The Facts of Inequality

The Facts of Inequality Review Essay for the Journal of Moral Philosophy, forthcoming 2010

Martin O’Neill

Department of Politics University of York York YO10 5DD U.K.

mpon500@york.ac.uk

Richard Wilkinson and Kate Pickett, The Spirit Level: Why More Equal Societies Almost Always Do Better (London: Penguin/Allen Lane, 2009), 352 pp. ISBN 9781846140396 (hbk). Hardback £20.00

John Hills, Tom Sefton and Kitty Stewart (eds.), Towards a More Equal Society? Poverty, Inequality and Policy since 1997 (Bristol: The Policy Press, 2009), 432 pp. ISBN 9781847422019 (pbk). Hardback/Paperback £65.00/£22.99

It is uncontroversial that aspects of the physical and social environment in which individuals live can have a profound influence on both the quality and the length of their lives. This can clearly be seen in the deep and systematic differences in life expectancy between rich and poor countries. For example, 2007 U.S. Census Bureau figures show that, while life expectancy at birth in Japan was 81.4 years, in Zimbabwe it was less than half of that, at only 39.5 years. Differences within countries tend not to be so large, but can nevertheless be both substantial and shocking. It is a disturbing fact that, in the UK, there is a difference in life expectancy at birth between those growing up in the most affluent and the most deprived areas of the country, amounting to a difference of 10 years for girls and 13 years for boys. (Boys’ life expectancy at birth in Glasgow is 70.8, but is 83.7 in Kensington & Chelsea; the figures for girls are 77.1 and 87.8 respectively.) The Glaswegian men who actually make it to the age of 65 can expect 13.8 years more life, taking them to 78.8, while their fellow senior citizens in Kensington & Chelsea can expect to enjoy their retirement for fully 64% longer, with a life expectancy at retirement that takes them all the way to 87.7.[1]

Although the differences between the richest and poorest areas of the UK are stark, large variations in health outcomes are not only a feature of the difference between the very highest and lowest strata of British society. Neither are they peculiarities only of either absolute deprivation or superabundant wealth. Rather, there is a steady, step-wise social gradient with regard to health outcomes and life expectancy, running from the richest to the poorest parts of society. Moreover, this social gradient can be seen with regard to those who live within the same area, just as much as it can be seen in the differences between different parts of the country. In his pathbreaking Whitehall studies, the epidemiologist Michael Marmot has shown that there is a social gradient for life expectancy and for rates of cardiovascular disease, in terms of individuals’ occupational social class. [2] In other words, other things being equal, bosses and senior managers are less likely to die young, or to suffer from heart problems, than are those who work beneath them in the same organizations. Marmot’s subjects were all employees of the UK civil service; therefore, everyone in the survey was in secure employment, and nobody was facing the kind of absolute deprivation that comes with unemployment or social exclusion. Perhaps most strikingly of all, only a small part of this social gradient in cardiovascular disease could be explained by differences in lifestyle between the different groups (such as the prevalence of smoking), or even by other standard risk factors (such as high blood pressure or cholesterol levels). The contribution of occupational class to the relative risk of death from cardiovascular disease was twice as great as the contribution of all other risk factors combined.

Social gradients in health outcomes of this kind obtain not only across occupational classes, but also across groups characterized in terms of differences in education, income, and wealth.[3] Needless to say, incomes, social class and education levels tend to be closely correlated with one another, and so claims of causal priority tend to be difficult to make, especially given the impossibility of conducting controlled experiments in the context of the epidemiological study of populations.  As Marmot puts it, “If the essence of science is experiments we are in real trouble with understanding the causes of diseases in human populations.”[4] Nevertheless, a number of plausible causal stories have been advanced in order to explain the existence of these stark social gradients. Michael Marmot’s view – as suggested by the title of his 2004 book, Status Syndrome – is that the central causal factor in explaining the social gradient in health is the effect on individuals of differences in their social status.

Individuals with a low social status seem to be subject to two broad varieties of (interacting) causal mechanisms. Firstly, low social status is associated with the erosion of self-respect, and with the undermining of individuals’ sense of themselves as valued members of society who have the secure standing necessary to conceive and execute an autonomous plan of life. Marmot (following Hiscock et al 6) characterizes this sort of inner erosion of status partly in terms of the disruption of what Anthony Giddens has called “ontological security”.[7] But it would seem just as natural to think of these inner effects of the loss of status terms of damage to individuals’ amour propre (in Rousseau’s sense), or in terms of the absence (or insecurity) of what John Rawls has described as “the social bases of self-respect” – that is, “those aspects of basic institutions normally essential if citizens are to have a lively sense of their worth as persons and to be able to advance their ends with self-confidence.”[8]

The second, more external, aspect of low social status concerns the character of the social relations that low-status individuals experience, especially in the workplace (and especially where such workplaces are hierarchical in character). The degree to which individuals’ daily lives are marked by their falling under the (more or less arbitrary) authority of others, as opposed to being expressions of their own autonomous decisions, varies fairly systematically with income and social class. In Marmot’s Whitehall II study, which aimed to go further in determining the salient aspects of occupational social class, it was discovered that low job control, in the sense of having little authority over decisions or discretion in developing and applying skills, could explain more of the gradient in coronary heart disease than all of the standard (i.e. physical and lifestyle) risk factors combined.[9]

Marmot’s suggestion (endorsed by many other researchers) is that both the inner and the external manifestations of low social status – that is, both the erosion of self-respect and the experience of relations of domination in day-to-day social relations – are associated with elevated long-term stress levels. Long-term stress leads to elevated levels of adrenaline and cortisol in the blood, which in turn causes higher levels of cholesterol, and an increase in other physiological risk factors for a variety of diseases, including heart disease, many forms of cancer, and susceptibility to various infections.

To put things succinctly, the picture that seems to be emerging from this recent work in social epidemiology is this. Social and economic inequalities generate a social gradient in status, with those closer to the bottom of the hierarchy experiencing diminished self-respect and social standing, as well as reduced scope for autonomous agency (both at work and more broadly), and finding themselves at the sharp-end of social relationships of domination. These phenomena, in turn, are associated with an unhealthy long-term activation of the body’s stress-response, thereby elevating susceptibility to a range of diseases, which are associated with significantly increased morbidity and mortality levels.

That the effects of socioeconomic inequality should have such a profound effect, even within relatively affluent societies, on the duration of people’s lives and the quality of their health, is disturbing enough. But it is not only in the spaces of life expectancy and health that such social gradients exist. In their challenging and potentially politically momentous new book, The Spirit Level: Why More Equal Societies Almost Always Do Better, Richard Wilkinson and Kate Pickett show that such social gradients also link the propensities of a broad variety of social problems – including mental illness, drug use, obesity, violence, imprisonment levels, and teenage pregnancy – to background levels of societal inequality.[10]

Wilkinson and Pickett’s central empirical finding – hammered home relentlessly with a plethora of graphs – is that it is not only the case that social gradients exist within particular societies, with regard to this range of unwelcome social outcomes. It is that there are also systematic differences between societies in terms of these outcomes, where those differences are explained by the levels of inequality within each society. Time and again, Wilkinson and Pickett show that relatively egalitarian societies, such as Japan, Sweden, Norway and Finland, perform much better with regard to this range of social ills than do inegalitarian societies such as the USA, UK and Portugal. As the subtitle of their book suggests, “more equal societies almost always do better” than less equal societies, and they do so in a broad variety of respects. Wilkinson and Pickett show that (at least in many cases) this is not just because, in egalitarian societies, there are fewer poor people to drag down the averages. Rather, “the benefits of greater equality spread right across society, improving health for everyone … In other words, at almost any level of income, it’s better to live in a more equal place” (p. 84). Wilkinson and Pickett have thus done the considerable empirical service of showing that greater equality often benefits those at the top as well as those at the bottom of society.

As well as presenting evidence to show that more equal societies perform better on a broad range of social indicators than do less equal societies, Wilkinson and Pickett also emphasize the revealing fact that, for developed countries, there is generally no meaningful correlation between overall societal wealth and social outcomes. As the 2006 UN Human Development Report has shown, life expectancy is closely correlated with overall economic performance for developing countries but, once countries reach a threshold national income per capita of around $20-25,000, there is no further correlation between the two.[11] Greater increases in national wealth above this level appear to make little or no difference to life expectancy, and thus a very rich country such as the US can have a lower average life expectancy than a country such as Greece, which has roughly half the level of per capita national income. Wilkinson and Pickett further show that (again, beyond a certain threshold level) differences between countries in expenditure on healthcare also seems to be unrelated to life expectancy. Greece’s per capita spend on healthcare is less than half of that in the US, and yet a baby born in Greece can expect to live 1.2 years longer than a baby born in the US, and has a 40% lower chance of dying in its first year of life (see pp. 79-80).

These findings are extraordinary, and Wilkinson and Pickett have done a splendid job of marshalling together such a broad range of empirical information about the apparent effects of inequality, and presenting it so clearly and forcefully. I have described their book as having the potential to be politically momentous, as it presents a clear case for the social corrosiveness of inequality that should give pause for thought to politicians who have any concern for what genuinely effective social policy might look like. Wilkinson and Pickett’s central thought about social policy is simple and arresting, and demands serious attention: is it that, given that the level of inequality in a society seems to do so much to create a range of social ills, we should move our efforts away from piecemeal attempts at tackling those social problems independently (i.e. with one policy on violence, one on teenage pregnancy, another on preventing heart disease, and so on), in favour of tackling their common root cause. A truly “evidence-based” approach to politics would, in the view of Wilkinson and Pickett, be centrally concerned with massively reducing the level of economic inequality in society, thereby dealing with these social problems in a much more fundamental way. (At p. ix, Wilkinson and Pickett say that they had initially thought of calling their book “Evidence-based Politics”.) Such a radical, “evidence-based” approach to social policy would surely have much to recommend it, and might have some real chances of success, but one hardly needs to emphasize the degree to which the adoption of explicitly egalitarian policies would represent a real revolution in social policy.

As well as having the potential to significantly inform the development of practical social policy, Wilkinson and Pickett’s findings are of potential significance for political philosophers who are interested in the value and justification of distributive equality. I have emphasized elsewhere that philosophical accounts of the value of equality should proceed in light of appreciation of the “deep social fact” that distributive inequalities tend to bring with them relationships of social domination, harms to individuals’ status, the breakdown of healthy fraternal social relations, and the erosion of self-respect (especially of the worst-off). A plausible account of the badness of inequality can best be framed when these relationships between distributive inequalities and the character of social relations are kept in view, for a large part of our normative objection to substantial inequalities depends on their problematic personal and social consequences.[12] What Wilkinson and Pickett have given us, in effect, is an overwhelmingly strong empirical case for accepting the claim that this “deep social fact” is, indeed, a fact. Moreover, it is not a characteristic only of a few particular societies that such relations exist; on the contrary, the relationship between material inequality and these various social ills seems to obtain worldwide, across very different kinds of society. Although this is, of course, an empirical rather than a normative or conceptual finding, it is nevertheless the kind of significant empirical finding that ought to inform our normative understanding of the value of equality.

Where a political philosopher might want to put more pressure on Wilkinson and Pickett’s findings is in the place that they give to economic inequality itself, as distinct from these related social factors. In a number of places, Wilkinson and Pickett seem unsure whether to say that income inequality is fundamental in the explanation of these various social ills, or whether it is a proxy for other considerations. In their discussions of “How Inequality Gets Under the Skin” (Ch. 3, pp. 31-45) and “Physical Health and Life Expectancy” (Ch. 6, pp. 73-87), they follow Marmot in suggesting that the explanation of what’s going on in the relationship between socioeconomic inequality and these various social ills can best be characterized in terms of the various effects of loss of social status (as in their discussion of the greater “social evaluative threat” in unequal societies (pp. 36-9)), connected with the erosion of self-respect and the experience of loss of control, leading to long-term stress (pp. 85-87), and to outcomes that are pathological both for the individual and for society at large. Indeed, Wilkinson and Pickett go so far as to suggest that what measures of income inequality are really telling us about is “differences in how hierarchical” different societies are, such that “where income differences are bigger, social distances are bigger and social stratification more important” (p. 27).

The picture suggested here is one whereby the central causal factor is not so much inequality, per se, but social status, and where measures of inequality are themselves best seen as proxies for measures of the social hierarchy that leads to the erosion of social status. But this plausible picture fits somewhat awkwardly with things that Wilkinson and Pickett say elsewhere about the explanatory priority of inequality itself, as (at p. 189) where they tell us that “it seems unlikely that … an important causal factor will suddenly come to light which not only determines inequality but which also causes everything from poor health to obesity and high prison populations”. One might want to suggest, as Wilkinson and Pickett themselves suggest elsewhere, that candidates for this “important causal factor” are, at least in some cases, social status itself, or the social hierarchy that creates stratification and status harms. They have, of course, demonstrated convincingly that hierarchy and status harmsoften go along with, and, indeed, are often caused by, socioeconomic inequality, but we presumably would not want to say that the existence of (a certain degree of) socioeconomic inequality is a necessary and sufficient condition for the existence of social hierarchy, or for the creation of social threats to individuals’ social status. Neither, presumably, would we want to say that it could never be the case that the direction of causation might be reversed, as when social stratification (for example, due to class) leads to income inequality, rather than being caused by it. It is also not difficult to think of cases where both the low levels of income inequality and the low levels of social hierarchy can have a common cause, generated by features of a national culture (as, in different ways, we see in Sweden or post-war Japan). Wilkinson and Pickett’s emphasis on income inequality per se (for example, the vast majority of their many graphs plot the prevalence of  some social problem against a specific inequality measure, determined by the income ratio between top and bottom quintiles on the income distribution), while undoubtedly justifiable in terms both of simplicity and the availability of relevant data, can nevertheless be somewhat misleading in that it has the capacity to oversimplify potentially complicated causal relationships.

The emphasis on income inequality, per se, may also cause a degree of misplaced emphasis when Wilkinson and Pickett turn to the positive part of their story – that is, their account of what “evidence-based” social policy should actually look like. They express a certain degree of scepticism about approaches to social problems that do not seek to attack underlying material inequalities directly, but instead treat problems in a piecemeal way (p. 234), but find themselves endorsing policies such as increased spending on early-years intervention or pre-school education that are not, in fact, best characterized in terms of direct effects on material inequality. Many of their positive suggestions are fascinating, and deserve serious attention, such as policies to encourage more economic democracy through employee self-ownership of firms, or policies to bolster the bargaining positions of unions (see Ch. 16, “Building the Future”, pp. 229-265). But their emphasis, in describing such policies, is perhaps excessively on the way in which these policies could lead to greater income equality through compressing wage differentials. An alternative approach that de-emphasized income inequalities to some degree could give more emphasis to the ways in which various forms of economic democracy can act in more direct ways to bolster individual status and self-respect, and to transform the character of social relations at work, through granting more power to individual workers.[13] Thus, while I would suggest that Wilkinson and Pickett are certainly on the right tracks with regard to egalitarian social policies, their approach prevents them from seeing the full range of egalitarian arguments that might be deployed in favour of such policies.

These considerations lead one to the question of whether or not the overall argument laid out by Wilkinson and Pickett should itself be seen as essentially egalitarian in character. In one sense, of course, any view that recommends the compression of socioeconomic inequalities is plausibly described as being basically egalitarian. But there are points in Wilkinson and Pickett’s discussion where one might question whether the underlying justification for the compression of inequalities is not just merely instrumental. After all, if one seeks to eradicate large inequalities because they lead to shortened life expectancies, and higher rates of violence, disease, and depression, then one’s reasons for favouring a more egalitarian society need not have anything to do with the valueof equality, as such. One might simply favour equality on grounds of maximizing aggregate welfare, or even simply on grounds of economic efficiency, given the massive social costs associated with the various social ills that large inequalities bring in their wake. In one sense, this explains the potential for political influence of Wilkinson and Pickett’s book. For, insofar as they have shown that greater distributive equality is in the self-interest even of those who belong to the better-off parts of society, they have laid the groundwork for a line of political argument that has the potential to convince (almost) everyone to support more egalitarian social policies, whether or not they have any attachment to equality at the level of fundamental values.

Nevertheless, I assume that, contrary to the occasional impression to that effect, Wilkinson and Pickett do not simply see it as a happy coincidence that we have independent, instrumental reasons to favour more egalitarian societies; and, indeed, nor should they, or we, see things in that way. On the “non-intrinsic egalitarian” view that I favour, what it is to be committed to the value of equality is to endorse a conception of how people might live together as equals, in a society free of forms of social domination, that protects the status and self-respect of its members, and in which healthy, fraternal social relations can flourish.[14] If this is the core normative commitment that informs our attachment to egalitarian politics, then Wilkinson and Pickett (and, indeed, Marmot) have given us good empirical evidence to believe that we have reasons (inter alia) to eradicate substantial inequalities of socioeconomic position, not because they are of fundamental importance in themselves, but because of their connection to status-related harms.

On this view, Wilkinson and Pickett, in showing the broader beneficial effects of greater equality, have also thereby provided what one might describe as a supplementary argument for egalitarian social policies. In other words, they have shown us that in addition to non-intrinsic egalitarian reasons to favour greater socioeconomic inequality there are supplemental instrumental, non-egalitarian reasons (concerned with maximizing health or welfare, or minimizing expensive social problems) to also favour such policies. Such an outcome is welcome for the egalitarian, as it removes a potential source of objection to the pursuit of egalitarian policies (i.e. such policies would be less attractive if they did not benefit everyone, or if they were socially costly in some other way), and makes the job of political persuasion in favour of egalitarian politics potentially more straightforward. Thus, although Wilkinson and Pickett generally concentrate on the attractions of a more equal distribution of income, there is nothing in their account that should lead us to think that the fundamental normative commitment of egalitarianism should be to an ideal of equal distribution, as opposed to an ideal of social relations. As they nicely put things at the end of the book, “greater equality is the material foundation on which better social relations are built” (p. 265), and, we might add, it is of the essence of egalitarianism to be committed to the realization of those better social relations.

Given its status as a popular book, aimed at a general audience, and designed to be a goad towards real political change, one can forgive Wilkinson and Pickett for skating over some details. Their sweeping claim that “we have got close to the end of what economic growth has done for us” (p. 5) is made without adequate support, apparently on the basis of the fact that there is no correlation of national income with either measures of life expectancy or of subjective measures of well-being. But it would take a much more sustained argument, which they do not provide, to show that the onward march of life expectancy among the developed countries over the past 30-40 years really has nothing to do with sustained economic growth. Similarly, readers may be left feeling somewhat unsatisfied with the lack of statistical detail. More exploration of different inequality measures would have been useful, as opposed to the potentially disproportionate emphasis on ratios between the top and bottom quintiles. It would also have been good to see more discussion of how one can disentangle the contributions of absolute and relative differences in income to the overall social effects of inequality. Perhaps most crucially, more could have been said in presenting a full and candid exploration of the question of whether greater equality really benefits each group of individuals (i.e. right across the income distribution), as opposed to (merely) benefitting societies taken in aggregate, with regard to each of the various social problems that Wilkinson and Pickett address.[15]

Nevertheless, these cavils are relatively minor. This is a fascinating book, and one that it would be heartening to see having a real effect on the contemporary political agenda. Moreover, it is a thought-provoking book for political philosophers to engage with, not least because it fills in some fascinating detail about the social world we inhabit, and clearly lays out the structure of certain empirical relationships that may, as we have seen, carry implications for normative theorizing about inequality.

Despite the grimness of some of their empirical findings, Wilkinson and Pickett’s dominant tone is hopeful rather than despairing.[16] The detailed empirical findings presented in John Hills, Tom Sefton and Kitty Stewart’s collection, Towards a More Equal Society?, might suggest a less optimistic outlook. This book is a detailed audit of the record of the UK Labour government since 1997, in terms of its effects on various aspects of equality. The picture it paints, through the accretion of detail on policy areas from child poverty, health, and education to pensions and immigration policy, is certainly not all bad, but it is far from encouraging.

The core findings can be stated simply. In terms of overall measures of inequality, the Gini coefficient in the UK has worsened by 2 points during the ten year period from 1997-2007. Changes to the tax and benefit system brought in during this time have probably slowed this growth in inequality (by around 1 point, in terms of the Gini coefficient), but have failed to prevent it. There have been modest but appreciable gains in areas that have been directly targeted by government policy (such as child and pensioner poverty), at least up to 2004/5, but many of these gains have started to reverse in the period since then. Most of the rise in inequality since 1997 can be explained by the stratospheric economic gains made during the past 10 years by the richest members of society, such that “between 1996-97 and 2004-05, the top 10% of the income distribution experienced faster growth in net incomes than the rest of the population, the top 1% experienced still faster growth and the top 0.1% the fastest growth of all.”[17] With regard to health, pace the claims made by Wilkinson and Pickett about the irrelevance of healthcare spending to health outcomes, the significant increase in NHS funding during Labour’s second term seems to have led to substantially better health outcomes. But health inequalities have actually increased in the UK during the same period.

Although this is a mixed picture, the reasons for pessimism are not difficult to discern. One might be tempted to think that the economic “take-off” experienced by the very richest parts of society is relatively innocuous, given that they constitute only a small section of society. But, if Wilkinson and Pickett are correct about the status-sapping effects of conspicuous consumption by the very rich (see, e.g., pp. 222-6), and about the “positional” character of much consumption, then we do not have good reason to be relaxed about the inegalitarian consequences of this development.[18] Instead, the zooming economic gains of the super-rich should fill egalitarians with trepidation.

Also discouraging is the finding by Hills et al that many of the small victories of the period since 1997 have gone into abeyance since 2004, as slower economic growth has led to more constrained social policy, and a loss of political momentum. This is shocking when one realizes that the figures given by Hills et al stop short of the credit crunch and the financial crisis of 2007-09. Needless to say, constraints on government spending in the years ahead will be much more acute than they were in the period immediately after 2004. Hills points out that the politics of redistribution is much easier to achieve during good economic times, when stealthy redistributions of relative wealth can effectively disappear under the rising tide of prosperity. It is much harder to achieve when the economic tide is going out rather than coming in, as the opportunities for stealthy redistribution disappear. Hence, a large part of the story told by Hills et al is a story of lost opportunities. The period after 1997 was unusually conducive to egalitarian politics, as a left-of-centre party found itself afforded the opportunities of a resounding political mandate, during a time of growing affluence. One can now see clearly how little was achieved when one considers the propitiousness of the circumstances.

One of the problems highlighted by Hills et al meshes nicely with the hopes of Wilkinson and Pickett. In his chapter on public attitudes, Tom Sefton points out that New Labour did little or nothing to challenge public perceptions regarding poverty and inequality, and that public attitudes have, in fact (and perhaps partly in consequence), tended to become less sympathetic to redistribution since 1997. He raises the question of the likely trajectory of public attitudes in the years ahead, in light of the recent financial crisis, and the fiscal crisis that has come in its wake: “Will it make people even more resistant to redistribution, because the resources available for this are so constrained? Or will it make people more sympathetic to those dependent on benefits as the structural causes of unemployment are now so evident?”[19] It remains an open question. Perhaps, with the help of valuable public interventions such as Wilkinson and Pickett’s The Spirit Level, one might at least entertain the hope that the answer to this question will be a welcome one.

1. For helpful discussion of the issues dealt with in this review essay, I am grateful to Mary Leng and Shepley Orr.
2. Office of National Statistics, Life Expectancy at Birth and at Age 65 by Local Areas of the United Kingdom, (London: HMSO, 2008).
3. See Michael Marmot et al, (1978), “Employment grade and coronary heart disease in British civil servants,”Journal of Epidemiology and Community Health, 32: 244-49. See also Michael Marmot, Status Syndrome: How Your Social Standing Directly Affects Your Health (London: Bloomsbury, 2004).
4. Presentation of the findings of relevant empirical research, together with references to the various epidemiological studies, can be found in Gopal Sreenivasan, (2009), “Justice, Inequality, and Health,” in theStanford Encyclopedia of Philosophy, (Spring 2009 edition), Edward N. Zalta (ed.), http://plato.stanford.edu/archives/spr2009/entries/justice-inequality-health
5. Michael Marmot, (2004), ibid., p. 198.
6. Hiscock et al’s study looked at the health effects of housing security. See Hiscock, R., et al, (2001), “Ontological Security and Psycho-social Benefits from the Home: Qualitative Evidence on Issue of Tenure,”Housing Theory and Society, 18: 50-66.
7. See Anthony Giddens, Modernity and Self-Identity: Self and Society in the Late Modern Age, (Cambridge: Polity Press, 1991).
8. See John Rawls, Justice as Fairness: A Restatement (Cambridge, Mass.: Harvard University Press, 2001), p. 59. Rawls describes self-respect as “perhaps the most important primary good” at p. 386 of A Theory of Justice(Cambridge, Mass.: Harvard University Press, revised edition, 1999). For Marmot’s discussion of the effects of insecurity and the loss of self-respect, see his wonderful and illuminating discussion of “The Missing Men of Russia”, Chapter 8 of his Status Syndrome, ibid. For Rousseau’s idea of amour-propre see “Discourse on the Origin and Foundations of Inequality Among Men, Or Second Discourse,” in The Discourses and Other Early Political Writings, ed. Victor Gourevitch (Cambridge: Cambridge University Press, 1997) and Emile, Or On Education, trans. Allan Bloom (New York: Basic Books, 1979). For Rawls’s discussion of Rousseau, and his egalitarian interpretation of the idea of amour-propre see John Rawls, Lectures on the History of Political Philosophy, (Cambridge, Mass.: Harvard University Press, 2007), pp. 189–248, esp. pp.231–35 and 244–48 .
9. See Michael Marmot et al, (1997), “Contribution of job control and other risk factors to social variations in coronary heart disease,” The Lancet, 350: 235-39
10. All page references occurring in parentheses in the text of this article refer to Wilkinson and Pickett’s book.
11. United Nations Development Program, Human Development Report, (New York: Oxford University Press, 2006).
12. See Martin O’Neill, (2008), “What Should Egalitarians Believe?”, Philosophy & Public Affairs, 36. 2: 119-56, esp. pp. 130 & 150-1.
13. See, for example, Martin O’Neill, (2008), “Three Rawlsian Routes towards Economic Democracy,” Revue de Philosophie Economique, 9.1: 29-55, and Nien-hê Hsieh, (2005), “Rawlsian Justice and Workplace Republicanism,” Social Theory and Practice, 31. 1: 115-42.
14. See O’Neill, “What Should Egalitarians Believe?”, ibid., esp. pp.  120-34.
15. These statistical questions point towards what may be some of the more significant weaknesses of Wilkinson and Pickett’s exploration of inequality. For much fuller discussion of related issues than I can pursue here, see Gopal Sreenivasan, (2009), “Ethics and Epidemiology: The Income Debate,” Public Health Ethics, 2. 1: 45-52; and Hugh Gravelle, (1998), “How Much of the Relation between Population Mortality and Unequal Distribution  of Income Is a Statistical Artefact?”, British Medical Journal, 316: 382-5.
16. Indeed, they have set up a campaigning website, designed to popularize the findings of their book, and to promote its political agenda. See http://www.equalitytrust.org.uk/
17. Hills et al, ibid., p. 26.
18. On the relationship between “positionality” and equality, see Harry Brighouse and Adam Swift, (2006), “Equality, Priority and Positional Goods,” Ethics, 116.3: 471-97.
19. Hills et al, ibid., p. 242.