home work project

CHAPTER 1

Social Exclusion and Social Policy Research:

Defi ning Exclusion

Jane Millar University of Bath, United Kingdom

ABSTRACT

This chapter examines recent social policy research on social exclusion, exploring the way in which the term has been conceptualized and defi ned. It discusses research on the mul- tidimensional measurement of social exclusion, on the process of social exclusion, and on the use of social indicators to monitor trends in social exclusion across countries over time. The empirical research that has captured and measured multidimensional disadvan- tage has identifi ed various dimensions, including material and situational circumstances, with the aim of exploring the extent of social exclusion. The research indicates that there is a degree of consensus in the social policy literature and that material poverty is not the same as social exclusion. Overall, the results from these studies tend to suggest that, although some people and groups experience relatively high rates of social exclusion as measured by these indicators, there is not a large degree of overlap across the various dimensions.

INTRODUCTION

Social exclusion is a concept that has increasingly entered the policy discourse in recent years. For example, the UK Government is committed to reducing social exclusion and has established a new department, the Social Exclusion Unit, to develop innovative policy proposals (also see Hale & FitzGerald, Chapter 8 and Gordon, Chapter 11). The European Union has agreed to tackle the issue of social exclusion through the development of national plans of action, and has identifi ed a set of statistical indicators for measuring social inclusion. Also, there has been a substantial investment in research, examining social exclusion and the relationship between social exclusion, poverty and citizenship. This chapter focuses on recent research in social exclusion within the fi eld of social policy. It has four main sections. The fi rst considers how the concept of social exclusion has been

Multidisciplinary Handbook of Social Exclusion Research. Edited by D. Abrams, J. Christian and D. Gordon. © 2007 John Wiley & Sons, Ltd.

CO PY

RI G

H TE

D M

A TE

RI A

L

2 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

defi ned in social policy research and outlines some of the various defi nitions of social exclusion that have been developed in the UK and elsewhere. These defi nitions stress the importance of taking a multidimensional approach to social exclusion and the need to understand the dynamics of social exclusion over time. The second section discusses the way in which the multidimensional nature of social exclusion has been operationalised in practice for research purposes, drawing on various UK and European research studies. The third section focuses on research which explores social exclusion as a process, exam- ining the ways in which people come to be at risk of social exclusion and the resources they can draw on to protect and sustain themselves. The fourth section turns attention to the way in which the UK Government and the European Union have sought to develop indicators of social exclusion, in order to monitor progress on the goal of tackling social exclusion.

DEFINING SOCIAL EXCLUSION

Whilst there is widespread, if by no means unanimous, agreement about the defi nition of poverty as a lack of material resources to meet needs, the defi nition of social exclusion remains a contested term defi ned in different ways by different authors (see Atkinson, 1998; Room, 1995; Tsakloglou & Papadopoulos, 2002). However, the various defi nitions have in common an understanding that social exclusion is not only about material poverty and lack of material resources, but also about the processes by which some individuals and groups become marginalised in society. They are excluded not only from the goods and standards of living available to the majority but also from their opportunities, choices and life chances.

There are many defi nitions of social exclusion to be found in the social policy literature. Room (1995), in one of the fi rst studies to specifi cally focus on social exclusion, argued that social exclusion implies a major discontinuity in relationships with the rest of society and points to fi ve key factors which he suggests are central to the defi nition of social exclusion:

• Multidimensional: social exclusion cannot be measured by income alone but should include a wide range of indicators of living standards.

• Dynamic: analysing social exclusion means understanding processes and identifying the factors which can trigger entry or exit.

• Collective: social exclusion is not just about individual living standards, but also about the collective resources (or lack of these) in the neighbourhood or community. This means insuffi cient or unsatisfactory community facilities, such as run-down schools, remotely sited shops, poor public transport networks and so on.

• Relational: the notion of poverty is primarily focused upon distributional issues, the lack of resources at the disposal of an individual or a household. In contrast, social exclusion focuses more on relational issues. In other words, it refers to inadequate social participation, lack of social integration and lack of power.

• Catastrophic: a catastrophic separation from society, as a consequence of long-standing and multiple deprivation across all the above.

Another defi nition is offered by Atkinson (1998) who proposes that social exclusion has three main elements:

DEFINING SOCIAL EXCLUSION 3

• Relativity: it implies exclusion from a particular society at a particular time and place. • Agency: it implies an act or acts, by an agent or agents, people are excluded by the

actions of others. • Dynamics: people are excluded not just because of their current situation, but also

because they have little prospect for the future.

Similarly, Tsakloglou and Papadopoulos (2002) suggest that there is a consensus around fi ve key attributes of the concept of social exclusion:

• Multidimensional: across a wide range of indicators of living standards, including neighbourhood or community resources.

• Dynamic: it relates not just to the current situation but also to prospects for the future.

• Relative: it implies exclusion in a particular society at a particular time. • Agency: it lies beyond the narrow responsibility of the individual. • Relational: meaning a major discontinuity with the rest of society.

These defi nitions are all relating social exclusion to the inability of people to participate in the society in which they live, and arguing that this applies across several dimensions, including the material but also the social and political. Thus, Burchardt, Le Grand and Piachaud (1999, 2002) from the Centre for the Analysis of Social Exclusion (CASE) at the London School of Economics, provide a succinct defi nition which focuses on partici- pation: “An individual is socially excluded if he or she does not participate in the key activities of the society in which he or she lives” (Burchardt, Le Grand & Piachaud, 2002, p. 30).

This implies that the opposite of social exclusion is not integration or inclusion, but rather that it is participation. This point is argued strongly by Steinert and Pilgram (2003, p. 6), who point out that the concept of inclusion implies that individuals must conform “to all social norms and demands . . . even having to prove such conformity under penalty of being excluded”. In multicultural and multi-ethnic societies, in which there is a variety of values, attitudes and beliefs, there is not necessarily a consensus about what people should have, about how people should live, or about what people should do. Their defi ni- tion of social exclusion thus focuses on participation, both as an individual and a social goal:

Social exclusion can thus be understood as the continuous and gradual exclusion from full participation in the social, including material and symbolic, resources produced, supplied and exploited in a society for making a living, organising a life and taking part in the development of a (hopefully better) future. (Steinert & Pilgram, 2003, p. 5)

The conceptualization of social exclusion, therefore, points towards empirical approaches that encompass various different elements of the extent to which people can participate in the society in which they live. The key elements are that research should be rela- tional—not simply treating people as separate individuals but locating them within the contexts of family, household, community and nation. It should include the range of resources available to people—not just income, but also access to goods and services, community facilities, political engagement, leisure and social activity. It should be dynamic—not just about current circumstances, but also about future opportunities and capabilities to take advantage of these. And it should recognise agency—that people are

4 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

excluded by the actions of other individuals and institutions but also that individuals have different ways of coping with the risk or actuality of social exclusion. These coping opportunities and strategies can be enhanced, or diminished, by government activity in respect of social policy.

The next two sections of the chapter discuss examples of research which have set out to explore the extent and experience of social exclusion. It is not an exhaustive or system- atic review of the literature, but rather it is focused on a number of recent UK and EU research studies which have been selected to illustrate two main approaches to researching social exclusion. First, we look at examples of research studies which seek to identify those who are socially excluded by establishing multidimensional indicators of social exclusion, and identifying those who are excluded according to these measures. These studies are concerned with the state, or the condition, of social exclusion: with being socially excluded. Second, we look at examples of research studies which explore the ways in which people experience social exclusion, the risks they face, the sorts of responses and coping strategies that they have, and the support offered by the institutions of the welfare state. These studies are concerned with the processes of social exclusion: with becoming socially excluded.

BEING SOCIALLY EXCLUDED: MULTIDIMENSIONAL INDICATORS

As discussed above, there is a general consensus in the social policy literature that social exclusion is a multidimensional phenomenon that cannot be captured by measures of income only, or even material resources more broadly, but must include a wider range of other factors. These other factors have been defi ned in various ways, and this section provides some examples of the ways in which these dimensions of social exclusion have been defi ned. There are two parts to this process of defi ning and measuring social exclu- sion: fi rst establishing the key dimensions and second deciding on the measurement of the indicators to apply to each. Burchardt, Le Grand and Piachaud (2002) included four main dimensions, each measured by particular indicators:

• Consumption: measured by the capacity to buy goods and services and by savings. • Production: measured by participation in economically or socially valued activities. • Political engagement: measured by involvement in local or national decision making. • Social: measured by regularity and frequency of meeting with family, friends and

neighbours.

The UK Poverty and Social Exclusion (PSE) Survey carried out in 1999, Gordon et al. (2000); see also Pantazis, Gordon and Levitas (2006), also included four dimensions:

• Impoverishment: measured by lack of material resources. • Labour market exclusion: measured by lack of work, poor conditions or quality of

work. • Service exclusion: measured by lack of access to public and private services. • Exclusion from social relationships: covering fi ve key areas, including non-participation

in common activities, the extent and quality of social networks, the support people can call upon routinely and in crisis, disengagement from political and civic activity, and confi nement resulting from fear of crime, disability of other factors.

DEFINING SOCIAL EXCLUSION 5

In an EU-funded study of social exclusion in six European countries (Austria, Germany, Greece, Norway, Portugal and the UK), using the European Community Household Panel, we compared across the domains of income poverty, household deprivation (lacking ame- nities, lacking necessary durables), subjective feelings of well-being, and social isolation (Apospori & Millar, 2003; Barnes et al., 2002). Tsakloglou and Papadopoulos (2002) extended this approach to 12 European countries (the EU-12 countries), and more formally grouped the indicators into four groups:

• Income poverty: measured by incomes below 60 % of median equivalised income per capita.

• Living conditions: measured by scoring below 80 % of the median, using a weighted scale based on 22 household items (e.g. lack of space, environment problems, consumer durables).

• Necessities of life: measured by scoring below 60 % of the median, using a weighted scale of activities (e.g. being able to afford heating, holidays).

• Social relations: measured by meeting friends, talking to neighbours, membership of clubs or groups including political parties.

There is clearly some, although not complete, agreement on the dimensions or domains of social exclusion. The studies cited above all include some measure of material depriva- tion (low income, lack of material resources, low expenditure, etc.) and some measure of social relations (contact with family and friends, membership of clubs, etc.). The latter is most comprehensively developed in the PSE study, which also includes political and civic engagement as part of social relations, while others make this a separate dimension. Barnes (2005) explored the relationship between different dimensions using factor analy- sis. He identifi ed “three distinct and integral elements of social exclusion”, which were “household economic deprivation” (income poverty, material deprivation, housing), “per- sonal civic exclusion” (neighbourhood perception, social relations) and “personal health exclusion” (physical and mental health). The inclusion of physical and mental health indicators provides a link to research into quality of life. Layard (2005), for example, stresses the central importance of mental health as a key factor affecting happiness among individuals and societies. There is also some attempt in these studies to include measures at different levels—individual, household and neighbourhood—although the latter are often based on subjective perceptions of neighbourhood facilities rather than being sepa- rately observed.

In practice, the choice of the actual indicators used in each of the identifi ed domains is often dependent on the nature of the data that are available to the researchers. As Levitas (2006) notes, much of the research into the multidimensional nature of social exclusion relies on the analysis of existing large-scale, and often multi-purpose, data sets, rather than involving the (expensive) collection of new data. This means that, rather than “moving, as social research ideally should, from defi nition to operationalisation to mea- surement, the process is reversed” (Levitas, 2006, p. 127). The choice of indicators may be driven more by the data available than by conceptual imperative. Thus, there is some- times a gap between the factors that the researchers would ideally like to measure and the data that are available to them. For example, social isolation may be measured by the amount of contact with neighbourhoods, with little regard for the nature and quality of that contact; political participation may be measured simply by voting or membership of political parties, and so on. These sorts of measures can be particularly diffi cult to

6 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

interpret cross-nationally because the way we engage with our neighbours, or with political institutions, will refl ect cultural norms, as well as national, regional and ethnic identities. Thus, for example, Tsakloglou and Papadopoulos (2002, p. 215) fi nd that social isolation tends to be very low in the southern countries of the EU, and rather higher in richer countries like Germany. The meaning of social isolation, as well as the relationship to social exclusion, is likely to be rather different in these different contexts.

Having identifi ed the dimensions, or domains, of social exclusion, these studies examine the levels of social exclusion, the risk of social exclusion according to socio-economic and demographic factors, whether and how material exclusion is related to other forms of exclusion, and (where longitudinal data are available) the persistence of social exclusion over time. The results show a complex picture of social exclusion varying across countries and according to social divisions of age, social class, gender and ethnicity. Three results in particular are noteworthy. First, there is a relationship between poverty and social exclusion such that people with low incomes tend to be at greater risk of social exclusion than people with high incomes. However, the extent of this varies with the different mea- sures of poverty and social exclusion used, and social relations tend to be less likely to be associated with poverty in this way than do measures of material deprivation. Second, there is not a substantial degree of overlap between the different dimensions and it is a relatively small minority of people who are excluded across a number of these dimensions. Third, although there are more people who have some experience of social exclusion over time than at a single point in time, persistent exclusion for the same people continuously over a period of years is very rare.

For example, Burchardt, Le Grand and Piachaud (2002), using data from the 1998 British Household Panel Survey (BHPS), fi nd that about 58 % of people of working age were not excluded across any of the four dimensions of consumption, production, political engagement and social interaction. About 30 % were excluded across one dimension, about 10 % across two, just over 2 % across three and only 0.1 % were excluded across all four. There was a clear relationship between income levels and other dimensions of exclusion. This was strongest for production, with just 6 % of the top quintile were excluded on this dimension (measured by employment status), compared with 46 % of the bottom quintile. For social interaction, 16 % of the top quintile were excluded compared with 28 % of the bottom quintile. Looking at social exclusion over time, they found that only a very small proportion of individuals remained permanently excluded over time, especially when multiple dimensions are considered. Thus, after 4 years of observations, the very small group (1.5 % in the fi rst year) of people excluded on all of the possible dimensions had ceased to exist.

The results were very similar to those in the study by Barnes (2005) who examined both UK and European data. For the UK in 1996, looking at seven dimensions (income poverty, access to material possessions, housing circumstances, perceptions of neighbour- hood conditions, social relations, physical health and mental health), he found 49 % with no social exclusion, 30 % excluded on one dimension, 14 % on two, 5 % on three and 3 % excluded on four or more. Over time, defi ning permanent exclusion as being deprived on at least one of these measures in seven out of nine possible observations, he estimates that just 2 % of individuals experience long-term persistent exclusion. Thus, there does not appear to be any strong evidence for any “underclass” of people, excluded across all dimensions and for long periods of time, or for Room’s (1995) “catastrophic” and irrevers- ible separation from society.

DEFINING SOCIAL EXCLUSION 7

However, this is not to suggest that there is no substantial problem of social exclusion. As Burchardt, Le Grand and Piachaud (2002, p. 31) point out in their discussion of the dimensions of social exclusion, each of these is important in its own right, so that “par- ticipation in every dimension is regarded as necessary for social inclusion; conversely lack of participation in any one dimension is suffi cient for social exclusion”. The relatively small proportions of people excluded across several dimensions over a period of time are drawn from a much larger pool of people who may move in and out of disadvantaged and diffi cult circumstances, but who may not be able to move very far and who continue to face a high risk of social exclusion. It is the lack of opportunity to escape the risk of social exclusion which presents an important challenge for policy: “People are excluded not just because they are currently without a job or income, but because they have little prospect for the future or for their children’s future” (Atkinson et al., 2005, p. 18). On this point, Byrne (1999, p. 128) argues that social exclusion is inevitable in modern post-industrial capitalist societies, with fl exible labour markets and very weak collective power of working people. The existence of social exclusion is “necessary and inherent” under such condi- tions and many people are thus at risk of social exclusion, because their range of mobility is restricted between unemployment and low-paid work.

BECOMING SOCIALLY EXCLUDED: RISKS AND RESOURCES

Understanding the processes that put people at risk of being socially excluded, or which protect them from it, is an important part of the research agenda for policy purposes. This section considers instances of research that focus on the processes of social exclusion. As discussed above, many defi nitions of social exclusion stress the issue of agency, arguing that social exclusion happens as a consequence of actions, or indeed non-actions, by indi- viduals, groups and institutions. Agency also refers to the active way in which people themselves respond to their situations, and in particular, their responses to risk events, or contingencies, and the resources that they are able to call upon. This section discusses some examples of research studies which aim to explore how people cope and manage when faced with risk events—such as losing a job, or disability, or migration—and the nature and effectiveness of the institutional support they receive. The approach to these questions often involves qualitative methods—case-studies, in-depth interviews, ethno- graphic studies—in order to try and capture the nature and experience of these processes.

The research reported in the edited collection by Steinert and Pilgram (2003) is an example of this sort of approach. The research was based in eight cities in seven countries (Austria, Denmark, Germany, Italy, Spain, Sweden and the UK), and carried out by a cross-national research team which placed a strong emphasis on the importance of recog- nising the agency of people at risk of, or experiencing, social exclusion. The aim of this research was to explore the ways in which people “develop their own strategies for manag- ing the routines of life . . . their own remedial strategies to re-gain control and routine after a problematic episode” (Steinert & Pilgram, 2003, p. 6) and how this relates to the mix of individual and family resources that people can call upon to sustain their living standards or protect themselves from adversity. Such resources might include incomes from earnings, resources in cash and kind from the people they live with and from wider family and friends, and resources from the state in the form of benefi ts and services. This

8 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

approach examines the nature and processes of social exclusion by exploring how people respond to particular events, such as losing their jobs, or having a child, or divorce, or becoming a single parent. What mix of resources can people call upon in such circum- stances? Do those resources enable them to cope with the particular event or contingency? What role do state services and provisions play in this?

In their project, the research team operationalise this empirically by means of commu- nity case studies in various different countries, focusing on events or episodes (as they call them) as the unit of analysis, rather than upon individuals or households. Thus, for example, there are chapters that compare across the cities the experience of social assis- tance recipients, of migrant workers and their families, of people on the margins of the labour market, and those on the margins of the labour market. There are also chapters that examine the role of local agencies, family and community in providing resources to enable people to cope with, or avoid, poverty and social exclusion. These analyses draw on data from in-depth interviews with people living in different cities, and with different experiences of employment, family and national and local state services. In the concluding chapter, the authors note that, “Situations in which no change seems possible and in which there is a vicious circle with a downward dynamic, are the extreme, not the average case” (Steinert & Pilgram, 2003, p. 255). They also set out a typology of resources, derived from their empirical work, which identifi es four main types of resources potentially avail- able to people facing adverse situations or risks: welfare state resources (benefi ts and services), access resources (getting information, advice and support), resources of mutual help and reciprocity (in the family, neighbourhood and other networks) and “getting together” resources (grass-roots and self-help organisations). Each of these can provide positive support, but not to everyone and not in all circumstances. Thus, for example, family may be a precarious and short-term resource, and access to state resources is restricted by various forms of conditionality. State rules that make it diffi cult for people to call on a range of resources—to combine wages, family and welfare—make it more diffi cult for people to cope.

In another cross-national European study, Chamberlayne, Rustin and Wengraf (2002) use what they describe as “socio-biographical methods” to explore how people deal with risk events. Their data are drawn from “life journey” interviews, that are “narratives related by individuals, which were focused on life journeys such as those brought about by enforced redundancy or early retirement, by diffi culty entering the labour market after formal education, or by exile and migration” (p. 3). These very detailed and in-depth personal biographies from seven countries (Britain, France, Germany, Greece, Italy, Spain and Sweden) provide a way into analysing the complex ways in which individual attitudes and motivations, personal and family circumstances, and institutional agencies interact when people face particular events and circumstances. Again, the research starts with a stress on the importance of understanding people as active agents in constructing their responses to situations and circumstances:

Even though damaging life contingencies often imposed themselves with unexpected suddenness and force, we found that many of our subjects had actively negotiated threats to their well-being, and were rarely passive in the face of them. Others transi- tions, such as migration depended on substantial individual and family commit- ments . . . Individuals make life choices in multi dimensional ways. They choose courses of action for emotional and moral reasons, as well as for material ones. Our

DEFINING SOCIAL EXCLUSION 9

biographical subjects were concerned, in different instances, with questions of dignity and recognition (especially when they were denied respect by the treatment they received); with meaning and satisfaction (for example in their work); and with sustain- ing relationships. (Chamberlayne, Rustin & Wengraf, 2002, p. 4)

In their discussion, Chamberlayne and her colleagues distinguish between “tactical” and “strategic” approaches to coping with risk events. Some people are in such situations that they can only “live life one day at a time”, responding to circumstances rather than being able to control these, they can only be tactical, they cannot be strategic. This resonates with UK research that highlights the problem of the “hardship trap”, a term that has been used to refer to people trapped out of work by the fact that their poverty and hardship is so time-consuming and so demoralising that they are unable to think about employment, let alone take steps to fi nd a jeb (Marsh et al., 2001). The same sort of division can be seen in studies of how people cope fi nancially on low incomes. Kempson (1996), for example, suggests that there are two main strategies that poor people adopt. On the one hand they place a very tight control over all aspects of their expenditure, and on the other hand they adopt a more contingent approach of paying whatever is the most pressing need or debt. She shows that people move between these two approaches. As Kempson con- cludes, it is not that there are different types of people who do different things, but that there are different strategies that people use at different times. Of course, ultimately, if there is not enough income it does not matter which strategy you use, you cannot make ends meet.

These two EU cross-national studies include very rich data from the individual accounts, and the qualitative approach provides a valuable focus on the relationship between every- day experiences and societal institutions. But, it is also possible to explore risk situations and circumstances and social exclusion using large-scale quantitative data. For example, in our EU study of social exclusion, we examined the situations of four “transitions” or “risk” groups, in order to consider the extent to which being in one of these groups was associated with social exclusion (Apospori & Millar, 2003; Barnes et al., 2002). Two of these transitions—the transition from childhood to becoming a young adult, and the transition from work to retirement—are life events that we all face and were chosen as important points of transition between being in and out of the labour market. The other two—becoming a single parent, and becoming long-term sick or disabled—are events that happen to some (and to an increasing proportion in many countries) but not to everyone, which makes them particularly interesting for cross-national comparison. Focusing upon these times of transition was thus intended to allow an exploration of the extent to which life course changes can make people particularly vulnerable to poverty and social exclusion, and to compare different national outcomes. The results showed that lone parents and sick and disabled people face a substantially higher risk of income poverty, of non-monetary material disadvantage and multidimensional disadvan- tage than the average for their country. The same is true for retired people in the southern countries and in the UK, although not in Austria, Germany or Norway. By contrast, young people did not face higher than average risks of social exclusion in Austria, Norway, Greece or Portugal, but did in Germany and the UK. Looked at over time, it was the young people who were most likely to be able to exit from poverty and deprivation, but otherwise there was quite a mixed picture in terms of changes over time. Thus, as the authors conclude:

10 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

Different people have different histories of income and purchases, and so experience a different effect of current poverty on current deprivation . . . deprivation is a dynamic experience and can only be understood as the result of a history of experiences. (Barnes et al., 2002, p. 150)

In summary, these three cross-national comparative studies used different methods to explore risk events and to consider how people respond to these events, and to explore the resources—both public and private—on which they can draw. As with the studies discussed in the previous section, these authors agree that there is no “socially excluded” underclass group, and conclude that the events and contingencies that potentially create social exclusion are both identifi able and amenable to policy intervention. This focuses attention in the role of the state in prevention and the protection of people from social exclusion, as well as alleviation of the problem. Room (2000) has argued that research which provides an analy- sis of the relationship between state, private and voluntary institutions and individual, family and household action is needed in order to understand the trajectories of social exclusion over time, and to assess whether welfare institutions provide “buffers” against risk events and/or “passports” to take advantage of opportunities (Room, 2000).

MONITORING SOCIAL EXCLUSION: SOCIAL INDICATORS

This fi nal section on social policy research into social exclusion looks at the development and use of social indicators for the purposes of monitoring progress on policies aimed at tackling social exclusion.

Monitoring Policy in the UK

The Social Exclusion Unit (SEU), established in 1997 and located in the Offi ce of the Deputy Prime Minister, acts as a cross-departmental centre for both specifi c projects (for example on issues such as rough sleeping, teenage pregnancy, school truancy) and more general assessments of policy trends and future issues.1 Two reports, Tackling Social Exclusion and Breaking the Cycle, provide an overview of Government policy since 1997 and discuss policy priorities for the future (SEU, 2004a, 2004b). These reports provide data on some key indicators, such as levels of child and pensioner poverty, unemployment and worklessness, educational attainment, homelessness, crime levels and teenage preg- nancy rates. The fi ve main priorities for future action are identifi ed by the SEU (2004b) as low educational attainment, economic inactivity and concentrations of unemployed, health inequalities, concentrations of crime and poor quality environments, and homelessness.

The Department for Work and Pensions (formerly the Department of Social Security) takes the responsibility for regular statistical monitoring of government progress through the annual report, Opportunity for All. This report also provides a set of quantitative social indicators intended to monitor progress in tackling poverty and inequality. The baseline for reporting is 1997 (the year that the Labour Party was elected to government) and the

1 Now the Social Exclusion Taskforce (www.cabinetoffi ce.gov.uk/social_exclusion/)

DEFINING SOCIAL EXCLUSION 11

indicators are divided into four headings including three population groups—children and young people, people of working age, older people—with the fi nal set focused on com- munities. Within each heading, policy priorities are identifi ed and specifi c indicators attached to these. For example, under the heading of working-age people, one of the goals identifi ed is “supporting vulnerable groups and those most at risk of discrimination and disadvantage”. Additionally, concerning the latter, there are four specifi c indicators to do this: (a) a reduction in the number of people sleeping rough, (b) a reduction in cocaine and heroin use by young people, (c) a reduction in adult smoking rates, and (d) a reduction in deaths rates from suicide and undetermined injury. The goal of “improving opportuni- ties for older people” also has four indicators: (a) a reduction in the proportion of older people affected by fear of crime, (b) an increase in healthy life expectancy at age 65, (c) a reduction in poor housing among household with someone aged over 75, and (d) an increase in the proportion of older people living independently.

The published annual report, Opportunity for All (see Department of Social Security, 1999 and Department for Work and Pensions for subsequent years), summarises the policy activity in each of the fi ve key areas, including information on expenditure on new and ongoing initiatives, and summarises the trends for the key indicators. In general the trend data show an improvement over time. The seventh annual report, for 2005, concluded that the trends for 41 key indicators were “moving in the right direction”; seven key indicators were “broadly constant”; and seven key indicators were “going in the wrong direction” (Department for Work and Pensions, 2005). The latter included educational outcomes for children “in care”, infant mortality rates, obesity for children aged 2 to 10, families in temporary accommodation, employment of disadvantaged groups (lowest qualifi ed), the number of people contributing to a non-state pension, and life expectancy at birth (see also Jackson, Chapter 7 and Clapham, Chapter 5).

These audits provide a valuable snapshot of the situation each year and the direction of change. The reports are, however, a compilation of indicators of different types from various data sources, covering different geographical areas and different time periods. The main sources are administrative data and social surveys. Barnes (2005) and Levitas (2000) discuss some of the limitations of this sort of audit list approach. First, the indica- tors seem rather arbitrary and are certainly not theoretically justifi ed. Second, some of the indicators—for example, smoking rates, drug use—do not seem to be directly related to poverty and social exclusion. Third, the indicators mix measures of risk factors (for example, lack of education and skills, teenage pregnancy) and output indicators (for example, health, poverty rates) without distinguishing between these. Also, the impact of government policy cannot be assessed from these fi gures—if trends are going in the right direction, it may be in spite of, not because of, the policy measures adopted (although if they are going in the wrong direction, that could be a valuable signal for a reassessment of current policy measures). Levitas (2000) also argues that these indicators are largely based in one particular model of the “causes of social exclusion”, in which social exclusion is seen as primarily a problem of exclusion from employment, and so there is a strong focus on indicators relating to labour market position and skills.

Other approaches have been adopted to overcome such problems. The Joseph Rowntree Foundation has also published a series of annual audits, Monitoring Poverty and Social Exclusion, based on the work of the New Policy Institute (NPI), with seven reports for Great Britain (1998–2004) and two for Scotland (2002 and 2004). The fi rst report (Howarth et al., 1998) sets out the baseline and the indicators, using offi cial statistics, cover many

12 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

of the same areas as the government reports (the main headings are income, children, young adults, adults 25 to retirement, older people and communities), but with some dif- ferent indicators. This, therefore, provides an additional measure of change across a range of policy areas. In 2004, looking back over 5 to 6 years, they judged 18 indicators to be improved, 17 to be steady, one to be mixed and eight to have worsened. In the report for Great Britain, Palmer, Carr and Kenway (2004a) conclude that there are four key issues to be tackled: the situation of working-age people without dependent children, the situation of the economically inactive people who would like to work, the poor quality of jobs at the lower end of the labour market and the situation of young adults with low, or no, educational qualifi cations. For Scotland, Palmer, Carr and Kenway (2004b) also add Scotland’s relatively high levels of poor health. These are consistent with indicators high- lighted by the European Council in 2000.

Social Inclusion and the European Union (EU)

Social indicators are a crucial element in the “Open Method of Coordination”, a process for co-ordinated policy development in EU member states, which was agreed by the European Council in 2000. The aim of the Open Method of Coordination is to promote policy learning across the EU by providing a means for countries to benchmark their progress against others, while still maintaining the capacity to develop specifi c policies that meet national circumstances and priorities.2 It has fi ve key stages: the agreement of common policy objectives, the establishment of common indicators to monitor progress and compare practice, the development of National Action Plans, the publication of reports on these plans, and the establishment of a Community Action programme to facilitate exchange and learning (Employment and Social Affairs Unit, European Commission).3 The process has been applied to the issue of social exclusion, with the indicators of social exclusion developed by leading experts in the fi eld (Atkinson et al., 2002).

Nolan (2003) summarises the key methodological factors that were applied in choosing these indicators, including for example, that they should address outputs rather than inputs, be robust, be responsive to policy interventions, be comparable across countries, and that the portfolio of indicators should be balanced across different dimensions. Three tiers of indicators were developed. The fi rst tier of primary indicators includes various measures of low income and the distribution of income, of employment and unemployment, of early school-leavers not in education or training, of life expectancy at birth and of self-defi ned health status. The secondary indicators cover the same areas but with more specifi c mea- sures. The third tier consists of specifi c national measures, refl ecting local circumstances and priorities.

Three joint reports have been published (European Commission, 2002, 2004, 2005), with the latter being a report specifi cally focused on the 10 new member states and social exclusion. These can be read alongside a review by Atkinson et al. (2005), which provides a comprehensive discussion of the key issues and progress since 2000. In the latter report,

2 For a more general discussion of European and EU policy initiatives, see Mayes, Berghman and Salais (2001) and Begg and Berghman (2002).

3 See the website at www.europa.eu.int/comm/employment_social/social_inclusion/index_en.htm.

DEFINING SOCIAL EXCLUSION 13

the researchers make various recommendations for modifying the measures of exclusion (dropping some, adding or revising others). The revised list of primary indicators includes the existing indicators (at-risk-of-poverty rate, at-risk-of-poverty threshold, income quin- tile ratio, children living in jobless households, adults living in jobless households, long- term unemployment rate and early school leavers not in education or training) and fi ve new indicators (premature mortality or life expectancy, aggregate index of four housing problems, homelessness, aggregate index of nine deprivation items and children-focused non-income-based indicator) (Atkinson et al., 2005; table 5.2a). The report also calls for the more systematic use of these, and other data, in policy analysis, and for a “children mainstreaming” to ensure that the circumstances and needs of children are visible and central to the social inclusion goals.

CONCLUSION

This review, although not comprehensive, has indicated that there is a degree of consensus about the meaning of social exclusion. It is multidimensional (including social and political, as well as material, dimensions), it is dynamic (including future options and opportunities, as well as current circumstances), and it is relational (located in specifi c social and eco- nomic contexts). The process of social exclusion implies agency. On the one hand, social exclusion is created in society, with people excluded by the actions of others. On the other hand, people respond to social exclusion, or to the risk of social exclusion, in active ways. They seek ways to manage and to cope, and if possible to improve their situations and, where they can, they call on a range of resources including state support to do this.

People move in and out of disadvantage over time, although they may not move very far, nor escape the risk of facing social exclusion in the future. Research focusing on the process of social exclusion, and on how people cope with risks, highlights the importance of the range of resources available to people. As with the quantitative analyses, these qualitative studies do not fi nd much evidence for extreme and long-term social exclusion, although the resources that people can draw upon when they face diffi cult situations may vary signifi cantly for different social groups and in different national and local contexts.

While social exclusion has certainly become an important concept in policy discourses and debates, and the research base is expanding rapidly, there are a number of challenges for future research. First, there is a need for the continuous improvement of the data avail- able, with more robust indicators, better longitudinal data and a wider range of analytical techniques to explore causal relationships. Second, there is a gap in research that links “information about institutional strategies, administrative processes, policy context and household-level experience” (Heady & Room, 2003, p. 177). Such research could provide a stronger link between the processes that lead to social exclusion and the outcomes for individuals over time. Finally, social inclusion, understood in terms of participation, also implies a more democratic approach to research and policy. The most effective way to understand the resources and opportunities available to people, the barriers to making use of these, and their advantages and disadvantages, is from people themselves. This implies putting the experiences of people, and their attitudes and behaviour, at the centre and starting the policy analysis from that perspective.

14 MULTIDISCIPLINARY HANDBOOK OF SOCIAL EXCLUSION RESEARCH

REFERENCES

Apospori, E. & Millar, J. (eds) (2003). The Dynamics of Social Exclusion in Europe. London: Edward Elgar Publications.

Atkinson, A.B. (1998). Social exclusion, poverty and unemployment. In A.B. Atkinson & J. Hills (eds), Exclusion, Employment and Opportunity. CASE Paper number 4, London: Centre for Analysis of Social Exclusion.

Atkinson, A.B., Cantillon, B., Marlier, E. & Nolan, B. (2002). Social Indicators: The EU and Social Exclusion. Oxford: Oxford University Press.

Atkinson, A.B., Cantillon, B., Marlier, E. & Nolan, B. (2005). Taking Forward the EU Social Inclusion Process. Luxembourg: Luxembourg Ministry for Family and Integration. www.fm. etat.lu

Barnes, M. (2005). Social Exclusion in Great Britain: An Empirical Investigation and Comparison with the EU. Aldershot: Ashgate Publishing Limited.

Barnes, M., Heady, C., Middleton, S., et al. (2002). Poverty and Social Exclusion in Europe. London: Edward Elgar Publications.

Begg, I. & Berghman, J. (eds) (2002). Special issue: social exclusion and reforming the European social model. Journal of European Social Policy, 12(3), 179–239.

Burchardt, T., Le Grand, J. & Piachaud, D. (1999). Social exclusion in Britain 1991–1995. Social Policy and Administration, 33(3), 227–244.

Burchardt, T., Le Grand, J. & Piachaud, D. (2002). Understanding Social Exclusion. Oxford: Oxford University Press.

Byrne, D. (1999). Social Exclusion. Buckingham: Open University Press. Chamberlayne, P., Rustin, M. & Wengraf, T. (2002). Biography and Social Exclusion in Europe.

Bristol: Policy Press. Department of Social Security. (1999). Opportunity for All: Tackling Poverty and Social Exclusion,

First Annual Report. Cm 4445. London: The Stationery Offi ce. Department for Work and Pensions. (2005). Opportunity for All: Seventh Annual Report. Cm 6673.

London: The Stationery Offi ce. European Commission. (2002). Joint Report on Social Inclusion. Luxembourg: Offi ce for Offi cial

Publications of the European Communities. European Commission. (2004). Joint Report on Social Inclusion 2004. Luxembourg: Offi ce for

Offi cial Publications of the European Communities. European Commission. (2005). Report on Social Inclusion in the Ten New Member States 2005.

Luxembourg: Offi ce for Offi cial Publications of the European Communities. Gordon, D., Adelman, L., Ashworth, K., et al. (2000). Poverty and Social Exclusion in Britain.

York: The Joseph Rowntree Foundation. Heady, C. & Room, G. (2003). Tackling poverty and social exclusion. In E. Apospori & J. Millar

(eds), The Dynamics of Social Exclusion in Europe (pp. 164–177). London: Edward Elgar Publications.

Howarth, C., Kenway, P., Palmer, G. & Street, C. (1998). Monitoring Poverty and Social Exclusion: Labour’s Inheritance. York: The Joseph Rowntree Foundation.

Kempson, E. (1996). Life on a Low Income. York: York Publishing Services Ltd. Layard, R. (2005). Happiness: Lessons from a New Science. London: Allen Lane. Levitas, R. (2000). What is social exclusion? In D. Gordon & P. Townsend (eds), Breadline Europe:

the Measurement of Poverty (pp. 357–384). Bristol: Policy Press. Levitas, R. (2006). The concept and measurement of social exclusion. In C. Pantazis, D. Gordon &

R. Levitas (eds), Poverty and Social Exclusion in Britain (pp. 123–162). Bristol: Policy Press. Marsh, A., McKay, S., Smith, A. & Stephenson, A. (2001). Low-Income Families in Britain: Work,

Welfare and Social Security in 1999. Department of Social Security Research Report No 138 Leeds: Corporate Document Services.

Mayes, D.G., Berghman, J. & Salais, R. (2001). Social Exclusion and European Policy. Chelten- ham: Edward Elgar Publishing.

Nolan, B. (2003). Indicators for social inclusion in the European Union. In P. Krause, G. Bäcker & W. Hanesh (eds), Combating Poverty in Europe: The German Welfare Regime in Practice (pp. 75–92). Aldershot: Ashgate Publishing Limited.

DEFINING SOCIAL EXCLUSION 15

Palmer, G., Carr, J. & Kenway, P. (2004a). Monitoring Poverty and Social Exclusion. London: The New Policy Institute and York: the Joseph Rowntree Foundation. Retrieved from http://www. poverty.org.uk/reports/mpse2004.pdf

Palmer, G., Carr, J., & Kenway, P. (2004b). Monitoring Poverty and Social Exclusion in Scotland 2004. York: The Joseph Rowntree Foundation.

Pantazis, C., Gordon, D. & Levitas, R. (2006). Poverty and Social Exclusion in Britain. Bristol: Policy Press.

Room, G. (ed.) (1995). Beyond the Threshold: the Measurement and Analysis of Social Exclusion. Bristol: The Policy Press.

Room, G. (2000). Trajectories of social exclusion: the wider context for the third and fi rst worlds. In D. Gordon & P. Townsend (eds), Breadline Europe: the Measurement of Poverty (pp. 407– 440). Bristol: Policy Press.

Social Exclusion Unit (SEU). (2004a). Tackling Social Exclusion: Taking Stock and Looking to the Future. London: Social Exclusion Unit.

Social Exclusion Unit (SEU). (2004b). Breaking the Cycle: Taking Stock of Progress. London: Social Exclusion Unit.

Steinert, H. & Pilgram, A. (2003). Welfare from Below: Struggles Against Social Exclusion in Europe. Aldershot: Ashgate Publishing Limited.

Tsakloglou, P. & Papadopoulos, F. (2002). Aggregate level and determining factors of social exclu- sion in twelve European countries. Journal of European Social Policy, 12(3), 211–226.