization s like the 4 ,0 0 0 or so C om m u n ity D evelopm ent C o rp o ra tio n s w h ich su pport local com m unities by setting up low -incom e housing schem es, p ro vid in g finance fo r local

businesses w hich they som etim es o w n and control. T h ere are 4 8 ,0 0 0 co-ops in the U S and some 1 2 0 m illion people are m em bers o f them. T h ere are around 1 0 ,0 0 0 credit unions, w ith assets to tallin g $ 6 0 0 billion, p ro vid in g financial services fo r 83 m illion A m ericans. A ro u n d 1,0 0 0 m utual insurance com panies are ow n ed by their po licy-h o ld ers, and 3 0 per cent o f A m erican farm products are m arketed through co-operatives.

In B ritain institutions like universities, hospitals and local go vern ­ ment are also often the largest local em ployers. Because m edical care and universities - like the rest o f education - are alm ost entirely pu b licly funded, they are governed by bodies accountable to the public. T h e govern in g bodies o f O x fo rd and C am brid ge colleges are dem o cratically com prised o f all fellow s. D espite a stam pede to cash

in the profits to be m ade by selling o f f frien d ly societies and m utuals, there are still 63 building societies (with over z ,o o o branches and 3 8 ,0 0 0 em ployees), 6 5 0 credit unions, 7 0 m utual insurance com ­ panies as w ell as 2 5 0 frien d ly societies in B ritain , pro vid in g variou s

financial services to their m em bers. T h ere are alm ost 17 0 ,0 0 0

charities w ith a com bined annual incom e o f over £ 4 4 billion. In 2 0 0 7 the C o -o p erative B an k , w ith £ 4 0 billion o f assets, w a s reco g­ nized as the m ost co rp o rately responsible co m p an y in the U K , acco rdin g to Business in the C om m u n ity, an influential charitable asso ciatio n o f British com panies. The recently revam ped 6 ,3 0 0

C o -o p shops still have a m arket share o f abo u t 5 per cent o f all food retailing and they rem ain the U K ’ s largest ‘ n eigh b o u rh o o d ’ retailer w ith a share o f alm ost 8 per cent o f that m arket. Even B ritain ’s experience o f nationalized industries (which once covered electricity, g as, w a ter, telephones, railw ays) w a s not all b ad. T h ro u gh o u t the

1 9 5 0 s and 19 6 0 s , as the econom ist and jo u rn alist W ill H u tto n has

pointed out, pro d u ctivity in the nationalized industries m atched or exceeded the private sector.379 H e says they began to get a b ad name w h en governm ents raided their profits and held do w n their prices to help reduce in flation ary pressures in the n ation al econom y.

T h e va riety and vast scale o f this organ izatio n al experience

leaves no doubt that profit-m akin g business is not the only effective

w a y people can w o rk together to pro vid e im p ortan t services. It is a

truism - but nevertheless an im portan t one - to say that the key difference between the kinds o f o rgan ization w e have listed and profit-m akin g co rp o ratio n s is sim ply w h eth er or not their prim ary purpose is to m ake m on ey or to provid e a service w hile rem aining econ om ically viable. A lth o u gh som e profit-m akin g businesses have

high ethical standards, the institutional fram ew o rk (and often cut­ th ro at m arket pressures) seem to invite them into an exp lo itative relationship w ith society - hence perhaps w h y w e have needed a ‘ fair trad e’ m ovem ent. Presu m ab ly because o f the m o tivatio n al

difference, there is a strong im pression that m an y o f the other form s o f organ izatio n a llo w institutions to develop a service ethic and to see their purpose as the furthering o f environm ental and com m unity interests. T h e fact that top salaries in the pro fit-m akin g sector are several hundred times top political, ju dicial or m ilitary salaries is no doubt p artly a reflection o f the profit-m akin g m otive.

W H A T C A N B E D O N E ?

So h o w can the inequality-generating forces in the profit-m akin g sector be contained and dem ocratized? H o w can they be adapted to fit in w ith the need to m ake ou r societies m ore equal? W hat can w e do w hich cannot be easily reversed by an incom ing governm ent w ith o ppo sin g interests? W hen thinking abo u t this w e should keep in m ind just h o w fundam ental a turning point w e have reached in hum an history. A s w e show ed in C hapters i and 2, further im p ro ve­ ments in the q u ality o f life no longer depend on further econom ic gro w th : the issue is n o w com m unity and h o w w e relate to each

other. O ne appro ach to tacklin g ru n a w a y p ay rates at the top m ight be

to plug lo opholes in the ta x system , lim it ‘ business expen ses’ , increase top t a x rates, and even legislate to lim it m axim u m p ay in a

co m p an y to som e m ultiple o f the average or lo w est paid. W hile such solutions m ay seem to be the only short-term o ption , they are very vulnerable to changes in governm ent: even if effective t a x changes

w ere devised and introduced, a n ew governm ent w ith different political allegiances could sim ply reverse them all. G iven the im port­ ance o f keeping in equality d o w n , w e need to find w a y s o f ensuring that greater eq u ality is m ore deeply rooted in the fab ric o f our societies and less vu ln erable to the w h im o f successive governm ents. W e need to address the. concentrations o f p o w e r at the heart o f the

econom ic life. A n ap pro ach w h ich w o u ld solve some o f the problem s is dem o­

cratic em ployee-ow nership. It not on ly avo id s concentrating p o w er in the hands o f the state, but evalu atio n s suggest that it has m ajo r econom ic and social advan tages over organ ization s ow n ed and

controlled by outside investors in w h ose interests they act.

In m any countries, governm ents use t a x concessions to encourage em ployee share-ow n ersh ip system s. T h ey do so because it is assum ed that share ow nership im p roves co m p an y perform an ce by reducing the o pposition o f interests betw een em ployers and em ployees. In

the U K , share-ow n ersh ip schemes n o w cover alm ost a q u arter o f all

em ployees and some 1 5 or 20 per cent o f all U K co m p an ies.380-81 In the U S , the 2.001 T a x L a w increased the t a x advan tages o f Em ployee Stock O w n ersh ip Plans (E S O P s), and they n o w co ver 8 m illion em ployees in 1 0 ,0 0 0 firms w ith an average em ployee-ow nership o f 1 5 - 2 0 per cent.382

H o w ev e r, m any share-ow n ersh ip schemes am ount to little m ore than incentive schemes, intended to m ake em ployees m ore com plian t w ith m anagem ent and som etim es to provid e a nest-egg fo r retire­ ment. A s a result, they are often seen as tokenism , rather than as a key to tran sform in g the structure o f em ploym ent. T h is is w h y

research show s th at em ployee share-ow n ersh ip, on its o w n , is not enough to m ake m uch difference to co m p an y perform an ce. Patrick R o o n e y , an econom ist at the universities o f Indiana and Purdue, found that em ployee share-ow n ersh ip did not necessarily m ean

that em ployees w ere m ore in volved in the running o f the com panies in w h ich they w o rk e d .383 H e com p ared the extent o f em ployee p articip ation in a w ide range o f decisions in com panies w ith and w ith o u t em ployee share-ow n ersh ip schemes. In general, em ployee involvem ent w a s lo w , but even in com panies w ith em ployee share- ow n ership schemes s ta ff m em bers w ere often not inform ed or

consulted, and the m a jo rity o f these com panies did not enable em ployees to have a significant input into decision m aking.

T o m ake a reliable difference to co m pan y perform an ce, share- ow n ership has to be com bined w ith m ore particip ative m anagem ent m ethods.384-5 T h ere h ave n o w been a num ber o f large and w ell- controlled studies - including those using b efore-an d-after p e rfo rm ­ ance data fo r several hundred m atched pairs o f com pan ies386 - w hich dem onstrate the econom ic benefits o f the com bin ation o f em ployee share-ow nership and p a rticip atio n .385-387 T h e studies sh o w repeatedly that substantial perform an ce benefits com e on ly w hen em ployee share-ow n ersh ip schemes are accom pan ied by m ore p articip ato ry m anagem ent m ethods.380’ 383> 388-9 R esearch that looked at a large num ber o f British com panies during the 19 9 0 s foun d that em ployee share-ow n ersh ip, profit-sharin g and p articipation each m ake an independent contribution to increased p ro d u ctivity .380 A review o f

research c o n c lu d e d :385

We can say with certainty that when ownership and participative management are combined, substantial gains result. Ownership alone and participation alone, however, have at best, spotty or short-lived results,

(p. n ) . . . the impact o f participation in the absence of (share) ownership is

short-lived . . . Ownership seems to provide the cultural glue to keep participation going, (p. 3)

Studies o f h o w w o rk affects health point in the sam e direction: as

w e sa w in C h apter 6 , people seem to thrive w here they h ave m ore control o ver their w o rk . H a v in g con trol at w o rk w a s the m ost successful single fac to r exp lain in g threefold differences in death rates betw een senior and ju n io r civil servants w o rk in g in the sam e g o v e rn ­

m ent offices in B ritain .64 In p ractice, this p ro b ab ly has a lot to do

w ith a sense o f auto n o m y and not feeling so directly subordinated. The im portance o f con trol at w o rk is n o w u nderstood to in volve a greater degree o f w o rk p la ce dem o cracy.390 T here is, in addition, gro w in g evidence that a sense o f unfairness at w o r k is an im p ortant risk facto r fo r p o o r h ealth .391

T h e concept o f a co m pan y being ow n ed by outside investors

has im plications w h ich lo o k in creasin gly anachronistic. A sm aller

and sm aller part o f the valu e o f a co m pan y is the valu e o f its b uild­

ings, equipm ent and m arketab le assets. It is instead the value o f its em ployees. W hen com panies are bought and sold , w h a t is actu ally being bought and sold is, abo ve all, its sta ff as a grou p o f people, w ith their assem bled skills, abilities, and kn o w led ge o f co m p an y system s and prod uction m ethods. O n ly they have the ability to m ake

the co m p an y tick. A n d o f course the concept o f a gro u p o f people being bought and sold , and belonging to anyone but its ow n m em bers, is a concept w h ich is the very opposite o f dem ocratic.

Should em ployees not have full con trol o ver their w o rk and the distribution o f its earnings? A n d should extern al shareholders

re ally receive unearned incom e beyond agreed interest on capital? P articipation , com m itm ent, con trol and profit-sharin g w o u ld be m axim ized if com panies w ere 1 0 0 per cent em ployee-ow ned. C o m p an ies could raise capital through loans or m ortgages, retaining con trol them selves. A t the m om ent, on ly a tiny p ro p o rtion o f the

m oney gam bled on the Stock E xch an ge m akes an y contribution to helping com panies buy prod uctive assets. Indeed, o ver time the paym en t o f dividends to extern al shareholders is a m ajo r drain on co m p an y profits w h ich m ight have been used to im p rove their techn ology and equipm ent.

R o b e rt O akesh ott, a B ritish auth o rity on em ployee-ow nership, says that em ployee-ow nership ‘entails a m ovem ent from business as a piece o f p ro perty to business as a w o rk in g com m u n ity’ .388’ p- 104 C o m pan ies change from being p ro perty to being com m unities when em ployees o w n a m ajo rity o f shares and so control the business. T h a t is w hen m anagem ent becom es responsible, not to outside shareholders w ith little interest in the co m p an y beyond returns on cap ital, but to the b o d y o f em ployees. Th en co m pan y meetings becom e occasion s w h en m anagem ent reports back to em ployees and has to deal w ith questions and discussion am on g peop le w h o have an intim ate k n ow led ge o f w h a t has gone right and w h a t has

gone w ro n g in the preceding period, and w h a t the rem edies m ight be. Th e tran sform ation after an em ployee buy-out fro m the usual to p-d ow n m entality can in volve a long slo w process o f peop le’ s em ancipation from the usual assum ptions rou n d class and ability

w h ich m ake those in m ore ju n ior positions feel them selves to be

in ferior hum an beings. W e discussed in C h ap ter 8 som e o f the experim en tal evidence using race and caste to sh o w h o w attributions

o f inferior status can affect perform an ce.

This process o f adjustm ent an d em ancipation is described in L o c a l H e ro e s, D avid E rd a l’ s account o f the em ployee buy-out o f Loch Fyne O ysters in S co tlan d .392 It is in part a process o f undoing the dam age o f class inequality, a process p resum ab ly m ade m ore difficult b y the fact that such assum ptions rem ain entrenched all aroun d peop le in the rest o f their lives. H o w ev e r, the structures in

w h ich w e w o r k are p ivotal.

C o -o p eratives and em ployee buy-outs have often originated as responses to desperate circum stances in w h ich tradition al systems o f ow n ership and m anagem ent have failed. E m ployees have used them to av o id closures and unem ploym ent in the m ost difficult m arket

circum stances. Even then they h ave som etim es succeeded beyond

expectation s - as did T o w e r C o llie ry in South W ales w h en , in 1 9 9 5 , m iners used their redun dan cy m on ey to buy the pit and ran it successfully until the co a l w a s w o rk e d out thirteen years later. M an y fu lly em ployee-ow ned com panies h ave a prou d record. E xam p les

include, or have included, the Lo n d o n S ym phon y O rch estra, C arl

Z e iss, United A irlin es, G o re -tex , the P o laro id C o rp o ra tio n , and the Jo h n L ew is Partnership (one o f B ritain ’s m ost successful retailers w ith 6 8 ,0 0 0 em ployee-partners and annual sales o f £6-4bn ). In the U S A , am on g the largest m a jo rity em ployee-ow ned com pan ies are

P u b lix Sup erm arkets, H y-vee Superm arkets, Science A p plicatio n s

International C o rp o ra tio n ( S A IC ) , the in tern ation al engineering and construction co m pan y C H z M H ill and T ribu n e w h ich , am on g other m edia operation s, publishes the L o s A n g eles T im es and C h ic a g o T rib u n e . These com panies average 5 5 ,0 0 0 em ployees each.

One o f the b est-know n co-operative grou p s is the M o n d ra g o n C o rp o ra tio n in the B asque region o f Spain. O ver h a lf a century it has developed into a gro u p o f o ver i z o em ployee-ow ned co-operatives w ith 4 0 ,0 0 0 w o rk e r-o w n e rs and sales o f $ 4 .8 billion U S dollars. M o n d ra g o n co-op eratives are tw ice as profitable as other Spanish firms and h ave the highest lab o u r pro d u ctivity in the co u n try .388 It is h ard to exp lain som e o f the successes unless a com bination

o f ow n ership and p articipation does indeed have the poten tial to

im p rove prod uctivity by reducing the conflict o f interests.

F o r m ost o f the em ployed po p u latio n it is at w o rk that they inter­ act m ost closely w ith people other than fam ily and have the potential to feel part o f a com m unity. In C h apter 3 w e sa w evidence o f the huge rises in a n xiety w h ich have taken place o ver the last fifty or so years as com m unity life has w eaken ed under the im pact o f gro w in g

geo graph ical and so cial m obility. W hile greater equality is associated w ith m ore cohesive com m unities and higher levels o f trust (see C h apter 4) and so m ay be expected to im prove life in residential neighbou rhood s, in the near future w e are u n likely to regain the

benefits o f the very close-knit residential com m unities o f the past.

B ut at w o rk there is the potential fo r people to find a nucleus o f friendships and to feel valued. T h is potential is u su ally underm ined by the h ierarch ical stratification o f people into vario u s grad ation s o f o rder-givers and ord er-takers, w h ich ensure that em ployees act not as a com m unity, but as p ro perty, b rought together and used to earn

a return on other peop le’s capital. One o f us recently visited tw o sm all com panies soon after they had been bought by their em ploy­ ees. W hen sta ff w ere asked w h a t difference it had m ade, the first thing office sta ff in both com panies said in reply w a s that, w h en they w en t on to the shop flo o r, ‘ people lo o k yo u in the eye’ . U nder the old

system , eye contact had been avoided. Em p loyee-ow n ersh ip has the advan tage o f increasing equality

specifically by exten din g lib erty and dem ocracy. It is bottom -up rather than to p-d ow n . A lth ou gh w e d o n ’ t k n o w w h a t scale o f incom e differences people w o u ld think fa ir, it seems likely that they

m ight agree that the ch ief executive o f the co m p an y they w o rk fo r should be p aid a sala ry several times as big as their o w n - m aybe three, or perhaps even ten, times as big. But it is unlikely that they w o u ld say several hundred times as big. Indeed, such huge differen­ tials can p ro b ab ly only be m aintained b y denying an y m easure o f econom ic dem ocracy.

A s long as the em ployee-ow ned sector rem ains on ly a sm all part o f the w h o le econom y, it can n ot use very different p a y scales from other com panies. If em ployee-ow ned com panies p aid ju n io r w o rk ers

m ore than other com pan ies, and the m ost senior s ta ff less, then the ju n io r sta ff w o u ld never leave an d senior ones w o u ld be harder to recruit. H o w ev e r, as the em ployee-ow ned sector becam e larger, peop le’s norm s and values abou t w h a t are ap p ro p riate rates o f p ay

fo r different jo b s, and w h a t differentials are acceptable, w o u ld change. W e m ight at least m ove to w a rd s the norm s o f the public and n on-profit sector. A n d if there w a s no longer a set o f hugely w ealth y private sector bosses inviting com parison s an d m akin g people think such salaries cou ld be justified, the n on-profit sector m ight itself becom e m ore egalitarian . Perhaps it is time w e m oved a w a y from

a w o rld in w h ich people regard m axim izin g p erson al gains as a lau d able aim in life.

D avid E rd a l, form er chair o f the Tu llis R u ssell G ro u p and D irector o f the B ax i Partnership, once studied the effects o f em p loy­ ment in co-operatives on the com m unities in w hich they w ere

situated.393 H e com pared three tow n s in northern Italy: Im o la, w hich

h as 25 per cent o f its w o rk fo rce em ployed in co-operatives, Faenza, w here 1 6 per cent w o rk in co -o peratives, an d S assu olo w here there are no co-operatives. O n the basis o f rather a sm all su rvey and lo w response rates, he concluded that health, ed ucation, crim e and social

participation w ere all better in the to w n s w ith a larger p ro po rtion o f the pop u latio n em ployed in co-operatives.

A s a w a y o f creating a m ore egalitarian society, em ployee- ow n ership and con trol h ave m an y advan tages. First, it enables a process o f so cial em ancipation as peop le becom e mem bers o f a team. Second, it puts the scale o f earning differentials ultim ately under

dem ocratic control: if the body o f em ployees w a n t big incom e differentials they co u ld choose to keep them. T h ird , it in volves a very substantial redistribution o f w ealth from extern al shareholders to em ployees and a sim ultaneous redistribution o f the incom e fro m that w ealth . In this co n text, that is a p articu larly im portan t advan tage.

Fourth , it im proves pro d u ctivity and so has a com petitive advantage. F ifth, it increases the likelih ood that people w ill regain the ex p e ri­ ence o f being part o f a com m unity. A n d sixth , it is likely to im p rove so ciab ility in the w id er society. T h e real re w a rd h ow ever, is not sim ply to have a fe w em ployee-ow ned com panies in a society still dom inated by a h ierarch ical id eology and status-seeking, but to have

a society o f people freer o f those divisions. A n d that can on ly be achieved by a sustained cam paign o ver several decades.

R a th e r than being com patible w ith just one system o f m an age­ ment and w o rk o rgan ization , em ployee-ow nership is h ighly flexible. It m erely puts ultim ate auth o rity in the hands o f em ployees to develop w h atever system s they find w o rk best. T h is enables systems to evolve to suit an y situation. System s o f w o rk team s, o f directors elected fo r longer or shorter periods, o f departm ental represen­

tatives, o f co m p an y trustees, o f anything fro m w e ek ly to annual co m p an y m eetings, cou ld all be tried from place to place. P ow er could be delegated, or exercised directly by the body o f votin g em ployees. G ra d u a lly people w o u ld learn the strengths and w e a k ­

nesses o f different structures and w h a t form s o f dem ocracy best

fitted the public and private sectors and h o w to represent the interests o f consum ers and lo cal com m unities.

H o w ev e r, to ensure that the num ber o f em ployee-ow ned w o r k ­ places increases, it is essential that they are constituted - as they easily can be - in w a y s w h ich prevent em ployees fro m selling their

com panies back to extern al shareholders. A lth o u gh m ost are a deq uately protected, there have also been cases o f sell-outs in w hich com panies have been lost to em ployee-ow nership and control.

A s a m eans o f tran sfo rm in g ou r societies, em ployee-ow nership has the advan tage that it can (and does) exist side b y side with con ven tion al business structures. N e w and o ld fo rm s o f business can

co exist: w ith the right legal su pport and t a x incentives the trans­

fo rm atio n o f society can start straight a w a y. It enables us to em bark on a fundam ental tran sfo rm atio n o f ou r society through an ord erly tran sition , m akin g the new society g ro w w ithin the old. G o v e rn ­ ments can give add ition al incentives and su pport to encourage

em ployee share-ow nership. C o m pan ies m ight be required to tran sfer a pro p o rtion o f shares each yea r, and retiring o w n ers m ight som etim es be w illin g to pass their com panies to em ployees.

A lth o u gh em ployee-ow ned and con trolled industry need not in volve local com m unity and consum er representatives on the

govern in g body, that is a fau lt w h ich can easily be rem edied. It m ight also be said in oppo sitio n to em ployee-ow nership, that it does nothing abou t the basic a m o rality o f the m arket. T h e desire to earn

a bigger profit w o u ld still lead com panies to act in anti-social w a y s, h ow ever they w ere con trolled . A s w ell as som e h ighly ethical com panies op eratin g in the m ark et su pportin g fair trade, the environm ent, givin g to local com m unities, etc., there are, at the sam e time, also com panies tryin g to exp an d m arkets fo r to b acco in the

developing w o rld in the sure k n ow led ge that they w ill cause m illions o f ex tra deaths. T h ere are com panies w hich have caused needless deaths by en couragin g m others in developing countries to buy pow d ered b ab y m ilk instead o f breast feeding, despite lack o f access to clean w a ter or basic hygiene. T h ere are others w h ich continue to

destroy ecosystem s, lan d and w a ter supplies, to e x p lo it m ineral

resources w h ere governm ents are to o w e ak o r co rrup t to stand up to them , and still others use their patents to prevent life-saving drugs being sold at affo rd a b le prices in p o o rer countries.

T here are reasons to think that em ployee-ow ned com panies m ight

m aintain higher stan d ards o f m o rality even w ith the profit m otive.

In co n ven tion al em ploym ent people are specifically hired to w o r k fo r purposes w h ich are not their o w n . T h e y are paid to use their exp ert­ ise to w h atever pu rpose their em ployer chooses. Y o u m ight disagree w ith the purpose to w h ich yo u r w o rk is being put, yo u m ight not even k n o w w h a t the pu rpose is, but you are not em ployed to have

opinions abo u t such things and certainly not to express them. Such issues are not y o u r concern. I f you are hired to advise on h o w yo u r co m p an y can exp an d its m arkets, im p rove profits, av o id press attention, the chances are that yo u are not being asked fo r an ethical opinion. Y o u are hired to put y o u r expertise to w o rk to serve som e­ one else’ s pu rpose. N o t on ly are the purposes not y o u r resp onsibility,

but as an em ployee you are lik ely to feel absolved fro m responsibility fo r them. T h is is w h y people have so often disclaim ed resp onsibility fo r w h a t they w ere doin g by sayin g that they w ere ‘ only carryin g out o rd ers’ . Th e fam ou s M ilg ra m experim ents sh o w ed th at w e have such a strong tendency to obey auth ority that it can result in us doing some

pretty a w fu l things. In w h a t w a s presented as a ‘learn in g’ experim ent, M ilg ra m show ed that people w ere w illin g to deliver w h a t they believed w ere not on ly very pain fu l, but also life-threatening electric shocks to a learning p artn er w h en ever the partn er gave the w ro n g an sw er to a question. T h ey did this at the request o f a m an in a w hite

co a t conducting the experim ent, despite hearing w h a t they thought w ere the scream s caused by the shocks they delivered.394

H o w ev e r, w ithin a fram ew o rk o f em ployee-ow nership and co n ­

tro l, people specifically regain ow n ership an d con trol o f the purposes o f their w o rk . If, fo r instance, yo u get to k n o w that som e aspect o f a design or m an u factu rin g process is harm ing ch ildren ’ s health, you w o u ld w a n t to change it and w o u ld p ro b ab ly start by finding out w h a t colleagues thought abou t it. T here w o u ld not be the sam e pres­ sure to keep y o u r doubts to yo u rself. N o r w o u ld yo u be able to shrug it o ff, dism issing it as none o f y o u r business. N either w o u ld yo u fear that y o u r job w o u ld be in jeo p ard y if yo u raised a w k w a rd questions. A lthough em ployee-ow ned firm s w o u ld not be abo ve all an ti-social beh aviou r, it is likely that they w o u ld succeed in m aking it at least a little less com m on.

F R E E D O M A N D E Q U A L I T Y

Th e idea that w e ca n ’t have both liberty and eq u ality seems to have

em erged during the C o ld W ar. W h at the state-ow ned econom ies

o f E astern E urope and the Soviet U nion seemed to sh o w w a s that greater equality cou ld only be gained at the expense o f freedom . A n im p ortan t id eological cost o f the C o ld W a r w a s that A m erica gave up its historical com m itm ent to equality. F o r the first A m erican s, as fo r T o m Paine, you co u ld n ’t have true liberty w ith o u t equality. W ithout

one you could not have the other. S lavery, as the sim ultaneous denial o f both, proved that rule. E q u ality w a s the bastion again st arb itrary po w er. T h is w a s expressed in the historical dem and fo r ‘N o ta xatio n w ith o u t representation’ , and ‘N o legislation w ith o u t represen tation ’ . Th e A m erican D eclaratio n o f Independence says that all men are born equal and en d ow ed w ith liberty as an inalienable right, just

as the French revolu tion aries dem anded liberty, eq u ality, fraternity. T h e com plem entarity o f liberty and equality has been proclaim ed in the w ritin gs o f m an y dem ocratic thinkers, including the social ph ilosoph er L. T . H o b h ou se, w h o believed that liberty depended, in all its dom ains, on eq u ality - equality before the la w , eq u ality o f op p o rtu n ity, eq u ality o f parties to a co n tract.395

The scale o f econom ic inequality w hich exists to d a y is less an exp ression o f freedom and dem ocracy as o f their denial. W h o , a p art from the super-rich, w o u ld vote fo r m ulti-m illion d o llar bonuses fo r the co rp o rate and financial elite w h ile denying adequate incom es to people w h o undertake so m any essential and som etim es unpleasant tasks - such as carin g fo r the elderly, collecting the trash, or w o rk in g

in em ergency services? The truth is that m odern inequality exists because dem ocracy is excluded from the econom ic sphere. It needs therefore to be dealt w ith by an exten sion o f dem ocracy into the w o rk p la ce . W e need to experim ent w ith every form o f econom ic

dem ocracy - em ployee o w n ership, prod ucer and consum er c o ­

operatives, em ployee representatives on co m pan y boards and so on.

R U N N I N G W I T H T H E T E C H N O L O G I C A L T I D E

In her b o o k , T h e W eightless W o rld, D ian e C o yle points out that although people in m ost industrialized countries experienced som e­ thing like a tw en tyfo ld increase in their real incom es during the twentieth century, the w eigh t o f all that w a s produced at the end o f the century w a s ro u gh ly the sam e as it had been at the beginning.396 She also says that the average w eigh t o f one d o lla r’s w o rth o f U S expo rts (adjusted fo r inflation) fell by a h a lf between 1 9 9 0 and 1 9 9 6 . W hile the trend to w ard s ‘w eightlessness’ is partly a reflection o f the g ro w th o f the service sector an d the ‘k n o w led g e’ econ om y, it is also a reflection o f changing techn ology and the trend to w ard s m iniaturization. T h a t so m uch o f m odern consum ption is actu ally lighter on the use o f m aterial resources than it w a s , is presum ably go o d new s fo r the environm ent. But the underlying nature o f the changes contributing to weightlessness m ay also have im portant im plications fo r equality.

In trodu ctory econom ics courses teach students the distinction between the ‘fixe d ’ costs o f pro d uctio n on the one hand, and

‘m a rgin a l’ or variab le costs on the other. Fixed costs are the costs o f the facto ry buildings and m achinery, and the v a ria b le costs are the add itio n al costs o f m ak in g one m ore unit o f output - trad ition ally

m ade up largely o f the costs o f the add itio n al lab o u r and m aterials

needed, on the assum ption that the plant and equipm ent are already

there. E con om ic th eory says that prices in a com petitive m arket should fall until they equal m argin al (or variable) costs. Prices higher than that w o u ld m ean that by prod ucin g and selling m ore, a m anu­ facturer could still earn a little m ore profit, w h ereas at a lo w er price m ak in g even one m ore item w o u ld add m ore to costs than it gained in incom e from sales.

Th ro u gh o u t large sw athes o f the m odern econ om y technological change is ra p id ly reducing va ria b le costs. Fo r everything that can be copied digitally, add ition al copies cost little o r n othing either to

produce or to distribute o ver the internet. T h is applies to all m usic,

to all com puter so ftw are and gam es, to film s, to all bo ok s and to the w ritten w o rd in an y form , to all in form ation and to pictures. T h at covers a large p art o f w h at is produced fo r entertainm ent and leisure, fo r education at all levels, and fo r m an y econom ic and profession al application s o f com puter so ftw are - w hether fo r stock

co n tro l, statistical an alysis or com puter-aided design. So lo w are m arginal costs o f digital products that there is a g r o w ­

ing ‘free’ sector. E ffo rts are m ade to enforce patents and copyright protection in an attem pt to restrict access and enable com panies to hold on to profits; but the logic o f techn ological progress is difficult to resist. System s o f co p y protection codes are cracked and goods

‘lib erated’ . In som e cases free access is supported by advertising, in others it is genuinely free, as w ith ‘ freew are’ or ‘ sh a re w a re’ co m ­ puter program m es. T h e internet has alread y p ro vid ed free access to alm ost unlim ited in fo rm atio n , not on ly b o ok s, en cyclopaedias,

dictionaries, n ew spapers, but increasingly to on-line journals. W hether legally or not, m usic and films are do w n lo ad ed free. Some

service provid ers n o w provid e unlim ited free storage space. Phone calls can cost only a fraction o f w h a t they used to and, w hen using com puter links, are increasingly free. E m ails and instant m essaging also provid e effectively free com m unications.

T h o u g h less dram atic than in the digital econ om y, the trend to w a rd s ra p id ly dim inishing variab le costs m ay also ap p ly to m any other areas o f techn ology, including the products o f n ano ­ techn ology, biotechnology, electron ically printed com ponents and

genetic engineering. Th ese n ew technologies hold out possibilities o f m ore efficient solar p o w e r, cheaper medicines and m ore econom ical n ew m aterials.

From the point o f v ie w o f m an y o f the com panies prod ucin g

digital products, the changes have not ap peared as n ew o p p o r­ tunities fo r enhancing hum an life and enjoym ent, but as pro fo un d threats to profits. Instead o f m axim izin g the benefits o f the new technologies, w e find ourselves w ith institutional structures w hich have fought to restrict this new potential. T h e dram atic lo w erin g o f variab le costs puts a ra p id ly w iden ing gap between the m axim ization o f profit and the m axim izatio n o f public benefit. In this situation it is im p ortan t that governm ents use their po w ers to aid the developm ent o f n ew institutional structures, not to p ro p up and defend the restrictions o f the o ld ones.

It used to be argued that good s fo r w hich the m argin al costs w ere

close to zero w ere inherently public good s and should be m ade

pu b licly available. B efore the digital era, bridges and ro ad s w ere com m on ly used exam p les. O nce society has incurred the capital costs o f m ak in g a bridge or ro a d , m axim u m benefit fro m the initial investm ent is gained only if use is unrestricted by charging. H ence,

people should be a llo w ed free access. Th e need to provid e un­

restricted free access in o rder to m axim ize the public benefit w a s offered as an econom ic exp lan atio n o f w h y ro ad s and bridges w ere in public ow nership - until governm ents began to try to recoup the costs o f ro ad building by ch argin g tolls.

O nce the capital cost has been incurred, the m ore people sharing

the benefits the better. W here m unicipal investm ent provides local internet access, there is no need to restrict access to it. W hen the V icto rian s established free public lib raries they recognized the same logic: a b o ok can be read repeatedly at no ex tra cost. Perhaps w e need public bodies and non-profits, funded from public revenue,

able to negotiate a price at w hich to buy access or copyrights fo r the nation. Perhaps w e need international bodies able to negotiate free access to education al and business resources th roughou t the w o rld . From the point o f vie w o f society as a w h o le, the tendency fo r techn ological change to reduce m argin al costs is rap id ly tipping the

balance o f advan tage a w a y from allo w in g p rofit-m axim izin g co r­

po ration s to con trol the distribution o f good s. In creasingly they can

on ly rely on the rem nants o f m onopolistic p o w e r provid ed by patents or copyright. W e need to find n ew w a y s o f p ayin g o rga n ­ izations and in dividuals fo r life-enhancing research, creativity and in n ovation - the geese w hich lay the golden eggs - w hich does not then restrict access to the benefits. Perhaps w e need charities to fund

the developm ent o f so ftw are fo r free w o rld w id e use. W e certainly

need a com plete revision o f co p yrigh t and patent law s so that those w h o produce valu ab le good s an d services can be paid in w a y s w hich do not restrict access to their products.

T h e question fo r politician s and the public is w hether it is possible

to find w a y s o f payin g co rp o ratio n s fo r their research and develop­

ment w ith o u t trying to police a pricin g system w h ich restricts access to the benefits o f w h a t they have produced - benefits w h ich m ay include life-saving drugs, agricultu ral innovations w h ich could feed the h un gry, and access to scientific and academ ic jo u rn als for

universities in the developing w o rld . I f it is correct to think that

n ew techn ology tends increasingly to lo w er va ria b le costs, then this problem w ill becom e increasingly pressing.

Perhaps the logic m oves us to w a rd s a society in w h ich access to an ever-increasing range o f go o d s is no longer tightly rationed by incom e, an d our possessions cease to p lay such an im portan t role in

social differentiation. W e m ight hope that w e w ill start to experience o urselves p rim arily as unranked m em bers o f the sam e society b rought together in different com bin ation s accordin g to ou r vario u s shared interests.

T H E F U T U R E O F E Q U A L I T Y

C au gh t up in d ay-to -d ay events, it is easy to forget that a longer view reveals an alm ost unstoppable historical trend to w a rd s greater eq uality. It runs like a river o f hum an progress fro m the first con ­

stitutional lim itations on the ‘ divin e’ (and arbitrary) right o f kings, and continues on through the slo w developm ent o f dem ocracy and the establishm ent o f the principle o f eq u ality before the law . It swells w ith the abolition o f slavery and is strengthened by the exten sion o f

the franchise to include non -property-ow n ers and w om en . It picks up pace w ith the developm ent o f free education, health services and system s o f minim um incom e m aintenance coverin g periods o f un­ em ploym ent and sickness. It runs on to include legislation to protect the rights o f em ployees and tenants, and legislation to prevent racial discrim ination. It includes the decline o f fo rm s o f class deference. The abo litio n o f cap ital and co rp o ral punishm ent is also part o f it. So to o is the gro w in g agitation fo r greater eq u ality o f opp ortu n ity - regardless o f race, class, gender, sex u al orien tation and religion. W e see it also in the increasing attention paid by lo bb y grou ps, social research and governm ent statistical agencies to po verty and inequality o ver the last fifty years; an d m ost recently w e see it in the attem pt to create a culture o f m utual respect.

A ll are different m anifestations o f g ro w in g equality. A n d , despite differences in political opin ion , there are few people w h o , when

lo o kin g back on these historical developm ents, w o u ld not regard them all as w elcom e. T h e historical forces u n derlying them ensure that these are changes w h ich a large m ajo rity w ill w a n t to continue. T h at this river o f hum an progress is o ccasio n ally briefly dam m ed up, o r w e experience eddying currents, should not blind us to its

existence. T h e relationships between in equality and the prevalence o f health

and so cial problem s show n in earlier chapters suggest that if the United States w a s to reduce its incom e in equality to som ething like the average o f the fo u r m ost equal o f the rich countries (Jap an ,

N o r w a y , Sw eden an d Fin lan d), the p ro p o rtio n o f the p o p u latio n

feeling they could trust others m ight rise by 7 5 per cent - presum ­ ab ly w ith m atching im provem ents in the q u ality o f com m unity life; rates o f mental illness and obesity m ight sim ilarly each be cut by alm ost tw o-th ird s, teenage birth rates cou ld be m ore than halved,

prison p o pu lation s m ight be reduced by 7 5 per cent, and people

could live longer w h ile w o rk in g the equivalent o f tw o m onths less per year.

Sim ilarly, if B ritain becam e as equal as the sam e fo u r countries, levels o f trust m ight be expected to be tw o-th irds as high again as

they are n o w , mental illness m ight be m ore than halved , everyone

w o u ld get an add ition al year o f life, teenage birth rates cou ld fall to

one-third o f w h a t they are n o w , hom icide rates could fall by 7 5 per

cent, everyone cou ld get the equivalent o f alm ost seven w eek s extra

h o lid ay a year, an d the governm ent could be closin g prison s all over the cou n try.

W h at is essential if w e are to bring a better society into being is to develop a sustained m ovem ent com m itted to doing that. Policy

changes w ill need to be consistently devoted to this end o ver several

decades and that requires a society w hich k n o w s w here it w an ts to go. T o help w ith this w e provid e - and w ill continue to provid e - our research findings, grap h s and other in fo rm atio n on the E q u ality T ru s t’ s w eb site (w w w .e q u a lity tr u st.o rg .u k ).

T h e initial task is to gain a w idespread public understanding o f

w h a t is at stake. But rather than allo w in g this to be just one more

idea that briefly gains attention before fash io n ab le o pin ion m oves on, w e need to build a social m ovem ent com m itted to its realization. It m ust be taken up and pursued by a n etw o rk o f eq u ality groups m eeting to share ideas and action everyw h ere, in hom es and offices,

in trade unions and po litical parties, in churches and schools.

It needs also to be pursued by the pressure gro u ps, charities and services concerned w ith the va rio u s issues w h ich are related to eq u ality, w hether health o r teenage births, prison p o p u latio n s or m ental health, drugs or edu cation al standards. A n d they need to be coupled w ith the urgent task o f dealing w ith g lo b al w arm in g.

In all these settings w e m ust speak out and exp la in the advan tages o f a m ore equal society.

N o r should w e a llo w ourselves to be co w ed by the idea that higher taxes on the rich w ill lead to their m ass em igration and econom ic catastrophe. W e k n o w that m ore egalitarian countries live w e ll, w ith high livin g standards and m uch better so cial environm ents. We

k n o w also that econom ic g ro w th is not the yard stick by w h ich every­ thing else m ust be judged. Indeed w e k n o w that it no longer contributes to the real q u ality o f ou r lives and that consum erism is a danger to the planet. N o r should w e a llo w ourselves to believe that the rich are scarce and precious m em bers o f a su perior race o f m ore

intelligent beings on w h o m the rest o f us are dependent. T h a t is m erely the illusion that w ealth and p o w e r create.

R a th e r than ado ptin g an attitude o f gratitude to w ard s the rich, w e

need to recognize w h a t a dam agin g effect they h ave on the social fab ric. T h e financial m eltd ow n o f late 20 08 an d the resulting reces­ sion sh o w us h o w dangerous huge salaries and bonuses at the top

can be. A s w ell as leading those in charge o f ou r financial institutions

to ado pt policies w h ich put the w ellb ein g o f w h o le p o pu lation s in je o p ard y , the very existence o f the super-rich increased the pressure to consum e as everyone else tried to keep up. T h e long speculative boom w h ich preceded the financial crash w a s fuelled substan tially by the g ro w th o f consum ers’ expenditure. Increased inequality

led people to reduce their savin gs, increase their ban k overdrafts

and credit card debt, and arrange second m ortgages to fun d co n ­ sum ption. B y adding to the speculative element in the cycles o f econom ic boom and bust, great inequality shifts our attention from the pressing environm ental and social problem s and m akes us w o rr y abo u t unem ploym ent, insecurity, and ‘ h o w to get the econom y

m ovin g a g a in ’ . R ed u cin g in equality w o u ld not o n ly m ake the

econom ic system m ore stable, it w o u ld also m ake a m ajo r co n tribu ­ tion to so cial and environm ental sustainability.

M o d ern societies w ill depend in creasin gly on being creative, ad ap tab le, inventive, w ell-in fo rm ed and flexible com m unities, able

to respond generously to each other and to needs w h erever they

arise. T h o se are characteristics not o f societies in hock to the rich, in w h ich peop le are driven by status insecurities, but o f p o p u lation s used to w o rk in g together and respecting each other as equals. A n d, because w e are tryin g to g ro w the n ew society w ith in the old, our

values and the w a y w e w o r k m ust be p art o f h o w w e bring a new

society into being. But w e m ust also try to bring abo u t a shift in

public values so that instead o f inspiring adm iratio n and en vy, co n ­ spicuous consum ption is seen as p art o f the prob lem , a sign o f greed and u nfairness w hich dam ages society and the planet.

M artin Luther K in g said, ‘T h e m o ral arc o f the universe is long,

but it bends to w a rd s ju stice.’ G iven that in hum an prehistory w e

lived in re m ark a b ly equal societies, m ain tain in g a steady state - or sustainable - w a y o f life in w h a t som e have called ‘the o rigin al affluent so ciety’ ,324 it is perhaps right to think o f it as an a rc, cu rvin g back to ve ry basic hum an principles o f fairness and eq u ality w h ich

w e still regard as g o o d m anners in a n y n orm al social in teraction .349

But at all stages, creating a m ore equal society in volves people speakin g their m inds, m ak in g the case, organ izin g and cam paigning.

It is im possible fo r governm ents not to influence incom e d iffer­

ences. N o t only are they the largest em ployer in m ost countries, but alm ost every area o f econom ic and social p o licy affects incom e distribution. T a x an d benefit policies are the m ost o b viou s w a y . O ther influential areas o f p o licy include m inim um w age legislation, education policies, the m anagem ent o f the n atio n al econom y, w h eth er unem ploym ent is kept to lo w levels, w hether different rates

o f V A T and sales ta xe s are applied to necessities and lu xu ries, p ro ­ visio n o f public services, pension policies, inheritance ta xes, negative incom e ta x , basic incom e policies, child support, p rogressive co n ­ sum ption ta x e s ,351 industrial p o licy, retraining schem es, and m any m ore. But in this ch apter w e have also suggested m ore fundam ental

changes to ensure that incom e differences are subject to dem ocratic con trol and greater eq u ality becom es m ore deeply rooted in the so cial fabric.

A t this stage, creating the po litical w ill to m ake society more equal is m ore im portan t than pinning ou r co lo u rs to a p articu lar set

o f policies to reduce inequality. Political w ill is dependent on the developm ent o f a vision o f a better society w h ich is both achievable and inspiring. W e hope w e have show n that there is a better society to be w o n : a m ore equal society in w hich people are less divided by status and h ierarch y; a society in w hich w e regain a sense o f com ­ m unity, in w hich w e overcom e the threat o f g lo b al w a rm in g, in w h ich w e o w n and con trol ou r w o rk d em o cratically as p art o f a

com m un ity o f colleagu es, and share in the benefits o f a gro w in g n o n ­ m onetized sector o f the econom y. N o r is this a u topian dream : the evidence show s that even sm all decreases in in equality, alread y a reality in some rich m arket dem ocracies, m ake a very im p ortant difference to the q u ality o f life. T h e task is n o w to develop a politics based on a recognition o f the kind o f society w e need to create and com m itted to m ak in g use o f the institutional and technological opportunities to realize it.

A better society w ill not happen au to m atically, regardless o f w hether or not w e w o rk fo r it. W e can fail to prevent catastrophic glo b al w a rm in g, w e can a llo w ou r societies to becom e increasingly

an ti-social and fail to understand the processes in volved. W e can fail to stand up to the tin y m in ority o f the rich w h o se m isplaced idea

o f self-interest m akes them feel threatened by a m ore dem ocratic an d egalitarian w o rld . T h ere w ill be problem s and disagreem ents on the w a y - as there a lw a y s have been in the struggle fo r progress - but, w ith a b ro ad conception o f w h ere w e are go in g, the necessary

changes can be m ade. A fte r several decades in w h ich w e have lived w ith the oppressive

sense that there is no alternative to the social and environm ental failure o f m odern societies, w e can n o w regain the sense o f optim ism w h ich com es from k n o w in g that the problem s can be solved. W e

k n o w that greater eq u ality w ill help us rein in consum erism and ease the in troduction o f policies to tackle g lo b al w a rm in g. W e can see h o w the developm ent o f m odern techn ology m akes profit-m akin g institutions ap pear increasingly an ti-social as they find themselves threatened by the ra p id ly exp an d in g potential fo r public go o d w hich n ew techn ology o ffers. W e are on the verge o f creating a q u alitatively better and m ore tru ly sociable society fo r all.

T o sustain the necessary po litical w ill, w e m ust rem em ber that it falls to ou r generation to m ake one o f the biggest tran sfo rm atio n s in hum an history. W e have seen that the rich countries h ave got to the end o f the really im portan t contributions w h ich econom ic gro w th can m ake to the q u ality o f life and also that ou r future lies in im p rovin g the qu ality o f the social environm ent in ou r societies. The

role o f this b o ok is to poin t out that greater equality is the m aterial fo un d atio n on w h ich better social relations are built.

Postscript - Research Meets Politics

T H E S P I R I T L E V E L D E B A T E

T h is b o o k w as first published in M a r c h 2 0 0 9 , abo u t six m onths after the start o f the w o rst financial crisis since the Second W o rld W ar. M u ch o f the blam e fo r the crisis w a s rightly attributed to the ex tra o rd in a ry risks taken by people in the financial sector w hose

excesses w ere m atched only by their grotesquely high salaries.

T h o u g h our research predates the crisis b y m any years, and its v a lid ­ ity is unaffected by the crisis, the b o o k ’ s gen erally positive reception clearly o w es som ething to its tim ing. M a n y people w h o , before the crash , had assum ed that huge salaries and bonuses reflected the

unique contributions and brilliance o f their recipients, changed their

m inds as they learned abou t the lack o f relationship betw een per­ form ance and re w a rd s.408

A N I D E A W H O S E T I M E H A S C O M E ?

But the reception o f the b o ok cannot be w h o lly attributed to the m om ent at w hich it ap peared. Since pu b licatio n , w e have - between us - given w ell o ver 3 5 0 lectures, in m an y different countries. W e have spoken to civil servants, health auth orities, academ ics, ch ari­ ties, faith grou p s, think tan ks, p rofession al association s, arts and lit­ erary festivals, trade unions, senior business people, com m unity g ro u p s, ro y al societies, international agencies and p olitical parties across the ideological spectrum . A lth ou gh w e have often been

invited to speak to grou p s w h ich w ere pre-disposed to be sy m p a­ thetic to the idea o f greater eq u ality, that has not a lw ay s been the

case. Y e t audiences have been so u n iform ly positive and appreciative that w e have sensed there is an intellectual vacu u m , a hunger fo r the

evidence w e present - as if under the surface the w o rld w a s full o f closet egalitarian s.

T hree things seem lik ely to have contributed to this. First is a desire fo r an exp lan atio n o f w h y , am idst such unprecedented afflu ­ ence, ou r societies are beset by such a w o rry in g a rra y o f so cial p ro b ­ lems. W h y are rates o f depression and a n xiety so high? W h y is there

such w idesp read reliance on drugs and alcohol? A n d w h y is violence

so com m on? The second com ponent is the evidence w ith w h ich w e started the b o ok - that a very large m ajo rity o f the p o pu latio n feels that ‘con sum erism ’ o r ‘m aterialism ’ is som ething w e get caugh t up in despite feeling it runs counter to our values and our desire fo r m ore

time w ith fam ily and friends o r in ou r com m unity. T h ird , our a n aly ­

sis seems to confirm peop le’ s intuition that in equality is divisive and so cially co rro sive. A g a in and again , people tell us they feel they have gained fro m the b o o k a picture o f the w o rld w h ich is both quite new to them and yet so m eh o w also im m ediately recognizable, a picture they feel they have been w aitin g fo r and w h ich changes h ow they see

w h a t is goin g on aroun d them. A recent report confirm s em pirically the im pression w e have

received so stron gly - that the general public is averse to the high lev­ els o f in equality in very unequal countries. In a ran dom sam ple o f over 5 ,5 0 0 A m erican s, researchers from D uke U n iversity and

H a rv a rd U n iversity investigated view s o f the distribution o f w ealth (rather than income) in society.409 People w ere sh o w n three pie charts illustratin g three different distributions o f w ealth - one in w h ich each fifth o f the pop u latio n go t the sam e, another w hich show ed (unlabelled) the distribution o f w ealth in the U S A and

another (also unlabelled) based on the distribution in Sweden.

N in e ty -tw o per cent said they w o u ld prefer to live in a society w ith the Sw edish distribution - and the percentage o n ly varied from 89 to 93 per cent depending on w h eth er they w ere rich o r po o r, D em ocrats or R epu blican s. W hen asked w h a t they thought the dis­

tribution o f w ealth is in the U S A , the average estim ate w a s that the

richest 20 per cent o f A m erican s con trol 59 per cent o f the w ealth . In

reality, they control 84 per cent. A sked w h a t they thought the ideal distribution w o u ld be, people preferred the top 20 per cent to have 3 2 per cent o f all w ealth.

N everth eless, as w ell as its v e ry positive reception, the b o ok has

attracted both thoughtful criticism and strident political attacks. The

m ain purpose o f this chapter is to respond to these before going on to discuss som e new research findings. But before doin g either, w e w o u ld like to address a criticism m ade by several com m entators w h ich seems to be based on a m isunderstanding.

W H O B E N E F I T S F R O M G R E A T E R E Q U A L I T Y ?

Som e review ers o f the b o ok w ere not convinced that w e had show n

that the vast m ajo rity o f the po pu latio n benefited fro m greater

eq u ality. T h ey seemed to think that the evidence did no m ore than establish that avera g e perform an ce across the w h ole pop u latio n is w o rse in m ore unequal societies.410

In a section running from p. 1 7 5 to p . 1 8 2 w e show no few er than

five sets o f data (and refer to another show n on p. 10 9 ) illustrating

that, w hether you classify people by education, social class or income, people in each category are healthier (or have higher literacy scores) if they are in a m ore equal society than people in the sam e category o f incom e, education or class in a less equal society. We also refer to studies w hich reach the same conclusions using statistical models which enable researchers to lo o k at the effects o f inequality after con­

trolling fo r the effects o f all individual incom es throughout society. W e do n o t argue that everyone in a m ore equal society does better

than everyone in a less equal one. W e are n ot sayin g that even the low est social class o r the least w e ll paid or educated catego ry in a m ore equal society does better than the highest catego ry in a less equal society. R a th e r, w e sh o w that w hen people in the sam e social class, at the sam e level o f incom e or education, are com p ared across

countries, those in m ore equal societies do better. So at an y given level o f person al incom e o r ed ucation, som eon e’s q u ality o f life w ill be higher if he or she has the sam e level o f incom e or education but

lives in a m ore equal society. T h a t is w h a t is show n in Figures 8.4,

1 3 . 2 , 1 3 . 3 , 1 3 . 4 and 1 3 . 5 . T h e con clusion is that greater equality

u sually m akes m ost difference to the least w ell o ff, but still produces some benefits fo r the w ell off.

A s w e pointed out on p .1 7 6 , the very large differences between m ore and less equal societies in the prevalence o f other social problem s - including m ental health, teenage births, trust, hom icide

and im prisonm ent - suggest that this picture is not confined to the areas o f health and literacy. Th e differences are u su ally m uch larger than w o u ld result if greater equality benefited on ly the least w ell o ff.

I N E Q U A L I T Y , C L A S S A N D S T A T U S

A cadem ic sociologists have som etimes been surprised that the b o ok focuses so exclu sively on incom e in equality and pays little attention to the vast am ount o f carefu l w o r k n o w availab le on social class

classification s.411 W e have great regard fo r m uch o f this w o rk , but it

does not feature here because social class classifications have tw o w eakn esses fo r the kind o f analyses this b o ok undertakes. First, because alm ost every cou n try uses a different socioeconom ic classifi­ cation system , it is difficult to m ake com p arison s between countries. Fo r exam p le, in early studies o f h o w health differences across the

social h ierarch y in Sw eden com pared w ith those in E n glan d and

W ales, Sw edish researchers had to re-classify the occupation s o f m any thousands o f Sw edes accordin g to the B ritish o ccu pation al class classification. W e show ed the results in Figures 1 3 . 3 and 1 3 .4 . But even if social com parison s could be m ade consistently across

m any different countries, there is a second m ore fundam ental p ro b ­ lem: few , if an y, social class classification s w o u ld a llo w an assess­ ment o f w h eth er the gaps betw een grou ps are bigger or sm aller in one co u n try than another. Incom e differences, on the other hand, allo w us not sim ply to categorize people into different classes, they also a llo w us to m easure the size o f the differences w ithin the

p o pu latio n . F o r all its im perfections as a m easure o f status differen ti­ ation, incom e in equality tells us a lot abo u t a society.

A n sw ers to a num ber o f other points w hich are often raised when w e give presentations on the b o ok (including questions on ethnicity, im m igration, the size o f countries, local inequalities and m an y more) are listed under Frequ en tly A sked Q uestions on T h e E q u ality Tru st w eb site at w w w .e q u a lity tru st.o rg .u k .

F A I R C R I T I C I S M A N D F O U L A T T A C K S

Public health, w ith epidem iology at its centre, has a long record o f political battle, from the nineteenth-century conflicts over the p ro v i­ sion o f sewers and clean w ater, to m odern legislation protecting peo­

ple from dangerous exposu res at w o rk or in the environm ent more w idely. The political battles com e when the scientific evidence runs up against vested interests o f m any different kinds - industrial, social and econom ic.

It is n o w alm ost universally accepted am ongst scholars and practi­

tioners o f public health that the most im portant determ inants o f health are social and econom ic circum stances. G e o ffrey R o se, w h o w a s one o f the m ost highly influential and respected epidem iologists o f the second h alf o f the twentieth century, said, ‘medicine and p o li­ tics cannot and should not be kept a p art’ . O ur gro w in g understand­

ing o f h o w hum an health and w ellbeing are so deeply affected by social structure inevitably pushes science into politics.

A cadem ics in every field o f course criticize each other’ s w o rk all the time: that is part o f the norm al process o f scientific progress. A ttacks clearly m ade fo r ideological reasons are quite different. R ath er than controversies abou t research methods or interpretations o f the evi­

dence being ham m ered out between colleagues w h o k n o w the subject area, suddenly people w h o do not k n o w the extensive research litera­ ture and have never made a contribution to it, use the m edia to try to convince the public that research findings are m isleading rubbish.

A ttem pts to o verth ro w large bodies o f scientific evidence that seem to have im portant political im plications are n o w a well-established phenom enon. T w o A m erican academ ics, N a o m i O reskes and E rik

C o n w a y , have recently described these tactics in their b o ok , M erch an ts o f D o u b t: H o w a H a n d fu l o f Scientists O b s c u re d the Truth on Issues fro m T o b a c co S m o k e to G lo b a l W arm ing.412 T h ey describe the techniques used - often by the sam e people operating in a

num ber o f quite different subjects - to give the im pression that crucial

areas o f science affecting public po licy are con troversial, long after the im plications o f the science w ere quite clear. A s a result, there have sometimes been substantial delays in the public response to pesticides, tobacco m arketing, acid rain, the hole in the ozone layer, exposu re to secondary sm oke, and o f course glo b al w arm in g. It is characteristic o f

the tactics o f these ‘m erchants o f dou b t’ , that one o f the attacks on

our b o ok w a s w ritten by som eone w h o had recently w ritten a d ia­ tribe against tobacco control and w h a t is n o w the w ell-established evidence that secondary sm oke is h arm ful to health. Fortunately the bans on sm oking in public places (implemented in Scotland, parts o f

the U S A and C an a d a , R o m e, Ireland, and England); w hich in each

case have been follo w ed by declines in death rates and have saved thousands o f lives, seem unlikely to be lifted in the light o f his w o rk .

W H Y T H E A T T A C K S F A I L

O n the w eb site o f T h e E q u ality T ru st w e have p rovid ed a point-by-

point rebuttal o f all the criticism s m ade by T h e T a x P ay e rs’ A llian ce, The D em o cracy Institute and the P o licy E xc h a n g e .413 A s m uch o f it is rather tedious, in this chapter w e w ill describe on ly the b ro ad o u t­ lines o f the criticism and ou r responses.

M u ch the most im p ortant strategy o f ou r critics has been to treat the relationships w e sh o w between inequality and social problem s as

if w e w ere the only people ever to suggest such links. T h ey then set abou t trying to cast doubt, one by one, on the relationships w e show in ou r graph s, suggesting that they are a com bination o f statistical flukes, the result o f a cunning selection o f countries, or o f choosing

problem s to suit ou r argum ent. F o r exam ple, they argue that the U SA

should be excluded from one an alysis, Ja p a n or the Scandinavian countries from another, or that another relationship disappears if you add in po o rer countries, and so on.

R egardless o f their merits in relation to our w o rk , this strategy

means that these criticism s are largely piecem eal, a d h o c, and irrele­

van t to the m any other dem onstrations o f sim ilar relationships in dif­ ferent settings published in academ ic journ als by other researchers. Because there are aroun d 20 0 papers in peer-reviewed academ ic jo u r­ nals testing the relationship between incom e inequality and health in m any different settings,10 m ore than 50 papers on violence and

in equality,210’ 211"414 and quite a num ber on inequality in relation to trust and social cap ital,400’ 415 it is n o w extrem ely difficult to argue credibly that these relationships d o n ’t exist. Indeed, those w h o do so are alm ost alw ay s those w h o are m akin g political attacks rather than an y kind o f academ ic criticism .

A cadem ic discussion am ong those w h o k n o w the literature in the field is n o w very largely confined to h o w the relationship s should be interpreted. T h at is w h y (chapter 1 3 , from p . 1 8 2 onw ards) w e dis­ cussed the strengths and w eakn esses o f possible alternative interpre­ tatio n s, before concluding that these relationships m ust reflect the

dam age done by inequality. W ith few exception s, w e have p revio u sly subjected to peer review and published in academ ic jo u rn als alm ost everything w e have show n in this book.

So w h a t are the criticism s and h o w do w e respond to them?

P I C K I N G A N D C H O O S I N G ?

S o m e critics h a ve su g gested that w e are selective in the ch o ic e o f health a n d so cia l p ro b le m s that w e ex a m in e .*16

The Spirit L evel does not claim to explain every kind of social problem: it is specifically a theory of problems that have social gradients, gradients which make them more common further down the social ladder. So, for example, we would not expect alcohol use to increase with inequality because, in most countries, alcohol use does not increase lower down the social ladder. In contrast, alcohol abuse (for example binge drinking and alcoholism) does have a social gradient, and deaths from alcoholic liver disease are more common in more unequal US states.8 We have also shown that death rates such as breast and prostate cancer, which do not tend to become more common lower down the social ladder, are not

related to inequality.8 This contrasts sharply with deaths from causes such as heart disease which do have a strong social gradient.

The reason we included (in chapter z) an analysis of the relationship between the U N ICEF Index of Child Wellbeing in Rich Countries and

income inequality was to show that our results were not a result of select­ ing problems to suit our argument. The U N ICEF Index combines 40 dif­ ferent aspects of child wellbeing which we played no part in selecting, yet it behaves exactly like our own Index o f Health and Social Problems - showing strong relationships with income inequality and none with aver­

age national income.

W H I C H C O U N T R I E S ?

C ritics h a v e also su g gested that w e selected coun tries a rb itra rily to

suit o u r arg u m en t a n d s h o u ld h a v e in c lu d e d m o re, a n d p o o re r,

co u n tries.416_7

The countries in our analyses result from the application of a strict set of criteria, applied with no departures or exceptions. Our source was the World Development Indicators Database, World Bank, April Z004. We took the richest 50 countries for which the bank publishes figures on

Gross National Income per capita, ranked according to the ‘Atlas method’ used by the World Bank to classify countries into Low , Medium and High Income categories. From that list we excluded countries with no interna­ tionally comparable data on income inequality and those with populations

with fewer than 3 million (to avoid tax havens). That gave us our final

dataset of Z5 rich countries. We looked exclusively at the richest countries not because these relationships only exist among them, but because these countries are on the flat part of the curve at the top right in Figure 1 . 1 on p.7, where life expectancy is no longer related to differences in Gross National Income per head (GNIpc) and it is therefore easier to distinguish

the effects o f relative and absolute levels of income. If we had also studied poorer countries two problems would have

arisen. First, comparable data on teenage births, mental illness, social mobility, social cohesion etc., are very rarely available for much poorer countries. Second, if we had included countries in which many people con­ tinue to have inadequate material resources so that increasing GNIpc is

still important, we would have had to control statistically for the log trans­ formation o f GNIpc in order to show the effect of inequality. In a book which we hope will be widely understood, this would have been a substan­

tial increase in complexity for little gain. Nevertheless, if we had included poorer countries it would have made little difference to our results. Studies of life expectancy, infant mortality and homicide - for which data is usu­ ally available for poorer countries - show that greater equality is beneficial at all levels of economic development.10’ 418

To have ‘cherry-picked’ countries - as our critics suggest we did - according to whether their data for this or that social problem did or did

not fit our thesis, would have made the book a pointless exercise. Instead we had an absolute rule that we used the most reputable data sources and took the data for as many of our 25 countries as were available - warts and all. For example, we include Singapore in our analysis o f infant mor­ tality although it is a very significant outlier: it claims the lowest infant

mortality in the world despite being the most unequal country in our dataset (see Fig 6.4 on p .82).

If we had shown graphs of data collected by other researchers we would often have been able to show even stronger and more dramatic associa­ tions with inequality than already appear in the book.207-419 But had we

done so we would necessarily have been referring to different groups of countries in studies which use different measures of income inequality and that would properly have raised questions of comparability. On the con­ trary, we wanted to show that there is a consistent pattern running through the data for quite different problems. Our purpose was to analyse every health and social problem using the same inequality measures with

reference to the same set of countries, and then, to make quite sure, to double-check our findings by repeating the analyses on the 50 states o f the USA.

The book therefore tries to show the relationships between income inequality and various health and social problems as simply and transpar­ ently as possible. The scatter graphs can be understood without any knowledge o f mathematics or statistics. We point out in each chapter that our findings cannot be attributed to chance. M ost readers should not feel any sense that there are mysterious areas where they cannot tell quite what is going on but, for those who want it, we provide more data and statisti­ cal detail at www.equalitytrust.org.uk

C U L T U R A L D I F F E R E N C E S ?

It has been su g gested that the relatio n sh ip s w e s h o w reflect d if fe r ­ ences in n a tio n a l cultu re rather than the effects o f in e q u a lity .416

This criticism has been made in two forms. One is the suggestion that

income differences are an expression of underlying cultural differences, and it is these which are the real determinants of the health and social problems we examine. The other is that particular countries should be excluded from one or other analysis because they are culturally different

from the others. Rather than varying from better to worse on a single scale as income

inequality does, national cultures differ in an infinite number of quantita­ tive and qualitative ways. In chapter 1 3 we mention the huge cultural dif­ ferences between Sweden and Japan which, despite both performing well in our analyses, are poles apart in other ways - including the extent of women’s participation in the labour force and in politics, the dominance of the nuclear family, and whether their narrower income differences result from redistribution or narrower differences in earnings before taxes

and benefits. In contrast, Spain and Portugal have many cultural similarities and were

both dictatorships until the mid 19 70 s. Yet as we have seen throughout the book, Portugal is now much more unequal than Spain and consistently suffers more from most of the health and social problems we discuss. So it looks as if cultural differences (Sweden and Japan) don’t necessarily make societies perform differently in our analyses, and cultural similarities (Spain and Portugal) do not necessarily make them perform similarly. What matters is the scale of income differences almost regardless of other

aspects of culture. In addition, we know that in the second half of the twentieth century

the USA and Japan came close to swapping positions in the international league tables of life expectancy and income inequality. In the 19 5 0 s the USA was more equal than Japan and had better health. But as the USA became more unequal, Japan became less unequal, and Japanese life expectancy outstripped that in the USA to become the best in the world. If what matters is culture rather than inequality, how is it that the plethora

of cultural changes which took place in these two nations did nothing to

alter the relation between each countries’ burden of health and social

problems and its income inequality? The more you think about the suggestion that the relationships we show

are merely a mirage created by some underlying dimension o f cultural dif­ ference, the less plausible it becomes. This unknown dimension o f culture would not only have to cause physical and mental ill-health, school bully­

ing, more punitive sentencing, obesity, teenage births and so on, but it would have to do so in proportion to the scale of income inequality.

The second use to which critics put the notion of cultural differences is to justify removing groups of countries at either the more or the less equal end of our analyses on the grounds that they are ‘ culturally different’ from

the others. For example, arguments have been made for removing the

Scandinavian or English-speaking countries, or sometimes even both groups.416 Such wholesale deletions would remove crucial information which looks as if it might explain why some English-speaking countries do better on some social measures than others - for example why the USA

has more mental illness and teenage pregnancies than the UK, which in turn has more of both than N ew Zealand (and of course all the US states are English speaking). More fundamentally, however, national cultures are themselves powerfully determined by inequality because of its corrosive effects on trust, cohesion and community life.

O U T L I E R S , D A M N O U T L I E R S A N D S T A T I S T I C S

S o m e critics h a v e su g gested that the relatio n sh ip s w e s h o w are d e p e n d e n t on ‘o u tlie rs ’ .*16

As well as excluding countries on the grounds of cultural differences, it has also been suggested that some countries should be removed for the purely statistical reason that they are ‘ outliers’ . In any of the graphs in our book, an outlier is a country or US state that is a long w ay from the line of best

fit between the countries or states. Good examples are Singapore in Figure 6.4 on p .82 (showing the relationship between income inequality and infant mortality) or Italy in Figure 5 .1 on p.67 (showing levels of mental

illness). There are established methods to calculate the influence a single data point has on the line of best fit on a graph, but no hard and fast sta­ tistical rules as to when outliers should be left out. We decided to keep all countries and states in our analyses for three reasons: first, they represent

real variation in population levels of health and social problems; second, because removing occasional data points would have invited the accusa­ tion that we were picking and choosing data points; and lastly because, as we said earlier, we wanted to show the consistency of the effects of inequality on different problems across the same group of countries.

Having mistakenly accused us of picking and choosing countries for

each analysis, our critics have also attempted to discredit some of the rela­

tionships we show by selectively removing countries on the grounds that they are outliers. For example, one critic said that the Scandinavian coun­ tries were outliers in our analysis of inequality and foreign aid, but did not suggest removing Japan which, on that graph (Fig 4.6, p .6 1), is actually

more of an outlier.416 Removing the Scandinavian countries would mean

the relationship was no longer statistically significant, but removing Japan as well would restore its significance.

The same thing happens in the case of obesity. A critic suggested that we should remove the USA from the analysis on the grounds that it is an outlier

and doing so reduces the relationship to non-significance.416 But Greece is a

more distant outlier: remove Greece as well and the significant link between inequality and obesity is restored. For overweight children, the same critic removes the USA, but not Canada, which is again a more distant outlier. Remove both and once again the relationship is statistically significant.

Our results are, of course, sometimes sensitive to exclusions simply

because we are looking at a limited number of countries, but the fact that

so many relationships with inequality are statistically significant, despite the limits to the data, is an indication of how powerful the underlying rela­ tionships actually are.

These criticisms of the evidence that inequality affects wellbeing fail on

two grounds: they do nothing to dent either the hundreds of other pub­

lished analyses of health and income inequality, nor our analyses o f the 50 American states; and ultimately the suggested exclusions do not remove the associations we show. In Figure 1 3 . 1 (p. 174 ) we summarize our inter­ national evidence by combining all the health and social problems into one

index. Sweden, Norw ay, Finland, Japan, USA and UK - all the most equal

and unequal societies - can be removed all at once, and still there is a

highly significant relationship between social health and inequality among

the remaining countries.

In short, we believe that our data are very robust - and, as we said ear­ lier, some of these relationships have been demonstrated many, many times before by other researchers in quite different settings. Research reports have shown for instance that income inequality is related to health

in the regions of Russia,420 the provinces of China421 or Japan ,422 the

counties of Chile,423 or among rich and poor countries combined.418 Those who dislike our conclusions would no doubt want to remove vari­ ous provinces, counties or countries from these reports too.

O T H E R F A C T O R S ?

It is som etim es su g gested that w e s h o u ld c o n tro l fo r , o r take a cco u n t o f, o th er facto rs that m ig h t e x p la in the asso ciatio n s b e tw ee n in co m e

in e q u a lity a n d health o r s o cia l p ro b le m s - such as N a tio n a l In c o m e p e r h ead, p o v e rty , eth n icity, o r w e lfa re se rvices.416

There were several reasons why we chose not to include other factors. First, we wished to present the simplest and most understandable pic­ ture of the correlation between income inequality and health and social problems, so that readers could see the strength of the relationship for

themselves. Second, it is a fundamental methodological principle of epidemi­ ological analysis that you should not control for factors which form part of the causal chain - in this case explaining h o w inequality causes a particu­ lar problem. For example, if we think - as we do - that societies with greater income inequality have worse health because poorer social rela­ tions increase chronic stress, then we would need to be cautious about how to analyse that particular causal sequence. Simply including meas­ ures of trust and social cohesion in a statistical model could remove the association between income inequality and health,400 even though it is likely that inequality actually leads to poorer health because it is socially divisive. Third, including factors that are unrelated to inequality, or to any particular problem, would simply create unnecessary ‘noise’ and be methodologically incorrect.

Nevertheless, many other studies of health and income inequality have controlled for poverty, average income, or each person’ s individual income. Still others have made careful explorations of the interplay between income inequality and public spending, social capital and the ethnic composition of populations. We discussed these in chapter 13 and will return to some of

them in the section below on recent advances in research.

A point w hich can h ard ly be m ade too often is that the relation ­ ships between inequality and vario u s health and social problem s are not reducible to the direct effects o f peop le’s m aterial living standards independent o f inequality. N o one doubts that health is com prom ised

when a substantial p roportion o f the popu lation o f poorer countries

lack basic necessities. But w hen it com es to explain in g the tendency fo r health to im prove all the w a y up the social hierarchy even in rich countries (see Figure 1.4 ) the causes are less clear. The concept o f ‘neom aterialism ’ w a s invented as a counter to psych osocial ex p la n a ­

tions o f this pattern. The idea w a s that health m ay continue to benefit

from higher levels o f com fort and lu xu ry all the w a y up the social

scale. Even if that w ere true, it w o u ld not exp lain w h y m ore equal societies are healthier. But Figure 1 . 3 show s life expectancy is no longer sensitive to living standards am ong the rich countries, Figure z.3 show s that this is also true o f our Index o f Health and Social Problem s,

and Figure 2 .7 show s the same is true o f the U N IC E F index o f child

w ellbeing. C hapter 6 references a review o f studies w hich have con ­ trolled not only fo r po verty but also fo r the effect o f each person ’s income on their health before going on to test fo r an effect o f in equal­ ity. A n im portant new review o f such studies is also mentioned

b elow . A lth ough it is easy to slip back into the conventional vie w that

m aterial standards are o f prim ary im portance and m ust som ehow explain all that w e attribute to inequality, w e must bear in m ind that this runs counter to a substantial body o f evidence.

T H E P R O O F O F T H E P U D D I N G

A n im portant strength o f the evidence w e present in this b o ok is its rem arkable consistency, w hich provides a coherent picture at tw o lev­ els. It is not sim ply that alm ost all health and social problem s w hich are related to social status sh o w the sam e tendency to get w orse when

Figures 1 7 . i and 1 7 . z The relationship between income inequality and mental illness in 8 rich countries (Fig i j . i ) was confirmed when new data become available fo r 4 additional countries (Fig 17 .2 ) .

there are bigger incom e differences: it is also that the data are consis­ tent w ith in each o f the health and social problem s w e look at. F o r exam ple, w h en w e found that m ore unequal societies have higher levels o f obesity (chapter 7) w e thought calorie consum ption levels should also be higher per person in those societies; and that indeed is

w h at the evidence sh o w ed .114 Sim ilarly, when w e found that educa­ tional scores are lo w er in m ore unequal societies, w e also found that m ore you ng people in those societies are dropping out o f school, are unem ployed and not engaged in further train in g.424

In chapter 1 1 w e said that the m ost im portan t reason fo r the m uch

higher im prisonm ent rates in m ore unequal societies is m ore punitive sentencing. Since pu b lication , w e h ave investigated w hether this also applies to children. W e h ave fo u n d that the age o f crim inal resp on si­ bility tends to be lo w er in m ore unequal societies, so children are m ore lik ely to be tried as adults.

The best test o f the v a lid ity o f a scientific th eory, indeed a th eory

o f an y k in d, is w hether or not it can m ake predictions o f things not already k n o w n that can later be verified. W e have alread y discussed (p. 1 9 5 ) exam ples o f successful predictions w h ich have com e out o f our th eory that greater inequality increases problem s w ith social g r a ­

dients. W e can d r a w attention to tw o m ore, both o f w h ich concern

relationships w here w e in itially had very little data. In 2.006, when w e first lo o ked at the relationship betw een incom e inequality and levels o f m ental illness, in tern ation ally co m p arab le estim ates o f m en­ tal illness fro m the W o rld H ealth O rgan izatio n w ere a v ailab le fo r

on ly eight rich, developed coun tries.425 W hen w e published a prelim ­

in ary an alysis in an academ ic jo u rn al, critics suggested ou r findings

w ere dependent on the high levels o f in equality and mental illness in the U SA . But by the time the first edition o f this b o o k w a s published, com p arab le data on m ental illness w ere a v ailab le fo r fo u r m ore countries. R a th e r than con tradictin g the tendency w e had suggested

fo r m ental health to be w o rse in m ore unequal countries, the new data filled in the gaps and confirm ed that relationship - see Figures 1 7 . 1 an d 1 7 . 2 . T h e add itio n al countries had rates o f m ental illness close to those w h ich their levels o f in equality w o u ld have predicted.

The same thing has happened in relation to measures o f social m obil­ ity. When T h e Spirit L e v e l w as first published, w e had com parable

Income Inequality

Income Inequality

Figures 17 .3 and 17 .4 The relationship between income inequality and social mobility in 8 rich countries (Fig 17 .2 ) was confirmed when new data became available fo r 3 additional countries (Fig 17 .4 ).

measures o f social m obility for only eight countries and some critics claimed that the relationship w e show ed between inequality and low levels o f social m obility w as spurious because w e had too few countries to justify an analysis. W e included it, despite initially having data from only eight countries, not only because it w as a statistically significant relationship, but also because research reports show that social mobility slowed while income differences widened, and it seemed plausible that w ider income differences w ould tend to reduce m obility. Since this book w as published, new data on social m obility has become available

which include a further three countries.426 A s Figures 1 7 .3 and 17 .4

show , the data for the additional countries provide a fuller and more robust estimate o f the impact o f inequality on social m obility, and con­ firm our original conclusions.

N E W E V I D E N C E

Death rates versus self-reported health

Since w e finished w ritin g T h e S pirit L e v e l in the spring o f 20 0 8 , there

have been m any m ore studies reporting relationships between inequal­ ity and health. N ine o f the new studies look specifically at rich, devel­ oped countries.427̂ 35 Seven find, as w e do, that health is w orse in more unequal societies. T h e tw o that differ both look at income inequality in relation to self-rep o rted health - where, instead o f death rates,

people are typically asked to say w hether their health has recently been

excellent, good , fair or p o o r.432-3 A lthough self-reported health is predictive o f longevity w ithin a country, com parisons between coun ­ tries find that self-rated health is actually better in countries where life expectancy is low er: so rather than contradicting the relation between

greater inequality and w orse health, the studies o f self-reported health

tell us something about the w a y people p erceive their health.436-7 But w h y is self-rated health not related to levels o f actual health in

cross-coun try com parison s? C o u ld it be that in m ore unequal so ci­ eties, w ith m ore status com petition, asserting that one has excellent or v e ry go o d health m ight be p art o f m aintaining a h ard ier self­

im age? O r perhaps peop le in m ore equal societies are less inclined to

rate themselves at the top o f a scale? W e can on ly guess. Such ques­ tions sh o w the im portance o f using objective m easures o f health and

so cial problem s in international com parison s.

Causal pathways

T h ree other im p ortan t n ew pieces o f evidence have appeared recently, w hich tell us m ore abou t h o w the relation between incom e in equality and health w o rk s. One is a review o f w h a t are called

m ulti-level studies published in the B ritish M e d ic a l J o u r n a l,438 M u lti-level studies lo o k at incom e and health in tw o stages: first in terms o f the relation betw een each person ’s in dividual incom e and health, an d second to see if there is an add ition al effect o f inequality across each society as a w h ole. T h is review com bines data from 26

m ulti-level studies w ith in dividual data fo r o ver 60 m illion people. It show s u n eq u ivocally that in equality has a dam agin g effect on health w h ich cannot be attributed to h o w rich or p o o r people are in absolute term s, and concludes that even after con tro llin g fo r in d ivid ­ ual incom es (including poverty) or education, reducing in equality in the O E C D countries alone w o u ld prevent u p w ard s o f 1 . 5 m illion

deaths per year (alm ost 1 in 1 0 deaths am ong adults aged 1 5 - 6 0 ) . T h is is likely to be a con servative estim ate because con trollin g for in d ividu al incom e w ill also con trol fo r in dividual social status w hich affects health an d is a key com ponent o f inequality.

Th e second n ew piece o f evidence, published in S o c ia l S cience a n d

M e d ic in e , adds to the sm all num ber o f studies lo o k in g at changes in in equality and health o ver time. A ll over the w o rld , life expectancy continues to increase, but this study sho w ed that those US states that had the biggest increases in incom e in equality between 1 9 7 0 and 2.000 had sm aller im provem ents in life expectan cy than other

states.439 T h e third piece o f evidence, published in T h e A m eric a n Jo u r n a l o f

P u b lic H ealth , is a study that tests h o w far the link betw een income in equality and health results from different levels o f trust or from different levels o f governm ent spending on health services.415 This study confirm ed w h a t w e suggested in chapter 6: that levels o f trust are indeed part o f the exp lan atio n , w hereas spending on health care

is not. W ider incom e differences seem to erode trust, w h ich in turn

seems to com prom ise healthy ageing.

Friendship and health

T h ro u gh o u t T h e S p irit L e v e l, w e discuss the vital im portance o f

social relationship s to hum an health and w ellbein g and sh o w that higher levels o f incom e inequality dam age the social fab ric w h ich contributes so m uch to healthy societies. N o w , a m ajo r n ew review o f the evidence fro m alm ost 1 5 0 studies confirm s the im portan t influence o f social relationship s on h ealth .440 People w ith stronger

social relationships w ere h a lf as likely to die during a stu dy’ s period

o f fo llo w -u p as those w ith w eak er social ties. Th e authors o f the rep o rt foun d that the influence o f social relations on su rvival w a s at least as im portan t as that o f sm okin g, and m uch m ore im p ortant than h eavy drin kin g, ph ysical activity or obesity. Th e effects were strongest w hen researchers com bined m easures o f different kinds o f relation ship s, such as m arital status, feeling lonely, size o f social net­ w o rk , participation in social activities, and so on.

Violence

A s w e described in C h ap ter 1 0 , there is a large and consistent body o f evidence on incom e in equality and violence. M o re recent studies continue to confirm this link. One study o f 3 3 countries, published in 2 0 1 0 , also show ed that social cohesion - as m easured by levels o f trust - seemed to pro vid e the cau sal link betw een incom e in equality and hom icide rates, w h ereas public spending on health and ed u ca­ tion did n o t.415

M artin D a ly at M c M a s te r U n iversity in C an a d a has published an an alysis o f w hether the relation betw een in equality and violence am on g the 50 states o f the U SA could - as som e have suggested - be due to ‘ southern cu ltu re’ or ethnicity, rather than to inequality. H e

show ed that on the co n tra ry , violence w a s related to inequality

am ong the southern and northern states considered sep arately, and that rates o f violence rose alon g w ith in equality am ongst both black and w h ite perpetrators o f violen ce.414

W hen w e w rote T h e S p irit L e v e l, w e k n ew o f no other studies o f

incom e inequality and child conflict, but a recent study o f 3 7 co u n ­

tries finds higher levels o f b ullyin g in m ore unequal countries. S upport from fam ily an d friends w a s associated w ith less bullying, but neither this nor differences in fam ily w ealth trum ped the effect o f incom e in eq uality.402

E Q U A L I T Y A N D S U S T A I N A B I L I T Y

The intertw ined issues o f eq u ality, social justice, su stain ability and econom ic balance are n o w receiving a great deal o f attention all over

the w o rld . E n viron m en tal o rgan ization s such as Friends o f the E arth

and the W o rld W ild life Fund n o w cam paign on in equality issues, including hum an rights and the fa ir distribution o f n atural resources, and the U K ’ s G reen P arty placed econom ic eq u ality at the heart o f its election m anifesto in 2 0 l o .

Evidence continues to accum ulate that m ore equal societies seem

to have lo w er carb o n footp rin ts and are in a better position to cope w ith the challenges o f clim ate change. M o re unequal countries have higher ecological fo otp rin ts, produce m ore w aste, consum e more w a ter and fly m ore air miles per ca p ita .441 T h is m ay be because m ore equal societies seem to foster a greater sense o f collective resp on sib il­ ity w h ich is crucial fo r political action to address clim ate change. Business leaders in m ore equal countries are m ore likely to agree that their governm ents should cooperate w ith international environm en­ tal agreem ents than those in less equal coun tries.442 L ead in g policy specialists n o w suggest that g lo b al inequalities stand in the w a y o f

co o peratio n on clim ate change.443

I N E Q U A L I T Y , T H E M A R K E T A N D D E M O C R A C Y

T h e w eight o f the evidence - our o w n and that o f m any others - and its continued rap id accum ulation, m ake the im portant link between incom e inequality and social dysfunction inescapable. But ill-founded

and politically m otivated criticism can m uddy the w aters and leave people w ith the im p ression that the evidence is less clear than it is. Im agine if som eone w ere to assert (with no justification w hatsoever) that clim ate science had not taken account o f, say, the effects o f v ariatio n s in the salin ity o f different oceans. U nable to evalu ate this

claim , the in exp ert listener m ight assum e that this w a s an im p ortant facto r, an d th at perhaps it had not been p ro p erly considered.

W h at often app ear to be ‘ b alan ced ’ discussions in the m edia can be m isleading. T h is happens even in areas o f science w h ere the accu ­ m ulation o f evidence leaves little legitim ate ro o m fo r doubt. Fo r

exam p le, if 98 per cent o f clim ate change scientists agree on an issue, and 2 per cent disagree, then inviting one person fro m each cam p to take p art in a new s pro gram m e o r public debate can leave people w ith an im pression that an issue is m uch m ore co n tro versial than it is. O n ly those view ers or readers w h o are p a rticu la rly diligent or

h ighly m otivated w ill be able to pursue the issues in detail. R ath er

than considering ou r replies to the po litical attack s on ou r w o rk , w e expect som e w h o are opposed to greater eq u ality w ill sim ply be co n ­ tent to im agine that the issue is ‘ co n tro v e rsia l’ and can n o w be safely ignored.

Perhaps the best tactic in this situation is to address the beliefs that

m otivate the attacks. In M erch an ts o f D o u b t, O reskes and C o n w a y suggest that the defence o f a kind o f free m arket fundam entalism is the m ost plausible ex p lan atio n o f w h y the sam e in dividuals and institutions are often in volved in attacks on research in areas as

diverse as to b acco con trol and the evidence on clim ate change. A s

w ell as defending the free m arket, they see them selves as countering

tendencies to big governm ent and protecting d em o cracy.412 The sam e beliefs are likely to guide the attacks on the evidence o f the so cially dam agin g effects o f inequality.

I f that is the m o tivatio n , then it is based on a serious m isconcep­

tion, one w h ich is alm ost the opposite o f the truth. G reater in eq ual­ ity a ctu ally increases the need fo r big governm ent - fo r m ore police, m ore p rison s, m ore health and social services o f every kind. M o st o f these services are expen sive and o n ly very p a rtia lly effective, but w e shall need them fo r ever if w e continue to have the high levels

o f in equality that create the problem s they are designed to deal

w ith. Several states o f the U S A n o w spend m ore on prisons than on higher education. In fact, one o f the best and m ost hum ane w a y s o f achieving sm all governm ent is by reducing inequality. Sim ilarly, the assum ption that greater eq u ality can on ly be achieved through higher ta xe s and benefits, w h ich presum ab ly led T h e T a x P ay e rs’ A llian ce to publish its criticism o f T h e S p irit L e v e l, is also a m istake.

W e have been at pains to point out (in chapter 1 3 ) that som e soci­ eties achieve greater eq u ality w ith u nusually lo w ta xa tio n because they have sm aller earnings differences before taxes.

There are few things m ore corrosive o f a properly functioning dem ocracy and o f the m arket than corrup tion and unbridled greed.

A lthough the international m easures o f corruption currently available w ere designed prim arily to assess levels o f corrup tion in po o rer coun ­ tries, they strongly suggest that one o f the likely costs o f greater inequality is increased corruption in governm ent and society m ore w id ely.444 In chapter 4 w e sa w that trust and the strength o f com m u­

nity life are w eakened by inequality, and this is true not o n ly o f inter­

personal trust, but also o f trust in governm ent - the difference between the attitude o f A m ericans and Scandinavians to their go vern ­ ments is w ell kn ow n . In addition, international data and data fo r the A m erican states suggest people trust governm ent less in m ore unequal

states.401’ 445 There is also evidence from societies w here voting is not

com pu lsory (as it is fo r instance in A ustralia) that voter turn-out m ay be lo w er in m ore unequal countries.446 W hether or not this reflects a greater separation o f interests and an increasing sense o f ‘us and them ’ between people at opposite ends o f the social ladder, it cer­ tainly suggests that too much inequality is a threat to dem ocracy.

Econom ists som etim es suggest that the m arket is like a dem o­

cratic vo tin g system : ou r expenditure pattern is, in effect, ou r vote on h o w prod uctive resources should be allocated between com peting dem ands. If this is true, som eone w ith tw enty times the incom e o f another effectively gets tw en ty times as m an y votes. A s a result

in equality seriously distorts the ability o f econom ies to p ro vid e fo r

hum an needs: because the p o o r cannot a ffo rd better housing, their dem and fo r it is ‘ in effective’ , yet the spending o f the rich ensures scarce prod uctive assets are devoted instead to the prod uction o f

lu xu ries.

I N E Q U A L I T Y , D E B T A N D T H E F I N A N C I A L C R A S H

A s w ell as these m ore general effects o f large incom e differences, there is n o w evidence that inequality played a central cau sal role in the financial crashes o f 1 9 2 9 and o f 20 0 8 .

W e suggested (p .2 2 8 , 27 0 ) that inequality leads to increases in debt. It turns out that they are intim ately related. U sing figures fo r the 4 0 years from 1 9 6 3 to 2 0 0 3 , M atteo Iaco viello , an eco n o ­ mist at the Federal R eserve B o a rd and B oston C o llege, has recently show n a v e ry close co rrelation betw een increasing debt and in­ creasin g inequality in the U S A and concludes that the longer term increases in debt can on ly be exp lain ed by the rise in in eq u ality.447 Using the latest international data fro m O E C D on debt, w e have

also foun d that both short-term h ousehold debt as a p ro po rtion o f h ousehold assets, as w ell as governm ent N a tio n a l D eb t as a

Figure 17 .5 The financial crashes o f 19 2 9 and 2008 took place at high points in inequality (continuous line on graph) and debt (two broken lines on graph).*51

pro p o rtion o f G ro ss D om estic P roduct, are higher in m ore unequal

coun tries.448 A ided by some o f the w o rld ’s most respected economists, the story o f

the w a y rising inequality and debt led to the financial crashes o f 19 2 9 and 2008 is well told in a docum entary film called T h e F la w .449 Both crashes happened at the tw o peaks o f inequality in the last hundred years after long periods o f rising inequality which had led to rapid increases in debt.450-1 A s Figure 1 7 . 5 show s, their trends over time are

strikingly similar. Rob ert W ade, professor o f political econom y at the

London School o f Econom ics, estimates that grow ing inequality meant that in the years before the 2008 crash about 1 . 5 trillion dollars per year were being siphoned from the bottom 90 per cent o f the US popu­ lation to the top ten per cent.449 A s a result, the richest people had more and more money to invest and to lend, but people outside the very

wealthiest category found it increasingly difficult to maintain their rela­

tive incomes or realize their aspirations. Both fo r speculators and for ordinary householders rising property prices made investment in prop­ erty look like a band w agon everyone had to get onto. People bought into the housing m arket wherever they could and remortgaged precari­ ously as prices rose. The financial sector handling and speculating on these debts found its share o f all US corporate profits rising from l 5 per cent in 19 8 0 to 40 per cent in 2 0 0 3 . As the bubble grew bigger, the w orse its eventual and inevitable burst became.

M A K I N G D E M O C R A C Y W O R K

R a th e r than being a threat to dem ocracy and the m arket, reductions o f inequality are surely an essential part o f their defence. G reater eq u ality w ill benefit even those w h o w o u ld deny the evidence.

N e a r the beginning o f this chapter w e m entioned research that

show ed that over 90 per cent o f the A m erican pop u latio n say they w o u ld prefer to live in a society w ith the incom e distribution w hich actu ally exists in Sw eden rather than that o f the U SA . R esearch in Britain also show s that people think incom e differences should be sm aller, even though they dram atically underestim ate h o w large they actu ally are. T h e w o rld really is full o f people w h o have m uch m ore egalitarian preferences and a stronger sense o f justice than w e tend

to assum e. Part o f the reason fo r this is that in recent decades m ost

people in the w o r ld ’ s richest societies have been persuaded to doubt the va lid ity and relevance o f egalitarian values. T h e rise o f n eo-lib­ eral political and econom ic thinking in the 19 8 0 s and 19 9 0 s meant that egalitarian ideas disappeared fro m public debate and those w ith a strong sense o f justice becam e - in effect - closet egalitarians.

It is n o w time egalitarians returned to the public arena. W e need to do so confident that ou r intuitions have been validated and foun d to be truer than most o f us ever im agined. Because the evidence show s that few people are aw are o f the actual scale o f inequality and in jus­

tice in our societies, or recognize h o w it dam ages the vast m ajo rity o f

the p o pu lation , the first task is to provide education and inform ation. U nderstanding these issues is alread y ch an gin g attitudes to

inequality am on g politicians. In B ritain T h e S p irit L e v e l has been endorsed across the political spectrum . In a m ajo r speech at the end o f 20 0 9 , D av id C am ero n , n o w the C o n servative prim e m inister, said

T h e S p irit L e v e l sh o w ed ‘that am on g the richest countries, it’ s the m ore unequal ones that do w o rse accordin g to alm ost every q u ality o f life in d icator . . . per capita G D P is m uch less significant fo r a co u n try ’ s life expectan cy, crim e levels, literacy and health than the size o f the gap betw een the richest and p oorest in the po pu latio n . . . We all k n o w , in ou r hearts, that as long as there is deep p o verty living system atically side b y side w ith great riches, w e all rem ain the

poorer fo r it.’453 In Septem ber 2 0 1 0 , in his first m ajo r speech as leader o f the L a b o u r P arty, Ed M ilib a n d said ‘I do believe this co u n ­ try is to o unequal and the gap between rich and p o o r doesn ’t just harm the p o o r, it harm s us a ll’454 and ‘ if yo u lo o k rou n d the w o rld -

at the countries that are healthier, h appier, m ore secure - they are

the m ore equal co u n tries.’455 A s Liberal D em o crat ministers in the

co alition governm ent, V in ce C ab le and Lyn n e Featherstone have signed a pledge com m itting them selves to reducing inequalities.456

W ord s are a start, but changing policies and po litics, changing the

w a y ou r societies organize them selves, w ill require the evidence to be recognized even m ore w idely. F ew tasks are m ore w o rth w h ile than this: as w e think T h e S p irit L e v e l sh o w s, the health o f our dem ocra­ cies, ou r societies and their people, is tru ly dependent on greater equality.

The Equality Trust

If reading this b o ok leaves yo u w an tin g to do som ething to help

reduce inequality, then please visit T h e E q u ality T ru st w eb site at w w w .e q u a lity tru st.o rg .u k . T here you w ill find d o w n lo ad ab le slides w h ich w e hope yo u w ill use, a d o w n lo ad ab le lecture on D V D , short sum m aries o f the evidence, answ ers to frequently asked questions, and suggestions fo r cam paigning.

H a v in g discovered h o w seriously societies are dam aged b y great in equality w e felt w e had to do w h a t w e cou ld to m ake the evidence better kn o w n . T h e T ru st w a s set up as a n ot-for-profit organization to educate and cam p aign on the benefits o f a m ore equal society. Its

w o rk depends on donations fro m in dividuals and organizations sharing ou r vision.

W e hope you w ill sign the E q u ality C h arter, put yo u r nam e dow n to receive the new sletter, m ake a don ation , give us yo u r ideas and join or fo rm a local eq u ality gro u p. M o st o f all w e hope you w ill use the evidence w e h ave started to put together to spread the w o rd and

convince others o f the need to reduce inequality. In p olitics, w o rd s are action.

T h e E q u ality T ru st is not a large organ izatio n able to im plement policies, run cam paigns and orchestrate things on yo u r behalf. In­ stead it aim s to m ake people better inform ed and provid e resources to stim ulate and strengthen their ow n po litical and educational

activities - w hether through talk in g to friends and colleagues, pass­ ing on our w eb address, w ritin g blogs, local cam paign in g, sending letters to n ew spapers and p olitician s, or raisin g the issues in the mass m edia.

O ur aim is to create a gro u n d sw ell o f opinion in fav o u r o f great

T H E E Q U A L I T Y T R U S T

equality. W ithout that politician s can do v e ry little. E galitarian

sentiments are hidden close to the hearts o f vast num bers o f people

o f all shades o f p olitical opinion. M o s t people k n o w h o w m uch w e sacrifice to consum erism and k n o w that there are fe w things nicer than re la x in g w ith friends and equals. T h ey also k n o w that it is fam ily, friends and com m unity that m atter to happiness and k n o w

that our present w a y o f life is ruining the planet. T h e culture o f the

last fe w decades has reduced us to closet egalitarian s: it is tim e w e cam e out o f the w o o d w o r k and set a course fo r sanity.

Appendix

H O W W E C H O S E C O U N T R I E S F O R O U R I N T E R N A T I O N A L C O M P A R I S O N S

First, w e obtained a list o f the 50 richest countries in the w o rld from the W o rld Bank. T h e report w e used w a s published in 20 0 4 and is based on data fro m 2 0 0 2 .

Th en w e excluded countries w ith po pu latio n s b elo w 3 m illion, because w e didn ’t w a n t to include t a x havens like the C aym an Islands and M o n a c o . A n d w e excluded countries w ith o u t co m p ar­ able data on incom e in equality, such as Iceland.

T h a t left us w ith 2 3 rich countries:

A ustralia Greece P ortugal A ustria Ireland Singapore Belgium Israel Spain C an ad a Italy Sweden D enm ark Ja p a n Sw itzerland Finland N etherlan ds United K in gd om France N e w Z e alan d United States o f A m erica G erm an y N o r w a y

T H E 5 0 A M E R I C A N S T A T E S

In ou r figures, w e label each A m erican state w ith the tw o-letter abb reviation used by the U S P ostal Service. A s these w ill be un­

fam iliar to some international readers, here is a list o f the states and

their labels:

A L A B A M A A L M O N T A N A M T

A L A S K A A K N E B R A S K A N E

A R I Z O N A A Z N E V A D A N V

A R K A N S A S A R N E W H A M P S H I R E N H C A L I F O R N I A C A N E W J E R S E Y N J C O L O R A D O C O N E W M E X I C O N M

C O N N E C T I C U T C T N E W Y O R K N Y

D E L A W A R E D E N O R T H C A R O L I N A N C

F L O R I D A F L N O R T H D A K O T A N D

G E O R G I A G A O H I O O H H A W A I I H I O K L A H O M A O K I D A H O ID O R E G O N O R

I L L I N O I S IL P E N N S Y L V A N I A P A

I N D I A N A I N R H O D E I S L A N D R I

IO W A IA S O U T H C A R O L I N A S C K A N S A S K S S O U T H D A K O T A SD K E N T U C K Y K Y T E N N E S S E E T N L O U I S I A N A L A T E X A S T X M A I N E M E U T A H U T

M A R Y L A N D M D V E R M O N T V T

M A S S A C H U S E T T S M A V I R G I N I A V A M I C H I G A N M I W A S H I N G T O N W A M I N N E S O T A M N W E S T V I R G I N I A W V M I S S I S S I P P I M S W I S C O N S I N W I

M I S S O U R I M O W Y O M I N G W Y

In this b o o k , fo r all international co m pariso n s, w e use the 2 0 :2 0 ratio m easure o f incom e in equality from the United N a tio n s D e­ velopm ent Program m e H u m an D evelopm ent In dicato rs, 2 0 0 3 - 6 . A s

su rvey dates v a ry fo r different countries (from 1 9 9 2 to 2 0 0 1 ) , and as the lag time fo r effects w ill v a ry fo r the different outcom es w e exam in e, w e to o k the average across the reportin g years 2 0 0 3 - 6 . For the US com parison s w e use the 19 9 9 state-level G in i coefficient based on household incom e prod uced by the U S C ensus Bureau.

Data Sources

United N a tio n s D evelopm ent Program . H um an developm ent report. N e w Y o r k : O x fo rd U n iversity Press, 2 0 0 3 , 2 0 0 4 , 2 0 0 5 , 20 0 6 .

U S Census Bureau. G in i ratios b y state. 19 6 9 , 1 9 7 9 , 19 8 9 , 19 9 9 . W ashin gton , D C : U S C ensus B ureau , 19 9 9 (table S4).

D E V E L O P I N G T H E I N D E X O F H E A L T H A N D S O C I A L P R O B L E M S

T h e I n t e r n a t io n a l I n d e x

Th e International In dex has 1 0 com ponents:

• L ife exp ectan cy (reverse coded)

• T eenage births • O besity • M en tal illness • H om icides • Im prisonm ent rates • M istru st • So cial m o b ility (reverse coded) • E du cation (reverse coded) • Infant m o rtality rate

Sixteen countries had at least nine o f these ten m easures. A further five countries had eight out o f ten. T w o countries (Israel and Singapore) w ith few er m easures w ere excluded from the in d ex but included in analyses o f in dividual m easures.

• C oun tries w ith data on all ten m easures: C a n a d a , G erm an y, U S A

• C oun tries w ith data on nine out o f ten m easures, but no data on social m obility: A u stra lia, B elgium , France, Italy, Ja p a n , N eth erlan d s, N e w Z e a la n d , Spain

• C oun tries w ith data on nine out o f ten m easures, but no data on m ental health: D en m ark , Fin lan d, N o r w a y , Sweden

• C oun tries w ith data on nine out o f ten m easures, but no data on education: U K

• C oun tries w ith data on eight out o f ten m easures, but no data

on social m o b ility o r mental illness: A u stria, G reece, Ireland, P o rtu gal, Sw itzerland

T h e In dex o f H ealth an d Social Problem s w a s created by tak in g the m ean o f the z-scores fo r each m easure (averaged o ver the num ber o f

m easures availab le fo r that p articu lar country).

The Index of Health and Social Problems for the 50 states of the USA

T h e U S In d ex has nine com ponents:

• T ru st (reverse coded) • L ife exp ectan cy (reverse coded) • Teen age births

• O besity

• H om icides • Im prisonm ent • E du catio n (reverse coded)

• In fan t m o rtality rate • M en tal illness

O f the 50 states, 40 have data fo r all eight m easures.

N in e states are m issing data on trust from the G eneral Social Survey:

• A la s k a , D elaw are, H a w a ii, Idaho , M ain e, N e b r a sk a , N e w M e x ic o , N e v a d a , South D ak o ta

W yo m in g has data on trust, but not on hom icides

T h e In dex o f H ealth and Social Problem s fo r the U S A w a s created

by tak in g the m ean o f the z-scores fo r each m easure (averaged over

the num ber o f m easures a vailab le fo r that particu lar state).

Sources of Data for the Indices of Health and Social Problems Component International data US state data

Trust Per cent of people who

respond positively to the

statement ‘most people

can be trusted’

1 9 9 9 - z o o i

World Values Survey1

Reverse-coded

Per cent of people who

respond positively to the

statement ‘most people can

be trusted’

1999

General Social Survey2

Reverse-coded

Life expectancy Life expectancy at birth

for men and women

Z004 United Nations Human

Development Report3

Reverse-coded

Life expectancy at birth for

men and women

zooo

US Census Bureau,

Population Division4

Reverse-coded

Infant mortality Deaths in the first year of

life per 1,000 live births zooo

World Bank5

Deaths in the first year of

life per 1,000 live births 2002 US National Center for

Health Statistics6

Obesity Percentage of the

population with BMI >30,

averaged for men and

women

zooz

International Obesity

TaskForce7-8

Percentage of the

population with BMI >30,

averaged for men and

women

1 9 9 9 -z o o z

Corrected estimates from

Prof Majid Ezzati, Harvard

University, based on

N H A N E S and BRFSS

surveys9

Component International data US state data

Mental illness Prevalence of mental

illness

2 0 0 1 - 2 0 0 3

W H O 10

Average number of days in

past month when mental

health was not good

1 9 9 3 -2 0 0 1

BRFSS11

Education scores Combined average of

maths literacy and reading

literacy scores of 15-year-

olds

2 0 0 0

O ECD PISA12

Reverse-coded

Combined average of

maths and reading scores

for 8th graders

2003

US Department of

Education, National

Center for Education

Statistics13,14

Reverse-coded

Teenage birth rate Births per 1 ,0 0 0 women

aged 1 5 -1 9 years

1998

U N IC E F 15

Births per 1 ,0 0 0 women

aged 1 5 -1 9 years

2 0 0 0

US National Vital Statistics16

Homicides Homicide rate per

1 0 0 ,0 0 0

Period average for

1 9 9 0 - 2 0 0 0

United N ations17

Homicide rate per

1 0 0 ,0 0 0

1999

FBI18

Imprisonment Prisoners per 1 0 0 ,0 0 0

United N ations17

Prisoners per 1 0 0 ,0 0 0

1 9 9 7 - 8

US Department of

Justice19

Social mobility Correlation between

father and son’s income

30-year period data from

8 cohort studies

London School of

Economics20

N/A

Data Sources

1 . European Values Study Group and World Values Survey Association. European and World Values Survey Integrated Data File, 1 9 9 9 -2 0 0 1, Release 1 . Ann Arbor, M I: Inter-university Consortium for Political and Social Research, 2005. 2. National Opinion Research Center. General Social Survey. Chicago: N O R C , 1999. 3. United Nations Development Program. Human D evelopm ent Report. New Y ork: Oxford University Press, 2004. 4. US Census Bureau. Population Division, Interim State Population Projections, Table 2. Internet release date: April 2 1 , 2005. 5. World Bank. World Development Indicators (WDI) September 2006: E SD S International, (M IM A S) University of Manchester. 6. US National Center for Health Statistics. Table 10 5 , Statistical abstract of the United States. Washington, D C : C D C , 2006. 7. International Obesity TaskForce. Obesity in Europe. London: International Obesity TaskForce in collaboration with the European Association for the Study o f Obesity Task Forces, 2002. 8. International Obesity TaskForce. O verweight and obese. London: International Obesity Taskforce, 2002. 9. M . Ezzati, H. M artin, S. Skjold, S. Vander Hoorn, C. J . M urray. ‘Trends in national and state-level obesity in the U SA after correction for self­ report bias: analysis of health surveys’ . / R Soc M ed 2006; 99(5): 2 5 0 -7 . 10 . K. Demyttenaere, R. Bruffaerts, J. Posada-Villa, I. Gasquet, V. Kovess, J . P. Lepine, et al. ‘Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys’ . Jam a 2004; 2 9 1( 2 1 ) : 2 5 8 1-9 0 . 1 1 . H. S. Zahran, R. Kobau, D. G. M oriarty, M . M . Z ack, J. Holt, R. Donehoo. ‘Health-related quality o f life surveillance - United States, 19 9 3 - 2 0 0 2 ’ . M M W R Surveill Summ 2005; 54(4): 1 - 3 5 . i a . O E C D . Education at a glance. O E C D Indicators, 2003. 1 3 . US Department of Education N C fE S . The N ation ’s R eport Card: Reading Highlights 2003. Washington, D C, 2004. 14 . US Department of Education N C fE S . The N ation ’s Report Card: Mathematics Highlights 2003. Washington, D C , 2004. 1 5 . U N IC E F Innocenti Research Centre. A league table o f teenage births in rich nations. Florence: Innocenti Report Card, 2 0 0 1. 16 . US Census Bureau. Statistical Abstract o f the United States: 2000 (120th Edition). Washington: Census Bureau, 2000.

1 7. United Nations Crime and Justice Information Network. Survey on Crime Trends and the Operations o f Criminal Justice Systems (Fifth, Sixth, Seventh, Eighth): United Nations, 2000. 18 . Federal Bureau of Investigation. Crime in the United States 19 9 9 . Washington, DC: US Government Printing Office, 1999. 19 . US Department of Justice BoJS. Incarceration rates for prisoners under State or Federal jurisdiction. File: corpop25.wki. 20. J. Blanden, P. Gregg, S. Machin. Intergenerational mobility in Europe and North America. London: Centre for Economic Performance, London School of Economics, 2005.

Statistics

Pearson C o rrelatio n C oefficients (r) and Statistical Significance

(p-value) fo r A sso ciatio n s w ith Incom e Inequality.

Indicator International data r p -v a lu e

U S data r p -v a lu e

T ru st - 0 .66 < 0 .0 1 - 0 .7 0 < 0 .0 1

Life expectan cy - 0 .4 4 0 .04 - 0 .4 5 < 0 .0 1

In fan t m o rtality 0 .4 2 0 .04 0 .4 3 < 0 .0 1

O besity 0 .5 7 < 0 .0 1 0 .4 7 < 0 .0 1

M en ta l illness 0 .7 3 < 0 .0 1 0 .1 8 0 . 1 2

Edu catio n score - 0 .4 5 0 .04 - 0 .4 7 .0 1

Teenage birth rate 0 .7 3 < 0 .0 1 0 .4 6 < 0 .0 1

H om icides 0 .4 7 0 .0 2 0 .4 2 < 0 .0 1

Im prisonm ent 0 .7 5 < 0 .0 1 0.48 < 0 .0 1

Social m obility 0 .9 3 < 0 .0 1 - -

In dex 0 .8 7 < 0 .0 1 0-59 < 0 .0 1

O verw eight children 0 .5 9 O.OI 0 .5 7 < 0 .0 1

D ru gs in d ex 0.6 3 < 0 .0 1

C alo rie intake 0 .4 6 0 .03

Public expenditure - 0 .5 4 O.OI on health care

C hild w ell-being - 0 . 7 1 < 0 .0 1 - 0 . 5 1 < 0 .0 1

T rip le education score - 0 .4 4 0 .04

In dicator International data U S data r p -v a lu e r p -v a lu e

C hild conflict 0 .6 2 < 0 .0 1

Spending on foreign aid - 0 . 6 1 < 0 .0 1

Recyclin g - 0 .8 2 < 0 .0 1

Peace index - 0 . 5 1 0 .0 1

Paid m aternity leave - 0 .5 5 0 .0 1

A dvertisin g 0 .7 3 < 0 .0 1

Police 0 .5 2 0 .04

Social expenditure - 0 .4 5 0 .04

W o m en ’ s status - 0 .4 4 0 .0 4 - 0 .3 0 0 .0 3

Patents per capita - 0-49 0 .0 2 Ju v en ile hom icides 0 .29 < 0 .0 5

H igh school drop-outs 0 .79 < 0 .0 1

C h ild mental illness 0 .3 6 0 .0 1

Pugnacity 0 .4 7 < 0 .0 1

References

i . The Harwood Group, Yearning fo r Balance: Views o f Americans on consumption, materialism, and the environment. Takom a Park, M D : Merck Family Fund, 19 9 5 . 2.. United Nations Development Program, Human Development Report. New York: Oxford University Press, 2006. 3. R. Layard, Happiness. London: Allen Lane, 2005. 4. World Bank, World D evelopm ent Report 19 9 3 : Investing in health. Oxford: Oxford University Press, 19 9 3 . 5. European Values Study Group and World Values Survey Association, European and World Values Survey Integrated Data File, 1 9 9 9 - 2 0 0 1 , Release 1 . Ann Arbor, M I: Inter-university Consortium for Political and Social Research, 2005. 6 . United Nations Development Program, Human D evelopm ent Report. N ew Y ork: O xford University Press, 2004. 7. G. D. Smith, J. D. Neaton, D. Wentworth, R. Stamler and J. Stamler, ‘Socioeconomic differentials in mortality risk among men screened for the Multiple Risk Factor Intervention Trial: I. White men’ , American Journ al o f Public Health (1996) 86 (4): 4 8 6 -9 6 . 8 . R. G. Wilkinson and K. E. Pickett, ‘Income inequality and socio­ economic gradients in mortality’ , American Jo u rn al o f Public Health (2008) 98 (4): 6 9 9 -7 0 4 . 9. L. M cLaren, ‘Socioeconomic status and obesity’ , Epidem iologic R eview (2007) 29: 2 9 -4 8 . 10 . R. G. Wilkinson and K. E. Pickett, ‘Income inequality and population health: a review and explanation of the evidence’ , Social Science and M edicine (2006) 62 (7): 17 6 8 - 8 4 . 1 1 . J. M . Twenge, ‘The age of anxiety? Birth cohort change in anxiety and neuroticism, 1 9 5 2 - 1 9 9 3 ’ , Jo u rn al o f Personality and Social Psychology (2007) 79 (6): 1 0 0 7 - 2 1 .

1 2 . M . Rutter and D. J. Smith, Psychosocial Disorders in Young People: Time trends and their causes. Chichester: Wiley, 19 9 5 . 1 3 . S. Collishaw, B. Maughan, R. Goodman and A. Pickles, ‘Time trends in adolescent mental health’, Jo urn al o f Child Psychology and Psychiatry (2004) 45 (8): 1 3 5 0 - 6 2 . 14 . B. Maughan, A. C. Iervolino and S. Collishaw, ‘Time trends in child and adolescent mental disorders’, Current O pinion in Psychiatry (2005) 18 (4): 3 8 1 - 5 . 1 5 . J. M . Twenge, Generation Me. N ew Y ork: Simon & Schuster, 2006. 16 . S. S. Dickerson and M . E. Kemeny, ‘Acute stressors and cortisol responses: a theoretical integration and synthesis o f laboratory research’ , Psychological Bulletin (2004) 13 0 (3): 3 5 5 - 9 1 . 1 7 . T. J. Scheff, ‘ Shame and conformity: the defense-emotion system’ , American Sociological R eview (1988) 5 3: 3 9 5 -4 0 6 . 18 . H. B. Lewis, The R ole o f Shame in Symptom Formation. Hillsdale, N J: Erlbaum, 19 8 7 . 19 . R. W. Emerson, Conduct o f Life. N ew Y ork: Cosimo, 2007. 20. A. Kalma, ‘Hierarchisation and dominance assessment at first glance’ , European Jo u rn al o f Social Psychology ( 19 9 1) 2 1 (2): 1 6 5 - 8 1 . 2 1 . F. Lim, M . H. Bond and M . K. Bond, ‘ Linking societal and psycho­ logical factors to homicide rates across nations’ , Jo u rn al o f Cross-Cultural Psychology (2005) 36 (5): 5 1 5 - 3 6 . 22. S. Kitayama, H. R. M arkus, H. Matsumoto and V. Norasakkunkit, ‘Individual and collective processes in the construction o f the self: self­ enhancement in the United States and self-criticism in Ja p an ’ , Jo u rn al o f Personal and Social Psychology (1997) 7 2 (6): 1 2 4 5 - 6 7 . 23. A. de Tocqueville, Democracy in America. London: Penguin, 2003. 24. National Opinion Research Center, General Social Survey. Chicago: N O R C , 19 9 9 -2 0 0 4 . 25. R. D. Putnam, B ow lin g Alone: The collapse and revival o f American community. N ew Y ork: Simon & Schuster, 2000. 26. R. D. Putnam, ‘ Social capital: measurement and consequences’ , I S U M A : Canadian Jo u rn al o f Policy Research (2001) 2 (1): 4 1 - 5 1 . 27. E. Uslaner, The M oral Foundations o f Trust. Cambridge: Cambridge University Press, 2002. 28. B. Rothstein and E. Uslaner, ‘All for all: equality, corruption and social trust’, World Politics (2005) 58: 4 1 - 7 2 . 29. J. C. Barefoot, K. E. M aynard, J. C. Beckham, B. H. Brummett,

K. Hooker and I. C. Siegler, ‘Trust, health, and longevity’, Jo u rn al o f Behavioral M edicine {1998) 2 1 (6): 5 1 7 - 2 6 . 30. S. V. Subramanian, D. J. Kim and I. Kawachi, ‘Social trust and self­ rated health in US communities: a multilevel analysis’, Jo u rn al o f Urban Health (2002) 79 (4, Suppl. 1): S 2 1 - 3 4 . 3 1 . E. Klinenberg, Heat Wave: A social autopsy o f disaster in Chicago. Chicago: University o f Chicago Press, 2002. 32. J . Lauer, ‘Driven to extremes: fear of crime and the rise of the sport utility vehicle in the United States’ , Crime, Media, Culture (2.005) 1 : 14 9 - 6 8 . 3 3. K. Bradsher, ‘The latest fashion: fear-of-crime design’ , N e w York Times, 23 Ju ly 2000. 34. M . Adams, Fire and Ice. The United States, Canada and the myth o f converging values. Toronto: Penguin (Canada), 2003. 3 5 . E. J. Blakely and M . G. Snyder, Fortress America: Gated commun­ ities in the United States. Washington, D C : Brookings Institute Press,

1997­ 36. 1. Kawachi, B. P. Kennedy, V. Gupta and D. Prothrow-Stith, ‘Women’s status and the health o f women and men: a view from the States’, Social Science and M edicine (1999) 48 (1): 2 1 - 3 2 . 37. H. J. Jun, S. V. Subramanian, S. Gortmaker and I. Kawachi, ‘ A multi­ level analysis of women’s status and self-rated health in the United States’, Jo urn al o f the American M edical W omen’s Association (2004) 59 (3): 1 7 2 - 8 0 . 38. O E C D , International Development Statistics Online. O E C D . Stat: http://www.oecd.org/dataoecd/5 0/17/503 7 7 2 1 .htm, 2005. 39. L. Clark and A. Dolan, ‘The disturbed generation’, Daily Mail, 20 June 2007. 40. C. Donnellan, M ental Wellbeing. Cambridge: Independence Educa­ tional Publishers, 2004. 4 1 . The G o o d C hildhood Inquiry. Evidence Summary 5 - Health. London: Children’s Society, 2008. 4 2 . J. M . Perrin, S. R. Bloom and S. L. Gortmaker, ‘The increase of child­ hood chronic conditions in the United States’, Jo urn al o f the American M edical Association (2007) 297 (24): 2 7 5 5 - 9 . 43. Child and Adolescent Health Measurement Initiative. National Survey o f Children’s Health. Data Resource Center on Child and Adolescent Health: http://www.childhealthdata.org (accessed 1 7 August 2006). 44. Office for National Statistics, Psychiatric M orbidity among Adults Livin g in Private Households, 2000. London: H M S O , 2 0 0 1.

45. Hansard (House of Commons Daily Debates). Written answers to questions. (2005) 439: 22 N ov. 2005: Column 1798W . 46. R. C. Kessler, W. T. Chiu, O. Dernier, K. R. Merikangas and E. E. Walters, ‘Prevalence, severity, and comorbidity of 12-month D SM - IV disorders in the National Comorbidity Survey Replication’, Archives o f General Psychiatry (2005) 62 (6): 6 1 7 - 2 7 . 47. T. L. M ark, K. R. Levit, J. A. Buck, R. M . Coffey and R. Vandivort- Warren, ‘Mental health treatment expenditure trends, 19 8 6 - 2 0 0 3 ’, Psychiatric Services (2007) 58 (8): 1 0 4 1 - 8 . 48. D. Rowe, H o w to Im prove your Mental Well-being. London: Mind, 2002. 49. K. Demyttenaere, R. Bruffaerts, J. Posada-Villa, I. Gasquet, V. Kovess, J. P. Lepine, M . C. Angermeyer, S. Bernert, G. de Girolamo, P. Morosini, G. Polidori, T. Kikkawa, N. Kawakam i, Y . Ono, T. Takeshima, H. Uda, E. G. Karam , J. A. Fayyad, A. N . Karam , Z . N . Mneimneh, M . E. Medina- M ora, G. Borges, C. Lara, R. de Graaf, J . Ormel, O. Gureje, Y . Shen, Y . Huang, M . Zhang, J. Alonso, J. M . Haro, G. Vilagut, E. J. Bromet, S. Gluzman, C. Webb, R. C. Kessler, K. R. Merikangas, J. C. Anthony, M . R. von Korff, P. S. Wang, T. S. Brugha, S. Aguilar-Gaxiola, S. Lee, S. Heeringa, B. E. Pennell, A. M . Zaslavsky, T. B. Ustun and S. Chatterji, ‘Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys’, Jo u rn al o f the American M edical Association (2004) 2 9 1 (21): 2 5 8 1 - 9 0 . 50. J. E. Wells, M . A. Oakley Browne, K. M . Scott, M . A. M cGee, J. Baxter and J. Kokaua, ‘Te Rau Hinengaro: the N ew Zealand Mental Health Survey: overview of methods and findings’, Australian and N e w Zealand Jo urn al o f Psychiatry (2006) 40 (10): 8 3 5 - 4 4 . 5 1 . Australian Bureau of Statistics. National Health Survey, Mental Health, 2 0 0 1. Canberra: Australian Bureau of Statistics, 2003. 52. W H O International Consortium in Psychiatric Epidemiology, ‘Cross­ national comparisons of the prevalences and correlates of mental dis­ orders’ , Bulletin o f the World Health Organization (2000) 78 (4): 4 1 3 - 2 6 . 53. Center for Disease Control and Prevention, ‘ Self-reported frequent mental distress among adults - United States, 1 9 9 3 - 2 0 0 1 ’ (2004) 53: 9 6 3 - 6 . 54. O. James, Affluenza. London: Vermilion, 2007. 55. A. de Botton, Status Anxiety. London: Hamish Hamilton, 2004. 56 . R. H. Frank, L uxury Fever. New York: Free Press, 1999. 57. United Nations Office on Drugs and Crime, World D rug Report. Vienna: UN Office on Drugs and Crime, 2007.

5 8 . Centers for Disease Control and Prevention. Compressed M or­ tality Files 19 9 9 - 2 0 0 2 . http://wonder.cdc.gov/mortSQL.html (accessed 9 September 2008) 59. D. M organ, K. A, Grant, H. D. Gage, R. H. M ach, J. R. Kaplan, 0 . Prioleau, S. FI. Nader, N . Buchheimer, R. L. Ehrenkaufer and M . A. Nader, ‘ Social dominance in monkeys: dopamine Dz receptors and cocaine self-administration’, Nature Neuroscience (2002) 5 (2): 16 9 - 7 4 . 60. M . Susser and E. Susser, ‘ Choosing a future for epidemiology: I. Eras and paradigms’ , American Jo urn al o f Public Health (1996) 86 (5): 6 6 8 -7 3 . 6 1 . M. Susser and E. Susser, ‘Choosing a future for epidemiology: II. From black box to Chinese boxes and eco-epidemiology’, American Jo u rn al o f Public Health (1996) 86 (5): 6 7 4 -7 . 62. M . G. Marmot, A. M . Adelstein, N. Robinson and G. A. Rose, ‘Changing social-class distribution o f heart disease’ , British M edical Jo urn al (19 78) 2 (6 145): 1 1 0 9 - 1 2 . 63. M. G. Marmot, G. Rose, M. Shipley and P. J. Hamilton, ‘Employment grade and coronary heart disease in British civil servants’, Jo u rn al o f Epidem iology and Community Health (1978) 32 (4): 2 4 4 -9 . 64. H. Bosma, M . G. Marmot, H. Hemingway, A. C. Nicholson, E. Brunner and S. A. Stansfeld, ‘Low job control and risk of coronary heart disease in Whitehall II (prospective cohort) study’ , British M edical Jo u rn al (1997) 3 1 4 (7080): 5 5 8 - 6 5 . 65. M. G. Marmot, G. D. Smith, S. Stansfeld, C. Patel, F. North, J. Head, 1. White, E. Brunner and A . Feeney, ‘Health inequalities among British civil servants: the Whitehall II study’ , Lancet ( 19 9 1) 3 3 7 (8754): 1 3 8 7 - 9 3 . 66. Council of Civil Service Unions/Cabinet Office, Work, Stress and Health: The Whitehall I I Study. London: Public and Commercial Services Union, 2004. 67. G. D. Smith, M . J . Shipley and G. Rose, ‘Magnitude and causes of socioeconomic differentials in mortality: further evidence from the Whitehall Study’ , Jo urn al o f Epidem iology and Community Health (1990) 44 (4): 2.65-70. 68. R. G. Wilkinson and M . Marmot, Social Determinants o f Health: The Solid Facts (2nd edition). Copenhagen: World Health Organization, Regional Office for Europe, 2006. 69. E. Durkheim, Suicide. London: Routledge, 19 5 2 . 70. L. Berkman and T. Glass, ‘Social integration, social networks, social support, and health’ , in L. Berkman and I. Kawachi, (eds), Social Epidem iology. New York: O xford University Press, 2.000.

7 1 . S. A. Stansfeld, ‘ Social support and social cohesion’ , in M . Marmot and R. G. Wilkinson (eds), Social Determinants o f Health. Oxford: Oxford University Press, 2006. 72. S. Cohen, ‘Keynote Presentation at the Eighth International Congress of Behavioral Medicine: The Pittsburgh common cold studies: psychosocial predictors o f susceptibility to respiratory infectious illness’ , International Jo u rn al o f Behavioral M edicine (2005) 1 2 (3): 1 2 3 - 3 1 . 7 3 . J. K. Kiecolt-Glaser, T. J. Loving, J. R. Stowell, W. B. M alarkey, S. Lemeshow, S. L. Dickinson and R. Glaser, ‘Hostile marital interactions, proinflammatory cytokine production, and wound healing, Archives o f General Psychiatry (2005) 62 (12): 1 3 7 7 - 8 4 . 74. W. T. Boyce, ‘Stress and child health: an overview’, Pediatric Annals (19 85) 14 (8): 5 3 9 - 4 2 . 75. M . C. Holmes, ‘Early life stress can programme our health’ , Jo urn al o f N euroendocrinology (2001) 1 3 (2): 1 1 1 - 1 2 . 76. R. H. Bradley and R. F. Corwyn, ‘Socioeconomic status and child development’ , Annual R eview o f Psychology (2002) 53: 3 7 1 - 9 9 . 77. M . Wilson and M . Daly, ‘Life expectancy, economic inequality, homicide, and reproductive timing in Chicago neighbourhoods’, British M edical Jo u rn al (1997) 3 1 4 (7089): 1 2 7 1 - 4 . 78. M . K. Islam, J. Merlo, I. Kawachi, M . Lindstrom and U. G. Gerdtham, ‘Social capital and health: does egalitarianism matter? A literature review’ , International Journ al fo r Equity in Health (2006) 5 : 3 . 79. I. Kawachi, B. P. Kennedy, K. Lochner and D. Prothrow-Stith, ‘ Social capital, income inequality, and mortality’, American Jo urn al o f Public Health (1997) 87 (9): 1 4 9 1 - 8 . 80. C. M cCord and H. P. Freeman, ‘ Excess mortality in Harlem’, N ew England Jo u rn al o f M edicine (1990) 3 2 2 (3): 1 7 3 - 7 . 8 1. R. G. Wilkinson, ‘Income distribution and life expectancy’, British M edical Jo urn al (1992) 304 (6820): 1 6 5 - 8 . 82. Editor’s Choice, ‘The Big Idea’, British M edical Jo urn al (1996) 3x2 (7037): o. 83. Department of Health, The N H S Plan: A plan fo r investment, a plan fo r reform. London: H M S O , 2000. 84. Office for National Statistics. ‘Trends in O N S Longitudinal Study estimates of life expectancy, by social class 1 9 7 2 - 2 0 0 5 ’ . http://www. statistics.gov.uk/StatBase/Product.asp?vlnk=846o&More=Y (accessed 9 September 2008). . 85. C. J . M urray, S. C. Kulkarni, C. Michaud, N. Tomijima, M . T. Bulzacchelli, T. J. Iandiorio and M . Ezzati, ‘ Eight Americas: investigating

mortality disparities across races, counties, and race-counties in the United States’ , Public Library o f Science M edicine (2006) 3 (9): e26o. 86. A. T. Geronimus, J. Bound, T. A. Waidmann, C. G. Colen and D. Steffick, ‘Inequality in life expectancy, functional status, and active life expectancy across selected black and white populations in the United States’ , Dem ography (2001) 38 (2): 2 2 7 - 3 1 . 87. G. K. Singh and M . Siahpush, ‘Widening socioeconomic inequalities in US life expectancy, 19 8 0 - 2 0 0 0 ’ , International Jo u rn al o f Epidem iology (2006) 35 (4): 9 6 9 - 7 9 ­ 88. P. M . Lantz, J. S. House, J . M . Lepkowski, D. R. Williams, R. P. M ero and J. Chen, ‘Socioeconomic factors, health behaviors, and mortality: results from a nationally representative prospective study of US adults’ , Jo urn al o f the American M edical Association (1998) 279 (21): 1 7 0 3 - 8 . 89. P. M akela, T. Valkonen and T. Martelin, ‘ Contribution of deaths related to alcohol use to socioeconomic variation in mortality: register based follow up study’ , British M edical Jo urn al (1997) 3 1 5 (7 10 2): 2 1 1 - 1 6 . 90. G. Rose and M . G. Marmot, ‘Social class and coronary heart disease’ , British Heart Jo urn al ( 19 8 1) 45 (1): 1 3 - 1 9 . 9 1 . R. G. Wilkinson, Unhealthy Societies: The afflictions o f inequality. London: Routledge, 1996. 92. R. Sapolsky, ‘ Sick of poverty’ , Scientific American (2005) 293 (6): 9 2 - 9 . 93. L. Vitetta, B. Anton, F. Cortizo and A. Sali, ‘M ind-body medicine: stress and its impact on overall health and longevity’, Annals o f the N ew York Academy o f Sciences (2005) 10 5 7 : 4 9 2 - 5 0 5 . 94. S. V. Subramanian and I. Kawachi, ‘ Income inequality and health: what have we learned so far?’ Epidem iologic R eview (2004) 26: 7 8 - 9 1 . 95. S. Bezruchka, T. Namekata and M . G. Sistrom, ‘Improving economic equality and health: the case of postwar Jap an ’, American Jo u rn a l o f Public Health (2008) 98: 2 1 6 - 2 1 . 96. P. Walberg, M . M cKee, V. Shkolnikov, L. Chenet and D. A. Leon, ‘Economic change, crime, and mortality crisis in Russia: regional analysis’ , British M edical Jo urn al (1998) 3 1 7 (7 15 4 ): 3 1 2 - 1 8 . 97. K. M . Flegal, M . D. Carroll, C. L. Ogden and C. L. Johnson, ‘Preva­ lence and trends in obesity among US adults’ , Jo u rn al o f the American M edical Association (2002) 288: 1 7 2 3 - 7 . 98. International Obesity TaskForce, Obesity in Europe. London: International Obesity TaskForce in collaboration with the European Association for the Study of Obesity Task Forces, 2002.

99. World Health Organization, R eport o f a jo in t W H O /F A O Expert Consultation. Diet, nutrition and the prevention o f chronic diseases. Geneva: W H O Technical Report Series no. 9 16 . W H O , 2002. 10 0 . C. L. Ogden, M . D. Carroll, L. R. Curtin, M . A. McDowell, C. J. Tabak and K. M . Flegal, ‘Prevalence of overweight and obesity in the United States, 19 9 9 - 2 0 0 4 ’, Jo u rn al o f the American M edical Association (2006) 295 (13): 1 5 4 9 - 5 5 . 1 0 1 . S. J. Olshansky, D. J. Passaro, R. C. Hershow, J. Layden, B. A. Carnes, J. Brody, L. Hayflick, R. N. Butler, D. B. Allison and D. S. Ludwig, ‘A potential decline in life expectancy in the United States in the 2 1st century’, N e w England Jo urn al o f M edicine (2005) 3 5 2 ( 1 1 ) : 1 1 3 8 - 4 5 . 102.. C B S News, ‘Teen slims down with gastric bypass: Surgery a growing trend among obese teenagers’ , C B S Broadcasting Inc. http://www. cbsnews.com /stories/2oo7/05/21/earlysh0w/health/m ain2830891.shtm / ?Source=Search-story (accessed 15 September 2008). 1 0 3 . G. Rollings, ‘ i4 st boy - is this child abuse?’ Sun, 26 February 2007. 10 4 . B. Ashford and V. Wheeler, ‘Sam, aged 9, is i4 st and size 1 8 ’, Sun, 28 February 2007. 10 5. A. Parker, ‘Daryl is 20 stone, aged just 1 2 ’ , Sun, 2 M arch 2007. 106. E. Brunner, M . Juneja and M . Marmot, ‘Abdominal obesity and disease are linked to social position’, British M edical Jo u rn al (1998) 3 16 : 308. 10 7 . A. Molarius, J. C. Seidell, S. Sans, J. Tuomilehto and K. Kuulasmaa, ‘ Educational level, relative body weight and changes in their association over 10 years: an international perspective from the W H O M O N IC A project’, American Jo urn al o f Public Health (2000) 90: 12 6 0 - 8 6 . 10 8 . P. Toynbee, ‘Inequality is fattening’, Guardian, 28 M ay 2004. 10 9 . International Obesity Taskforce, O verweight and Obese. London: International Obesity Taskforce, 2002. 1 1 0 . U N IC E F Innocenti Research Centre. Child Poverty in Perspective: An overview o f child well-being in rich countries. Florence: Innocenti Report Card, 2007. h i . H. S. Kahn, A. V. Patel, E. J. Jacobs, E. E. Calle, B. P. Kennedy and I. Kawachi, ‘ Pathways between area-level income inequality and increased mortality in U.S. men’, Annals o f the N e w York Academy o f Sciences (1999) 896: 3 3 2 - 4 . 1 1 2 . A. V. Diez-Roux, B. G. Link and M . E. Northridge, ‘A multilevel analysis of income inequality and cardiovascular disease risk factors’ , Social Science and M edicine (2000) 50 (5): 6 7 3 - 8 7 . 1 1 3 . M . Ezzati, H. Martin, S. Skjold, S. vander Hoorn and C. J. M urray,

‘Trends in nadonal and state-level obesity in the U SA after correction for self-report bias: analysis of health surveys’ , Jo u rn al o f the Royal Society o f M edicine (2006) 99 (5): 2 5 0 - 7 . 1 1 4 . K. E. Pickett, S. Kelly, E. Brunner, T. Lobstein and R. G. Wilkinson, ‘Wider income gaps, wider waistbands? An ecological study o f obesity and income inequality’, Jo u rn al o f Epidem iology and Community Health (2005) 59 (8): 6 7 0 -4 . 1 1 5 . K. Ball, G. D. Mishra and D. Crawford, ‘Social factors and obesity: an investigation of the role o f health behaviours’ , International Jo u rn al o f Obesity and Related M etabolic Disorders (2003) 27 (3): 3 9 4 -4 0 3 . 1 1 6 . E. J. Brunner, T. Chandola and M . G. Marmot, ‘ Prospective effect of job strain on general and central obesity in the Whitehall II Study’, American Jo urn al o f Epidem iology (2007) 16 5 (7): 8 2 8 - 3 7 . 1 1 7 . L. R. Purslow, E. H. Young, N. J. Wareham, N. Forouhi, E. J. Brunner, R. N . Luben, A. A. Welch, K. T. Khaw, S. A. Bingham and M . S. Sandhu, ‘Socioeconomic position and risk of short-term weight gain: prospective study of 14 ,6 19 middle-aged men and women’, B ioM ed Central Public Health (2008) 8: 1 1 2 . 1 1 8 . S. P. Wamala, A. Wolk and K. Orth-Gomer, ‘Determinants of obesity in relation to socioeconomic status among middle-aged Swedish women’ , Preventive M edicine (1997) 26 (5 Pt 1): 7 3 4 - 4 4 . 1 1 9 . P. Bjorntorp, ‘Do stress reactions cause abdominal obesity and comorbidities?’ Obesity Review s (2001) 2 (2): 7 3 - 8 6 . 12 0 . V. Drapeau, F. Therrien, D. Richard and A. Tremblay, ‘Is visceral obesity a physiological adaptation to stress?’ Panminerva Medica (2003) 45 (3): 1 8 9 - 9 5 . 1 2 1 . J . Laitinen, E. Ek and U. Sovio, ‘Stress-related eating and drinking behavior and body mass index and predictors of this behavior’, Preventive M edicine (2002) 34 (1): 2 9 - 3 9 . 12 2 . M . F. Dallman, N. Pecoraro, S. F. Akana, S. E. La Fleur, F. Gomez, H. Houshyar, M . E. Bell, S. Bhatnagar, K. D. Laugero and S. M analo, ‘Chronic stress and obesity: a new view of “ comfort food’” , Proceed­ ings o f the N ational Academy o f Sciences USA (2003) 10 0 (20): 1 1 6 9 6 - 7 0 1 . 1 2 3 . A. M . Freedman, ‘ Deadly diet’ , Wall Street Journ al, 1 8 - 2 0 December 1990. 12 4 . C. C. Hodgkins, K. S. Cahill, A. E. Seraphine, K. Frost-Pineda and M . S. Gold, ‘Adolescent drug addiction treatment and weight gain’, Jo urn al o f Addictive Diseases (2004) 23 (3): 5 5 - 6 5 . 12 5 . G. A. James, M . S. Gold and Y . Liu, ‘Interaction of satiety and

reward response to food stimulation’, Jo urn al o f Addictive Diseases (2004) 23 (3): 2 3 - 3 7 . 126 . K. D. Kleiner, M . S. Gold, K. Frost-Pineda, B. Lenz-Brunsman, M . G. Perri and W. S. Jacobs, ‘Body mass index and alcohol use’, Journal o f Addictive Diseases (2004) 23 (3): 1 0 5 - 1 8 . 1 2 7 . J. H. Gao, ‘Neuroimaging and obesity’, Obesity Review s (2001)

9 ( 1 1 ) : 72-9 - 3°- 12 8. K. Sproston and P. Primatesta (eds), Health Survey fo r England 2003. Vol. 2: Risk Factors fo r Cardiovascular Disease. London: H M S O , 2004. 12 9. C. Langenberg, R. Hardy, D. Kuh, E. Brunner and M . Wadsworth, ‘Central and total obesity in middle aged men and women in relation to lifetime socioeconomic status: evidence from a national birth cohort’ , Jo urn al o f Epidem iology and Community Health (2003) 57 (10): 8 1 6 - 2 2 . 130 . R. M . Viner and T. J. Cole, ‘Adult socioeconomic, educational, social, and psychological outcomes of childhood obesity: a national birth cohort study’ , British M edical Jo urn al (2005) 330 (7504): 13 5 4 . 1 3 1 . S. L. Gortmaker, A. Must, J. M . Perrin, A. M . Sobol and W. H. Dietz, ‘ Social and economic consequences of overweight in adolescence and young adulthood’, N e w England Jo u rn al o f M edicine (19 9 3) 329 (14): 1 0 0 8 - 1 2 . 1 3 2 . J. D. Sargent and D. G. Blanchflower, ‘ Obesity and stature in adoles­ cence and earnings in young adulthood. Analysis o f a British birth cohort’ , Archives o f Pediatric and Adolescent Medicine (1994) 148 (7): 6 8 1 - 7 . 1 3 3 . D. Thomas, ‘Fattism is the last bastion of employee discrimination’ , Personnel Today, 25 October 2005. 13 4 . J. Wardle and J. Griffith, ‘Socioeconomic status and weight control practices in British adults’ , Jo urn al o f Epidem iology and Community Health (2001) 55 (3): 18 5 - 9 0 . 1 3 5 . J. Sobal, B. Rauschenbach and E. A. Frongillo, ‘M arital status changes and body weight changes: a US longitudinal analysis’, Social Science and M edicine (2003) 56 (7): 1 5 4 3 - 5 5 . 13 6 . T. Smith, C. Stoddard and M . Barnes, ‘Why the Poor Get Fat: Weight Gain and Economic Insecurity’ , School of Economic Sciences Working Paper, Washington State University: http://ideas.repec.0rg/p/wsu/wpaper/ tgsmith-2.html (accessed 15 September 2008). 13 7 . B. Fisher, D. Dowding, K. E. Pickett and F. Fylan, ‘Health promotion at N H S breast cancer screening clinics in the U K ’ , Health Promotion International (2007) 22 (2): 1 3 7 - 4 5 . 13 8 . J. R. Speakman, H. Walker, L. Walker and D. M . Jackson,

‘Associations between B M I, social strata and the estimated energy content of foods’, Jo urn al o f Obesity and Related M etabolic Disorders (2.005) 29 (10): 1 2 8 1 - 8 . 13 9 . N. E. Adler, E. S. Epel, G. Castellazzo and J . R. Ickovics, ‘ Relation­ ship of subjective and objective social status with psychological and physiological functioning: preliminary data in healthy white women’, Health Psychology (zooo) 19 (6): 5 8 6 -9 2 . 14 0 . E. Goodman, N. E. Adler, S. R. Daniels, J. A. Morrison, G. B. Slap and L. M . Dolan, ‘Impact of objective and subjective social status on obesity in a biracial cohort of adolescents’, Obesity Review s (2003) 1 1 (8): 10 18 - 2 .6 . 1 4 1 . B. M artin, ‘Income inequality in Germany during the 1980s and 19 9 0 s’, R eview o f Income and Wealth (2000) 46 (1): 1 - 1 9 . 14 2 . V. Hesse, M . Voigt, A. Salzler, S. Steinberg, K . Friese, E. Keller, R. Gausche and R. Eisele, ‘Alterations in height, weight, and body mass index of newborns, children, and young adults in eastern Germany after German reunification’ , Jo u rn al o f Pediatrics (2003) 1 4 2 (3): 2 5 9 - 6 2 . 14 3 . S. Baum and K. Payea, Education Pays: The benefits o f higher education fo r individuals and society. Washington, D C : College Board, 2004. 14 4 . Bureau of Labor Statistics, Weekly and H ourly Earnings Data from the Current Population Survey. Washington, D C : US Department of Labor, 2007. 14 5 . M . Benn and F. M illar, A Comprehensive Future: Quality and equality fo r all our children. London: Compass, 2006. 14 6 . J . D. Teachman, ‘ Family background, educational resources, and edu­ cational attainment’, American Sociological R eview (1987) 52: 5 4 8 - 5 7 . 14 7 . O E C D and Statistics Canada, Literacy in the Inform ation Age: Final report o f the International Adult Literacy Survey. Paris: Organization for Economic Co-Operation and Development, 2000. 14 8 . R. Wilkinson and K. E. Pickett, ‘ Health inequalities and the UK Presidency of the E U ’ , Lancet (2006) 367 (9 517 ): 1 1 2 6 - 8 . 14 9 . R. G. Wilkinson and K. E. Pickett, ‘The problems o f relative depriv­ ation: why some societies do better than others’, Social Science and Medicine (2007) 65 (9): 19 6 5 - 7 8 . 15 0 . J. D. Willms, ‘ Quality and inequality in children’s literacy: the effects of families, schools, and communities’, in D. P. Keating and C. Hertzman (eds), D evelopm ental Health and the Wealth o f Nations. N ew York: Guilford Press, 1999. 1 5 1 . J. D. Willms, ‘Literacy proficiency o f youth: evidence of converging

socioeconomic gradients’ , International Jo urn al o f Educational Research (2003) 39: 2 4 7 - 5 2 . 15 2 . A. Siddiqi, I. Kawachi, L. Berkman, S. V. Subramanian and C. Hertzman, ‘Variation of socioeconomic gradients in children’s develop­ mental health across advanced capitalist societies: analysis of 22 O EC D nations’ , International Jo u rn al o f Health Services (2007) 37 (1): 6 3 - 8 7 . 1 5 3 . Centre for Longitudinal Studies, Disadvantaged Children up to a Year Behind by the Age o f Three. London: Institute of Education, 2007. 15 4 . R. H. Frank and A. S. Levine, Expenditure Cascades. Cornell University mimeograph. Ithaca: Cornell University, 2005. 1 5 5 . G. W. Evans and K. English, ‘The environment of poverty: multiple stressor exposure, psychophysiological stress, and socioemotional adjustment’, Child D evelopm ent (2002) 73 (4): 12 3 8 - 4 8 . 15 6 . P. Garrett, N . N g ’andu and J. Ferron, ‘Poverty experiences o f young children and the quality of their home environments’, Child Developm ent (1994) 65 (2, Spec, no.): 3 3 1 - 4 5 . 15 7 . V. C. M cLoyd, ‘The impact of economic hardship on black families and children: psychological distress, parenting, and socioemotional development’ , Child D evelopm ent (1990) 6 1 (2): 3 1 1 - 4 6 . 15 8 . V . C. M cLoyd and L. Wilson, ‘Maternal behavior, social support, and economic conditions as predictors of distress in children’ , N e w Directions fo r Child and Adolescent Developm ent (1990) 46: 4 9 -6 9 . 15 9 . A. Lareau, ‘ Invisible inequality: social class and childrearing in black families and white families’ , American Sociological R eview (2002) 67: 7 47- 7 6 . 16 0 . J. Currie, Welfare and the Well-being o f Children. Reading: Harwood Academic Publishers, 19 9 5. 1 6 1 . L. G. Irwin, A. Siddiqi and C. Hertzman, Early C hildhood Developm ent: A po w erful equalizer. Geneva: World Health Organization Commission on Social Determinants of Health, 2007. 16 2 . U N IC E F Innocenti Research Centre, A League Table o f E d u ­ cational Disadvantage in Rich Nations. Florence: Innocenti Report Card, 2002. 16 3 . K. H off and P. Pandey, B e lie f Systems and D urable Inequalities: An experimental investigation o f Indian caste. Policy Research Working Paper. Washington, D C : World Bank, 2004. 164. C. M . Steele and J. Aronson, ‘Stereotype threat and the intellectual test performance of African-Americans’ , Jo urn al o f Personality and Social Psychology (1995) 69: 7 9 7 - 8 1 1 . 16 5 . S. J. Spencer, C. M . Steele and D. M. Quinn, ‘Stereotype threat and

women’ s math performance’ , Jo u rn al o f Experim ental Social Psychology

(1999) 35 (1): 4 - 2.8­ 1 66. W. Peters, A Class D ivided: Then and now. N ew Haven: Yale University Press, 19 8 7 . 16 7 . J. Zull, The A rt o f Changing the Brain: Enriching the practice o f teaching by exploring the biology o f learning. Sterling: Stylus Publishing, 2002. 16 8 . G. Evans. Educational Failure and Working Class White Children in Britain. Basingstoke: Palgrave, 2006. 16 9 . L. Atkinson, ‘Sorry, Mum, we’re all pregnant!’ Sneak (2005) Issue no. 16 2 . 17 0 . J. Askill, ‘Meet the kid sisters’ , Sun, 23 M ay 2005. 1 7 1 . S. Carroll, ‘These girls’ babies are the real victims’, D aily M irror, 25 M ay, 2005. 17 2 . Committee on Adolescence AAoP, ‘Adolescent pregnancy - current trends and issues’, Pediatrics (1998) 10 3 : 5 1 6 - 2 0 . 1 7 3 . Social Exclusion Unit, Teenage Pregnancy. London: H M S O , 1999. 17 4 . D. A. Lawlor and M . Shaw, ‘Too much too young? Teenage pregnancy is not a public health problem’ , International Jo u rn al o f Epidem iology (2002) 3 1 (3): 5 5 2 - 4 . 1 7 5 . A. T. Geronimus, ‘The weathering hypothesis and the health of African-American women and infants: evidence and speculations’, Ethnicity and Disease (1992) 2 (3): 2 0 7 - 2 1 . 17 6 . A. T. Geronimus, ‘Black/white differences in the relationship of maternal age to birthweight: a population-based test o f the weathering hypothesis’ , Social Science and M edicine (1996) 42 (4): 5 8 9 -9 7 . 1 7 7 . J. Hobcraft and K. Kiernan, ‘ Childhood poverty, early mother­ hood and adult social exclusion’, British Jo urn al o f Sociology (2001)

52- (3): 4 9 5 - 5I 7- 17 8 . J . Rich-Edwards, ‘Teen pregnancy is not a public health crisis in the United States. It is time we made it one’, International Jo urn al o f Epidem iology (2002) 3 1 (3): 5 5 5 - 6 . 17 9 . S. Cater and L. Coleman, ‘Planned’ Teenage Pregnancy: Views and experiences o f young people from p o o r and disadvantaged backgrounds. Bristol: Policy Press for the Joseph Rowntree Foundation, 2006. 18 0 . K. Luker, Dubious Conception. The politics o f teenage pregnancy. Cambridge, M A : Harvard University Press, 1996. 1 8 1 . J . Ermisch and D. Pevalin, Who Has a Child as a Teenager? IS E R Working Papers, Number 2 0 0 3 - 3 0 . Institute for Economic and Social Research, University of Essex, 2003.

18 2 . U N IC E F Innocenti Research Centre, A League Table o f Teenage Births in Rich Nations. Florence: Innocenti Report Card, 2 0 0 1. 18 3 . S. J. Ventura, T. J. Mathews and B. E. Hamilton, ‘Teenage births in the United States: trends, 1 9 9 1 - 2 0 0 0 , an update’, N ational Vital Statistics Reports (2002) 50 (9). 18 4 . Alan Guttmacher Institute, US Teenage Pregnancy Statistics Overall Trends, Trends by Race and Ethnicity and State-by-state Information. N ew Y ork: A G I, Z004. 18 5 . K. E. Pickett, J. Mookherjee and R. G. Wilkinson, ‘Adolescent birth rates, total homicides, and income inequality in rich countries’, American Jo u rn al o f Public Health (2005) 95 (7): 1 1 8 1 - 3 . 18 6 . R. Gold, I. Kawachi, B. P. Kennedy, J. W. Lynch and F. A. Connell, ‘Ecological analysis of teen birth rates: association with community income and income inequality’, Maternal and Child Health Jo u rn al (2001) 5 (3): 1 6 1 - 7 . 18 7 . S. Ryan, K. Franzetta and J. M anlove, Hispanic Teen Pregnancy and Birth Rates: L o o k in g behind the numbers. Washington, D C : Child Trends, 2005. 18 8 . M . Dickson, Latina Teen Pregnancy: Problems and prevention. Executive summary. Washington, D C : Population Resource Center, 2 0 0 1. 18 9 . J . Bynner, P. Elias, A. McKnight, H. Pan and G. Pierre, Young P eo ple’s Changing Routes to Independence. Y ork: Joseph Rowntree Foundation, 2002. 19 0 . H. Graham and E. McDermott, ‘ Qualitative research and the evi­ dence base of policy: insights from studies o f teenage mothers in the U K ’ , Jo u rn al o f Social Policy (200J) 35: 2 1 - 3 7 . 1 9 1 . J. Belsky, L. Steinberg and P. Draper, ‘ Childhood experience, inter­ personal development, and reproductive strategy: an evolutionary theory of socialization’ , Child D evelopm ent ( 19 9 1) 62 (4): 6 4 7 -7 0 . 19 2 . R. Gold, B. Kennedy, F. Connell and I. Kawachi, ‘Teen births, income inequality, and social capital: developing an understanding of the causal pathway’ , Health and Place (2002) 8 (2): 7 7 - 8 3 . 19 3 . D. A. Coall and J. S. Chisholm, ‘ Evolutionary perspectives on pregnancy: maternal age at menarche and infant birth weight’ , Social Science and M edicine (2003) 57 (10): 1 7 7 1 - 8 1 . 19 4 . T. E. Moffitt, A. Caspi, J. Belsky and P. A. Silva, ‘Childhood ex­ perience and the onset of menarche: a test of a sociobiological model’, C hild Developm ent (1992) 63 (1): 4 7 - 5 8 . 19 5 . ‘American Academy of Pediatrics Committee on Adolescence: Adolescent pregnancy’ , Pediatrics (1989) 83 (1): 1 3 2 - 4 .

196. B. J. Ellis, J . E. Bates, K. A. Dodge, D. M . Fergusson, L. J . Horwood, G. S. Pettit and L. Woodward, ‘Does father absence place daughters at special risk for early sexual activity and teenage pregnancy?’ Child Developm ent (2003) 74 (3): 8 0 1 - 2 1 . 197. J . Borger, ‘ Gunned down: the teenager who dared to walk across his neighbour’ s prized law n’, Guardian, 22 March 2006. 198. J. Allen, Worry about Crime in England and Wales: Findings from the 2003/04 and 2004/05 British Crime Survey. London: Research Development and Statistics Directorate, Home Office, 2006. 199. C. Hale, ‘Fear of crime: a review of the literature’, International R eview o f Victimology (1996) 4: 7 9 - 1 5 0 . 2,00. H. Cronin, The Ant and the Peacock. Cambridge: Cambridge University Press, 1 9 9 1 . 2 0 1. J. Gilligan, Preventing Violence. New Y ork: Thames & Hudson, 2 0 0 1. 202. J. Gilligan, Violence: O ur deadly epidemic and its causes. N ew York: G. P. Putnam, 1996. 203. R. Wilkinson, ‘Why is violence more common where inequality is greater?’ Annals o f the N e w York Academy o f Sciences (2004) 10 3 6 : 1 - 1 2 . 204. M . Wilson and M . Daly, Homicide. Piscataway, N J : Aldine Transaction, 1988. 205. M . Daly and M . Wilson, ‘Crime and conflict: homicide in evolution­ ary psychological perspective’ , Crime and Justice (1997) 22: 5 1 - 1 0 0 . 206. M . D aly and M . Wilson, ‘Risk-taking, intrasexual competition, and homicide’ , Nebraska Symposium on Motivation (2001) 47: 1 - 3 6 . 207. M . Daly, M . Wilson and S. Vasdev, ‘Income inequality and homicide rates in Canada and the United States’, Canadian Jo u rn al o f Public Health - Revue canadienne de criminologie (2001) 43 (2): 2 1 9 - 3 6 . 208. M. Wilson and M . Daly, ‘Competitiveness, risk-taking and violence: the young male syndrome’, Ethology and Sociobiology (1985) 6: 5 9 - 7 3 . 209. D. M . Buss, The Evolution o f Desire: Strategies o f human mating. N ew York: Basic Books, 1994. 2 10 . P. Fajnzylber, D. Lederman and N. Loayza, ‘Inequality and violent crime \ Jo u rn al o f L a w and Econom ics (2002) 45: 1 - 4 0 . 2 1 1 . C.-C. Hsieh and M . D. Pugh, ‘Poverty, income inequality, and violent crime: A meta-analysis of recent aggregate data studies’, Criminal Justice R eview (19 9 3) 18 : 1 8 2 - 2 0 2 . 2 1 2 . United Nations Crime and Justice Information Network, Survey on Crime Trends and the Operations o f Criminal Justice Systems (Fifth, Sixth, Seventh, Eighth). N ew York: United Nations, 2000.

2 1 3 . Federal Bureau of Investigation, Crime in the United States. Washington, D C : US Government Printing Office, 19 9 0 -2 0 0 0 . 2 1 4 . M . Killias, J. van Kesteren and M . Rindlisbacher, ‘Guns, violent crime, and suicide in 2 1 countries’, Canadian Jo urn al o f Criminology (2001) 43: 4 2 9 - 4 8 . 2x 5 . U N Commission on Crime Prevention and Criminal Justice, ‘Criminal justice reform and strengthening of legal institutions measures to regulate firearms’, in Secretary-General Report E /C N .15/19 9 7/4. Vienna: United Nations, 19 9 7. 2 1 6 . M . Miller, D. Hemenway and D. Azrael, ‘ State-level homicide victimization rates in the US in relation to survey measures of household firearm ownership, 2 0 0 1 - 2 0 0 3 ’, Social Science and M edicine (2007) 64 (3): 6 5 6 -6 4 . 2 1 7 . ‘ Behavioural Risk Factor Surveillance Survey. Survey Results 20 01 for Nationwide: Firearms’ . North Carolina State Center for Health Statistics. http://www.schs.state.nc.us/SCHS/brfss/2001/us/firearm3.html (accessed 9 September 2008). 2 1 8 . D. Popenoe, L ife 'Without Father. New Y ork: Free Press, 1996. 2 1 9 . H. B. Biller, Fathers and Families: Paternal factors in child develop­ ment. Westport, C T : Auburn House, 19 9 3 . 220. S. R. Jaffee, T. E. Moffitt, A. Caspi and A. Taylor, ‘Life with (or without) father: the benefits of living with two biological parents depend on the father’s antisocial behavior’, Child D evelopm ent (2003) 7 4 (1): 1 0 9 - 2 6 . 2 2 1 . M . Anderson, J. Kaufman, T. R. Simon, L. Barrios, L. Paulozzi, G. Ryan, R. Hammond, W. Modzeleski, T. Feucht and L. Potter, ‘School- associated violent deaths in the United States, 1 9 9 4 - 1 9 9 9 ’ , Jo urn al o f the American M edical Association (2001) 286 (2 1): 2 6 9 5 -7 0 2 . 22 2. M . R. Leary, R. M . Kowalski, L. Smith and S. Phillips, ‘Teasing, rejection, and violence: case studies of the school shootings’ , Aggressive Behavior (2003) 29: 2 0 2 - 1 4 . 2 2 3 . C. Shaw and H. M cK ay, Ju ven ile D elinquency and Urban Areas. Chicago: University of Chicago Press, 19 4 2 . 224. R. Sampson, S. Raudenbush and F. Earls, ‘Neighborhoods and violent crime: a multilevel study of collective efficacy’ , Science (1997) 27 7: 9 1 8 - 2 4 . 2 2 5 . W. J . Wilson, The Truly Disadvantaged: The inner city, the under­ class, and public policy. Chicago: University of Chicago Press, 19 8 7 . 226. Federal Bureau of Investigation, Crime in the United States 2006. Washington, D C : US Government Printing Office, 2006.

2 2 7 . H. Boonstra, Teen Pregnancy: Trends and lessons learned. Guttmacher Report on Public Policy. Washington, D C : Alan Guttmacher Institute, 2002. 228. B. E. Hamilton, J . A. Martin and S. J. Ventura, ‘Births: preliminary data for 2006’, National Vital Statistics Report (2007) 56 (7). 229. R. V. Burkhauser, S. Feng and S. P. Jenkins, ‘Using the P90/P10 Index to Measure US Inequality Trends with Current Population Survey Data: A View from Inside the Census Bureau Vaults’, I Z A Discussion Paper N o. 2839, available at Social Science Research Network: http://ssrn.com/abstract=998222 (accessed 9 September 2008). 230 . C. Cantave, M . Vanouse and R. Harrison, Trends in Poverty. Washington, D C : Center for Political and Economic Studies, 1999. 2 3 1 . Child Trends DataBank, Children in Poverty. Washington, D C : Child Trends, 2003. 2 3 2 . A. Blumstein, F. P. Rivara and R. Rosenfeld, ‘The rise and decline of homicide - and w hy’, A nnual R e v ie w o f Public Health (2000) 2 1 : 5 0 5 - 4 1 . 2 3 3 . Annie E. Casey Foundation, KidsCount D atabook. Baltimore, M D . Annie E. Casey Foundation, 19 9 5. 234. C. G. Colen, A. T. Geronimus and M . G. Phipps, ‘Getting a piece of the pie? The economic boom of the 1990s and declining teen birth rates in the United States’ , Social Science and M edicine (2006) 63 (6): 1 5 3 1 - 4 5 . 2 3 5 . D. Dorling, ‘Prime suspect: murder in Britain’ , in P. Hillyard, C. Pantazis, S. Tombs, D. Gordon and D. Dorling (eds), Criminal Obsessions: Why harm matters more than crime. London: Crime and Society Foundation, 2005. 236. R. Walmsley. ‘An overview of world imprisonment: global prison populations, trends and solutions’, United Nations Programme Network Institutes Technical Assistance Workshop. Vienna, 2 0 0 1. 23 7 . R. Walmsley, World Prison Population List (6th and 7th editions). London: International Centre for Prison Studies, King’ s College, 2005 and 2006. 238. A. Blumstein and A. J. Beck, ‘Population growth in US prisons, 1 9 8 0 - 1 9 9 6 ’, Crime and Justice (1999) 26: 1 7 - 6 1 . 239. E. Chemerinsky, ‘Life in prison for shoplifting: cruel and unusual punishment’ , Human Rights (2004) 3 1 : 1 1 - 1 3 . 240. M . Hough, J. Jacobson and A. Millie, The Decision to Imprison: Sentencing and the prison population. Rethinking crime and punishment. London: Prison Reform Trust, 2003.

2 4 1. D. Downes, ‘The buckling of the shields: Dutch penal policy 1 9 8 5 - 1 9 9 5 ’, in R. P. Weiss and N . South (eds), Comparing Prison Systems: Towards a comparative and international penology. Amsterdam: Gordon & Breach Publishers, 1998. 2 4 2 . M . M auer, Comparative International Rates o f Incarceration: An examination o f causes and trends. Washington, D C : Sentencing Project, Z003. 2 4 3. US Department of Justice, Bureau of Justice Statistics, ‘ Incarceration rates for prisoners under State or Federal jurisdiction’ , File: co rp o p25.w k i. http://www.ojp.usdoj.gov/bjs/data/corpopz5.wk1 (accessed 30 March 2006). 2.44. W. S. Wooden and A. O. Ballan, ‘Adaptation strategies and transient niches of one middle-class inmate in prison’ , Psychological Reports (1996) 78 (3, Pt 1): 870. 24 5. The Sentencing Project, State Rates o f Incarceration by Race. Washington, D C : Sentencing Project, 2004. 246. R. Councell and J. Olagundoye. The Prison Population in 2.001: A statistical review. Home Office Findings 19 5 . London: Home Office, 2003. 247. Annie E. Casey Foundation, KidsCount D atabook. Baltimore, M D : Annie E. Casey Foundation, 2008. 248. Leadership Conference on Civil Rights and Leadership Conference on Civil Rights Education Fund, Justice on Trial: Racial disparities in the American criminal justice system. Washington, D C : L C C R / L C C R E F , 2000. 249. E. H. Johnson. ‘The Japanese experience: effects of decreasing resort to imprisonment’ , in R. P. Weiss and N. South (eds.), Comparing Prison Systems: Towards a comparative and international penology. Amsterdam: Gordon & Breach Publishers, 1998. 250. J. O. Haley, ‘Confession, repentence and absolution’, in: M . Wright and B. Galow ay (eds), M ediation and Criminal Justice. Newbury Park, C A : Sage, 1989. 2 5 1 . Amnesty International, Annual R eport - United States o f America. London: Amnesty International, 2004. 2 5 2 . Human Rights Watch and Amnesty International, The Rest o f Their Lives: L ife without parole fo r child offenders in the United States. New York: Human Rights Watch, 2005. 2 5 3 . Human Rights Watch, C old Storage: Super-maximum security confinement in Indiana. N ew Y ork: Human Rights Watch, 19 9 7 . 254. Human Rights Watch, R ed O nion State Prison: Super-maximum

security confinement in Virginia. N ew Y ork: Human Rights Watch, 1999. 255. United Nations Committee against Torture, Conclusions and Recommendations o f the Committee against Torture: United States o f America. Geneva: United Nations, 2006. 256. J. Irwin, The Warehouse Prison: Disposal o f the new dangerous class. Cary, N C : Roxbury Publishing Company, 2005. 2 5 7. Amnesty International, Ill-treatment o f Inmates in M aricopa County Jails, Arizona. London: Amnesty International, 19 9 7. 258. E. James, ‘A life again’, Guardian, 5 September 2005. 259. L. A. Rhodes. ‘Can there be “ best practices” in supermax?’ in D. Jones (ed.), Humane Prisons. Oxford: Radcliffe Publishing, 2006. 260. The Commission on Safety and Abuse in America’s Prisons, Confronting Confinement. N ew Y ork: Vera Institute of Justice, 2006. 26 1. P. Carter. Managing Offenders, Reducing Crime. Correctional Services R eview . London: Prime Minister’s Strategy Unit, 2003. 262. Home Office, E xplaining Reconviction Rates: A critical analysis. Home Office Research Study 13 6 . London: Home Office, 19 9 5. 263. S. Henry, ‘On the Effectiveness of Prison as Punishment. Incar­ ceration Nation: The warehousing of America’s poor’ . Ivy Tech State College, South Bend, Indiana: http://www.is.wayne.edu/stuarthenry/ Effectiveness_of_Punishment.htm, 2003. 264. E. Currie, Crime and Punishment in America. N ew York: Henry Holt & Co, 1998. 265. Youth Justice Board, Anti-social Behaviour Orders (B289). London: Youth Justice Board for England and Wales, 2006. 266. N C H , Tackling Anti-social Behaviour: Have w e got it right? London: N C H Children’s Charities, 2006. 267. K. Beckett and B. Western, ‘Governing social marginality’, in D. Garland (ed.), Mass Imprisonment: Social causes and consequences. London: Sage, 2 0 0 1. 268. D. Downes and K. Hansen, Welfare and Punishment: The relation­ ship between welfare spending and imprisonment. London: Crime and Society Foundation, 2006. 269. J. Silverman, ‘Does prison w ork?’ E S R C Society Today: Spotlights. http://www.esrc.ac.uk/ESRCInfoCentre/about/CI/CP/Our_Society_Today/ Sp0tlights-2oo6/prison.aspx?ComponentId= 1 6448&SourcePageId=i 6475 (accessed 9 September 2008). 270. M . Tonry, ‘Why are US incarceration rates so high?’ Crime and D elinquency (1999) 45: 4 1 9 - 3 7 . 2 7 1 . J. Blanden, P. Gregg and S. Machin, Intergenerational m obility in

Europe and North America. London: Centre for Economic Performance, London School of Economics, Z005. 2 7 2 . L. Mishel, J. Bernstein and S. Allegretto, The State o f Working America 1006/7. An Economic Policy Institute Book. Ithaca, N Y : IL R Press, an imprint of Cornell University Press, 2007. 27 3 . O E C D , Education at a Glance Z003. O E C D Indicators. Paris: O E C D , 2004. 274. D. S. M assey, ‘The age of extremes: concentrated affluence and poverty in the twenty-first century’, Demography (1996) 3 3: 3 9 5 - 4 1 2 . 27 5. P. A. Jargow sky, ‘Take the money and run: economic segregation in U.S. metropolitan areas’, American Sociological R eview (1996) 6 1 (6): 9 8 4 -8 . 276. P. A. Jargow sky, Poverty and Place: Ghettos, barrios and the American city. N ew Y ork: Russell Sage Foundation, 19 9 7. 2 7 7 . P. A. Jargow sky, Stunning Progress, H idden Problems: The dramatic decline o f concentrated poverty in the 1990s. The Living Cities Census Series. Washington, D C : Brookings Institution Press, 2003. 278. D. Dorling, ‘Why Trevor is wrong about race ghettos’, Observer, 25 September 2005. 279. D. Dorling, Human Geography o f the U K. London: Sage Publications, 2005. 280. D. Dorling and P. Rees, ‘A nation still dividing: the British census and social polarization’, Environm ent and Planning (2003) 35: 1 2 8 7 - 3 1 3 . 2 8 1. A. Berube, M ixed Communities in England. York: Joseph Rowntree Foundation, 2005. 2 8 2 .1. Kawachi, ‘Income inequality and economic residential segregation’, Jo u rn al o f Epidem iology and Community Health (2002) 5 6 (3): 1 6 5 - 6 . 28 3. S. M ayer, H o w the G rowth in Income Inequality Increased Econom ic Segregation. The Joint Center for Poverty Research Working Paper 235. Chicago: Northwestern University/University of Chicago, 2 0 0 1. 284. P. Lobmayer and R. G. Wilkinson, ‘Inequality, residential segrega­ tion by income, and mortality in US cities’ , Jo u rn al o f Epidem iology and Community Health (2002) 56 (3): 1 8 3 - 7 . 285. N. J. Waitzman and K. R. Smith, ‘ Separate but lethal: the effects of economic segregation on mortality in metropolitan America’, Milbank Quarterly (1998) 76 (3): 3 4 1 - 7 3 . 286. P. Bourdieu, Distinction: A social critique o f the judgem ent o f taste. London: Routledge, 1984. 287. K. Fox, Watching the English: The hidden rules o f English behav­ iour. London: Hodder 8c Stoughton, 2004.

288. J. Epstein, Snobbery: The American version. N ew Y ork: Houghton Mifflin Company, 2002. 289. R. Sennett and J. Cobb, The Hidden Injuries o f Class. N ew York: Alfred A. Knopf, 197Z. 290. S. J. Charlesworth, P. Gilfillan and R. G. Wilkinson, ‘Living inferior­ ity’, British M edical Bulletin (Z004) 69: 49-60. 2 9 1 . A. Marcus-Newhall, W. C. Pedersen, M . Carlson and N. Miller, ‘Displaced aggression is alive and well: a meta-analytic review’, jo u rn a l o f Personality and Social Psychology (zooo) 78 (4): 670-89. 292,. R. A. Baron, J. H. Neumann and A. Geddes, ‘Social and personal determinants of workplace aggression: evidence for the impact of per­ ceived injustice and the Type A behavior pattern’ , Aggressive Behavior (1999) 25 (4): 2 8 1 - 9 6 . 29 3. D. L. Horowitz, ‘Direct, displaced and cumulative ethnic aggression’, Comparative Politics (1973): 6 (1): 1 - 1 6 . 294. H. Crawley, Evidence on Attitudes to Asylum and Immigration: What w e know , d o n ’t kn ow and need to know. Working Paper No. Z3. Oxford: Centre on Migration, Policy and Society, University of Oxford, zoo 5. 295. J. L. Ireland, Bullying among Prisoners: Evidence, research and intervention strategies. Hove: Brunner-Routledge, zooz. 296. P. Earley, The H ot House: L ife inside Leavenworth prison. New York: Bantam, 19 9 2. 297. J. Sidanius and F. Pratto, Social Dominance. Cambridge: Cambridge University Press, 1999. 298. K. E. Pickett and R. G. Wilkinson, ‘ People like us: ethnic group density effects on health’, Ethnicity and Health (zoo8) 13 (4): 3Z I- 34 . 299. J . Boydell, J. van Os, K. McKenzie, J. Allardyce, R. Goel, R. G. McCreadie and R. M . M urray, ‘Incidence of schizophrenia in ethnic minorities in London: ecological study into interactions with environ­ ment’, British M edical Jo urn al (zo o i) 323 (7325): 1 3 3 6 - 8 . 300. J. Neeleman and S. Wessely, ‘Ethnic minority suicide: a small area geographical study in south London’ , Psychological M edicine (1999): 29 (2): 429 -36 . 3 0 1 . J . Neeleman, C. Wilson-Jones and S. Wessely, ‘Ethnic density and deliberate self-harm: a small area study in south east London’ , Jo urn al o f Epidem iology and Community Health (2001) 55: 85-90. 302. J. Fang, S. M adhavan, W. Bosworth and M . H. Alderman, ‘Residential segregation and mortality in N ew Y o rk City’, Social Science and M edicine (1998) 47 (4): 469-76. 303. L. Franzini and W. Spears, ‘Contributions o f social context to

inequalities in years of life lost to heart disease in Texas, U S A ’, Social Science and Medicine (2003) 57 (10): 1 8 4 7 - 6 1 . 304. C. M . Masi, L. C. Hawkley, Z . H. Piotrowski and K. E. Pickett, ‘Neighborhood economic disadvantage, violent crime, group density, and pregnancy outcomes in a diverse, urban population’, Social Science and M edicine (2007) 65 (12 ): 2 4 4 0 -5 7 . 305. K. E. Pickett, J. W. Collins, Jr ., C. M . M asi and R. G. Wilkinson, ‘The effects of racial density and income incongruity on pregnancy outcomes’, Social Science and M edicine (2005) 60 (10): 2 2 2 9 - 3 8 . 306. E. M . Roberts, ‘ Neighborhood social environments and the dis­ tribution of low birthweight in Chicago’, American Jo urn al o f Public Health (1997) 87 (4): 5 9 7 -6 0 3 . 307. L. C. Vinikoor, J. S. Kaufman, R. F. MacLehose and B. A. Laraia, ‘Effects of racial density and income incongruity on pregnancy outcomes in less segregated communities’, Social Science and M edicine (2008) 66 (2): 2 5 5 - 9 . 308. A. M . Jenny, K. C. Schoendorf and J . D. Parker, ‘The association between community context and mortality among Mexican-American infants’, Ethnicity and Disease (2001) n (4): 7 2 2 - 3 1 . 309. J. R. Dunn, B. Burgess and N. A. Ross, ‘Income distribution, public services expenditures, and all cause mortality in US States’ , Jo urn al o f Epidem iology and Community Health (2005) 59 (9): 7 6 8 -7 4 . 3 10 . A. Deaton and D. Lubotsky, ‘M ortality, inequality and race in American cities and states’ , Social Science and M edicine (2003) 56 (6): 1 1 3 9 - 5 3 . 3 1 1 . D. K. McLaughlin and C. S. Stokes, ‘Income inequality and mortality in US counties: does minority racial concentration matter?’ American Jo urn al o f Public Health (2002) 92 (1): 9 9 - 1 0 4 . 3 1 2 . R. Ram, ‘Income inequality, poverty, and population health: evi­ dence from recent data for the United States’ , Social Science and Medicine (2005) 6 1 (12): 2 5 6 8 -7 6 . 3 1 3 . S. V. Subramanian and I. Kawachi, ‘The association between state income inequality and worse health is not confounded by race’, International Jo urn al o f Epidem iology (2003) 3 2 (6): 1 0 2 2 - 8 . 3 1 4 . R. Ram, ‘Further examination of the cross-country association between income inequality and population health’, Social Science and M edicine (2006) 62 (3): 7 7 9 - 9 1 . 3 1 5 . J . Banks, M . Marmot, Z . Oldfield and J . P. Smith, ‘Disease and disadvantage in the United States and in England’ , Jo u rn al o f the American M edical Association (2006) 295 (17 ): 2 0 3 7 - 4 5 .

3 16 . J. Banks, M . Marmot, Z . Oldfield and J. P. Smith, ‘The SES Health Gradient on Both Sides o f the Atlantic’ . N B E R Working Paper 12 6 7 4 . Cambridge, M A : National Bureau of Economic Research, 2007. 3 1 7 . D. Vagero and O. Lundberg, ‘Health inequalities in Britain and Sweden’, Lancet (1989) 2 (8653): 3 5 - 6 . 3 18 . D. A. Leon, D. Vagero and P. O. Olausson, ‘Social class differences in infant mortality in Sweden: comparison with England and Wales’, British M edical Jo urn al (1992) 305 (6855): 6 8 7 - 9 1 . 3 1 9 . S. V. Subramanian and I. Kawachi, ‘Whose health is affected by income inequality? A multilevel interaction analysis of contemporaneous and lagged effects of state income inequality on individual self-rated health in the United States’, Health and Place (2006) 1 2 (2): 1 4 1 - 5 6 . 320. M . Wolfson, G. Kaplan, J. Lynch, N. Ross and E. Backlund, ‘Relation between income inequality and mortality: empirical demonstra­ tion’, British M edical Jo urn al (1999) 3 1 9 (7 2 15 ): 9 5 3 - 5 . 3 2 1 . S. J. Babones, ‘Income inequality and population health: correlation and causality’, Social Science and M edicine (2008) 66 (7): 1 6 1 4 - 2 6 . 3 2 2 . C. A. Shively and T. B. Clarkson, ‘ Social status and coronary artery atherosclerosis in female monkeys’, Arteriosclerosis and Thrombosis (I 9 9 4 ) 14 (5>: 7 2 i ~ 6. 3 2 3 . C. A. Shively and T. B. Clarkson, ‘Regional obesity and coronary artery atherosclerosis in females: a non-human primate model’ , Acta M edica Scandinavica, Supplement (1988) 7 2 3 : 7 1 - 8 . 324. M . Sahlins, Stone Age Economics. London: Routledge, 2003. 3 2 5 . T. Hobbes, Leviathan. Oxford: O xford University Press, 1998. 326. K. Jensen, J . Call and M . Tomasello, ‘ Chimpanzees are rational maximisers in an ultimatum game’, Science (2007) 3 18 (5847): 1 0 7 - 9 . 32 7 . J. Henrich, R. Boyd, S. Bowles, C. F. Camerer, E. Fehr, H. Gintis and R. McElreath, ‘Overview and synthesis’ , in J . Henrich, R. Boyd, S. Bowles, C. F. Camerer, E. Fehr and H. Gintis (eds), Foundations o f Human Sociality. O xford: Oxford University Press, 2004. 328. F. B. de Waal and F. Lanting, B on obo: The forgotten ape. Berkeley: University of California Press, 19 9 7. 329. E. A. D. Hammock, L. J. Young, ‘Microsatellite instability generates diversity in brain and sociobehavioral traits’, Science (2005) 308 (5728): 1 6 3 0 - 3 4 . 330 . J. B. Lassner, K. A. Matthews and C. M . Stoney, ‘Are cardiovascular reactors to asocial stress also reactors to social stress?’ Jo u rn al o f Personality and Social Psychology (1994) 66 (1): 6 9 - 7 7 . 3 3 1 . R. I. M . Dunbar, ‘Brains on two legs: group size and the evolution

of intelligence’ , in F. B. de Waal (ed.), Tree o f Origin: What primate behavior can tell us about human social evolution. Cambridge, M A : Harvard University Press, 2 0 0 1. 3 3 2 . C. Boehm, Hierarchy in the Forest: The evolution o f egalitarian behavior. Cambridge, M A : Harvard University Press, 1999. 3 3 3 . D. Erdal and A. Whiten, ‘Egalitarianism and Machiavellian intelli­ gence in human evolution’, in P. Mellars and K. Gibson K. (eds), M odelling the Early Human M ind. Cambridge: McDonald Institute Monographs,

1996. 3 34 . R. G. Wilkinson, The Im pact o f Inequality. N ew Y ork: N ew Press, 2005. 3 3 5 . J. Woodburn, ‘Egalitarian societies’ , Man (1982) 17 : 4 3 1 - 5 1 . 3 36. I. C. G. Weaver, N. Cervoni, F. A. Champagne, A. C. d’Alessio, S. Sharma, J. R. Seckl, S. Dymov, M . Szyf and M . J. Meaney, ‘Epigenetic programming by maternal behaviour’ , Nature Neuroscience (2004)

7: 8 4 7 -5 4 . 3 3 7 . S. M orris, ‘Women laughed as they forced toddlers to take part in “ dog fight’” , Guardian, 2 1 April 2007. 338. G. Rizzolatti and L. Craighero, ‘The mirror-neuron system’, Annual R eview o f Neuroscience (2004) 27: 1 6 9 - 7 2 . 339. M . Kosfeld, M . Heinrichs, P. J. Z ak , U. Fischbacher and E. Fehr, ‘Oxytocin increases trust in humans’ , Nature (2005) 4 35: 6 7 3 - 6 . 340. P. J. Z ak , R. Kurzban and W. Matzner, ‘The neurobiology of trust’, Annals o f the N e w York Academy o f Sciences (2004) 10 3 2 : 2 2 4 - 7 . 3 4 1. J. K. Rilling, G. A. Gutman, T. R. Zeh, G. Pagnoni, G. S. Berns and C. D. Kilts, ‘A neural basis for social cooperation’ , Neuron (2002)

35: 3 9 5 -4 0 5 . 342. N . I. Eisenberger and M . D. Lieberman, ‘Why rejection hurts’, Trends in Cognitive Science (2004) 8: 2 9 4 - 3 0 0 . 3 4 3. J. W. Ouwerkerk, P. A. M . van Lange and M . Gallucci, ‘Avoiding the social death penalty: ostracism and cooperation in social dilemmas’ , in K. D. Williams, J. P. Forgas and W. von Hippel (eds), The Social Outcast: Ostracism, social exclusion, rejection and bullying. N ew York:

Psychology Press, 2005. 344. World Bank, World Developm ent Indicators (W D I) September 2 006. Economic and Social Data Service International, Manchester: Mimas. 345. World Wildlife Fund, Livin g Planet R eport 2.006. Gland, Switzerland: W W F International, 2007. 346. R. M . Titmuss, Essays on the Welfare State. London: Unwin, 1958-

347. H. Daly, Steady-state Economics. Washington, D C : Island Press, 19 9 1.

348. M . Bookchin, The Ecology o f Freedom. Oakland, C A : A K Press, 2005. 349. R. G. Wilkinson, Poverty and Progress. London: Methuen, 19 7 3 . 350. H. C. Wallich, ‘Zero growth’, N ew sw eek, 24 January 19 7 2 . 3 5 1 . R. H. Frank, Falling Behind: H o w rising inequality harms the middle class. Berkeley, C A : University of California Press, 2007. 3 5 2 . S. Bowles and Y . Park, ‘Emulation, inequality, and work hours: was Thorsten Veblen right?’ Econom ic Jo u rn al (2005) 1 1 5 : F 3 9 7 - F 4 12 . 35 3 - D. Neumark and A. Postlethwaite, ‘ Relative income concerns and the rise in married women’ s employment’, Jo u rn al o f Public Econom ics (1998) 70: 1 5 7 - 8 3 .

354. Y . Park, Veblen Effects on L abo r Supply. M ale earnings inequality increases w om en ’s labor force participation. New London, C T : Depart­ ment o f Economics, Connecticut College, 2004. 3 55- S. J. Solnick and D. Hemenway, ‘Is more always better? A survey on positional concerns’, Jo u rn a l o f Econom ic B ehavior & Organization (1998) 37: 3 7 3 - 8 3 .

3 5 6 . T. Veblen, The Theory o f the Leisure Class. Oxford: Oxford University Press, 2007. 3 57 . Planet Ark, The Recycling O lym pic Report. Sydney: Planet Ark Environmental Foundation, 2004.

3 58. Vision of Humanity, G lo ba l Peace Index: M ethodology, results and findings. Cammeray, N SW : Vision of Humanity, 2007. 359 . G. B. Shaw, The Intelligent Woman’s G uide to Socialism and Capitalism. Edison, N J: Transaction Publishers, 2007. 360. M . Bloom, ‘The performance effects of pay dispersion on indi­ viduals and organizations’ , Academ y o f Management Jo urn al ( 19 9 1) 42: 2 5 - 4 0 .

3 6 1. J. P. Mackenbach, ‘Socio-economic inequalities in health in Western Europe , in J. Siegrist and M . M armot (eds), Social Inequalities in Health. Oxford: Oxford University Press, 2006. 362. O E C D , Social Expenditure - Aggregated Data. Vol. 2008, O E C D . Stat, 2 0 0 1.

363 T a x Foundation, State and L o ca l Tax Burdens Com pared to Other US States, 19 7 0 - 2 0 0 7 . Washington, D C : T ax Foundation, 2007. 364. Justice Policy Institute, Cellblocks or Classrooms? Washington, D C : Justice Policy Institute, 2002. 365. L. J. Schweinhart and D. P. Weikart, ‘Success by empowerment:

the High/Scope Perry Preschool Study through age 2 7 ’ , Young Children

(19 9 3) 49: 54~8. 366. World Bank, The East Asian Miracle. Oxford: Oxford University

Press, 1993- . 367. J. M . Page, ‘The East Asian miracle: an introduction’ , World

D evelopm ent ( 1 9 9 4 ) 2 2 ( 4 ) : 6 1 5 - 2 5 . 368. R. M . Titmuss, ‘War and social policy’ , in R. M . Titmuss (ed.), Essays on the Welfare State (3rd edition). London: Unwin, 19 7 6 . 369. L. M cCall and J. Brash, What do Americans Think about Inequality?

Working Paper. N ew Y ork: Demos, 2004. 370. J. Weeks, Inequality Trends in Some D eveloped O E C D Countries. Working Paper No. 6. N ew Y ork: United Nations Department of

Economic and Social Affairs, 2005. 3 7 1 . J. Benson, ‘A typology of Japanese enterprise unions’, British jo urn al o f Industrial Relations (1996) 34: 3 7 1 - 8 6 . 3 7 2 . L. Osberg and T. Smeeding, ‘ “ Fair” inequality? Attitudes to pay differentials: The United States in comparative perspective’ , American

Sociological R eview (2006) 7 1 : 4 5 0 - 7 3 . 3 7 3 . J. Finch, ‘The boardroom bonanza’, Guardian, 29 August 2007. 374. International Labour Organization, ‘Income inequalities in the age of financial globalization’ , World o f Work Report 2008. Geneva: ILO , 2008. 3 7 5 . Institute for Policy Studies, Annual C E O Compensation Survey. Washington, D C : Institute for Policy Studies, 2007. 376. United Nations Conference on Trade and Development, Are transnationals bigger than countries?’ Press release: T A D /IN F/P R /4 7- Geneva: United Nations Conference on Trade and Development, 2002. 3 77. T. Paine, The Rights o f Man. London: Penguin, 19 84 . 378. G. Alperovitz, America beyond Capitalism. Hoboken, N J: Wiley,

2004. , 379. W. Hutton. ‘Let’ s get rid of our silly fears of public ownership ,

Observer, 6 April 2008. 380. M . J. Conyon and R. B. Freeman, Shared Modes o f Compensation and Firm Performance: U K Evidence. N B E R Working Paper W8448. Cambridge, M A : National Bureau of Economic Research, 2 0 0 1. 3 8 1. A. Pendleton and C. Brewster, ‘Portfolio workforce’, People Management (2001) July: 3 8 - 4 0 . ^ 382. G. Gates, ‘Holding your own: the case for employee capitalism ,

Demos Quarterly (1996) 8: 8 - 1 0 . 38 3. P. M . Rooney, ‘Worker participation in employee owned firms’, Jo u rn al o f Econom ic Issues (1988) X X II (2): 4 5 1 - 8 .

384. J. L. Cotton, Em ployee Involvem ent: M ethods fo r im proving performance and work attitudes. Newbury Park, C A : Sage, 19 9 3 . 385. National Center for Employee Ownership, Em ployee Ownership and Corporate Performance: A comprehensive review o f the evidence. Oakland, C A : National Center for Employee Ownership, 2004. 386. J. Blasi, D. Kruse and A. Bernstein, In the Company o f Owners. New Y ork: Basic Books, 2003.

387. P. A. Kardas, A. Scharf and J. Keogh, Wealth and Income Consequences o f Em ployee Ownership. Olympia, W A : Washington State Department of Community, Trade and Economic Development, 199 8 . 388. R. Oakeshott, Jo b s and Fairness: The logic and experience o f employee ownership. Norwich: Michael Russell, 2000.

389. M . Quarrey and C. Rosen, ‘H ow well is employee ownership w ork­ ing?’ H arvard Business R eview (1987) Sep.-Oct.: 1 2 6 - 3 2 . 39°- T. Theorell, Democracy at work and its relationship to health’, in P. Perrewe and D. E. Ganster (eds), Em otional and Physiological Processes and Intervention Strategies: Research in occupational stress and w ell being, Volume 3. Greenwich, C T : J A I Press, 2003.

391- R. de Vogli, J. E. Ferrie, T. Chandola, M . Kivimaki and M . G. Marmot, ‘Unfairness and health: evidence from the Whitehall II Study’ , Jo u rn al o f Epidem iology and Community Health (2007) 6 1 (6): 5 1 3 - 1 8 . 392. D. Erdal, Local Heroes. London: Viking, 2008. 393- D. Erdal, ‘The Psychology o f Sharing: An evolutionary approach’ Unpublished PhD thesis, St Andrews, 2000.

394- S. Milgram, O bedience to Authority. N ew York: Harper, 1969. 395- L. T. Hobhouse, Liberalism. London: Williams &c Norgate, 1 9 1 1 . 396. D. Coyle, The Weightless World. Oxford: Capstone, 19 9 7 .' 397- K. E. Kiernan, F. K. Mensah. ‘Poverty, maternal depression, family status and children’s cognitive and behavioural development in early childhood: A longitudinal study’ . Jo u rn al o f Social Policy 2009; doi: 10.1017/S0047279409003250: 1-20. 398. J. Bradshaw, N. Finch. A Comparison o f C hild Benefit Packages in 22 Countries. Table 2.2. London: Department for Work and Pensions, 2002. 399- O E C D . Society at a glance 2009: O E C D Social Indicators. O E C D 2009.

400. I. Kawachi, B. P. Kennedy, K. Lochner, D. Prothrow-Stith. ‘ Social capital, income inequality, and mortality’ . Am J Public Health 1 9 9 7 - 87(9): 1 4 9 1 - 8 . ’

40 1. E. Uslaner. The m oral foundations o f trust. Cambridge: Cambridge University Press, 2002.

402. F. J. Elgar, W. Craig, W. Boyce, A. M organ, R. Vella-Zarb. ‘ Income inequality and school bullying: Multilevel study of adolescents in 37

countries’ . Journ al o f Adolescent Health 2009; 45(4): 35 1 _ 9 - 403. World Intellectual Property Organization. Intellectual property statistics, Publication A . Geneva: W IPO , z o o i. 404. Personal communication, R. De Vogli, D. Gimeno 2009. 405. R. G. Wilkinson, K. E. Pickett. Equality and sustainability. London: London Sustainable Development Commission, 2009. 406. J. Hills, T. Sefton, K. Stewart (eds). Towards a more equal society? Poverty, inequality and policy since 1 997. Bristol: Policy Press, 2009. 407. P. Krugman. The Conscience o f a Liberal: Reclaim ing America from the right. London: Penguin, 2009. 408. L. Bamfield, T. Horton. Understanding attitudes to tackling eco­ nomic inequality. York: Joseph Rowntree Foundation, 2009. 409. M . I. Norton, D. Ariely, ‘Building a better America - one wealth quintile at a time’ , Perspectives on Psychological Science: in press. 4 10 . D. Runciman, London R eview o f Books 2009, N o. 29; 22 Oct. 2009. 4 1 1 . J. H. Goldthorpe, ‘Analysing Social Inequality: A Critique of Two Recent Contributions from Economics and Epidemiology’, European Sociological R eview , 2009; doi: I0.i093/esr/jcp046 4 1 2 . N . Oreskes, E. M . Conway, Merchants o f D oubt: H o w a H andful o f Scientists Obscured the Truth on Issues from Tobacco Sm oke to Global Warming. New Y ork: Bloomsbury, 2 0 10 . 4 1 3 . The Equality Trust. The authors respond to questions about The Spirit L evel’s analysis, http://www.equalitytrust.org.uk/resources/ response-to-questions 2 0 10 . 4x4. M . Daly and M . Wilson, ‘ Cultural inertia, economic incentives, and the persistence of “ southern violence’” , in Evolution, culture, and the human mind. Edited by M . Schaller, A. Norenzayan, S. Heine, T. Yamagishi and T. Kameda. New York: Psychology Press. (2010) Pp. 2 2 9 -2 4 1. 4 1 5 . F. J. Elgar, N . Aitken, ‘ Income inequality, trust and homicide in 33 countries’ , European Jo urn al o f Public Health 2 0 10 ; d o i:io .i0 9 3 / eurpub/ckqo68. 4 16 . P. Saunders, Beware o f False Profits, Policy Exchange, London 2 0 10 . 4 1 7 . C. J . Snowdon, ‘ The Spirit L evel D elusion’ , Democracy Institute / Little Dice, London 2 0 10 . 4 18 . S. Hales, P. Howden-Chapman, C. Salmond, A. Woodward, J. Mackenbach, ‘National infant mortality rates in relation to gross national product and distribution of income’ , Lancet 19 9 9 ; 354 :2047.

4 19 . N . A. Ross, M . C. Wolfson, J . R. Dunn, J. M . Berthelot, G. A. Kaplan, J. W. Lynch, ‘ Relation between income inequality and mortality in Canada and in the United States: cross sectional assessment using cen­ sus data and vital statistics’ , British M edical Jo u rn al 2000; 320: 898-902. 420. P. Walberg, M . McKee, V. Shkolnikov, L. Chenet, D. A. Leon, ‘Economic change, crime, and mortality crisis in Russia: regional analysis’ , British M edical Jo u rn al 19 9 8; 3 1 7 (7 15 4 ) : 3 1 2 - 8 . 4 2 1 . X . P e i, E. Rodriguez, ‘Provincial income inequality and self-reported health status in China during 1 9 9 1 —7, Jo u rn al o f Epidem iology and Community Health 2006; 6 0 :10 6 5 -9 . 422. Y . Ichida, K. Kondo, H. Hirai, T. Hanibuchi, G. Yoshikawa, C. M urata, ‘Social capital, income inequality and self-rated health in Chita peninsula, Japan: a multilevel analysis o f older people in 25 communities’, Social Science & M edicine 2009; 69(4)1489-99. 423. S. V. Subramanian, I. Delgado, L. Jadue, J. Vega, I. Kawachi, ‘Income inequality and health: multilevel analysis o f Chilean communities’, Journ al o f Epidem iology and Community Health 2 o o 3 ;5 7 (u ):8 4 4 -8 . 4 24 - K- E. Pickett, R. G. Wilkinson, ‘ Child wellbeing and income inequal­ ity in rich societies: ecological cross sectional study’, British M edical Jo u rn al 2007; 3 3 5 (7629 ):io8o. 4 25. K. E. Pickett, O. W. Jam es, R. G. Wilkinson. Income inequality and the prevalence of mental illness: a preliminary international analysis. J Epidem iol Community Health 2006; 60(7): 646-7. 426. J . Blanden, How much can we learn from international comparisons of intergenerational mobility’, London: Centre for the Economics of Education, London School of Economics, 2009. 427. S. J. Babones, ‘ Income inequality and population health: Correlation and causality’, Social Science & M edicine 2008; 66(7): 1 6 14 - 2 6 . 428. D. Collison, C. Dey, G. Hannah, L. Stevenson, ‘Income inequality and child mortality in wealthy nations’, Jo urn al o f Public Health 2007; 29(2): X 14 -7 . 429. F. J . Elgar, ‘Income Inequality, trust, and population health in 33 countries’, American Jo u rn al o f Public Health, doi: 10 .2 10 5 / A JP H .2 0 0 9 .18 9 13 4 .

430. V. Hildebrand, P. Van Kerm, ‘ Income inequality and self-rated health status: evidence from the European Community Household Panel’, D em ography 2009; 46(4): 80 5 -25 . 4 3 1 . A. J. Idrovo, M . Ruiz-Rodriguez, A. P. Manzano-Patino, ‘Beyond the income inequality hypothesis and human health: a worldwide explo­ ration’ , Revista Salude Publica 2 0 10 ; 44(4): 69 5-70 2.

432. M . H. Jen, K. Jones, R. Johnston, ‘Compositional and contextual approaches to the study of health behaviour and outcomes: using multi­ level modelling to evaluate Wilkinson’s income inequality hypothesis’, Health & Place 2009; i5 ( i) :i9 8 - z o 3 . 4 3 3 . M . H. Jen, K. Jones, R. Johnston, ‘Global variations in health: evalu­ ating Wilkinson’s income inequality hypothesis using the World Values Survey’, Social Science & M edicine 2009; 68(4): 6 4 3 -5 3 . 434. M . Karlsson, T. Nilsson, C. H. Lyttkens, G. Leeson, ‘Income inequal­ ity and health: importance of a cross-country perspective’, Social Science & M edicine 2 0 10 ; 70(6): 8 7 5 -8 5. 435- D. Kim, I. Kawachi, S. V. Hoorn, M . Ezzati, ‘Is inequality at the heart of it? Cross-country associations of income inequality with cardio­ vascular diseases and risk factors’ , Social Science & M edicine 2008; 66(8): 1 7 1 9 - 3 2 . 436. A. Barford, D. Dorling, K. E. Pickett, ‘ Re-evaluating self-evaluation: A commentary on Jen, Jones, and Johnston (68: 4, 2009). Social Science & M edicine 2 0 10 ; 70(4): 49 6 -7. 4 3 7 . D. Dorling, A. Barford, ‘The inequality hypothesis: thesis, antithesis, and a synthesis?’ Health & Place 2009; 15(4): 1 1 6 6 - 9 . 438- N. Kondo, G. Sembajwe, I. Kawachi, R. M . van Dam, S. V. Subramanian, Z . Yam agata, ‘Income inequality, mortality, and self rated health: meta-analysis of multilevel studies’ , British M edical Jo u rn al 2009;

339;b4 4 7 I - 439. A. Clarkwest, ‘Neo-materialist theory and the temporal relationship between income inequality and longevity change’ , Social Science & M edicine 2008; 66 (9): 1 8 7 1 - 8 1 . 440. J. Holt-Lunstad, T. B. Smith, J. B. Layton, ‘Social relationships and mor­ tality risk: a meta-analytic review’. PLoS Medicine 2 0 10 ; 7(7 ):e io o o 3i6 . 4 4 1. D. Dorling, ‘Is more equal more green?’ Lecture to the Royal Geographical Society; 2 0 10 ; London, http://sasi.group.shef.ac.uk/ presentations/rgs/. 442. R. G. Wilkinson, K. E. Pickett, R. De Vogli, ‘A convenient truth’, British M edical Jo urn al in press. 443. J. T. Roberts, B. C. Parks, Climate o f Injustice: G lo ba l Inequality, North-South Politics and Climate Policy. Boston: The M IT Press, 2006. 444. J-S. You, S. Khagram, ‘ Comparative study of inequality and corrup­ tion’ , American Sociological R eview 2005; 70: 13 6 - 5 7 . 445. I. Kawachi, B. P. Kennedy, ‘Socioeconomic determinants of health : Health and social cohesion: why care about income inequality?’ British M edical Jo urn al 19 9 7 ; 3 1 4 : 10 3 7 .

446. B. Geysa, ‘Explaining voter turnout: A review o f aggregate-level research’ Electoral Studies, 2006; 25(4): 6 3 7 -6 6 3. 447. M . Iacoviello, ‘Household debt and income inequality 19 6 3 to 20 0 3 ’, Jo u rn al o f M oney, Credit and Banking 2008; 40 :9 29-6 5. 448. O ECD StatExtracts 2008, 2009 http://stats.oecd.org/Index.aspx? D ataS etC o d e= S N A h ttp://w w w .o ecdw ash .o rg/P U B S/E LE C T R O N IC / SAMPLES/natac_vol3_guide.pdf 449. D. Sington, The Flaw: Dartmouth Films, 2 0 10 . 450. B. M ilanovic, ‘Income inequality and speculative investment by the rich and poor in America led to the financial meltdown’, Tw o Views on the Cause of the Global Crisis - Part I. Y aleG lobal Online, 4 M ay 2009. 4 5 1 . D. M oss, ‘An Ounce of Prevention: Financial Regulation, M oral Hazard, and the End of “ Too Big to Fail” ’, H arvard Magazine, 2009; September-October 2009. 452. P. Krugman, ‘Inequality and crises: coincidence or causation?’ http:// www.princeton.edu/~pkrugman/inequality_crises.pdf 453- D. Cameron, the Guardian Hugo Young Lecture, 10 November 2009. 454. E. Miliband, Victory speech, Labour Party Conference, Manchester 25 September 2 0 10 . 4 55 . E. Miliband, BBC Radio 4, Today, 1 2 Ju ly 2 0 10 . 456 . The Equality Trust, http://www.equalitytrust.org.uk/pledge/ signatories

Index

abdominal fat see obesity, central abortion 12 3 -4 , Z4Z> I 43 Adam s, M ichael 58 addictions 67

see also alcohol abuse; drug abuse

A dler, Alfred 40 adolescents:

and body image 43 and crime 3 5 ,1 2 9 , 1 3 1 - 2 , 13 7 -8 ,

150, 154 depression and behavioural

problems 35, 43 education and inequality 10 5-8,

126 fast and slow lanes to adulthood

12 5 , 128 killings o f and by 129, 137 levels o f distress 43, 63 mental health 63, 68 obesity 98-9, 10 1 parenthood see teenage births and self-esteem 36, 43 sexual activity 1 19 , 1 2 1 , 12.6-7,

13 7 , 142; see also teenage births

unemployment 143 see also teenagers

adrenaline 1 1 5

advertising 40, 228 causing dissatisfaction 40, 70,

230 and inequality 40, 70, 228 and luxury goods 70, 227 and social com parisons 40, 230

affiliative strategies 207, 210 , 232 affluence:

‘ affluenza’ virus 69 diseases o f 10 original ‘ affluent society’ 270

African-Am ericans: arrest and detention rates 150 in Chicago heatwave 57 and crime 57, 150 de Tocqueville on 5 1- 2 imprisonment rates 150 and income inequality 185 interstate com parisons 185-6 life expectancy 80, 84, 185 maternal age 120 poverty concentration 162 proportion in populations 185-6 and student ability tests 1 1 3 , 194,

215 teenage births and pregnancy

120, 124, 142, 143 and thinness 99 trust, levels o f 57

aggression: in children 2 1 1 - 1 2 displaced 159, 166-8 downward flow of 166 and health 76 and inequality 140 and social capital 140

agriculture 224, 2 3 1 co-operatives 253 developing countries 30 and global warming 2 3 1 and inequality 208

innovations 267 aid see foreign aid AIDS 8 1, 99 Alabama 149, 174 Alan Guttmacher Institute 123 Alaska:

homicide rate and inequality 13 7 taxation 246

alcohol abuse and addiction 3, 19, 35, 65, 75

Alperovitz, Gar 252 altruism:

‘altruistic punishment’ 203 enforcing 237 see also others

animals 95, 194, 202 adapting to environment 210 and co-operation 201 epigenetic processes 2 1 1 ranking systems and social

organization 199-200, 204-5 anti-depressant drugs 65 anti-social behaviour 5, 1 3 1 , 138,

140 Anti-Social Behaviour Orders

(ASBOs) 154 anxiety 3 3-7 , 4 1, 168-9

and inequality 67-8

and narcissism 37, 39 rise in 3 3-7 , 39, 43, 67-8, 259

aristocracy 45, 25 1 Arizona 152 Arpaio, Joe 152 ASBOSs 154 Attention Deficit Hyperactivity

Disorder 63 Austen, Jane 29 Australia:

fear of crime 1 3 1 imprisonment rates 148 income differences 244 mental illness 67, 183 parental leave 1 1 2

Austria 145 autonomy see control

Bangladesh 80 Bangladeshis in UK 124 bankruptcy 227 behaviour changes 102 Belgium, literacy and equality

108-9, x8o belonging, sense of 97 Belsky, Jan 126 benefits 19 3, 245-6, 27 1

see also welfare Benn, Melissa, and Fiona Millar

105 Bermondsey 1 1 5 - 1 6 bicycling reaction 166-8, 207 biofuels 23 1 birth 214

see also infants birthweight:

and group density effect 168 and inequality 81 and stress in pregnancy 100

Bismarck 242

black Americans see African- Americans

Blanco, Kathleen 50

Blanden, J o 159 blood pressure 75

see also hypertension Blumstein, Alfred, and Allen Beck

147 B M I see Body M ass Index

body image 99-100 Body M ass Index 89, 91 bonding 214 bonobos 2.03-4 Bookchin, M u rray 224

Boston 165, 166 Botton, Alain de 69 Bourdieu, Pierre 163, 164 Bow les, Sam, and Y . Park 228,

229-30

brain: and group size 205-8 and social interaction 205-8,

2 1 3 - 1 6 and stress 86-7, 87

Brandt Report 2 17 breast cancer 27, 84 breastfeeding 262 British Crime Surveys 1 2 9 - 3 1 British M edical Jo u r n a l 81 ‘ broken heart’ 215 ‘ broken society’ 5 building societies 2 5 1, 253 bullying 90, 139 Buss, D avid 134 buying and selling 203

C alifornia 147 , 152 calorie intake 95 Canada:

education 109

equality levels 58, 183 homicide rates 135 imprisonment rates 147 income differences 244 levels o f trust and vehicle choice

58 mental illness 67, 183 social m obility 159

cancer 10, 73

breast 27, 84 causes and types o f 75, 84 and class 84 death rates 79, 84 and education 177

prostate 84 cap and share 224 capital punishment see death

penalty capitalism:

alternatives to 252-4 early 2 5 1- 2 and economic grow th 224

carbon emissions 2 17 equality and 2 2 1- 2 , 224, 2 3 1 rationing 2 2 1- 2 reducing n , 2 2 1- 2 , 232 see also global warm ing

cardiovascular disease 10, 194 and social integration 76 and thrifty phenotype 100 see also heart disease

C aribbean communities (UK) 124 C arl Zeiss 258 cars:

changing values 58, 226 and class and segregation 57-8,

164-5 fuel-efficient 223-4 ownership 25 as status markers 40, 57-8

cars: - cont. SUVs 57-8

C arver, George W ashington 97 caste:

and perform ance 1 1 3 - 1 4 , 194 and social m obility 15 7

causality, inequality and: biological plausibility 179-82 confounding 17 5 -7 , 190—91 education 10 8 -10 , 1 1 0 - 1 1 , 1 1 3 - 1 5 health and social problems

16 5-8 , 18 3 -7 , 194 imprisonment 14 5 - 8 ,15 0 - 5 3 mental illness 69-70, 7 1 - 2 obesity 95-8, 100 predictive pow er 195 reverse causality 54-6, 6 1- 2 ,

184-8, 1 9 1 - 3

social evaluative threat 3 7 -4 1 social m obility 159-69 stress and health 84, 85-7 teenage births 13 3 - 4 trust 54-6, 6 1- 2

violence 12 7 -8 celebrities 18 1

see also rich people charities 253

and inequality 60 Charlesworth, Simon 165

Chicago: heatwave (1995) 57 homicide and other death rates

77- 8, 1 3 1 neighbourhoods and ethnic

groups 140 Chicago School sociologists 140 child development see children Child Wellbeing, U N I C E F

index/report 23-4 , 92, 192 on children’s aspirations 1 1 5 - 1 6

incomes 23-4 inequality 23-4, 25, 1 1 5 - 1 6 , 139,

192, 212 obesity 92 violence and conflict 139

childbirth 2 14 children:

abuse (physical and emotional) h i , 138

anxiety levels 33-4 , 63 ‘dam aged’ 2 1 1 development (emotional, social

and cognitive) 7 7 , 1 1 0 - 1 2 early years learning and

development 5, 1 1 0 - 1 1 , 2 1 0 - 1 1 , 247

and fairness 208 and fam ily breakdown 63, 96,

h i grow th 77 imprisonment o f 152 infectious diseases 10 mental health 63, 68

neglect 1 1 1 , 138 obesity 89-90, 9 2-3, 96, 10 1,

2 12 pride and shame in 41 relative poverty 23 social relations 2 1 1 stress see stress, in early life and violence and conflict i n ,

*37> 13 9 - 4 °, 2 1 1 - 1 2 see also educational

perform ance; infant m ortality; parenting

Chile 251 chimpanzees 202, 203-5 China:

Communist government 241 2008 earthquake 50

inequality 142, 179 modesty 45

civic engagement 55, 126, 249 see also community life

civil servants, health o f 75-6 , 84, 194, 256

class: based on material inequality

26-9 ‘classless society’ aim 242 cultural differences 28, 16 3-6 ,

185 and death rates 84, 178, 182 decline o f deference 267-8 and education 1 1 3 - 1 7 , 1 2 5 and employee buy-outs 258 and employment 16 5-6 , 258 and health 10, 84, 177—8, 215 hostility 129, 16 5-6 identity 28 and imprisonment 1 4 9 - 5 0 ,154m and life expectancy 8 1-4 markers 28, 16 3-6 , 185 and obesity see obesity and parenting m prejudices 45, 52, 16 3 -6 , 185 and social status 40 see also social mobility

climate change see carbon emissions; global warm ing

clothes 27, 1 4 1 , 218 co-operation:

and the brain 2 1 4 - 15

experiments in 20 2-3, 2'I 4- I 5 and global warm ing 2 3 1- 3 human potential for 201 maintaining 215 and self-realization 20 9 -10 w ithout contact 214 see also trust

Co-operative Bank and stores (U K) 253

co-operatives 25 2 -3, 258, 260 cocaine 70, 72, 9 6 ,19 4

cognitive behavioural therapy 238 cold virus 76

see also immune system Colen, Cynthia 143 C olorado 92-3 communism 190, 244

see also state ownership communities:

decline and break-up o f 5, 42 with high trust levels 57 replaced by mass society 42

Com m unity Development C orporations 253

community life: and consumerism 229-30 and inequality 45, 70, 195, 199,

230 involvem ent in 45, 56-7, 78-9,

10 3, 199, 2 15 , 230; see also social capital; voluntary activities

and mental health 70 companies see corporations com parisons, invidious 70, 230 competition:

international 193 for resources 200-201, 204-5 fo r status see status

Confederate states 188 confidence see self-esteem conflict:

avoiding 2 0 1-2 , 204-5 children and see children and health see stress and resources 202, 204 see also status, competition

Connecticut 13 7 consumption and consumerism:

competitive 2 1 7 - 1 9 , 228, 230, 270 ‘conspicuous’ 43, 230 and financial crash 270 and inequality 2 17 , 224 and new technology 264 restraining 269-70 and social needs 230 and social stratification 230 spending addiction 3, 43, 226,

230 taxation and 27 1

see also goods contraception 142 contraction and convergence 224 control, sense of:

and behaviour changes 102

and health 75-6 , 256 conviction rates 154 copying, digital 265 corporal punishment 268 corporations 233, 249-52

democracy and 256-8 and the environment 262 ethical issues 254, 2 6 1-2 and new technologies 264-7 size o f 25 1 see also executive incomes

corruption 252, 262 cortisol 38-9, 9 5 ,1 1 5 counter-dominance strategies

208 C oyle, Diane 264 credit cards see indebtedness credit unions 2 5 1, 253 crime:

deterrence and reduction 15 5 -6 , 238

drug-related 147

education and 1 1 2 environmental differences 13 2 ethnic minorities and 150 and fatherlessness 137 fear o f 54, 57-8, 129 -30 , 156 fluctuations in 85, 13 2 , 147, 190 high-school massacres 139

and inequality 4 5 ,1 3 4 - 7 , 14 0 -4 1,

142, 14 3 -4 , i 47, i 5°> 155, 194, 199

organized 1 5 1 property 5 4 , 1 2 9 , 1 3 5 , 1 3 7 , 1 4 7 ,

150 repeat offending and recidivism

1 5 4 - 5 and social cohesion 85 and trust 57 in w orld w ars 85 youth see adolescents

Cronin, Helena 1 3 1 C uba 220 cultural markers 16 3 -6 , 185 Currie, Elliott 155

Daim lerChrysler 251 D aly, Herm an 224 de Tocqueville, A lexis:

on equality 49, 50-52, 56, 209 on slavery 5 1 - 2 , 209

de W aal, Frans, and Frans Lanting 204-5

death penalty: abolition 268 and inequality 149, 150, 153

death rates: economic segration and 163 and inequality 27, 60, 178-80,

182, 195 international com parisons 176,

178-9

and lo w status 75-6, 84, 178 related to income and

disadvantage i 2 - r 3 , 163, 178-80

and social needs 5 w om en’s 60 see also suicide

debt: bankruptcy 228 and consumption 270 credit-card 270 financial 228

Declaration o f Independence 263

delinquency, juvenile see adolescents

democracy: and capitalism 2 5 1- 2 development o f 267-8 institutions 197 and large corporations 25 1 w orkplace 256-8

denationalization 250 Denm ark 25 1

advertising 228 foreign aid 60 prison population 145 see also Scandinavian countries

depression 35, 71 causes o f 65, 75 children and adolescents 35, 43,

63 and inequality and low status 35,

65, 75, 81 research 35, 68 see also anxiety; mental illness

deprivation/disadvantage: and high-quality education

1 1 2 problems related to 176, 239

and teenage motherhood 1 2 1 - 5 , 128

see also neighbourhoods developed countries 30, 35

see also rich countries developing/less developed countries

30 equal countries and 6 1, 2 3 5 -7 heart disease 189 inequality in 30, 189 and new technologies 267 obesity 9 9 ,18 9 social class 99 U K and 61 violence 189 see also foreign aid; poorer

countries development aid see foreign aid diabetes 100

and education 177 stress and 86, 100, 2 12

Dickerson, Sally, and M argaret Kem eny 38-9, 4X

diet: and inequality 1 0 1 - 2 modern 75, 9 0-9 1, 96-8 and obesity see food and social problems 25 and status 9 7 - 8 ,16 3 and stress 95-6 in w orld w ars 84-5 see also food

digital products 264-5 digitization 225 disadvantage see deprivation;

neighbourhoods disasters, reactions to:

C hicago 57 China 51 early Americans 51

disasters, reactions to: - cont. Hurricane K atrina 5 0 - 5 1, 54

discrimination: class see snobbery damage to wellbeing 168 in employment 43 and ethnicity 185 opposition to '268

taste 165 see also racism ; social exclusion

distinction, social 227 division o f labour 20 1, 210 divorce:

children and 96, i n and inequality i n , 137 and obesity 99

domestic conflict i n see also fam ily

dominance: among animals 200, 202-5 hierarchies see hierarchies and inequality 168 strategies 207

Donne, Jo h n 17 3 , 18 1 dopamine 7 1- 2 , 1 1 5 , 194 D orling, Danny 143 D ostoevsky, Fyodor 145 D ouglas, Frederick 129 D ownes, D avid 147, 1 5 1 D ownes, D avid, and Kirstine

Hansen 15 5 -6 drop-outs (school) 1 0 7 - 8 , 1 3 8 ,1 4 3 drug abuse and addiction 3, 19, 65,

70-72, 96, 99 adolescents 35 celebrities 18 1 dealing and trafficking 140, 1 5 1 death from overdose 70 drug-related crimes 147, 150 governments and 19 3, 238, 239

and inequality 7 0 - 7 1, 246 rehabilitation programmes 150,

238 stimulation o f the brain 96 and violence 142

drugs (medical) 262, 266, 267 drugs (recreational) see drug abuse

and addictions D unbar, Robin 205 Durkheim , Emile 76

Earley, Pete 167 early human evolution and

societies 26, 2 0 1- 2 , 204-5,

206-7, 2°6 - 7 , 2 I 5> z7 ° East Asia 241 Eastern European countries:

inequality in 190 life expectancy 190 state ownership 252, 263

eating disorders 63, 95-6 see also food

ecological footprints 220, 221 economic development:

and happiness 6 -10 and inequality 189, 237 and innovation 225 and life expectancy 6-7

economic downturn see financial crisis

economic experiments 202-3 economic growth:

benefits and costs o f 8 -10 , 2 1 7 - 1 8 , 228—30, 269-70

as determinant 189 end o f benefits o f 5 - 14 , 254, 272 environmental issues 8, 2 3 1 and inequality 228 limits on 1 1 in poorer countries 2 17

and quality o f life 1 1 - 1 4 , 2 3 1- 3 and trust 55

economic institutions see financial institutions

economists, and inequality 193 ecosystems, destruction o f 262 education:

benefits o f 10 3, 177 early childhood programmes

1 1 2 , 247 equality see educational

opportunity and health 103, 177 higher 59-60, 16 1 importance o f 10 3, 1 1 5 and new technologies 262-6 public spending on 159, 16 1, 247 school drop-out rates 10 7-8 , 138,

143 and social mobility 159, 16 1 and status 16 1 wom en and 59-60

educational opportunity: access to higher education 59,

16 1 equality o f 10 5-8 , 1 1 2 international com parisons 10 5 -7 ,

179 and social problems 25

educational performance 19, 1 0 3 - 1 7 and fam ily background 10 3-4 ,

138 and imprisonment 149 and inequality 19, 30, 10 8 -10 ,

14 3 , 16 1 , 2 12 , standards o f 10 8 -10

egalitarian policies and societies

197 alternatives to capitalism 2 5 1- 3 early 20 1, 207-8, 270

evolution and 20 6 -10 , 237, 239 and global warm ing 2 3 1 wartim e Britain 222 see also equal/more equal

societies Eisenberger, N aom i 2 14 elderly people 143

in Chicago heatwave 57 fear o f strangers 1 3 1

electronic communications 225, 265 elitism 164 Elliott, Ja n e 1 1 4 Ellis, Bruce 127 embarrassment 4 1, 133

see also shame and humiliation Emerson, R alph W aldo 3 1, 43-4 empathy:

in childhood 2 1 1 , 2 13 and equality 168 and inequality 5 1, 56 and m irror neurons 2 13 strategies for 210 with those not ‘like us’ 5 1, 56,

209 employee-ownership schemes

255-63 and employee buy-outs 257-8,

261 ethical standards in 2 6 1-3 and local communities 261 regulating sale o f shares 261 types o f 260-61

employment see unemployment; w ork

emulation 228-30 England 177-80

see also United Kingdom English-speaking countries:

inequality 183 mental health 183

environmental issues see carbon emissions; global warm ing

epidemiological transition 73 epidemiology xiii epigenetic processes 2 1 1 Epstein, Joseph 164-5 equal/more equal societies 26,

173-7 achievement in 237-6 benefits for all inhabitants 84,

176, 179-80 and the environment 232, 235

inclusiveness and empathy in 168 international policies 235 poor people in 176 see also hunter gatherers

equality: attainability 197, 2 1 3 , 267-8 benefits from im proving 29-30,

33, 62, 1 0 2 ,1 7 5 - 8 2 ,1 9 2 o f conditions 5 0 - 5 1, 209 and counter-dominance

strategies 208 early societies and 200 and economic grow th 226 and government changes

2 - 5 4 - 5

groups committed to x v , 242, 269 and human differences 237 as ideal 4 5 ,19 7 o f incomes 18, 55 and liberty 263 and merit 237 o f opportunity see equality o f

opportunity responses to 236-7 routes to 184, 197, 2 1 3 , 236-7 and sustainability 2 1 7 - 3 1 unstoppable trend 267-8

equality o f opportunity 263, 267-8

ideology of/distant prospect 157, 164, 169

and social m obility 15 7 ,16 9 Equality Trust x v , x x i, 269 Erdal, D avid 258, 260 ethnic differences 1 1 3 , 150 ethnic minorities 143

discrimination and social exclusion 18 5 -6

exclusion from education and employment 185

and group density effect 168-9

income differences 18 5 -7 and inequality 185-6 and mental illness 68-9, 168

ethnicity: and health 186 and inequality 1 7 9 ,1 8 5 - 7 and status 18 5 -7 and teenage births 124

Euripides 3 European and W orld Values

Survey 53

Evans, Gillian 1 1 5 - 1 7 evidence-based medicine/politics xiii evolution see early human

evolution exchange, forms o f 202-3 executive incomes 24 9 -51, 254,

259-60, 269-70 exercise/physical activity, lack of

7 5

and low status 75 and obesity 90, 95, 102

experiments 1 1 3 - 1 4 , 194 economic 20 2-3, 2.09 in obedience 262 on students 1 1 3 - 1 5 , 194, 215 in trust 214

Ezzati, Professor M ajid 93m

face, loss o f see shame and humiliation

fair trade 254, 261 fairness:

basic human principle 242, 270 evolution o f 208-9

family: breakdow n 63, 96, i n , 13 7 and education see educational

performance ‘extended’ or ‘ nuclear’ systems

209 and inequality 1 1 0 - 1 2 , 126, 138,

2 12 support measures 1 1 2 , 247 values 238 and violence i n , 137

fathers absent 12 6 -7 , I 37- 8 biological compared with step­

parents 138 important roles o f 13 7 -8 negative roles 138 and social mobility 159 teenage 12 7 -8

fear o f crime see crime reducing 237 see also mistrust

feudalism 45 and social mobility 157

fight-or-flight response 85 financial crisis (2008) 5 ,1 6 2 , 228,

270 financial institutions 5, 249

non-profit 250, 252-3 privatization 2 5 1, 253

Finland: child wellbeing 212 equality 180, 268

gun ownership 136 homicide rate 136 literacy 10 8 -10 , 180 penal policies 145, 156 stress and eating 95 teenage births 125 see also Scandinavian countries

food: and biofuel crops 2 3 1 cheap, energy-dense 9 1, 97 as comfort 89, 95, 95-6 insecurity 30 malnutrition 189 modern habits 3, 75, 9 0-9 1, 96-8 and obesity 9 0-91, 10 0 -10 2 rising prices 2 3 1 sharing and eating together 96,

200, 20 2-3, 2°8 j 209 social benefits and drawbacks

97 and status 199 see also agriculture; diet

foraging societies 208 see also hunter gatherers

foreign aid 6 1, 232 and inequality 6 0 -6 1, 232, 235

Fortune 500 companies 250 Fox, Kate 164 France 145 franchise, extension of 268 Frank, Robert 70, 226, 227 Frank, Robert, and Adam Levine

i n fraternity 45, 52, 209, 263 ‘free’ sector 265 freedom 252, 263

of speech 197 French Revolution 45, 209, 263 friendly societies 2 5 1, 253 friendship 3, 65, 197-202, 20 7-10

friendship: - cont. and death rates 76 and health 39, 45, 65, 7 6 ,16 9 ,

202, 2.15 preference for near equals 5 1,

199 and social status 199-202 strategies for 206, 2 0 7 -10 and w o rk 259 see also trust; social relations

Gallup polls 249 gated communities 48, 58 gender differences see sex/gender

differences G en eral Social Survey:

on death rates 78-9 on trust levels 53-4

geographical m obility 42, 259 geographical segregation 159,

16 2-3 Germany:

body-mass index changes 10 1 imprisonment rates 147 inequalities after reunification

10 1 mental illness 67 prison population 145 social insurance 242 social mobility 159 unification 242

gifts: and equality and peace-keeping

2 0 1-2 , 208-9 and friendship 197, 200, 209

Gilligan, Jam es 1 3 3 ,1 4 8 , 154 Gini coefficient o f inequality 17 - 1 8 ,

56 G lobal Peace Index 232, 235 global warm ing 2 17 - 3 2 , 27 1

attitudes to 2 3 0 -3 1 concern about 223 defiance o f regulations 2 3 0 -3 1 and economic grow th n , 2 1 7 - 3 1 equality movement and 269, 271 and inequality 270 see also carbon emissions

G old, Rachel 12 3 -4 , Good Childhood Inquiry 63 goods:

changing evaluation o f 227, 230 economy o f cultural 164 luxury 70, 227 perceived as second-rate 30, 227 striving for 70, 227 valued for status and prestige

30, 70, 218 , 227, 2 3 0 -3 1 visibility of 230 see also consumption and

consumerism Gore-tex 258 governments:

and corporations 262 and early childhood programmes

1 1 2 and employee ownership 261,

270 as employers 263 and global warming 226, 2 3 0 -3 1 and new technologies 265-6 and poverty 268

gradients: class 27 health 1 2 - 1 3 , 27 social 1 2 - 1 3 , 24- 55 z7t 68, 76, 9 1,

98, 99, 108, 149, 17 5 -8 2 , 195 graffiti and litter 140 Graham, Hilary, and Elizabeth

McDermott 12 5 , 128 Graham, Sheila 89

graphs: cross-sectional 190 interpreting x ix - x x i

‘ G reat C rash ’ (1929) 243 Greece:

health care and incomes 79 infant m ortality 80 life expectancy 6-7, 79-80 literacy n o teenage births 124

greed and avarice 199 group conflict 168 group density effects 159-60, 168-9 group membership 79 group size 205-8 grow th see height and growth Guatem ala 251 gun ownership 13 6 -7 , 142

Hansson, Albin 242 happiness levels:

changes over time 8 cultural differences 8 economic grow th and 6 -10 and health 76 and incomes see income

inequality see also wellbeing

Harlem 80, 9 7 ,17 5 H aw aii 150 H D I see United N ations Human

Development Index health:

and class see class determinants o f 73-7 and education 10 3, 177 and income see income

inequality inequalities xiii, 28-9, 73-87,

17 7 -8 0 , 238-9

and social relationships 76-7, 195, 202

and stress see stress and trust see trust

Health Behaviour in School-age Children survey 92

health care 26, 79 -8 1 doctors’ incomes 79-80 free 268 and health and social problems

238 high tech 79 -8 1 inequalities in 7 9 -8 1, 8 1-4

health and social problems index 19 -2 2 , 18 5-6 , 192, 267-8

and average incomes 19 1 and inequality i j y , 195 for nations 19 - 2 1 and public social expenditure

184, 19 1 US states 19, 2 1 - 2

Health Survey for England 98 heart disease:

and class or status 10, 7 3-6 , 85,

177 and death rates 79, 84 and education 177 and group density effect 168 and stress 75, 86, 212 see also cardiovascular disease

height and grow th failure and inequality 81 and stress 86

hierarchies: among animals see primates at w o rk 249-50, 259 and confidence 40 dominance 2 0 3 -10 employers and 249, 259 evolution and 206-7

hierarchies: - cont. importance o f 27 lack o f data 27 lo w position in 69, 70, 75-6 in prison 167 and social m obility 163 and social systems 203

Hispanic people 177 and community life 57 and food 97-8 and judicial processes 150 teenage births and poverty 124 trust, levels o f 57

H obbes, Thom as 20 1, 204 Hobhouse, L. T . 263 H o ff, K arla, and Priyanka Pandey

1 1 3 homeless people 97 homeostasis 85 homicide rates 19 , 77-8, 1 3 1 - 2

age and sex 1 3 1 - 2 and class 84 fluctuations 1 3 1 , 142 and inequality 13 5 - 7 , 142, 192 international com parisons 1 3 1 - 2 ,

135-7 suicide and 175 use o f guns 13 7

H ong Kong 241 hormones see stress housing problems 25, i n Hsieh, C .- C ., and M . D. Pugh 135 human brother/sisterhood see

fraternity human nature and inequality 199 human resource professionals 98-9 Hum an Rights Watch 152 hum iliation see shame and

humiliation hunter gatherers 26, 2 0 1-2 , 2 0 7 -10

see also early human evolution Hurricane K atrina 49-50, 54, 57 H utton, Will 253 hypertension 75, 86, 177

Iceland 267 identity:

class 28 and in-groups 209 social 42

illegal drugs see drug abuse and addiction

immigrants 140, 167, 248

immune system 85-6 imprisonment 19, 145-56

as deterrent 15 3 -6 factors affecting 14 5-8 , 15 5 , 192,

246 international com parisons

145-50 , i 54- 5> 176 limited effectiveness o f 15 4 -5 ,

247 purposes o f 1 5 3 - 4 ^ rates o f see prison populations re-offending rates 154 -5 social gradient 149

inclusiveness, and equality 168 income:

aspirations for 13 , 229 and education 103 grow th in 5 - 10 and health and social problems

2 1 - 2 , 1 9 1- 2 and obesity 100 see also executive incomes;

income differences; pay; wealth

income differences and inequality 179, 184, 19 2 - 3 , 249-52, 259, 27 1

and basis o f social hierarchy z 6-9

changes in 142, 160, 169, 184, 234-6, 239, 249, 2 5 0 -5 1

and children’ s aspirations 1 1 6 - 1 7 de Tocqueville on 5 1 - 2 employers and 242 governments and 5, 18 3-4 ,

19 2 - 3 , 241 and health and happiness 8,

1 1 - 1 4 . I 5, 20-22, 27 -30 , 87, r68, 17 7 , 183-6 , 182, 2 12 , 227, 229-30

measures o f 15 - 18 and psychosocial wellbeing 5,

238 routes to narrow er difference

18 3 -4 , 24 1—2 and trust 45, 5 1-6 2 varying patterns 14 2 -3 in wartim e 85, 233 within societies vs. between

societies n - 1 4 , 229, 235 see also inequality

income mobility see social mobility indebtedness, sense o f 209

see also gifts India:

caste and performance 1 9 1 , 194, 215

and emissions 224 individualism:

and consumerism 230 and SU V s 58

Indonesia 241 inequality:

changes in 35, 69, 14 2 -4 , 162, 228, 239 -41

and class 26-7 economic grow th and 226

and the environment 2 3 2 -3 , 325 ethnic divisions and 18 5 -6 and happiness 227 historical explanations fo r 188-9 in human history 207-8 and international policies 235-6 and police numbers 155 and recycling 232-3 and w om en’s status 58-60 and w orking hours 228 see also income differences

infant m ortality 19, 8 1- 4 , 178-9 and carbon emissions 219 and class 178-9 and inequality 8 1, 178 , 2 12 , 219 international com parisons 80,

8 1-4 , 17 8 -9 , 220 infants:

birth and breeding vs. wealth 29 and stress 77, 2 1 0 - n see also birthweight; children;

infant m ortality infectious diseases:

as cause o f death 10, 73 poverty and social status and 10,

73, 85 inferiority:

feelings o f 40, 43, 1 1 3 - 1 5 , 165-6, 2 1 5 - 1 6 ; see also shame

feudal shackles as 45 and possessions 227

inflation: and inequality 55 and nationalized industries 253

insecurity: and consumption 230 personal and social 37 reducing 237-6, 270

Institute for W om en’ s Policy Research 58

I N D E X

interdependence 209 -10 International Adult Literacy Survey

107, 108 international comparisons xix

data for 190, 267 internal and international

inequalities 235 see also u nder topics

International Labour Organization 250

International Obesity Task Force

91 international relations 227-8 internet 265 interpersonal tensions 205 Iowa 137 Ireland:

prison population 145 teenage births 124

Irwin, John 15 2 Israel 267, 268 Italy:

co-operatives 260 mental illness 67 teenage births 124 women’s status 59-60

Jam es, Oliver 69 Jap an :

births outside marriage 183 crime rate 190 equality 1 5 - 1 7 , 44, 59~6o, 87,

174, 18 3, 184, 197, 2 4 1-2 , 245-6, 268

executives’ pay 250 and foreign aid 61 happiness levels 8 homicide rates 136 imprisonment and judicial

system 145, 148, 1 5 1 - 2

improvements in economy and health 87, 190

infant m ortality 80 life expectancy 176, 190 mental illness 67 modesty and blame 4 4 - 5 ,17 5 obesity 91 postw ar reconstruction 62, 87,

242 public social expenditure 183 redistribution o f wealth and

power 87, 18 3 -4 , 24I , 245 teenage births and marriages

12 3 -5 w om en’s status 59-60, 183

Ja rg o w sk y , Paul 162 job status:

and confidence 40 and self-worth 43

Jo h n Lewis Partnership 258 Johnson, Lyndon B. 19 judicial systems:

international com parisons 1 5 1 - 3 and minorities 150

K aw ach i, Ichiro 78 Kennedy, Jo h n F. 10 3, 18 1 Kennedy, Robert 18 1 K erry, Bill 236 King, M artin Luther 270 K ipling, Rudyard 15 7 Kissinger, Henry 207 Klinenberg, Eric 57 ‘knowledge’ economy 264 K rugm an, Paul 242-3

land-ownership 30, 25 1 language 206

and social class ^ 3 - 5 , 185 Lareau, Annette n 1

I N D E X

Lauer, Jo sh 57-8 law:

equality before 237, 263, 267 respect for 5 6 source o f inequality 52

Layard, Richard 8, 70, 227 learning by imitation 2 13 learning disabled people 152 Lennon, Jo h n 18 1 Lew is, Helen 41 liberty 45, 5 1, 263 life expectancy xiii, 6-7, n - 1 3 , 19,

73-87, 220 and childhood obesity 89 and economic development 6-7 and health care 80-81 and homicide rates 78-9 and income differences 190 and inequality 79-85, 176 international com parisons 79-84,

220 medical and social factors in 26,

73- 7, 79- 8i premature death and inequality

73 in richest countries n - 1 3 in w orld w ars 84-5, 87

lifestyle 73, 75-6 , 84 Lister, Joseph x v literacy:

adolescents 10 9 -10 , 180 adults 107, 10 8 -10

living standards: and economic grow th 6 -10 ,

29-30 as goods in use 224 and population health 77 pressure to maintain 25-6 , 219 and sustainability 219 , 224 and trust 62

in world wars 85 Loch Fyne Oysters 258 London Symphony Orchestra

258 Louisiana:

homicide rate 137 imprisonment rate 149 inequality and health and social

problems 149, 174 social capital levels 79

low-skilled work 1 1 6 - 1 7 Luker, Kristin 1 2 1 luxury 3

goods 70, 227 taxing 222, 27 1 see also consumption and

consumerism

MacArthur, General 242 McCord, Colin, and Harold

Freeman 80 Malaysia 241 male dominance 204-5, 2°8

see also dominance management methods,

participative 255-6 market economy/system 252

and inequality 199 mechanisms 210

Marmot, Michael 176 marriage:

choices 29, 44 and class 12 5, 163 in equal and less equal societies

44 and health 76 and obesity 99 romantic vs. financial

considerations 29, 44 and women’s work 228

M artin, Charles 129 M arylan d 153 mass society 42 M assachusetts:

health care for prisoners 153 homicide rate 13 7

material differences, scale of: and social divisions 56 and social problems 25

materialism 3-4, 249 maternity leave 1 1 2 m aturity, early 12 6 -7 M au er, M arc 147 M ax w ell Poll on C ivic Engagement

249

media 156 medical care see health care mediocrity, and equality 237 men see fathers; male dominance;

sex/gender differences menarche 189 mental health see mental illness;

wellbeing

mental illness 19, 63-72, 183 chemical changes in 65, 7 1 - 2 and child development i n cultural differences in 66 and group density effect 168 and inequality 6 3-72, 67-9, 176,

183, 246 international com parisons 66-9,

176, 183 measures o f 66 in prisons 15 2 -3 rises in 65 see also depression

M erck Fam ily Foundation 3 middle class:

and criminals i54n. and education i n , 1 1 5

and imprisonment 149-50 income levels in world w ars 85 and inequality 226 and parenting i n status m arkers 16 3-4

middle-income countries 6-7 M ilgram , S. 262 M iliband, D avid 222 militarization/militarism 47-8, 235

demonstrations against 248 M I N D (National Association for

M ental Health) 65 miniaturization 264 minimum wage legislation 243, 27 1 M innesota:

equality in 153 humane prison regime 153

imprisonment rate 149 inequality and health and social

problems 174 social capital levels 79

minorities see ethnic minorities m irror neurons 2 13 M ississippi:

inequality and health and social problems 174

teenage pregnancies 124 trust, levels o f 53-4

mistrust 57-8 and child development i n , 210 and fear o f crime see crime and pugnacity 14 0 -4 1 and reproductive strategies 126 and violence 140 see also trust

modesty 44-5 M on dragon Corporation 258 monkeys:

maternal protection 215 status and stress among 7 1 - 2 , 194

mood disorders 67-8 mood-altering drugs 65, 72 mortgages 270 mothers see parents; teenage births multinationals 251 murder see homicide music, as class and status indicator

27, 164 mutuals 2 5 1, 253

nanotechnology 265 N ation al Center for Education

Statistics 107 N ational Health Service 65 N a tio n al H ealth Service Plan 81 N ation al Policy Association 236 N ational Survey o f Children’ s

Health 68, 93 N ational Survey on D rug Use and

Health 68 nationalized industries 253 N ative Americans, de Tocqueville

and 51 neglect see children neighbourhoods:

cohesive 140, 259 concentration o f poverty 162 deprived/disadvantaged 28, 84,

1 2 1 , 12 5 , 13 2, 140, 16 3, 168 and ethnic groups 140 trust levels 57, 259 and violence 13 7 , 140

neighbourliness 57 ‘neo-liberal’ thinking, and

inequality 193 neocortex 206 Netherlands, The:

crime 1 5 1 equality 236 foreign aid 61

obesity 92 prison system 147, 150 trust, levels o f 52-3

neurology, and learning environments 1 1 5

N evada 79 N e w Brunsw ick, University o f 108 N e w Hampshire:

homicide rate 137 inequality and health and social

problems 174 , 184, 246 taxation 246

N ew Jersey 150 N e w Orleans 49-50, 54, 57 N e w Y o rk 13 7

see also Harlem N e w Zealand:

advertising 228 imprisonment rates 148 income differences 244 life expectancy 6-7 literacy n o mental illness 67, 183 teenage births 12 4 -5 , I27

N igeria 251 N obel prizes 2 3 1, 237 non-monetized sector 27 1 non-profit organizations 252-4

pay in 259 N ordic countries 197

see also Scandinavian countries N orth D akota:

inequality and health and social problems 174

trust, levels o f 53-4 N orth K orea 241 N o rw ay:

advertising 228 equality 1 6 1 , 268 foreign aid 60

N o rw ay: - cont. literacy n o parental leave 1 1 2 public spending on w elfare and

education 16 1, 2 12 teenage births 125 trust, levels o f 54 see also Scandinavian countries

nutrition see diet

O akeshott, Robin 257 obedience 262 obesity (and overweight) 10, 19,

89-102 causes o f 9 0-9 1, 10 0 -10 2 central (abdominal fat) 86, 95,

194 childhood see children and class 10 , 9 1, 95-6, 9 9-102 emotional and social

consequences 89-90 and employment/earnings 98-9,

1 0 0 , 1 0 1 - 2 , 1 9 1 and genetic factors 90 illnesses associated with 89, 95 and imprisonment rates 192 and income 9 1, 99, 1 0 1 - 2 inequality and 9 1- 5 , 98-9, 176,

189, 238, 239, 240 international com parisons 9 1- 3 ,

98, 176 and lo w status 75, 98-9 and marriage 98

and self-discipline 99 sex differences in relation to

inequality 27, 98-9 social distribution 91 social factors in 96-8 social gradient and 27 and thrifty phenotype 100

treatment and prevention 1 0 1- 2 , 238

in wom en see wom en see also poor people; rich people

obsessive-compulsive disorder

63 occupational classes 17 9 -8 1 O E C D see Organization for

Economic Cooperation and Development

oligarchs, Russian 28 optimism 55, 76 Organization for Economic

C ooperation and Development (O E C D ) 23

foreign aid com parisons 61 imprisonment rates in 155 public social expenditure data

184 working hours in 228

others, attitudes to: and health 76 and lo w social status 76 and stressors 38-9, 42-3, 57-8

oxytocin 2 1 3 - 1 4

Pachauri, Rajendra 231 Page, Jo h n 241 pain, social and emotional like

physical 2 14 -5 Paine, Tom 2 5 1- 2 , 263 Pakistan 2 5 1 parental leave 1 1 2 parenting:

and education see educational performance

and experience in early years 126 ‘p o o r’ 2 1 1 providing experience o f adversity

2 1 1 - 1 2

I N D E X reflecting altruism 210 and trust 55 see also parents

parents: attitudes to poverty h i feelings o f inadequacy n 1 and inequality m - 1 2 level o f education 10 8 - 10 ,1 8 0 maternal protection 2 15 postponing parenthood 120,

12 6 -7 single 12 7 , 179, 192 step-parents 138 and stress m - 1 2 see also fathers; parenting;

teenage births parole 1 5 1 , 154 participation, employee 255-6, 258 Pasteur, Louis xv patents 262 pay:

limiting 254 top 245, 249, 254 see also income

peace-keeping 2 0 1-2 sexual activity as 204

pecking orders 19 9 -20 1 and health and social problems

13 see also social status

peer pressure 43, 63, 138 personal relationship problems:

and inequality 45 and self-esteem 37

pessimism 76 petroleum 2 3 1 Philip M o rris 25 1 Philippines 241 physical activity see exercise physical education 90

P IS A see Programme for International Student Assessment

Polaroid C orporation 258 policing:

inequality and 15 5 , 246 and social problems 238 and violence 142

policy makers, and health and social problems 26

political w ill 247-9, 264 politicians:

and a better society 4 and ‘change’ 249 and crime i54n ., 156 and equality 15 7 , 197, 247-9, 269,

2 7 1 and inequality 5, 197, 238

polls (public opinion) 249 poor people:

blamed for broken society 5 causes o f their problems n , 24-6 and education n , 1 1 5 - 1 7 ‘genteel p o o r’ 29 and ill-health 10, 73, 1 7 7 , 1 8 1 - 2 ,

239 and imprisonment 149 and life expectancy 80-85 obesity am ong 5, 10 and parenting m - 1 2 proportion o f all incomes 17 spending choices 25 and violence 1 1 , 1 3 1 - 2 , 194 see also poorer countries;

poverty poorer countries:

benefits o f equality in 30 drug companies and 262 and life expectancy 6-7, 8 needs o f 235

poorer countries: - cont. and wellbeing 8 see also developing/less

developed countries Popenoe, D avid 13 7 Popper, Sir K arl 195 population health 77, 202

and health care 81, 238 and social factors 76-7

Portugal: health and social problems 183 inequality 15 , 183, 186, 179 infant m ortality 80 literacy n o police numbers 155 teenage births 124-5 trust, levels o f 52-3, 62

position, indicators o f 27 see also hierarchies

possessions: coveting 30, 69-70 as markers o f status 30 see also goods

poverty: and child development and

wellbeing m , 212 concentration o f 162 defined 15 fluctuations in 143 inner-city 16 2 -3 and pregnancy 120 and school drop-out rates 108 in Second W orld W ar 85 and trust 57, 62 see also poor people; poorer

countries power:

as aphrodisiac 207 arbitrary 263

pre-school program mes 1 1 2 , 247

prehistoric man see early human evolution; hunter gatherers

prejudices 45, 52, 168 dow nw ard 164, 166, 185, 207 see also class; racial prejudice;

snobbery pride 13 3 - 4 , I 4 I primates 198, 20 3-5, 2°8

aggression and hierarchies among 16 7 -8 , 203-5

prison see imprisonment; prison populations; prisons

prison populations 19, 14 5-50 and fatherlessness 13 7 fluctuations 145, 192 and inequality 148-50, 1 9 1 - 2 international com parisons

14 5-50 juveniles 167 sex offenders 167

prisons: abuses and bullying in 15 2 - 3 , 167 conditions 150 -53 education, treatment and

rehabilitation in 1 5 1 - 3 , 15 5 -6 over-crow ding 145 public expenditure on 247 sexual abuse in 167

privatization 25 1 productive pow er, ownership o f

productivity 10 3, 258, 260 profit, and new technologies

265-7 profit-making organizations 2 5 2 -3,

272 reducing inequality in 254-63

Programme for International Student Assessment (PISA) i o 5- 7, 109

progress, human 197, 267-8 provident societies 25 1 psychosocial factors:

definition 77 and health 87 and inequality 195

psychosocial wellbeing: for individuals and whole

populations j , 238 need for 4-5 politicians and 4, 238

psychotherapy 33 puberty 12 6 -7 , x89 public goods, and new

technologies 266 public sector pay 250 public social expenditure 18 3-4 ,

246 and inequality 184, 237

Publix Supermarkets 258

punishment 15 4 -5 , 203> Putnam, Robert 54-5, 56, 140

quality o f life see economic growth

racial differences: in imprisonment rates 150 perceptions o f 1 1 3 - 1 4

racism and racial prejudice 185 and inequality 5 1- 2 , 168 racial discrimination 43, 185, 268 and self-esteem 37 see also violence (racially

motivated) ranking see animals; hierarchies rape 129, 138

and fatherlessness 13 7 and inequality 135

rationing 222-3 Reagan, Ronald 244

recessions: 1980s 143 2008 see financial crisis

recidivism see imprisonment reciprocity 209, 210 , 214 recycling, and inequality 232-3 redistribution o f wealth and power

87, 18 3 -4 , 2 4 1-2 , 245 in Ja p a n 8 7 ,18 3 , 2 4 1- 2 , 245 in Sweden 18 3 ,18 4

regression line x ix rejection see social exclusion religion:

and death rates 79 decline o f discrimination 43,

268 and equality 269 and food sharing 205 and health 76 and psychosocial wellbeing 238 sharing in 209 and social m obility 15 7 and social status 185 and teenage births 124

reproductive difficulties see stress reproductive strategies 12 6 -7 , J 37>

204-5 research:

experimental see experiments health 27-8 and inequality 27-8, 193, 267 poverty 267 responses to 17 5 -6

resources: exploitation o f 262 limiting consumption 224-5 rivalry for 20 1, 204-5

respect 40, 13 3 - 4 , I 4I , 2 I 5, reverse causality see causality Rhodes, Lorna 15 2 -3

rich countries 100, 267 environmental policies 224-5,

235 and happiness 8 -10 , 69-70 health care 8 1-4 health and social problems 25-6,

z7 life expectancy 6 -7, n - 1 3 , 8 1-5 mental health 35, 64-70 teenage births 123 trust, levels o f 52-3 and wom en 57-60

rich people: benefits from equality 180 -2 cost to society o f 227, 269-70 and global w arm ing 222 and inequality 69-70, 1 8 1 - 2 , 229 m otivation o f 227, 229 obesity am ong 5 , 1 0 proportion o f all incomes 17 and reducing living standards

219 standing up to 269-70, 272 ta x proposals 227, 269 see also luxury

rights, fo r employees and tenants 268

R io de Jan eiro 135 risk-taking 5, 77, 207 rivalry see competition Robin H ood Index 19 Rom an Catholics 124 Rooney, Patrick 255 Roosevelt 244 Rothstein, Bo 55, 62

see also Uslaner, Eric R o yal Dutch/Shell 25 1 Russia:

inequality and history 189 life expectancy 87

new income differences and class structure 28, 87, 189

Soviet 252

Sahlins, M arshall 15 , 197, 200, 201 Sampson, Robert 140 Sanitary M ovem ent 73 Sartre, Jean-Paul 205 satisfaction see happiness savings, decline in 228, 270 Scandinavian countries:

equality 174, 183

income inequality 1 5 , 1 7 4 obesity 91 social m obility 159 trust, levels o f 52-3

scapegoating i54n. Scheff, Thom as 41 schizophrenia 168 Scholz, Tom 235 schools:

and obese children 90 and violence 13 7 , 139 see also education; educational

performance Science Applications International

Corporation 258 sea level, rises in 1 1 self-criticism 44-5 self-deprecation see modesty self-esteem 36-7, 38-9, 65

and esteem o f others 42-3 healthy and unhealthy 37, 45, 65,

207 result o f co-operation 20 9 -10 and social status 4 0 ,16 4 types o f 37, 44-5

self-harm 43 self-image, preoccupation with 37,

69, 99-100

self-interest: and consumerism 230, 232 experiments in 202-3 as governing principle 199, 202 restraining 202, 208, 2 10 , 214

self-realization from co-operation 209-10, 215

self-reported health problems 75, 81

self-worth see self-esteem Semmelweiss, Ignaz xv Sennett, Richard, and Jonathan

Cobb 165 sentences (penal):

length o f 14 5 -7 , 154 minimum m andatory 147 no-remission 147 non-custodial 147 see also death sentence

serotonin 7 1 , 1 1 5 sex discrimination 43, 268 sex education 120 sex equality among primates 204 sex offenders 167 sex/gender differences:

homicide rates 1 3 1 - 2 mental illness 68-9 obesity 95, 98-100 social gradient in health and

social problems 27 and ‘stereotype threat’ 1 1 3

sexism and inequality 5 1, 168 sexual activity:

competing for mates 13 4 , 210 and food 204 and status 134 see also adolescents

shame and humiliation 4 1, 129, 13 3 - 4 , 208

loss o f face 40, 13 3 - 4 , 2 1 5

and parenting 138 as social emotion 41 and status 69

and violence 1 3 3 - 4 , I 4°~4I ) x44> 215

w orking-class men 165 young men 1 3 3 - 4 , I 38) 1 4 0 - 4 1 ,

144 share-ownership schemes 255-6 shared values 4 ,5 6 shareholders, external:

limiting unearned income o f 257 redistribution aw ay from 260,

261 shareware 265 Shaw, George Bernard 237 shoplifting 147 shopping, obsessive 3, 70, 226, 230

see also consumption sickness:

benefits 268 and low status 75

Sidanius, Jim , and Felicia Pratto 168

Siddiqi, Arjum and 10 9 -10 Silverman, Jo h n 156 Singapore 241

gun ownership 136 homicide rate 136 income inequality 15 prison population 148-9

single parents see parents slavery 5 1- 2 , 15 7 , 188, 209, 267 Slovakia 25 1 Smith, A dam 25 smoking:

as cause o f disease 75 and low status 75, 87

snobbery 16 3 -6 , 207 as zero-sum game 165

sociability: increasing 237, 260 measures o f 45, 56, 78-9, 103,

199; see also trust social capital:

and aggression 140 defined 79 and health 78 and inequality 54-6, 126

social class see class social cohesion 45, 78, 84, 215

governments and 193 and inequality 195, 259 and teenage births 12 6 -7 in w orld w ars 8 5

social com parisons, invidious 70, 229

Social Democratic Party (Sweden) 242

social differentiation and inequality

i 95 social distinctions 28-9 social evaluation/social evaluative

threats 37, 38-9, 4 1- 2 , 43, 45 social exclusion 2 1 4 - 15

snobbery as 165 see also ethnic minorities

social expenditure see public social expenditure

social gradients see gradients social insurance 242 social interaction:

hum ans’ preoccupation with 205-6

oxytocin and 2 14 social involvement see community

life social life, and evolution o f brain

206 social m obility 19 , 24-5, 157-6 9

and ability 43 changing rates o f 16 0 -6 1, 259 and health 13 , 259 and income mobility 15 7 -9 and inequality 15 9 - 6 1, 16 3-9 inter- and intra-generational

157-9 international com parisons

15 9 -6 1 measuring 19, 15 7 -9 and obesity 98-9 see also sorting system

social needs 1 - 3 , 230 social networks 76-7 social organization 200

social strategies to deal with 20 6 -10

social problems: changing with changes in income

distribution 193 and material conditions 25 search for solutions 240 see also health and social

problems social relations:

built on material foundations

4-5 and child development m - 1 2 and health see health inequality and 5 1-6 2 , 195 and mental health 70

social security transfers 246 social self, preservation o f 38-9 social status 13 , 3 9 -4 1, 19 9 -20 1

competition see status and confidence levels 4 0 -4 1, 43-4 and ethnicity 18 5 -6 and health 7 5 -7 , 85-7, 169,

199-200, 207 importance o f 39, 40 -4 1, 7 1 - 2

I N D E X

and inequality 13 , 188, 199-200 and literacy 180 stratification 199-200 threats to 38-9 see also status

social strategies see social organization

social stratification 229, 242 social support 76, 85, 169 social theorists x iv - x v social trust see trust social w ork 238 socialists 52 societies:

as failures 18 types o f 26

solar pow er 266 solitary confinement 15 2 -3 sorting system, society as 24-5,

43 South K orea 242 Soviet Union (former) 252, 263

see also Russia Spain 258

mental illness 67 teenage births 124

Sports Utility Vehicles 57-8 state ownership 244, 252, 263 State o f W orking A m erica 2006/7,

The 160 status:

anxiety and insecurity 44, 69, 97-8, 100, 10 1

competition 4 4 - 5 ,16 4 ,19 4 , 207-8, 208, 209, 210 , 227, 257

deprivation o f 166 and health 194, 199, 207 and inequality 134, 13 7 , 1 4 1 , 199,

2 15 , 225, 227, 257 markers 28, 30, 40, 57-8, 134,

1 4 1 , 16 3 -6 ; see also goods men and 134, 14 0 -4 1 social see social status

steady-state economies 224-5, 2 7 ° Steele, Claude, and Joshua

Aronson 1 1 3 Steinbeck, John 73 stereotypes and stereotype threats

1 1 3 - 1 4 , 164 stigma 16 8-16 9

see also shame and humiliation stress:

chronic and acute 85-7, 96 in early life 39, 63, 77, 85, 100,

126, 2 1 0 - 1 1 , 2 1 1 and health 37-8, 4 1, 75, 85-7,

2 0 1, 205, 2 12 hormones 38-9, 95, 100, 1 1 5 , 2 1 1 levels see anxiety and obesity and food intake 95-6 and pregnancy 77, 100, 120,

2 1 0 - 1 1 and reproductive issues 8 6 ,12 6 ,

205 responses to 37-9 and risk-taking 77 sources of 37-9, 4 1, 63 stress-related diseases 86, 2 12 stressors, research into 38-9, 41 in unborn children 77 and weight loss 95

students: anxiety levels 33-4 experimental research on 19 1,

194 and narcissism 37

success 40, 69 see also status

suicide 175 international comparisons 76, 175

suicide: - cont. and lo w status 75, 76 social gradient 175

Summer, V olker 167-8 ‘ superm ax prisons’ 15 2 -3 support networks see social

networks Survey on Crime Trends and the

Operations o f Criminal Justice Systems 13 5 , 148

sustainability 2 1 7 - 3 1 , 252, 270 SU V s 57-8 Sweden:

births outside marriage 183 child wellbeing 212 death rates 178 equality 178-80, 183, 188, 24 1,

242, 268 executives’ pay 250 foreign aid 60 imprisonment and re-offending

i 45, i 55, i 57 infant mortality 80, 178-9 literacy n o parental leave 1 1 2 police numbers 155 teenage births 125 trust, levels o f 52-4, 62 welfare and redistributive taxes

183 w om en’ s status 59-60, 183 see also Scandinavian countries

Taiw an 45, 241 taste 164 taxation:

avoidance 2 3 1 and employee ownership 255,

261 inheritance 27 1

progressive 222, 271 and redistribution X83, 19 3, 2 7 1 and reducing inequality 254-5 and representation 263 o f rich people 227, 254, 269 V A T and sales taxes 264-7, Z7 J-

technologies 220-24, 272 changes and innovations 193, 225 new 220-24, 272 and sustainability 220

teenage births 19, 118 - 2 8 , 138 , 193 absent fathers and 127 benefits o f 1 2 0 , 1 2 1 , 1 2 8 data com parisons 19, 1 2 1 - 5 ,

12 6 -7 , I 42 disadvantages o f 1 2 0 - 2 1, 124-8 fluctuations in rates o f 142, 143 and inequality 27, 120-28, 142,

18 9 ,19 2 , 2X2, 239 and m arriage 124 young fathers 12 7 -8

teenagers see adolescents; teenage births

T exas 93 Thatcher, M argaret 244 thinness 99-100 ‘three-strikes’ laws 147 ‘ thrifty phenotype’ 100 Titm uss, R . M . 222, 242 tobacco 262 T o w er C olliery 258 Toynbee, Polly 91 trade unions 79, 243, 245, 269 tradeable carbon quotas 222 transnational corporations (T N C s)

2.51 trust 19 , 5 1-6 2 , 2 13 - 4 ,

in 19th-century America 51 benefits o f 56-7 causes o f change in 54-6

and choice o f vehicle 57-8 experiments in 214 and fear o f crime see crime and health 57 inequality and 4 5 ,5 1 - 6 2 , 195,

203, 259, 268 levels o f 19, 52-6, 77-8 lost w allet test 54 and oxytocin 2 1 3 - 1 4 and reproductive strategies 12 6 -7 unlocked doors 54 see also mistrust

‘truth-in-sentencing’ law s 147 Tunisia 25 1 Twenge, Jean M . 33-4

on anxiety and depression 33-4,

42- on narcissism 37, 42 on self-esteem 36

U K see United Kingdom U K M illennium C ohort Study i n ultimatum game 202-3 U N C T A D see United N ations

Conference on Trade and Development

underweight 189 unemployment:

benefit 268 and education 103 and employee buy-outs 258 fluctuations 143 and inequality 55, 12 5 , 192, 263,

27 1 policies 271 and weight gain 100 see also employment

unfairness 206, 256, 270 U N I C E F see United N ations

Children’ s Fund

United Airlines 258 United Kingdom:

crime and violence 12 9 -3 2 , 147, 156

depression in 35 drug abuse and addiction 65, 71 education and child development

10 5, 107, 1 1 0 - 1 2 employee participation schemes

256 ethnic minorities and

imprisonment 150 executive pay 250 foreign aid and colonial ties 61 happiness levels 8 health inequalities 8 1- 4 ,17 7 - 8 0 ,

238-9 homicide rates 136, 144, 269 inequality 15 , 14 3 -4 , I ^ i, i74>

17 7 -8 0 , 188, 242, 245, 249-52, 269

life expectancy 8 1-4 , 85, 268 literacy rates 10 8 -10 mental health 35, 6 3-5, 67, 183,

268 nationalized industries 253 non-profit sector 250, 251 obesity and overweight 90, 95,

96, 98-9, 10 1 . penal policies 145, 147, 154, 156,

268 residential segregation o f rich

and poor 16 2 -3 social class markers 164 social m obility 15 9 -6 1 teenage births and parenthood

119 - 2 0 , 12 2 -7 , 268-9 trust levels 52, 268 w orking class sense o f injury

16 5-6

United Kingdom : - cont. w orking hours 268 in w orld w ars 84-5, 242 see also England

United Nations: data 23, 175 foreign development aid targets

60-61 see also Survey on C rim e Trends

United N ations Children’s Fund (U N IC E F ):

on teenage births 123 see also Child Wellbeing

United N ations Committee against Torture 152

United N ations Conference on Trade and Development 251

United N ations Human Development Index 22 0 -2 1

United N ations Office on Drugs and Crime 70

United States o f America: advertising 228 alternatives to capitalism in

252—4 anxiety levels 33-5 and biofuels 2 3 1 death rates 1 2 - 1 3 , 178-9 and emissions 224 equality, historical commitment

to 49, 54-5, 263 executive pay 250 fear o f crime 1 3 1 federal poverty line 25 geographical segregation in 16 2-3 happiness levels 8 health 7 9 - 8 1 ,19 0 high-school massacres 139 historical explanations for

divisions 188

homicide rates 1 3 1 , 142 income differences 1 2 - 1 3 , I 9°>

220, 244, 249, 268 inequality 15 , 174, 18 5-6 , 190,

220, 239-40 infant mortality 80, 8 1- 5 , 220 and Ja p an 242

life expectancy 6-7, 79-80, 8 1- 5 , 17 6 ,19 0 , 220, 268

literacy rates 10 8 -10 mental health 6 3-5, 15 3 , 183 narcissism and blame 44-5, 175 non-profit sector 250, 252-3 obesity and overweight 89, 9 1,

95-6, 96, 9 8 - 10 1, 268 parental leave 1 1 2 prison and judicial system

MJ-SQ* i 5i - 3> 155- 6, 247,2169 public spending on education

1 6 1, 247 self-promotion 44-5 shared values 56 social expenditure 246 social mobility 15 9 -6 1 suicide rates 175

teenage births 12 2 -7 , 143, 2.68 top salaries 250 trust, levels o f 52-6, 268 w orking class sense o f failure 165 working hours 228, 268

universities and higher education 59-60, 16 1, 252, 267

U S Census Bureau 18 U S Department o f Justice, Bureau

o f Justice Statistics 149 US N ational V ital Statistics System

123 U S A see United States o f America Uslaner, Eric 55-6, 62 Utah 124

utopias 197, 27 1

Veblen, Thorstein, and Veblen effect 230

Venezuela 25 1 Vermont:

inequality and health and social problems 174, 246

phone calls for prisoners 153 social capital levels 79 taxation 246

violence 129-44 bullying as predictor o f 139 and celebrities 18 1 environmental influences on

13 9 - 4 1 and fatherlessness 137 fear o f 5 8 ,1 2 9 - 3 1 fluctuations in 14 3 -4 in foraging societies 208 governments and 193 and imprisonment 154 and inequality 45, 13 4 -7 , 14 0 -4 1,

14^ 1 4 3 - 4 , 190, i 9 5 , i9 9 international com parisons 1 3 1 - 2 ,

134- 7 , 13 9- 40, 1 5 0 preventing or controlling 134 -6 ,

142 and punishment 154 racially motivated 1 3 1 random shootings 129 and self-esteem 37 symbolic 164 triggers o f 40, 215

virtual systems 225 Visions o f Hum anity 235 voluntary activities 56, 78-9, 103

wage labour 210 W ake Forest School o f M edicine 7 1

Wales 178-90 W allich, H enry 226 waste see recycling water supplies 30, 2 3 1, 262 Wattleton, Faye 119 wealth:

attitudes to 4, 30, 69-70 differences in 27, 164-6, 199 in foraging societies 208 gain or loss o f 29, 43,

69-70 and merit 237

‘weathering’ 120 ‘weightlessness’ 225, 264 w elfare spending:

and child wellbeing 2 12 and education 1 1 0 - 1 2 and imprisonment rates 155 and teenage births 143, 193

welfare states 183, 241 wellbeing:

children’ s see Child Wellbeing economic 8, 55 indicators 8 international com parisons 220 mental health 65 and physical health 76, 205 and sustainability 2 18 -9 see also psychosocial wellbeing

West Germ any 147, 159 see also Germ any

West, Kanye 50 W hitehall Studies 75-6 , 84 W illms, D ouglas 108-9 Wilson, M argo , and M artin D aly

77, 133-4 W ilson, W illiam Ju liu s 163 women:

body image 99-100 death rates and inequality 60

wom en: - cont. differences in treatment see

sex/gender differences earnings 58-9 employment 58-9, 75, 228 homicide rates 1 3 1 - 2 and husbands’ income 228 and men’s status X34 mental illness and inequality 68 obesity and overweight 27, 9 1,

95; 98-100, io r political participation 58-9, 268 preference for high-status males

134, 204 and prospective partners 134 social and economic

empowerment 58-60 status and inequality 58-60, 203,

228, 268 and violence 1 3 1

w ork: aggression at 167 demands o f 205 and health 256 relationships at 259 safety improvements 73 w orking hours 228 w orkplace democracy 256 see also employment; sickness

absence w orking class:

and education i n , 1 1 5 - 1 7 feelings o f failure and inferiority

16 5-6 and middle-class values

1 1 5 - 1 6 and parenting i n in w orld w ars 85

W orld Bank 23, 13 5 , 241 World Drug Report 70

World Health Organization 66 on cardiovascular diseases 91 and mental illness 66 and obesity 89, 98

World Mental Health Survey Consortium 66

World War, First 84-5, 87 World War, Second 84-5, 87, 242 world wars, co-operation and

equality in 85, 242 World Wildlife Fund (W W F) 220 wound healing 76

Yearning for Balance 3-4, 249 young people see adolescents

Richard Wilkinson has played a formative role in international research on the social determinants of health. He studied eco­ nomic history at the London School of Economics before training in epidemiology and is Professor Emeritus at the University of Nottingham Medical School, Honorary Professor at University College London and Visiting Professor at the University of York.

Kate Pickett is Professor of Epidemiology at the University of York and a National Institute for Health Research Career Scientist. She studied physical anthropology at Cambridge, nutri­ tional sciences at Cornell and epidemiology at the University of California - Berkeley.

They live in North Yorkshire.

R E V I S E D A N D U P D A T E D

" An im portant bo o k . . . [W ilk in so n and Pickett] a rg u e that g ro ss in eq u a lity

te ars at the h u m a n psyche, creatin g anxiety, distrust a n d a n a rra y of

m ental an d p h ysical a ilm en ts— a n d they cite m o un ta ins o f data to suppo rt

th eir a rg u m e n t." — NICHOLAS D. KRISTOF, N E W YORK TIMES "In o u r time, [the] id ea of u n iv e rs a l opportu nity is o n ce a g a in u n d e r a ssa u lt

fo r w o rk in g p e o p le of ev ery ra ce . . . If you w an t m o re ev id en ce,

get y o u r h a n d s on this book, The Spirit L evel."

— BILL MOYERS O N BILL MOYERS JOURNAL "T h e evid en ce , h ere p a in sta k in g ly m a rsh a le d , is h a rd to d isp u te ."

. — EC O N O M IS T Why do A m ericans mistrust their fellow citizens more than the Ja p a n e se do? Why do we have higher rates of teenage pregnancy than the French? Why

do we have more hom icides than the S panish, A ustralians, and D anes put

together? The answ er may be this sim ple: We have more inequality.

This gro und b reaking book, based on years of research, provides hard evidence to show:

• How almost every measure of w ell-being-from life expectancy to mental illness, violence to illiteracy—is affected less by how wealthy a society is than by how unequal it is

• W h y societies with a bigger g ap between rich and poor are worse for everyone in them -including the well-off

• How we can find positive solutions and move toward a happier, fairer future

Urgent, provocative, and genuinely uplifting, The Spirit Level has been h eralded as providing a new way of thinking ab o u t ourselves and ou r comm unities. It will ch an g e the way you see the world.

R IC H A R D W IL K IN S O N has played a form ative role in international research on inequality, and his work has been published in ten la n g u a g e s. He is professor emeritus at the University of Nottingham M e d ica l School.

KATE P IC K E T is a professor of epidem io logy at the University of York and a N atio n al Institute for Health Research c a re e r scientist.

C U R R E N T A F F A IR S /U .S . $ 1 8 .0 0

C o v e r desig n: P atti R atchford C o v e r p h o to g ra p h : G a n d e e V asan

C /1 Available as aneBook V

ISBN:III III III 5 1 8 0 0 781608 193417 lUDIIIIII

w w w .b loom sb u ryp ress.com