This paper starts out from a puzzle. During the past thirty years, incomes have grown more unequal, a small group at the top has captured a much greater share of resources and poverty has increased. Despite this, most people are markedly less likely to want government to redistribute income or tackle poverty and are less sympathetic towards those without jobs. The greater insecurity of many people's lives in the current crisis renders the issue more perplexing. This paper describes trends in inequality, poverty and unemployment; presents new data on attitudes, media discussion and political platforms; discusses theoretical approaches from social psychologists, political scientists, sociologists and other commentators; and considers how a more generous welfare state might be pursued.
This article starts out from a puzzle. During the past thirty years, incomes have grown more unequal, a small group at the top has captured a much greater share of resources, and poverty has increased. Despite this, most people are markedly less likely to want government to redistribute income or tackle poverty and are less sympathetic towards those without jobs.
The greater insecurity of many people's lives in the current crisis renders the issue more perplexing. In order to examine how a more generous welfare state might yet be pursued in this unpromising context, I describe trends in inequality, poverty and unemployment; present new data on attitudes, media discussion and political platforms; and discuss theoretical approaches from social psychologists, political scientists, sociologists and other commentators.
Inequality, poverty and unemployment
Economic growth in the UK during most of the post-war period has led to higher living standards and the rise of a mass-consumption society. During the period up to the mid-1970s incomes rose at roughly the same rate for most population groups and rather more slowly for those at the top, leading to a slight overall convergence. Since then rates of increase have tended to diverge, with the most rapid growth among the highest income groups and those at the bottom falling behind.
This pattern may be summarised in terms of three overall trends:
For the mass of the population, economic inequalities have tended to ‘fan out’ as relatively higher income and lower income groups move away from the median (Figure 1).
For small minorities of the very rich (the top 1 or 0.1 per cent), incomes and wealth have grown very much more rapidly than those of the mass: the top 10 per cent have increased their share of total incomes by a third and the top 5 per cent by a half, while the top 0.1 per cent have improved theirs by a factor of three (Figure 2).
For those at the very bottom, poverty shows a different pattern—initially stabilising, then rising very rapidly in the mid and late 1980s, followed again by stability and a slight fall in the early 2000s, a rapid rise from 2006 to 2007 and a slight decline in 2010–11, as wages fell faster than benefits (Figure 3).
Inequalities in wealth also appear to have grown, although data is imprecise. Personal wealth declared for tax purposes roughly doubled from 4 to 8 per cent of incomes between 1977 and 2007, and the holdings of the most wealthy increased much faster. Dorling and colleagues estimate that the percentage of ‘exclusively wealthy’ households (those who use only private schools and health care, and own more than two properties and other high-value items such as boats) rose from 3.5 to 5.6 per cent between 1990 and 2000, while the proportion of poor households grew from 21.3 to 27 per cent. Wealthy and poor households became increasingly segregated.1
The trend toward inequality in the UK has proceeded more rapidly than that elsewhere in Europe, but slightly behind that in the US and in some large Asian countries. Between 1999 and 2006 a range of measures focused on lower income groups (especially tax credits, the minimum wage, housing benefits andthe guaranteed minimum pension), together with occupational pensions, somewhat reduced the incidence of poverty.2
Commentators have identified four main drivers of the trend to economic inequality: first, the increasing return to skill, resulting from technical advances, higher education levels and greater competitive pressures on the less skilled in a more globalised labour market; second, the trend since the late 1970s for an increasing proportion of the value added in national economies each year to go to capital and a smaller proportion to employees; third, the shift towards ‘winner-takes-all’ politics as increasingly wealthy minorities are able to exert more power over distribution and tax policies and gear up inequalities; and fourth, the high positional rents extracted by small groups in areas such as finance.3 These drivers may operate simultaneously. Opinion differs on their relative importance.
Contributory factors include the rise of the service sector, which replaced the adequate working-class incomes available from skilled and semi-skilled manual work with a much broader distribution between low-paid (retail, personal care, call-centre) and high-skilled (financial, legal, education and health service) work; population ageing, which reinforces the lifetime component in wealth inequalities; the trend for women to move into full-time work and to marry men in the same income band, compounding the effect of labour market inequalities; and the decline in taxes on capital and at the top end as governments seek to attract investment and skilled labour.
The 2007–8 crisis produced a real fall in GDP of some 6 per cent, of which about half had been made up by June 2012. Current OBR projections indicate that the rest will not be recouped until 2016–7. The IFS estimates that this will translate into a 7.1 per cent fall in net household incomes at the median between 2009–10 and 2014–5, with a slight narrowing for the mass, a fall in benefit incomes for unemployed families, disabled people and single parents and an increase of some 400,000 in the number of children in poverty.4 Benefits were uprated in line with prices so that they outstripped earnings in 2010–11, and the poverty statistic fell. This effect will be temporary as reforms cut benefits and target them more strictly. The impact at the very top is unclear. Of the long-term drivers of inequality mentioned above, at least the first two will continue, and the established trends toward greater inequality and higher poverty will persist.
Unemployment remained relatively low during the period of growth in the 1950s and 1960s, rising in the crises of the late 1960s and mid-1970s and then reaching a peak of 13 per cent in 1983 (using the IMF definition so that measurement is not affected by the various redefinitions during the period). It then fell to about 7 per cent by 1990, rising in the 1993 crisis to over 10 per cent, followed by a gradual decline to below 5 per cent by 2005. It accelerated during the 2007–8 crisis to exceed 8 per cent by 2010. It is not expected to fall below this level until 2014–5 at the very earliest. Many commentators believe current levels do not yet reflect the full severity of recent recessions. The proportion of the labour force unable to find full-time work and involuntarily working part-time more than doubled between September 2005 and June 2012, going from 8 per cent to 18 per cent.5
Public attitudes and sympathy for the poor
Trends toward inequality, poverty and unemployment are all rising, exacerbated by the current austerity measures. Public opinion, media discourse and political platforms in this area have become less sympathetic to those at the bottom in recent years, limiting the range of feasible policy responses.
Support for higher spending on the poor reached a peak in 1989, reflecting the increase in poverty during the 1980s. The subsequent trend has been decisively downwards, with some fluctuation between 2004 and 2009 (Figure 4). Sympathy for unemployed people shows a similar pattern. During the period of high unemployment throughout the 1980s and early 1990s, the proportion of the population expressing their concern for unemployed people by stating that benefits were ‘too low and caused hardship’ gradually rose, and the proportion believing benefits were ‘too high and discouraged working’ remained low (Figure 5). As unemployment fell from 1993 to 2004, sympathy declined and censure increased, so that the two attitude graphs cross over. However, as unemployment escalates after 2004, sympathy does not recover. Public attitudes turn against unemployed people, with over 60 per cent in 2011 believing benefits are too high and only 20 per cent that they are too low.
Attitudes to inequality and government policy are more stable. Between 75 and 90 per cent of the population believe income differences are too large, although the trend in attitudes is gradually downwards since the mid-1990s despite the growing gap between top, middle and bottom incomes. Rather fewer believe it is the government's job to address this issue. The proportion agreeing that government should ‘spend more … on the poor, even if it leads to higher taxes’ has steadily declined since 1993, from over half to just 27 per cent in 2009, with a 2 per cent increase by 2011.
The knowledge-base of attitudes provides a context for these beliefs. Most people are unaware of the scale of inequality or the speed of change. For example, the largest group among respondents to the academic International Social Survey in 2009 put the earnings of the chief executive of a large national company at about ten times the salary of a shop assistant, when a more accurate ratio would be in excess of 100 times. In addition, many people appear unaware of the proportion of those on low incomes whose main source of income is paid work. A total of 59 per cent of adults of working age who live in households where incomes fall below 60% of the median (after housing costs) are in work.6 Attitude surveys do not bear directly on reasons for poverty among this group, but answers to a British Social Attitudes survey question in 2009 indicate that only 9 per cent of respondents attribute child poverty to low pay, as against 20 per cent specifying parental addiction to alcohol or drugs, 15 per cent parental unwillingness to work, 10 per cent family break-up and 10 per cent lack of education.
Public opinion and media discourse indicate a decline in sympathy for the poor and a growing willingness to condemn unemployed people as scroungers. Political discourse shows an interesting pattern, in some ways reflecting and in others anticipating trends in opinion.
Analysis of media discourse also indicates a decline in sympathy for the working-age poor. One measure is the number of times that the derogatory term ‘scrounger’ is used to refer to benefit claimants in UK national broadsheet newspapers. Since these are broadsheets, the count reflects reporting of use by others rather than comments by the papers’ journalists. Usage remains low between 1993 and 2003, doubles from 200 to 400 per year by 2009 and continues to rise.
A convenient index of the main currents is provided by content analysis of manifestos. A considerable body of political science research indicates that manifestos are a better guide to the general direction of party policy than other sources, with over 70 per cent of the winning parties’ pledges passing into policy. The Comparative Manifestos project provides a robust, consistent and widely used summary of all post-war party manifestos for fifty-five countries. The study measures the emphasis given in the manifestos to the themes of social justice and welfare spending. ‘Social justice’ is defined in terms of ‘equality; the need for fair treatment of all people; special protection for underprivileged; need for fair distribution of resources; removal of class barriers; end of discrimination such as racial or sexual discrimination, etc.’. ‘Welfare state expansion' is ‘favourable mentions of need to introduce, maintain or expand any social service or social security scheme; support for social services such as health service or social housing (but excluding education)’, and ‘Welfare state limitation’ the reverse.7
Analysis of manifestos for the two main UK parties from 1987 to 2012 shows three patterns:
As might be expected, Labour manifestos score higher than Conservative ones on references to social justice and pro-welfare content. Conservative interest in social justice as equality or redistribution is limited, and is virtually non-existent for the 1987, 2001 and 2005 elections.
Pro-welfare content for both parties rises as both poverty and inequality rise for 1987 and 1992 but falls back sharply in 1997, with a higher level in 2001 and 2005 and then a decline (despite rising poverty and inequality) in 2010, so that the 2010 level is below that for 1987. The 1997 pattern is at first sight surprising, although it is to some extent explained by the emphasis on the social impact of education in the New Labour programme displacing explicit references to welfare. The vaguer concept of social justice follows this pattern but peaks in 1997.
References to social justice as equality do not follow the rising 90–10 ratio across the mass of the population or the very rapid increases in income for small minorities at the top. Instead they rise through 1987, 1992 and 1997, fall back to low levels in 2001 and 2005 and then reassert themselves somewhat in 2010.
Taken together, the attitude, media and manifesto data indicate a shift in the pattern of public and political discourse, starting sometime in the mid-1990s but becoming more marked in recent years. People living in a society which was becoming more and more unequal became less concerned about poverty and about the advantages of those at the top.
The disjunction between social trends and social attitudes is puzzling. It is hard for someone who glances at the property pages of a local newspaper, studies car prices or looks at salaries on employment websites not to be aware of inequalities. Inequality has traditionally been seen as a major political force. Why then do substantial and continuing shifts in inequality not generate a proportionate response in popular attitudes, discourse or party politics? A total of four themes emerge in social science explanations, stressing group exclusion, the narrow range to which most people confine their social comparisons, individualism and political disengagement.8 These factors promote the stigmatisation of the poor.
Social psychologists see group membership as basic to society. People define their own identities by comparison with others, and this leads to differentiation from those categorised as members of other groups: ‘self-definition in a social context’, as Tajfel puts it. Identity may feed back to comparison processes, to the extent that individuals choose comparisons that help them to maintain a particular self-identity. The sociology of group closure developed by Parkin and others and the anthropology of group membership, in which ethnicity often plays a strong role, places a comparable emphasis on group boundaries and self-evaluation by reference to others.
The sociologists Runciman and Merton argued that the primary element in identification was a sense of ‘relative deprivation’. Individuals articulate grievances by comparing their circumstances with those of others. These are typically social groups with whom there is social contact, not the distant wealthy. Reference-group comparisons are further constrained by a straitened awareness of the extent of inequalities and of the proportion of poor people who live in working households.
Broader accounts of social change suggest that identity for many people is becoming more individualised. Post-materialism, post-modernism and risk society theories point to shifts in the bases of social identity away from nation and class and more towards individual choice. They emphasise the apparent capacity for individuals to define themselves and to choose the social groups with which they identify through the choices they make from an unprecedented range of opportunities in life style and life course. According to the authoritative British Social Attitudes survey, the majority of the population sees themselves as ‘working class’, with the remainder ‘middle class’ (less than 1 per cent see themselves as upper class), but as ‘middle income’, with most of the remainder ‘low income’ (less than 5 per cent see themselves as ‘high income’. This trend is reflected in voting: the proportion voting by class has fallen steadily, from 63 per cent in 1964 to 41 per cent in 2005. Group membership and group boundaries are increasingly about the choices people make and their behaviour, rather than about the class positions in which they are placed.
Political scientists such as Boix point to the role of inequality in political behaviour. This leads to the discussion of how alliances between upper, middle and working-class groups are constructed in order to command majorities in modern capitalist democracies. More recently, researchers such as Pontussen and Rueda have stressed the importance of mobilisation: while inequality may increase party polarisation, it also causes some individuals to become less engaged in politics, indicating severe limitations to the impact of rising inequality on public and political discourse. The problem is compounded in the UK by the declining trust in politicians and rejection of current institutions sometimes termed as ‘anti-politics’. Differences of group interest are less likely to be articulated as the basis of political conflict.
Taken together, the long-standing processes whereby social identity is linked to group membership, and the newer developments that lead to a breakdown of traditional class identities and a greater emphasis on choice and individual responsibility for outcomes, promote the moralisation of social divisions, especially that between the middle mass and the poor. The widening of inequalities, the political disengagement of the poor and the lack of understanding on the part of many people that nearly two-thirds of those officially categorised as poor are primarily dependent on wages facilitates this division. Social policy experts have pointed to a growing tendency to stigmatise unemployed and homeless people, single parents and, very recently, sick and disabled groups as undeserving. The moral division between the middle mass and a discredited underclass has become more robust.
The current government's welfare state policies fit closely with the dominant strands in public opinion. They make a sharp division between deserving and undeserving groups and focus on strengthening work incentives for claimants. The work-centred approach is at the heart of the government's Social Justice Strategy: ‘income through benefits maintains people on a low income, and can even risk bolstering welfare dependency … Work, on the other hand, and the income it brings, can change lives.’9 The new Universal Credit scheme, designed to incorporate almost all state support for those of working age, reinforces the work ethic by confining out-of-work support to the poorest, imposing a benefit cap, time-limiting the housing element, imposing strict entitlement conditions relating to training and job-seeking and backing these up with a three-month benefit loss for any breach.10 The reforms to disability benefit are designed to cut spending and place paid work at the centre of policy. These policies follow opinion. A recent Ipsos-MORI poll showed more than three-quarters of the sample endorse stricter conditions for disability benefits and benefit penalties for able-bodied claimers who refuse work.11
This raises the question of whether other directions in policy are politically feasible and whether it is possible to counteract the trend toward stigmatising poverty. Three possible approaches draw on the key themes in public opinion in different ways and try to lead discourse in a progressive direction.
Reciprocity: working with public attitudes
This approach is concerned with overcoming the barriers between in-groups and out-groups by building empathy between individuals. Work in social anthropology (Mauss), political science (Ostrom), decision theory (Gintis) and social policy (Mau, Titmuss) demonstrates the importance of reciprocity in social relationships. It also emerges as a theme in attitude survey findings: those who are seen to make a contribution in return for entitlement are generally favoured as deserving. It seeks to address the lack of sympathy on the part of the middle mass for those at the bottom by presenting benefit recipients as members of groups who have made or currently make contributions to society or are likely to contribute in the future. This makes it harder to see them as simply a dependent out-group.
Horton and Gregory seek to reconcile commitment to universalism with the realities of deserving/undeserving distinctions by incorporating as many groups as practicably possible. They propose an expanded tax-credit system, effectively a citizen's income for all low-income people. This would be reinforced by a combined housing benefit that covers subsidies to both lower-income tenants and home-owners.12 They describe the psychological connection between contribution and entitlement in the National Insurance system as ‘gold dust’. The notion of insurance contribution is broadened to social participation. Benefit entitlement for all working-age claimants would be based (for workers) on work participation and paid contributions or (for non-workers) on social participation, including unwaged social care for children or frail older or disabled family members, voluntary work and training that makes a more valuable future contribution to society possible. There would be real sanctions for claimants who refuse opportunities to engage in these activities, demonstrating that the proposals are about building reciprocity rather than subsidising voluntary dependency.
Bell and Gaffney propose a rather less ambitious approach which would extend and strengthen the contributory principle embodied in the existing National Insurance system.13 The virtue of the insurance approach is that the benefits it offers are clearly ‘something for something’, as Ed Miliband has put it on a number of occasions. This does not confront public attitudes which stigmatise the dependent poor directly, but offers the opportunity to define a much larger group of claimants not as dependents but as contributors. The coverage of an enhanced National Insurance scheme could be broadened by crediting parental leave and ‘childcare, caring for sick or disabled relatives and training’. The current Lower Earnings Limit which excludes low-paid workers from national insurance would be abolished.
The authors point out that such a scheme could only be one part of a social security programme, and that it would depend on changes to the labour market to promote better-paid employment opportunities for those at the bottom to enable them to pay more in contributions. They argue that the contributory principle offers a way of addressing the decline in public confidence in the welfare state and point out that the European countries that rely on social insurance were rather more successful in encouraging labour market participation in the early 2000s.
Solidarity: shifting public discourse
Commentators on the centre left, from Titmuss in the Gift Relationship onwards, have promoted the idea of shared welfare provision that actively promotes inclusion and solidarity. This involves confronting current attitudes which stress divisions and moral boundaries between the mass of the population and the stigmatised poor. Individualism valorises personal choice, but solidarity rests on shared services that bring people from a range of social groups together and make clear to everyone their common interests in high standards of provision. The approach rests on a universal health service, non-specialised comprehensive schooling, national pensions and other benefits. Group differences are eroded and comparisons are not restricted to a particular reference group. Proponents assume that success in maintaining solidarity will facilitate the taxation required to finance a high level of universal services, a challenging assumption in the current context.
One approach that pursues this logic in the field of welfare is the Citizen's Income Scheme. The core idea is to provide a basic benefit to all citizens, replacing current systems and financed by a proportional tax which combines income tax and National Insurance contributions. At a certain level of income, tax paid would exceed benefit entitlement, so that the scheme would be redistributive. Below that level individuals are net gainers, above it net contributors. Finance through a progressive tax would enhance the redistributive effect. The advantages are individual autonomy, since the benefit would meet basic living costs; transparency, since the levels of contribution and receipt are obvious; and solidarity, since the scheme includes all citizens.14
Problems arise in gaining public support for the taxation required (a flat rate of at least 40 per cent of income to finance a benefit for the whole population) and overcoming the initial public disquiet at treating workless people in the same way as other citizens. For these reasons, most advocates of Citizen's Income propose partial schemes which would require a lower overall tax rate and would incorporate benefits for lower-paid and unemployed people. It is hoped that this would develop broader support for the scheme since workers and unemployed people would be in the same system, breaking down the attitudinal divisions.
Reciprocity and solidarity approaches take the problems of stigma and the lack of awareness of inequalities seriously. The proposed reforms promote greater inclusiveness, in the one case by presenting claimants as former or future contributors rather than dependents, and in the other by incorporating all citizens within a common solidaristic benefit framework. Of the two, the first seems more compatible with current trends in public opinion. It addresses the stigmatisation of dependency, following the logic of contribution and then expanding it to include many more claimants. Building solidarity through Citizen's Income rests on the assumption that a new tax and benefit structure will stimulate a shift in public attitudes.
Other approaches seek to alter the context within which public opinion applies, for example by moving outside the discourse of ‘hard-working citizen’ versus ‘dependent non-contributor’. One example is predistribution as an approach to labour market inequalities.
Predistribution: changing the rules of the game
Predistributive policies tackle inequalities at source by using the state's regulatory power to address some of the social changes that have led to wider social divisions. The long-term objective is to ‘transform our economy so it is a much higher skill, higher wage economy’, as Ed Miliband put it in a recent speech.15 Predistribution differs from most of the recent approaches to policy which seek to redistribute through various combinations of taxation and spending on the poor, both of which are unpopular and harder to deliver in tough economic times. As argued earlier, changes in the occupational structure and the increase in low-paid jobs have contributed to inequality. Predistributive policies might include a strengthening of the bargaining position and influence of workers through stronger trade union rights and representation on works councils, a higher minimum wage or living wage, measures to curb incomes at the top end through reforms to remuneration systems and possibly maximum wage legislation, and interventions to control the prices of items of common consumption such as utilities, transport or food.16
Such policies are designed to change the rules of the game and narrow the gaps between top and middle, and middle mass and stigmatised poor. They will also reduce reliance on divisive and targeted welfare for the latter group. Problems lie in constructing a system that is genuinely inclusive and does not simply redraw social boundaries. So far the predistribution approach has not been developed into a detailed scheme of social provision, and is perhaps better seen as a backdrop to the proposals discussed above.
Inequalities have been widening since the 1970s. The trend towards higher levels of poverty was slowed by benefit reforms in the early 2000s, but has now reasserted itself. Unemployment and labour market insecurity are also increasing. The mass public appears less sympathetic to the poor, less generous to the unemployed and less concerned about inequalities. These patterns are reflected in newspaper discourse and the programmes of political parties. The upshot is that the pressures on the poor and on those parts of the welfare state that provide for them are increasingly severe, while there is little support for policies to address these issues.
This paper reviews relevant approaches in social psychology, political science, sociology and social policy. These emphasise the social and psychological processes that reinforce group identities, limit the comparisons people make and deepen and moralise divisions between the middle mass and the poor, as a discarded underclass. Current reforms work with the grain of public opinion by defining claimants primarily as dependents and deepening the moral division between claimants and those in paid work.
Several commentators have proposed policies designed to shift opinion in a more progressive direction. One approach stresses reciprocity between claimants and others by giving the actual or potential contributions to society of those claiming benefits a much stronger place. A second approach seeks to weaken divisions between those seen as dependent claimers and those seen as self-sufficient workers. A third is intended to bolster the institutions that limit the growth of inequalities in the first place.
Economic stagnation appears likely to continue for some time. If the postwar tradition of cradle-to-grave state welfare is to be maintained, policies must take account of current trends in public attitudes. Growing inequality and the spread of insecurity may trigger a ‘tipping point’ in the patterns of social division that could lead in contrary directions: one path is a strengthening of the moral boundaries between middle mass and excluded poor, while another might lead to stronger perceptions of a common interest between insecure middle and disadvantaged minorities. The impact of policy on how people experience and think about their social institutions may influence which way people turn. Coalition policies draw support from the current stigmatisation by stressing dependency. The exclusion of the poor will reinforce these processes. Alternative approaches that emphasise reciprocity, solidarity and inclusion are possible. Under these circumstances, real opportunities exist for progressive political leadership to contribute to a shift in public attitudes in a more inclusive and less stigmatic direction.
J. Bamfield and T. Horton, Understanding Attitudes to Inequality, JRF, http://www.jrf.org.uk/publications/attitudes-economic-inequality, 2009; A. Giddens, The Consequences of Modernity, Cambridge, Polity, 1990; R. Inglehart, The Silent Revolution, Princeton, Princeton University Press, 1997; L. Kenworthy and L. McCall, Inequality, Public Opinion, and Redistribution, LIS WP459, http://www.lisproject.org/publications/liswps/459.pdf, 2007; R. Merton, Social Theory and Social Structure, New York, Free Press, 1968; J. Pontussen and D. Rueda, ‘The Politics of Inequality’, Comparative Political Studies 43(6), 675-705, 2010; W. Runciman, Relative Deprivation and Social Justice, University of California Press, 1966; T. Sefton, ‘Give and Take’, in A. Park, et al (eds) British Social Attitudes, 22nd Report, London, Natcen, 2005; G. Stoker, Building a New Politics? British Academy, 2012; H. Tajfel, Differentiation between Groups, London, Academic Press, 1978; W. van Oorschott, ‘Making the Difference in Social Europe: Deservingness Perceptions Among Citizens of European Welfare States’,Journal of European Social Policy 16(1), 23-42, 2006.