See discussions, stats, and author profiles for this publication at: [Link]

net/publication/236981206

Explaining Social Exclusion; A Theoretical Model Tested in The Netherlands

Book · January 2007

CITATIONS READS

143 21,383

2 authors, including:

J. Cok Vrooman
Sociaal en Cultureel Planbureau
148 PUBLICATIONS 1,143 CITATIONS

SEE PROFILE

All content following this page was uploaded by J. Cok Vrooman on 02 June 2014.

The user has requested enhancement of the downloaded file.

Explaining Social Exclusion
Explaining Social Exclusion

A theoretical model tested in the Netherlands

Gerda Jehoel-Gijsbers
Cok Vrooman

The Netherlands Institute for Social Research/scp

The Hague, July 2007
The Netherlands Institute for Social Research (SCP) was established by Royal Decree of March 30, 1973
with the following terms of reference:
a. to carry out research designed to produce a coherent picture of the state of social and cultural
welfare in the Netherlands and likely developments in this area;
b. to contribute to the appropriate selection of policy objectives and to provide an assessment of the
advantages and disadvantages of the various means of achieving those ends;
c. to seek information on the way in which interdepartmental policy on social and cultural welfare is
implemented with a view to assessing its implementation.
The work of the Netherlands Institute for Social Research focuses especially on problems coming under
the responsibility of more than one Ministry. As Coordinating Minister for social and cultural welfare, the
Minister for Health, Welfare and Sport is responsible for the policies pursued by the Netherlands Institute
for Social Research. With regard to the main lines of such policies the Minister consults the Ministers
of General Affairs; Justice; Interior and Kingdom Relations; Education, Culture and Science; Finance;
Housing, Spatial Planning and the Environment; Economic Affairs; Agriculture, Nature and Food Quality;
and Social Affairs and Employment.

© The Netherlands Institute for Social Research/scp, The Hague 2007

scp-special
Graphic design: Textcetera, The Hague
Figures: Mantext, Moerkapelle

isbn 978 90 377 0325 2

nur 740

This report is printed on chlorine-free paper

Insofar as reprographic reproduction of parts of this publication is permitted under Article 16h of the
Copyright Act 1912 the fees payable by law therefore should be remitted to Stichting Reprorecht (p.o.
Box 3060, 2130 kb Hoofddorp, The Netherlands, [Link]). Permission for reproducing
(a) selection(s) of this publication for anthologies, readers and other compilations (Article 16
Copyright Act 1912) may be sought from Stichting pro (Stichting Publicatie- en Reproductierechten
Organisatie, p.o Box 3060, 2130 kb Hoofddorp, The Netherlands, [Link]/pro)

The Netherlands Institute for Social Research/scp

Parnassusplein 5
2511 vx The Hague
Telephone (070) 340 70 00
Fax (070) 340 70 44
Website: [Link]
E-mail: info@[Link]
Content

Abstract 7

1 Introduction 9

2 Social exclusion: a conceptualisation 11

2.1 Risk factors: an indirect definition of social exclusion 11
2.2 Social exclusion and poverty 12
2.3 A conceptual model 16
2.4 Operationaliziation 19

3 Research findings 21
3.1 Collection of new data 21
3.2 Indices for the dimensions of social exclusion 22
3.3 A general index for social exclusion 24
3.4 Profiles of risk groups 28
3.5 Processes of social exclusion 30

4 Conclusions 34

Notes 36

Appendix A 38

References 40

5
Abstract

Although social exclusion has become a key issue on the European policy agenda in
recent years, both the social phenomena the term refers to and the best way to moni-
tor these remain unclear. In response to this, we developed a conceptual model for
social exclusion and a methodology for its empirical assessment. Social exclusion
is conceived as a multidimensional concept. It is operationalized as a combination
of material deprivation; insufficient access to social rights; a low degree of social
participation; and a lack of normative integration.
In a survey among 860 Dutch households we found a valid scale which expresses
the degree of social exclusion in a single figure. This measure indicates that about
11% of the adult population may be regarded as socially excluded.
A causal analysis subsequently showed that having a bad health is the most impor-
tant risk factor. Other main determinants of social exclusion are a low income, benefit
dependency, limited Dutch language skills, and living in a one-parent household.
The outcomes suggest that it is worthwhile to strive for a specific measurement of
social exclusion as such; and that the concept should not be equated with shortfalls in
income and labour participation, as the current European policy debate tends to do.

7
1 Introduction

According to a report of the European Union’s statistical office, the degree of social
exclusion in the 10 new member states and some candidate countries is remarkably
1
low. On the most important common indicator for social inclusion , the ‘at-risk-of-
poverty rate’, Romania and Bulgaria score only slightly above 15%, the average of the
fifteen old eu-members2. In terms of this officially adopted criterion, the problem of
poverty and social exclusion in these two countries would thus be less severe than in,
for instance, the United Kingdom, Italy, and Ireland, where the poverty rate ranges
from 17% to 21% (Eurostat 2004a, 2004b).
This is intriguing – but is it true? An obvious explanation is that the outcome is
a consequence of the relative poverty thresholds the eu uses, 60% of the national
median income3. In Romania this amounts to only 14% of the eu15-average, whereas
in the uk the national threshold exceeds it by 28%. If the eu15-norm were applied
to both countries, poverty and social exclusion in Romania would be consider-
ably higher, while the uk figure would drop somewhat. However, those are not the
policy norms the eu and its member states act upon: each country has to be judged
according to its own standards. One starts to wonder if the current eu indicators are
suitable starting points for the development of a policy ‘to fight poverty and social
exclusion’, the central goal that was adopted at the European Council meetings in
Lisbon and Nice in 2000. The logical conclusion from the Eurostat findings would be
that this social issue requires less, or less intense, policy interventions in Romania
than in the uk; and somehow this seems unlikely. From a policy point of view, it
could be wise to re-consider the way poverty and social exclusion within the eu are
monitored.

There is also a scientific interest at stake here. In the literature there seems to be a
growing awareness that poverty and social exclusion require a ‘multi-dimensional
approach’ (a.o. Saraceno 2001; Short 2005). Yet there is no consensus on which
dimensions are involved; how these relate to one another; which risk factors may
lead to low rankings on these dimensions; and to what extent the latter may be
influenced by specific policy measures. This publication tries to shed light on some
of these topics, mainly from the point of view of social exclusion.

Although reducing social exclusion has become a key objective of national and eu
government policy, the best way of measuring the degree of success in achieving this
objective is unclear. The literature refers to ‘underdevelopment’ of the concept from
the point of view of policy, theory and methodology (Saraceno 2001). While the sci-
entific debate has been extensive and some empirical analyses have been performed
(cf. Atkinson et al. 2002, 2005) no full consensus on the nature of social exclusion
has been achieved yet.

9
The eu’s Laeken-indicators serve as proxy measures for social exclusion from a policy
4
point of view, aiming to foster comparability between countries . The monitoring
activities in relation to these eu-standards provide information on the individual
risk factors that increase the chance of being socially excluded, but make it hard to
gain insight in the social exclusion phenomenon as such.

Against this background we set up a study which aimed to arrive at a more precise
definition of the concept of social exclusion, and tried to develop a methodology
for measuring the phenomenon empirically. Subsequently, a survey was conducted
among a representative sample of the Dutch population, through which the method
was tested. In this publication we first address the various theoretical issues and the
conceptual model. Subsequently we present our research findings, and discuss the
implications of our study5.

10 Introduction
2 Social exclusion: a conceptualisation

Although the term ‘social exclusion’ has come into widespread use only recently,
this does not imply that the social phenomena to which it refers are novelties as well.
Already in the 1960s social exclusion became the subject of debate in France, but only
after the economic crisis of the 1980s and the introduction of the Revenu Minimum
d’Insertion (the national assistance law) the concept was widely used here (Silver 1994:
532). Once social exclusion had become a prominent item on the eu’s policy agenda
in the second half of the 1990s, attention began to focus on defining and specifying
the concept more closely. The policy to combat social exclusion has to be evaluated,
and to do this it is necessary to establish what social exclusion entails, what indica-
tors can be used to establish its existence, and which factors influence it. While this
has intensified the scientific debate on the meaning of social exclusion considerably,
up till now policy-makers have not been provided with a generally agreed scientific
conceptualisation.
An assessment of the way social exclusion has been operationalized shows that
most current definitions are indirect ones, while– in our view- a more direct definition
would be preferable for policy evaluation purposes. Such an approach has also been
advocated by other researchers; for instance, Levitas (2006) also proposes a direct
measurement of social exclusion, based on the British Poverty and Social Exclusion
survey.

2.1 Risk factors: an indirect definition of social exclusion

The difficulty of providing an adequate characterization of social exclusion is illus-

trated by a definition given by a United Kingdom government agency (Social Exclu-
sion Unit 2001): ‘a short-hand term for what can happen when people or areas suffer
from a combination of linked problems such as unemployment, poor skills, low
incomes, poor housing, high crime environment, bad health and family breakdown’.
Social exclusion is thus seen as a potential consequence of a number of risk factors,
without that consequence being spelled out. What may be understood by the term
social exclusion is left implicit: in several studies preference is given to an ‘indirect’
definition, by indicating which factors or indicators influence the risk of social exclu-
sion (e.g. Robinson & Oppenheim 1998; Paugam 1996, Edwards and Flatley 1996,
Howarth et al. 1998, in: Burchardt et al. 2002: 5-6). In other words, these studies do
not observe social exclusion itself, but rather its potential causes or predictors, with
the focus being mainly or exclusively on individual risk factors. In the Poverty and
Social Exclusion Survey of Britain, however, a direct measurement of several aspects of
social exclusion was attempted (Levitas 2006: 135).
Earlier policy documents from the European Commission do not provide a ‘direct’
definition of social exclusion as a separate concept either. These offer an indirect

11
demarcation, mostly by referring to the rights of social citizenship: ‘The extent of
social exclusion calls on the responsibility of society to ensure equal opportuni-
ties for all. This includes equal access to the labour market, to education, to health
care, to the judicial system, to rights and to decision-making and participation’ (cf.
Saraceno 2001: 3).
For the framing of their National Action Plans for social inclusion, the Member
States have agreed that social exclusion will be defined on the basis of a number of
social indicators, or ‘risk factors’. These risk factors, which are assumed to exert
a negative influence on the prospect of social inclusion, are low income, unskilled
labour, poor health, immigration, low education level, school dropout, gender
inequality, discrimination and racism, old age, divorce, drug abuse, alcoholism and
living in a ‘problem accumulation area’ (European Commission 2002: 10;). Concrete
agreements have been reached for a number of these indicators, the so-called Laeken
indicators (resulting from the 2001 eu Summit in Laeken-Brussels, Belgium). To
date, this consensus predominantly relates to indicators concerned with income and
6
employment; low income and lack of labour participation are generally seen as the
main risk factors for social exclusion (see, e.g., the Kok-report, which argues that ful-
filment of the social objectives will result from economic and employment growth
and that primacy should be given to job creation (European Communities 2004)).
In the literature, however, reference is also made to research showing that the
correlation between a low income and unemployment on the one hand and features
of social exclusion on the other is not particularly strong (Saraceno 2001: 5, 9). The
relationship varies substantially among social groups and across countries, depend-
ing on differences in the social security system, family arrangements, cultural set-
tings, et cetera (Saraceno 1997; Gallie & Paugam 2000). A low income or absence of
paid work does not by definition lead to social exclusion, and conversely people may
be socially excluded without having a low income or being unemployed (De Koning
& Mosley 2001: 7; Bailey 2006: 180; Levitas 2006: 155).
If this limited correlation holds for the two risk factors ‘income’ and ‘labour par-
ticipation’, it is likely that it also applies for the other – probably less dominant – risk
factors cited by the eu. Monitoring such factors may provide some information on
the evolution of the risk on social exclusion, but cannot be regarded as an adequate
measurement of the development of social exclusion as such. The proxy variables
that are commonly used in the indirect approach simply are not close enough. This
is why we think social exclusion should be defined in a more direct way. In doing so,
it is appropriate to pay some attention to the conceptual distinction between poverty
and social exclusion.

2.2 Social exclusion and poverty

The debate on what social exclusion actually means in concrete terms was fed mainly
by scientists and researchers who were concerned with the problem of poverty in the
1990s. Towards the end of that decade, policy goals shifted from combating poverty

12 Social exclusion: a conceptualisation

to reducing social exclusion. This led to the use of two different concepts in both
literature and research, although they are often used in one and the same breath.
The meaning of each concept is controversial, which can be traced back to
differences between the French and the Anglo-American scientific tradition (Gough
1997: 82; Room 1997: 256-257; Saraceno 2001: 6; Todman 2004: 1). The French school
builds upon the theories of Durkheim (1897) on social cohesion and solidarity,
the importance of collective values and norms, and the risk of social alienation
(anomie). This perspective tends much more towards the concept of social exclusion
than poverty, the core issue in the Anglo-American literature. Here scientific
research took its lead from theories of social inequality and relative deprivation
which regard unequal access to income, basic goods, public services and citizenship
rights as the starting point for research into poverty and social exclusion. The
work of Runciman (1966) and Townsend (1979) can be seen as the most prominent
exponents of this tradition. The wider social dimension received little attention in
Anglo-Saxon research (Levitas 2006: 133), although this has been changing in recent
years (Hills et al. 2002, Pantazis et al. 2006:7, Levitas 2006: 135).
While some authors say that there is hardly any difference between poverty
and social exclusion (e.g. Somerville 1998; Bhalla and Lapeyre 1997; Nolan and
Whelan1996), others argue that the two concepts differ fundamentally from each
other in a number of respects (a.o. Room 1995; Berghman 1995; Vrooman & Snel
1999; Saraceno 2001; Papadopoulos & Tsakloglou 2001; Abrahamson 1997; 2001;
Todman 2004). The following distinctions are often mentioned:

• Static condition versus dynamic process

Poverty refers to a static condition, relating to a given income situation or stand-
ard consumption pattern at a certain moment. Social exclusion is dynamic and
has to do with the process through which people become excluded.
• Absolute versus relative concept
Poverty may be conceived as an absolute lack, e.g. persons who do not attain the
income level required for the fulfilment of their basic needs. For social exclusion
there is no such absolute demarcation point. It can only be assessed in a relative
way, by comparing a persons circumstances vis-à-vis others in the same socio-his-
torical context.
• Unidimensional versus multidimensional disadvantage
Poverty relates to a single dimension: a shortage of financial or material
resources, or income deprivation. Social exclusion involves deficiencies in several
dimensions, which are associated with ‘full citizenship’: paid work and income,
education, housing, health care, legal assistance, accessibility of public provi-
sions.
• Distributional versus relational focus
Poverty relates to the distribution of economic aspects of disadvantage in income
or consumption. Social exclusion also concerns relational and socio-cultural
aspects, such as solidarity, social bonds and participation, integration, engage-

Social exclusion: a conceptualisation 13

 ment, discrimination, and norms of social citizenship (e.g. reciprocity, mutual
obligations). This difference is also often described as the material versus non-
material nature of the two concepts.
• Endogenous versus exogenous agency
Agency refers to the individual or collective actors that bring about shortages.
Poverty is typically analyzed at the individual or household level. The agency
lies mainly in the characteristics of the disadvantaged themselves, and may be
regarded as endogenous. Social exclusion, on the other hand, also derives from
a lack of ‘communal resources’: a person’s neighbourhood and social network,
social security agencies, and the social infrastructure. The excluded may have
little or no control over such exogenous factors.

However, this sharp juxtaposition of poverty and social exclusion also attracted criti-
cism. First of all, the distinction between static poverty and dynamic exclusion may
be questioned. Silver (1994: 545) argues that exclusion is not only a dynamic process,
but it also points to the outcomes of historical developments. It may therefore very
well be regarded as a static condition, or a state, sometimes referred to as ‘being
socially excluded’ or ‘excludedness’. Poverty, on the other hand, can be regarded in a
dynamic fashion, as happens in empirical research on the process of becoming poor
and terminating periods of poverty (see e.g. Goodin et al., 1999; Jäntti & Danziger,
2000: 353-362;).

The contrast between absolute poverty and relative social exclusion may also be
questioned. Poverty is sometimes measured in a purely relative fashion, as in the
familiar 60% of median income threshold used in many comparative studies. But
even ‘absolute’ poverty measures have a relative aspect. While they refer to the
realisation of certain ‘absolute’ minimum standards, the means this requires may
vary over time, location and social setting. This point has been repeatedly made by
Sen (1985: 669-671; 1992: 115-116), and implies the need for a sensible poverty line to
evolve, to some extent, in line with changing standards of living and social percep-
tions of necessities.
Vrooman & Snel (1999) state that poverty may very well be analysed in a broad sense.
An early definition used by the Council of European Communities (1985) provides
a good example: ‘individuals or families whose resources are so small as to exclude
them from the minimum acceptable way of life in the Member State in which they live’;
resources being defined as ‘goods, cash income plus services from public and private
sources’. Alcock (1991) also uses a wider approach of the poverty concept, and tends to
regard poverty as a multidimensional phenomenon. At first sight poverty, thus con-
ceived, may even seem to be synonymous with social exclusion. However, some authors
note an essential difference: although deficiencies other than financial shortages are
included in the broad definition of poverty, the reason for those deficiencies is mainly
financial (see Nolan & Whelan 1996). In the case of social exclusion, by contrast, there
may be other causes than a lack of financial means, such as illness, old age, neighbour-

14 Social exclusion: a conceptualisation

hood factors, and discrimination. Thus, one might be socially excluded without being
financially poor (Burchardt et al. 2002: 5-6; Uunk & Vrooman 2001: 144; Saraceno 2001:
4; see also Abrahamson 1997: 130; Room 1997: 256; De Koning & Mosley 2001).

The agency issue is regularly discussed in the theoretical scientific literature (a.o.
Jordan 1996), but is not really prominent in the policy debate or in the National
Action Plans, nor is it treated extensively in empirical research. Analytically, exclud-
ing actors can be defined at the micro-, meso- and macro-level, both for poverty and
social exclusion. There is no reason why an individual cannot be, at least to a certain
extent, an agent of his own social exclusion. Drug-addiction or school dropout, for
instance, may be important causes for shortages on several dimensions; and these
are partly based on choices made by the individual person. On the other hand,
poverty cannot always be attributed to its victims; the actions of benefit and job
agencies, and government policy on benefit levels and entry conditions may seriously
affect poverty rates and should be taken into consideration. Thus, distinguishing
poverty and social exclusion through differences in agency does not seem very fruit-
ful.
The proper way to analyse both is probably to take into consideration the actions
of various agents which may increase the risk of poverty and social exclusion. This
would include actions (or negligence) of the afflicted persons themselves, or of other
individual or corporate actors. Schuyt & Voorham (2000) note that fellow citizens
may cause exclusion, by morally rejecting people who are different; discrimination
in hiring and firing by employers on the basis of ethnic origin, age or health status
provides another example. Intermediate organisations that are charged with car-
rying out government policy in social security, health, welfare and education may
also be agents of poverty and social exclusion, through unclear goal-definitions, an
inefficient work process, a high case load, the preconceptions and preferences of
individual employees, et cetera. The municipalities and the national government may
also be regarded as actors, if their policies enhances the risk on poverty or social
exclusion (e.g. by denying certain groups access to a sufficient level of education),
or whose measures to combat these phenomena may be ineffective. And finally, at
a more abstract level the welfare state itself may even be regarded as an ‘actor’ that
causes poverty and social exclusion. This follows the well-known neo-liberal cri-
tique, which assumes that the welfare state does not in fact help people, but makes
them dependent and passive instead (e.g. Murray 1984, 1997). From this perspec-
tive, social exclusion is an inevitable outcome of the institutions of the modern
welfare state, which takes away the incentive for people to shape their own lives,
both through the safety net they provide and through the incentives administrative
organisations have in sustaining a passive attitude of their clients.

In addition to the possibility that actors at various levels can function as exclud-
ers, social exclusion may also result from more general socio-economic develop-
ments. Examples of these are a rising unemployment due to a recession or structural

Social exclusion: a conceptualisation 15

changes in labour supply and demand, demographic transitions (immigration of
low-skilled labourers and refugees), and cultural changes (e.g. a slackening of the
work ethic, the rise and fall of certain subcultures).

2.3 A conceptual model

Elaborating on these theoretical notions we developed a conceptual model, with the

aim to test it empirically. As the foregoing makes clear, the concept of social exclu-
sion is defined partly by contrasting it with the concept of poverty; but the distinc-
tive criteria are not very sharp. Social exclusion need not relate solely to the process
of being socially excluded (dynamic), but can also denote the condition of being
socially excluded (static). Social exclusion can relate both to non-material charac-
teristics (relational dimension) and to material aspects (distributional dimension).
The causes of social exclusion and of being socially excluded may lie at a collective
level, but individual characteristics and behaviour can theoretically be important as
well. The only distinction that could remain intact is that social exclusion involves
different dimensions, while poverty relates only to the financial/material aspect;
but even this difference only holds for poverty in a strict sense, and not for broader
definitions.

Against this background, we think it worthwhile to try to combine the two scientific
traditions, in order to enhance theoretical and methodological development. We
consider social exclusion as a concept denoting two main aspects:
1) economic-structural exclusion, which refers to a distributional dimension in line
with the Anglo-American approach;
2) social-cultural exclusion, which refers to the relational dimension emphasized in
the French school.

Within the first dimension we identify two distributional aspects: a material

(income and goods) and a non-material (social rights) one. The second dimension
is also divided in two different aspects: social integration, which point to social
relations and networks; and cultural integration, which regards values and norms.
Our approach thus combines the idea that poverty and social exclusion are mainly
the result of structural factors (e.g. W.J. Wilson 1987, 1997; Katz 1989) with the thesis
that they are mainly based in specific social settings and subcultures (e.g. Lewis
1968, 1969; J. Wilson, 1994).
In insert 1 these dimensions are described more explicitly, and we give an idea of
the type of indicators one could use.

16 Social exclusion: a conceptualisation

 Insert 1 Characteristics of social exclusion

A. Economic/structural exclusion (distributional dimension):

1. Material deprivation:
Deficiencies in relation to basic needs and material goods; ‘lifestyle deprivation’; proble-
matic debts; payment arrears (a.o. housing costs).
2. Inadequate access to government and semi-government provisions (‘social rights’):
Waiting lists, financial impediments and other obstacles to: health care, education (espe-
cially of children), housing, legal aid, social services, debt assistance, employment agen-
cies, social security, and certain commercial services (such as banking and insurance);
insufficient safety.

B. Socio-cultural exclusion (relational dimension):

3. Insufficient social integration:
A lack of participation in formal and informal social networks, including leisure activities;
inadequate social support; social isolation.
4. Insufficient cultural integration:
A lack of compliance with core norms and values associated with active social citizenship,
indicated by a weak work ethic; abuse of the social security system; delinquent behaviour;
deviating views on the rights and duties of men and women; no involvement in the local
neighbourhood and society at large.

The last sub-dimension, cultural integration, may require some qualification. The
eu’s focus in assessing social exclusion is on whether or not the rights of social
citizenship (equal access to education, employment, housing et cetera) are at stake.
The duties of social citizenship receive little attention; these may include, among
other things, complying with the moral or legal duty to work, having a sense of
responsibility towards one’s fellow citizens, social engagement, and behaving in
accordance with applicable legislation and regulations. Failure to observe such
duties of citizenship can be seen just as much as a characteristic of social exclusion
(or self-exclusion) as an inadequate access to rights of social citizenship. This issue
has been explored in the Netherlands in several qualitative studies focusing on the
coping strategies of benefit recipients (cf. Kroft et al. 1989; Engbersen et al. 1993;
Engbersen and Staring 2000). The line of research revealed that in many cases the
reaction to benefit dependency was a ‘traditional’ one, and in line with formal rules
and the norms and values that are considered dominant in Dutch society. In other
groups, however, the key aspect was strategic and rational behaviour. Among other
things, these groups had deviant views on the acceptability of breaking rules, on
the government, and on the advantages and disadvantages of working. This was
reflected in their behaviour: these groups were more inclined to apply for benefits,
and to continue drawing them, even when their entitlements were doubtful.

Social exclusion: a conceptualisation 17

Of course, one may question the possibility of assessing a ‘dominant culture’, espe-
cially in a society with a great degree of variation in terms of ethnic origin, religious
7
denomination or life styles. Moreover, who is entitled to identify core norms and
values; and how perfect has the assimilation into the dominant culture to be? These
reservations may be justified, but should, in our view, not lead to an ultra-relativistic
approach. We think it may be possible to identify some central values and norms
empirically, e.g. those which are enforced by law; and that these should be included
if one wants to assess the degree of social exclusion.

These considerations have led us to three basic assumptions for the development of
our conceptual model:
– social exclusion is a multidimensional phenomenon, which refers to both eco-
nomic-structural and socio-cultural aspects of life. Theoretically it consists of
material deprivation, insufficient access to social rights, deficient social participa-
tion and a lack of cultural/normative integration;
– a distinction can be made between traits which describe the actual state of social
exclusion (status characteristics) and risk factors that increase the chance of
social exclusion (process);
– the risk factors operate at the micro-level of the individual, at the meso-level of
formal and informal organisations and social settings, and at the macro-level of
government and society at large.

Figure 2.1 shows the conceptual model. The various aspects of social exclusion as a
state, or being socially excluded, are the variables to be explained. The risk factors
are displayed as determinants of these phenomena.

Based on the distinction between risk factors and features of social exclusion as a
state, the development in the degree of being socially excluded ought to be measured
directly, on the basis of ‘deficiencies’ in the four dimensions identified. For example,
the model does not equate being socially excluded with having a low income but with
material deprivation, which shows in the inability to meet basic needs, having prob-
lematic debts, payment arrears, et cetera. Having a limited income as such, however,
is not regarded as an indicator of social exclusion, but as a potential cause of it: a risk
factor.
The model essentially presumes a one-sided causality: the risk factors are con-
sidered to increase the chance of being socially excluded. Empirically the relation-
ships between some variables may in fact be reciprocal. Being socially excluded
can be a consequence of a poor health, but it can also cause a deterioration in one’s
physical or psychological well-being. In fact, most risk factors that are considered
to be amenable to policy interventions in figure 1 may empirically show a two-way
causal relation. Because the aim here is to identify the theoretical causes of social
exclusion, such feedback mechanisms are not included in the conceptual model. In
empirical research, however, this is a serious issue which cannot be neglected, but

18 Social exclusion: a conceptualisation

not easily be solved either. Detailed longitudinal data are needed to create enough
time distance between causes and consequences. Since the data we use are cross-
sectional, we cannot estimate two-way causal effects in our analysis. Therefore, the
results presented below are interpreted as if the direction of causality is a one-sided,
as is assumed in the theoretical model.

Figure 2.1
Conceptual model: risk factors and characteristics of being socially excluded

micro: persons/households

background characteristics risk factors amenable to characteristics of being socially excluded:

as risk factors interventions
- economic/structural deficiencies:
- age - coping abilities a. material deprivation
- sex - health b. insufficient access to ‘social rights’
- civil status - education
- family composition - labour market position - sociocultural deficiencies:
- social background - income a. insufficient social participation
- etnicity - physical and social setting b. insufficient normative integration

risk factor: risk factor: risk factor:

social developments government official bodies, business, citizens
- economic recession - inadequate policy - inadequate implementation
- individualisation - inadequate availability - waiting times
- bureaucratisation provisions - financial obstacles
- urbanisation - insufficient access to - risk selection (by employers,
- migration provisions banks, etc.)
- population ageing - discrimination, stigmatisation

macro macro meso

contextual risk factors

Source: Jehoel-Gijsbers (2004)

2.4 Operationaliziation

Subsequently, each theoretical dimensions of social exclusion was operationalized

in terms of questionnaire items. If available, we used indicators which were used and
tested in previous research on similar topics. In order to establish the degree of mate-
rial deprivation, respondents were asked about difficulties with making ends meet,
8
about the adequacy of their income for basic everyday and some other needs, about
arrears with rent, electricity and water company bills, insurances et cetera. Another
indicator for material deprivation was the opportunities people saw for obtaining

Social exclusion: a conceptualisation 19

financial support from a bank, family and friends et cetera in the event of serious
financial problems.

In the questionnaire the realisation of social rights was subdivided into access to edu-
cation, health care, social security agencies, housing and safety (e.g. not being able
to pay school fees, long waiting periods for medical treatment, being treated badly
when applying for benefits). Business services –such as being refused services by
banks, insurance companies or mail-order companies over the past two years- were
also added to this list, even though this does not concern a truly ‘social’ right, since
it does not depend on collective regulation.
The social integration-dimension was divided into several components: societal par-
ticipation (defined as having paid and/or unpaid work), informal social participation
(contacts with family, friends et cetera) and organised social participation (member-
ship of associations, church attendance et cetera).
On the dimension of cultural integration we focused on dominant values and
norms, that are supposed to be rather imperative; thus, in its operationalization
the dimension narrows down somewhat to one of that could be called ‘normative
integration’. Not trespassing the law we judged as the most undisputed norm; such
rules are agreed at the macro-level, and every citizen has to comply with these, or
risk a legal penalty. We used views on breaking the law and on the misuse of social
security as indicators for this. Besides this, there are imperative social norms, which
generally do not lead to legal punishment in case of defection, but may have serious
negative social sanctions if they are violated. As indicators of this type we used views
on the equal treatment of men and women, behaving in line with the work ethic, and
involvement in society.

20 Social exclusion: a conceptualisation

3 Research findings

The conceptual model containing the risk factors and features of social exclusion
served as a starting point for the design of a questionnaire which was used in a
verbal survey, conducted in 2003. In this empirical part of our project, we confine
ourselves to an analysis of the concepts and the relations at the micro level (the
upper part of figure 2.1). A multilevel approach would have been more appropriate,
but is not feasible with the current dataset. Key data on the meso- or macro-level
in our models are either not available (such as the behaviour of municipalities and
other intermediate actors), or do not show the temporal or spatial variation one
would need in order to perform such an analysis (e.g. policy measures of the central
government).

In analysing the data, we first constructed an index for each dimension of social
exclusion. Subsequently, we developed a social exclusion index which summarizes
the scores on the separate dimensions. This was followed by a multivariate analy-
sis, through which we assessed whether the various risk factors have independent
effects. The outcomes were used to construct a number of risk profiles. Finally, we
performed a structural causal analysis, which made it possible to examine how the
process of social exclusion unfolds through the risk factors.

3.1 Collection of new data

Our study intended to map out the extent of social exclusion, the size of the risk
groups and the relationship between risk factors and social exclusion. It focused on
the blocks of variables at the micro-level (above the dotted line in figure 2.1). This
requires an integrated dataset containing all information at the individual or house-
hold level; and because this was not available, a new survey was needed.
A major problem in collecting such data is that vulnerable groups may be less
inclined to participate in the survey. People with poor education, benefit claimants,
members of ethnic minorities and the residents of deprived areas are less easy to
reach in a survey, whereas it is precisely such groups that theoretically are at risk of
social exclusion. Selective non-response could thus lead to a serious underestimation
of the degree of social exclusion in the population.
A second problem is that social exclusion probably affects a relatively limited
share of the population, making it difficult to observe the phenomenon reliably in a
standard national sample of limited size.
We decided to address these problems in the following way:
– Respondents were selected in residential neighbourhoods that were stratified
9
by the scp’s standard status score of neighbourhoods. Half the interviews (431)
were carried out in 20 districts from the lowest status score quintile, the other

21
 half (429) were conducted in 30 districts with a higher status score. To obtain
results that are representative for the Dutch population over 25 years and living in
independent households, the data were weighed according to the status score.
– The interviews were carried out verbally in the respondent’s home; the interviewer
handed the potential respondent a short written letter in several different lan-
guages (Dutch, Turkish, Arabic, English) explaining the purpose of the survey
and offering the prospects of a reward of 10 euros for participating in the survey.
– The interviews in low status neighbourhoods were carried out as far as possible
by bilingual interviewers (Turkish/Dutch, Moroccan/Dutch) who had a question-
naire translated into Turkish or Arabic; in addition, all interviewers had a copy of
the questionnaire in Dutch and English.
– Care was taken to ensure that the percentage of the ethnic minority respondents
was representative for the district concerned; the data collection agency was
required to interview a predetermined percentage of members of ethnic minori-
ties in each district.

The data were collected in the summer of 2003, the final report was published in
Dutch little over a year later (Jehoel-Gijsbers 2004).

3.2 Indices for the dimensions of social exclusion

One of the main objectives of our study was to investigate whether it would be possi-
ble to construct a general index for being socially excluded. With this aim in mind, the
four sub-dimensions of social exclusion were each operationalised through a set of
indicators, and subsequently submitted to Categorical Principal Component Analysis
10
(Catpca), an optimal scaling procedure. Appendix A contains more detailed infor-
mation about the indicators and the results of the scaling procedure. The analysis led
us to conclude that reliable scales could be constructed for the dimensions of material
deprivation (nine indicators; α=0.89) and of social participation (eight indicators;
α=0.76). The ‘access to social rights’ empirically showed two sub-dimensions:
1) access to adequate housing and a safe environment (seven indicators; α=0.71)
2) access to institutions and provisions (five indicators; α=0.82).

The scale-construction for the dimension ‘cultural/normative integration’ caused

some problems. Only four of the eleven included indicators formed a consistent
scale, but with a lower reliability than the other dimensions (α=0.61). The four items
all had to do with trespassing the law.

The full report discusses the results on these five scales and the associated risk
factors extensively (Jehoel-Gijsbers 2004: 59-106). Here, we confine ourselves to
some general outcomes (see table 3.1).

22 Research findings
Table 3.1
Index score for each of the dimensions of social exclusion, to risk groups; higher score means
more exclusion (mean score for each dimension for the total group= 0) a

Dimension 1 Dimension 2 Dimension 3 Dimension 4

a. access to b. access to
material good social social normative
Risk factors deprivation housing institutions participation integration
Not amenable to intervention
≥ 65 year - 0,22 - 0,30 - 0,18 0,33 - 0,19
Woman 0,09 0,04 - 0,02 0,01 0,02
Single 0,14 0,02 0,04 0,29 - 0,10
Lone parent child <18 yrs 1,22 0,59 0,37 0,47 0,08
Couple with children <18yrs 0,03 0,01 0,02 - 0,13 0,04
Low education of father
(maximum grammar school) 0,14 - 0,07 - 0,04 0,22 - 0,10
Non-western ethnic minority 0,86 0,42 0,30 0,34 - 0,07

Amenable to intervention
Low education respondent
(<mbo) 0,26 0,00 - 0,03 0,43 0,03
Poor mastery of Dutch language 0,63 0,24 0,48 0,47 0,12
Poor ict-capabilities 0,14 - 0,08 - 0,15 0,57 - 0,15
Poor administrative capabilities 0,13 - 0,08 0,04 0,39 - 0,02
Not a good health 0,41 0,11 0,30 0,54 - 0,03
Chronical disease 0,25 0,06 0,21 0,39 -0,12
Low psychological well-being
(MHI-score <60) 1,04 0,40 0,73 0,99 - 0,17
In past 5 yrs: long term illness 0,38 0,12 0,34 0,27 - 0,05
HH <65yr b : no working hh-members 0,59 0,21 0,42 0,45 - 0,06
HH <65 yr b : on benefi t (unemployment,
disability, social assistance) 0,70 0,19 0,66 0,58 0,22
HH <65 yr b : in past 5 yrs: unemployed
at least once 0,66 0,33 0,63 0,41 0,10
<105% of social minimum income 1,12 0,27 0,71 0,95 -0,09
Living in low status area 0,31 0,37 0,15 0,15 -0,13

Total 0 0 0 0 0
a Significant differences are printed in bold (p <0,05). Significance is tested referred to the non-risk group.
b HH <65 yr: household of which the head and/or partner is younger than 65 years.

Source: SCP (Social Exclusion 2003)

Nearly all risk factors mentioned in the conceptual model (figure 2.1) show a
significant correlation with the dimensions of material deprivation and social
participation. To a somewhat lesser extent this also applies to the two ‘social right’

Research findings 23
dimensions. However, hardly any risk factor is empirically related to cultural/norma-
tive integration. This may be due to the methodological problems we encountered in
constructing this scale.
Taking the separate risk factors into consideration, it seems the proper ones were
selected in the model. Only two theoretical relationships are not corroborated by the
empirical evidence. Gender has little to do with any of the dimensions of social exclu-
sion, while age correlates with four of five dimensions in the ‘wrong’ direction: in
the Netherlands elderly people turn out to be less excluded than younger ones, except
where social participation is concerned.
We also counted whether certain groups are above a threshold value on the dimen-
sions, which provided a first idea on the accumulation of the different aspects of
11
social exclusion as a state. Thus measured, nearly two third of our respondents is
not excluded on any dimension at all. Roughly a quarter is above the threshold value
on one aspect, while about 12% experiences an accumulation of social exclusion (a
high score on at least two dimensions).

3.3 A general index for social exclusion

We will not dwell on the outcomes on the separate dimensions any further here,
but prefer to focus on our more general index of social exclusion instead. In order
to construct this we used Overals, which performs a nonlinear canonical correlation
analysis. The programme is part of a separate module of the spss software pack-
age, and its theoretical underpinnings are described by Gifi (1990: 204). Overals is
especially well-suited to our purposes, since it allows us to test simultaneously if
the various indicators actually fall into the coherent subdimensions we theoretically
expect, and whether by combining these subscales a good measure for the general
concept of social exclusion can be obtained. The outcomes once again show four
valid subscales: material deprivation, two subscales for access to social rights, and
social participation. Moreover, these subdimensions scale in the same direction,
which suggests they actually are different aspects of one and the same underlying
12
phenomenon.
The indicators for normative integration, however, did not fit in very well (see
table 3.2). This may lead to two different conclusions. The first is that normative
aspects of the kind we measured refer to a different kind of exclusion: people who
are, e.g., deprived in material terms, empirically do not have more deviating views on
normative issues than the well-to-do. If this were the case, one could conclude that
we have actually found two types of being socially excluded (the general underlying
dimension, and a lack of normative integration), which need to be analysed sepa-
rately. The second interpretation could be that there is no such thing as a consistent
set of indicators for ‘normative integration’ among our respondents, or if there is,
that we have not succeeded in measuring it adequately. If one looks at the scaling
results, we are inclined to hold the latter view. In the separate analysis of this dimen-
sion discussed above, we had to drop most of the indicators that were intended to

24 Research findings
measure this aspect of being socially excluded, and even then the reliability for the
scale of the remaining four normative issues was not quite high. The same happens
in performing Overals: the total fit of the solution increases considerably if the nor-
mative items are excluded (the proportion of explained variance rises from 0.46 to
0.56). Because people have not responded in a very coherent way on the survey ques-
tions which were meant to measure normative integration, we decided to drop the
normative integration issues from the general index. In future research we intend to
investigate whether the measurement of the ‘lack of cultural/normative integration’
aspect of social exclusion may be improved.

Table 3.2
Fit with general index for social exclusion (results of Overals procedure)

Fit with general index Fit with general index

including dimension 4 excluding dimension 4
Dimension 1: material deprivation
8 indicators 0.66 0.68
Dimension 2: access to social rights
a: access to good housing(conditions),
7 indicators 0.47 0.48
b: access to social institutions and provisions,
5 indicators 0.53 0.53
Dimension 3: social participation
8 indicators 0.53 0.55
Dimension 4: normative integration
4 indicators 0.13 --

Total fit (proportion of explained variance) 0.46 0.56

Source: SCP (Social Exclusion 2003)

The remaining dimensions allowed us to construct an index which may be used to

express the degree of being socially excluded in a single figure. It should be borne in
mind that this figure indicates a relative position. The distribution of the respond-
ents on this general index is presented in figure 3.1: the higher the score, the more
socially excluded a person is.

Most of the respondents score around zero, none below minus 1, and some above 1.
Although the index essentially is a relative measure, the latter score implies a rather
extreme position, statistically. If we apply this as a threshold value, 11% of the total
population of adults above 25 years would belong to the socially excluded. For cer-
tain groups, however, the percentage turns out much higher.

Research findings 25
Figure 3.1
Frequency distribution of the general index for social exclusion

not socially excluded socially excluded

number of respondents

250

200

150

100

50

0
Below -1
-1 to -0,75

-0,75 to -0,50

-0,50 to -0,25
-0,25 to 0,00

0,00 to 0,25

0,25 to 0,50
0,50 to 0,75

0,75 to 1,00

1,00 to 1,25
1,25 to 1,50

1,50 to 1,75

1,75 to 2,00
2,00 to 2,25

2,25 to 2,50

2,50 to 2,75
2,75 to 3,00

3,00 to 3,25

3,25 to 3,50
3,50 or higher
index for social exclusion

Source: SCP (Social Exclusion 2003)

Figure 3.1 shows the main risk groups. These people suffer a high degree of social
exclusion, at least to current Dutch relative standards.
It is rather remarkable that people with a poor health in general show the highest
score on the index for social exclusion, whereas not having paid work (in households
aged under 65 years) appears only in ninth place. It has to be borne in mind here
that a third of the non-working group is in receipt of early retirement or surviving
dependant’s benefit, while two-thirds are on unemployment, social assistance or
disability benefit. These latter categories of benefit claimants achieve a higher score
on the social exclusion index (they are in fifth place) than the total group of house-
holds without paid work. Atkinson (1998) and Paugam (1997) give another possi-
ble explanation for the fact that (not) having a paid job is not a particularly strong
predictor of being socially excluded. They argue that some jobs might in fact be more
socially excluding than being on social assistance. In the same line of reasoning,
Saraceno (2001) states that the idea that any job is better than no job if one wants to
avoid social exclusion is at least over-simplistic.

26 Research findings
Figure 3
Main risk groups for social exclusion in the Netherlands

index score % with index score >1

1,4 50

45
1,2
40
1 35

30
0,8
25
0,6
20

0,4 15

10
0,2
5

0 0
(very) <105% single- frequently on lowest non- poor no long total
*
poor minimum parent unem- benefit income Western command working illness average
health income ployed quintile ethnic of Dutch hh. in past
in last 5 yr* minority members* 5 yr

index score % with index score >1

* households < 65 years

Source: SCP (Social Exclusion 2003)

Another somewhat unexpected finding is that pensioners do not appear in the above
list of risk groups, whereas they are in poorer health and have a lower income. This
result contradicts the common view of pensioners as a very vulnerable group. In
terms of social exclusion, this view is based mainly on social isolation and poverty.
And indeed, when it comes to the dimension ‘social participation’, the research
findings show that over-65s do achieve a higher score than the rest. However, the
empirical results also indicate that they are better of on the other dimensions: they
suffer from material deprivation significantly less often and also have fewer prob-
lems with social institutions and with issues relating to safety and housing. Once
again, it should be stressed that this concerns their relative position compared to
the other age groups. There are of course pensioners who do encounter these forms
of exclusion, but the percentage of them is lower than in the rest of the popula-
tion. The explanation for the fact that old age seems a protective factor rather than
a risk booster must probably be sought in the different behaviour, different needs
and different spending patterns of older persons. The general coverage of the older
Dutch population in the pension and health systems at a rather high level may also
be relevant here.

Research findings 27
One comment should be made here about the fact that the socio-economic status of
the neighbourhood does not appear on the list of main risk factors. At the start of
our study we assumed that social exclusion would be a phenomenon mainly to be
found in deprived neighbourhoods. To some degree this turns out to be the case, but
the difference with non-deprived neighbourhoods is not that large. Moreover, after
controlling for the other risk factors in an additional multilevel analysis, the effect of
this factor disappears altogether. This implies that the somewhat greater degree of
being socially excluded in disadvantaged neighbourhoods is caused by the popula-
tion profile of the neighbourhood (a large share of people with risk characteristics),
not by the neighbourhood as such.

Something similar is found for people of non-Western ethnic origin. Although

they appear at number six in the above list, after controlling for other risk factors
ethnicity has no direct effect on their degree of being socially excluded. Non-West-
ern immigrants are more excluded because they have many risk characteristics, not
because of their ethnic origin. This implies that an immigrant from Morocco is no
more excluded than a native Dutch person if other characteristics (level of education,
age, income, health et cetera) are held constant. The same conclusion emerges from
a more elaborate causal analysis, presented in section 3.5 below.

3.4 Profiles of risk groups

Figure 3.2 showed the index score for separate risk categories. Risk factors often
occur in combination and may reinforce each other. Multivariate analysis was used
to identify the combination of variables that provides the best explanation for the
differences in the index for social exclusion. Since a number of risk factors are not
relevant for the elderly, separate analyses were carried out for the group aged under
65 years and for the over-65s.

The factor general health shows by far the strongest correlation with being socially
excluded. A relatively high proportion of the differences in degree of social exclusion
is explained by this one factor: for those below the age of 65 this amounts to 16% of
the total variance in the dependent variable, for the over-65s the figure is even higher
(25%). The results indicate that the combination of poor health, frequent unemploy-
ment in the last five years and low income has a major impact on the degree of social
exclusion (index score=2.2). Where this combination occurs, other risk factors add
little: the three risk factors explain 31% of the total variance in the index, whereas
adding six more explaining variables add 5 percentage points. Among the over-65s,
too, the combination of risk factors with the highest explanatory power for social
exclusion is also poor health and low income. Instead of unemployment (a factor
that is not relevant for this group), not performing unpaid work is a significant factor
here, albeit with a limited additional effect. These three risk factors explain 35% of
the total variance in the index.

28 Research findings
A number of other accumulations that are relevant for policy were also examined.
The left-hand section of table 3.3 shows the predicted index for a number of combi-
nations of risk factors.

Table 3.3
The impact of the accumulation of risk-increasing and risk-reducing factors a (< 65 years) b

Index for Index for

Accumulation of risk-increasing social Accumulation of risk-increasing social Change
factors only exclusion and risk-reducing factors exclusion in index
1a. very poor health + benefit 1.6 1b. very good health + benefit 0.1 - 1.5
2a. low income + single-parent 2b. high income +
family 1.2 single-parent family - 0.1 - 1.3
3a. low income + ethnic minority 3b1. high income + ethnic
minority - 0.3 - 1.4
1.1 3b2. low income + indigenous 0.7 - 0.4
4a. ethnic minority + single-parent 4b. indigenous + single-parent
family 1.1 family 0.7 - 0.4
5a. low income + couple with 5b. high income + couple with
children 1.0 children - 0.5 - 1.5
6a. ethnic minority + poor 6b1. indigenous + poor
command of Dutch command of Dutch 0.6 - 0.3
6b2. indigenous + good
0.9 command of Dutch 0.2 - 0.7
a Predicted index values; risk-reducing factors are shown in bold.
b Average index value < 65 years = 0.1.

Source: SCP (Social exclusion 2003)

People with poor health who are in receipt of a benefit score highly on the social
exclusion index compared with the other risk combinations shown. Single-parent
families with a low income are slightly more socially excluded than members of
ethnic minorities with a low income. Couples with children and a low income score
slightly lower on the index than single-parent families with the same low income.
Evidently having a complete family offers the prospect of a better position in society
even where the financial capacity is low. The combination of ‘non-Western ethnic
minority’ with ‘poor command of Dutch’ takes the degree of social exclusion above
the average, but to a more limited extent than the other combinations.
The right-hand side of table 2 gives the predicted index value if one of the two risk-
increasing factors is replaced by a risk-reducing, or protective, factor. This allows us
to quantify the risk-reducing effects; for example, what is the impact on social exclu-
sion if one is in receipt of benefit, but not combined with a very poor but with a very
good health. Comparing the two halves of the table shows that for benefit claimants,
being in very good health greatly reduces the degree of social exclusion (from 1.6 to
0.1). Furthermore, replacing a low income with a high income is a key factor in low-

Research findings 29
ering social exclusion (a reduction with 1.3 to 1.5 points). Ethnic origin has a much
smaller influence: indigenous people with a certain risk factor score only 0.3 – 0.4
lower on the index than members of ethnic minorities with the same risk factor.

Subjective indicators and the index

To construct the index for social exclusion we selected indicators which are as objec-
tive as possible. We asked the respondents about their actual situation, obviously
as perceived through their cognitive process. The goal was to measure the extent
of being socially excluded in terms of peoples circumstances, not whether they
felt isolated, unhappy or dissatisfied. Nevertheless it is interesting to examine the
relation between our measure of social exclusion and a number of subjective indica-
tors. The correlation coefficients in table 3.4 are rather high, especially those with
income satisfaction (-0.56) and with psychological well being (-0.48). Those who are
socially excluded generally are less satisfied, especially with their income, and have a
poorer mental health than the non-excluded. Once again, the causal direction is not
quite clear here; but the fact that feelings of dissatisfaction and a low psychological
well being are so closely connected with social exclusion provides some subjective
evidence for the validity of the index.

Table 3.4
Correlation between subjective indicators and the index for social exclusion (correlation
coeffi cient; maximum=1)

Correlation with index for social exclusion

Satisfaction with life in general - 0.47
Satisfaction with the number of social relationsa - 0.38
Satisfaction with housing – 0.33
Satisfaction with living conditions - 0.37
Satisfaction with income - 0.56
The mental health indexb - 0.48
a Based on the question if one prefers more or less social contacts than he/she has at the moment.
b he mental health index used consists of 5 items from the General Health Questionnaire (GHQ).

Bron: SCP (Social exclusion 2003)

3.5 Processes of social exclusion

The conceptual model in figure 1 implies that the various risk factors do not by defi-
nition, or at least not only, have a direct effect on the degree of social exclusion. It may
be assumed that they also influence each other, and thus also have an indirect impact
on the dependent variable. The direct and indirect effects of the various risk fac-
tors were tested empirically through a structural equation model developed in Amos

30 Research findings
(Arbuckle & Wothke, 1999). The empirical model is derived from the theoretical rela-
tions at the micro-level noted earlier, and fits the data well (cf. figures 2.1 and 3.3).

Figure 3.3
Empirical determinantsa of social exclusion in the Netherlands (for the group 25-65 years)

lone parent
health
family

single
person benefit
recipiency

couple with
children
social exclusion index
level of household - material/financial deprivation
education income - low access to social rights
age - lack of social participation

ethnic
mastery
minorities
of Dutch

father’s
education
ICT- Goodness of fit:
capabilities χ2/df= 2,14
gender NFI=0,99
RFI=0.98
RMSEA=0,04

a arrows in thick lines denote main casual effects

Source: Jehoel-Gijsbers (2004)

Table 3.5 presents the coefficients belonging to this graph. The outcomes show
that the general health situation is found to have the strongest direct effect: social
exclusion rises considerably when people are in poor health. Income takes second
place, followed by being in receipt of unemployment, social assistance or disability
benefit and having a poor command of the Dutch language. Single-parenthood also
has a substantial direct effect, but its indirect effect, via income, is almost as great
– hardly surprising, given that single-parent families often have a low income. ict
skills have a limited direct effect: the lower these competences, the more people are
excluded. This variable includes not only ‘computer skills’, but also activities that
may be counted as basic skills required in a modern society, such as the ability to use
cash machines, to buy public transport tickets from automatic vending machines et
cetera.
The effect of age turns out to be rather complicated. We found a strong negative
direct effect (the older people in the 25-65 age group are, the less they are excluded),

Research findings 31
which is partly cancelled out by an opposing indirect effect (more exclusion with
increasing age). This indirect effect is caused by the fact that older persons are less
healthy and possess fewer ict-related skills, increasing their score on the index. On
balance, however, old age proves to have a protective effect against exclusion. This is
found to apply not only for those in the 25-65 age group, but also for the elderly (not
presented here): among the over-65s, too, the degree of social exclusion is found to
reduce with increasing age.

Table 3.5
Indirect, direct and total effects on social exclusion in the Netherlands (standardised
coeffi cients; 25-65 years)

indirect effects direct effects total effects

Risk factors not open to policy intervention
couples with children (0=no; 1=yes) -0.02 - -0.02
single persons (0=no; 1=yes) 0.16 - 0.16
lone parent families (0=no; 1=yes) 0.12 0.10 0.22
gender (0=women; 1=men) -0.01 - -0.01
age (in years) 0.07 -0.13 -0.06
non-western ethnic minorities (0=no; 1=yes) 0.12 - 0.12
father’s education (Æ higher) -0.08 - -0.08

Risk factors amenable to policy intervention

health (Æ worse) 0.07 0.26 0.33
ICT skills (Æ less) - 0.08 0.08
command of Dutch language (Æ worse) 0.03 0.16 0.20
respondent’s education (Æ higher) -0.16 - -0.16
a,b
unemployment/disability/social assistance benefit
(0=no; 1=yes) 0.07 0.17 0.24
household incomec (Æ higher) -0.08 -0.23 -0.31
a benefit received by head of household and/or partner
b if the variable ‘in receipt of benefit’ is replaced by ‘paid work’ the effects are: – 0.09 (indirect), – 0.08 (direct) and
– 0.16 (total effect)
c net income of head household plus partner (if any)
- left out of analysis after assessing that the effect was non-significant

Source: SCP (Social exclusion 2003)

One striking finding is the fact that two risk characteristics that are considered
important, namely education level and ethnic origin, have no direct influence on the
degree of social exclusion. Instead, their effect runs entirely through other risk fac-
tors. The influence of a poor education operates via a low income and poor health;
among non-Western ethnic minorities, the causal chain goes via a low education
level, poor command of Dutch and poor ict-related skills. This means ethnic origin

32 Research findings
and level of education are key background variables for explaining the process of
social exclusion.
Couples with children are not a group with a high risk of social exclusion; this is
mainly because they generally have at least one income from employment and they
have social contacts through their children. The same applies for gender: there is
only a very slight indirect effect, through women’s lower score on ict skills. Gener-
ally speaking, women below the retirement age experience hardly any more social
exclusion than men, though it must of course be borne in mind that this is a ‘pure’
effect, after correction for the influence of other factors. The strong effect of being a
single-parent family noted above largely concerns women.

Three remarkable elements emerge from this empirical model:

1. First, contrary to expectations, income is not the main determinant of social
exclusion, at least in the way that the concept was measured in this study. People’s
health is more important; and social exclusion most certainly cannot be equated
to their income position, a notion that corresponds with the theoretical literature
discussed earlier.
2. Secondly, the effect of being in receipt of benefit is primarily a direct one. It might
be assumed, as with single-parent families, that the influence of this risk factor
would operate mainly via the level of income, but in fact this indirect effect proves
to be limited. This means that the relatively high degree of social exclusion of
benefit claimants can be attributed to only a small extent to their income. Being
in receipt of benefit, regardless of the level of income, itself increases the degree
of being socially excluded. Compared with being in receipt of benefit, having paid
13
work has a relative small direct effect (- 0.08). This means that having a job
alone is not a main explanatory factor for reducing exclusion.14
3. Finally, it is worth noting that many of the more important determinants of social
exclusion are in principle amenable for policy intervention. This applies to health,
income, benefit recipiency, and command of the Dutch language. One-parent
families are the only risk group where the direct effect is strong, and where social
exclusion will be hard to influence through policy measures aiming to reduce
single parenthood.

The causal chains described above relate to the population aged under 65 years.
As a number of the risk factors that are relevant for them (such as being in receipt
of unemployment, social assistance or disability benefit and having a history of
unemployment) are not important for the over-65s, a separate causal analysis was
performed for the latter group. It was found that, here too, health has the strongest
direct effect, followed by income. Performing unpaid work also has a direct (though
limited) influence in combating social exclusion.

Research findings 33
4 Conclusions

Social exclusion is a wide-ranging and complex phenomenon, and it is therefore not

easy to represent it with a single figure. In our study we made an attempt to develop
a numerical index for social exclusion as a state. This does not allow an absolute
threshold to be drawn above or below which social exclusion can be said to exist.
What it does do, however, is to assess the relative position of various risk groups in
terms of being socially excluded, the impact of risk factors, and the causal mecha-
nism through which they operate. This may be an important policy tool, which can
be used to set priorities and to develop policy strategies which tackle the risk factors
that produce social exclusion. This could provide more focus to the national action
plans.

This study also demonstrates clearly that the main policy indicators currently
employed (income, work and unemployment), can only partly explain the degree of
being socially excluded – the correlation is fair, but by no means complete. A number
of other individual characteristics turn out to play a key role, independently of these
factors; in particular health, but also command of the national language, are impor-
tant determinants of social exclusion. Also striking is the strong correlation with
subjective indicators such as psychological well-being.

To date Dutch policy in combating poverty and social exclusion focused strongly
on the reduction of unemployment and benefit dependency. To achieve this, an
important change was made in the access to social benefits and in the execution of
reintegration activities: it switched from public to private organizations. Moreover,
combating school-dropout is considered important to promote social inclusion, as
well as the recent policy measure to oblige ethnic minorities to learn Dutch. In our
view too little attention is paid to health factors, which are mostly regarded from a
medical point of view, and far less as a problem which may have serious socio-rela-
tional consequences. Moreover, the last few years the dominant Dutch policy line
shifted from compensating and helping the disadvantaged, to stressing their own
responsibility. Although this may be a tenable policy maxim for the privileged and
the well-educated citizens, one may question the ability of socially excluded people
to live up to these expectations.

The theme of social exclusion is set to remain on the policy agenda in the years
ahead. All Member States have an obligation to report to the eu on this matter; in
doing so, they make use of a number of policy indicators agreed at the eu level. Our
study indicates that social exclusion can also be assessed in another way, with a
stronger emphasis on the characteristics of the phenomenon as such, and with spe-

34
cific attention for the process of exclusion. The two perspectives should, in our view,
be complementary. The Laeken indicators are important benchmarks against which
the relative success of different types of policy may be assessed. However, if the aim
is to ascertain the degree to which social exclusion actually exists in the Member
States, consideration also needs to be given to a direct measurement of the phe-
nomenon. In addition, if the object is to study the causes of social exclusion and to
examine the extent to which they may be amenable to policy intervention, an attempt
should be made to model the actual impact of risk factors. The present study has,
for instance, shown that income is not such an all-important determinant of social
exclusion as is often assumed and that, as a minimum, the autonomous impact of
health, benefit recipiency and command of the national language should be added.

Of course, this study represents no more than a first step. Even in the Dutch case,
the approach is in need of refining, for example as regards the ‘cultural/normative’
dimension of social exclusion. In other eu Member States and Candidate Countries
it is possible that other causal factors will emerge, and the relative weights of the
various determinants may be quite different. First steps have to be taken, however;
and in our view supplementing the Laeken indicators is a worthwhile exercise, both
conceptually and in order to increase our empirical understanding of the phenom-
enon of social exclusion.

Conclusions 35
Notes

1 In recent policy documents at the European level the concept of ‘social exclusion’
has gradually been replaced by ‘social inclusion’. The difference between the two is
utterly vague. ‘Inclusion’ suggests a process through which people are ‘brought back
into society’ from a position of backwardness, preferably through wilful and effective
governmental interventions. However, both in policy and research social inclusion often
is treated as a lack of social exclusion – the Laeken-indicators, for instance, pretend to
measure both. In this publication we regard the two concepts as complements, and
throughout use the term social exclusion.
2 In the Czech Republic and Hungary the at-risk-poverty rate is even much lower: 8% and
9%.
3 Another explanation is that ‘income-in-kind’ was included in the total income defi nition
of the new member states and candidate countries, whereas it is left out of consideration
in the eu-15. Eurostat (2004a) motivates this by mentioning that such income compo-
nents (e.g. own production of food, hunting and fishing, withdrawal from stocks by
tradespeople, government provision of housing, meals, crèches) account for a substan-
tial part of total income in the new countries. Furthermore Eurostat (2004a) notes that
inequality is low in the new member states and candidate countries (due to historical
circumstances, the lack of information on the hidden economy, and misrepresentation
of very poor and very rich people). If one uses a relative poverty threshold, poverty tends
to decline if inequality decreases.
4 The primary reference point in the Laeken-indicators is the ‘at-risk-of-poverty rate’,
defined as 60% of the median income. Other poverty indicators include long-term
poverty, poverty based on the 60% income threshold anchored in time, the poverty rate
before and after social transfer and the poverty gap. Alternative poverty thresholds use
40%, 50% and 70% of median income.
Other Laeken-indicators are non-monetary: e.g. the share of long-term unemployment
(12 or 24 months) and people living in households where no one has paid work; regional
cohesion, indicated by the regional dispersion of employment at the so-called nuts2-
level; the share of early school-leavers and of people in the 25-64 age bracket having
completed lower secondary school or less; and the health situation, mainly measured by
life expectancy at birth.
5 The full report is published in Dutch (Jehoel-Gijsbers, 2004). A preliminary version of
this summarizing publication was presented at the European Union’s Third Round Table
on Poverty and Social Exclusion (Rotterdam, 18-19 October 2004).
6 It should be mentioned that since the Laeken indicators were agreed upon in 2001, they
have been refined and extended somewhat (e.g. with the indicator ‘literacy performance
of 15-year old pupils’). Moreover, within the set of indicators the position of children and
the elderly has been elaborated. Nevertheless, precedence is still given to income and
employment (see European Commission 2004).
7 The idea that assimilation into dominant culture is a prerequisite for social inclusion
is, of course, central to Durkheim’s theory, for instance in his suicide typology. Silver
(1994: 542) states that post-modernist uses of the term ‘dominant culture’ incorporates
multicultural notions about how the basis of solidarity is, or should be, reconfigured.
8 The items do not only regard the most essential goods and services, but also whether
one has enough money for attending a sporting club, buying newspapers, an Internet
connection et cetera. If such less essential items are not present, respondent are asked
whether this is due to financial reasons.

36
9 The status score is based on the mean income and the percentages with a low income,
a low education level and unemployed in a given postcode area.
10 Catpca was formerly known as Princals (Principal components analysis by alternating
least squares); see Gifi (1990).
11 In assessing accumulation, persons with a score exceeding one standard deviation from
the scale mean were considered socially excluded.
12 The Overals-procedure gives a better result if the dimension of ‘access to social rights’
is split up in two subsets: one for ‘access to formal institutions’ and one for ‘adequate
housing and safe living conditions’.
13 The analysis was also performed with the variable ‘having paid work’ instead of ‘being
in receipt of benefit’. As the two risk factors (work and benefits) show a high correla-
tion they cannot be included in the analysis model simultaneously (a multicollinearity
problem). In the model with ‘paid work’ included, both the total and the direct effects on
social exclusion are lower than was the case with ‘benefit’ (–0.16 (versus 0.24 for
‘benefit’) and –0.08 (versus 0.17 for ‘benefit’)).
14 This is because about one third of the non-working households is in receipt of early
retirement or surviving dependant’s pensions. This group was not included in the
benefits variable (which refers to social assistance, unemployment and disability
benefits) and is less socially excluded.

Notes 37
Appendix A

Table A.1
Results of optimal scaling procedure a for each dimension: component loadings and reliability
(Cronbach’s alpha)

Component loading Cronbach’s alpha

Dimension 1: Material deprivation 0.89
1 payment of fi xed expenditures is very hard 0.77
2 has payment arrears 0.58
3 worries often/continuously about fi nancial situation 0.77
4 has diffi culty in making ends meet 0.88
5 fi nds it more diffi cult to make ends meet than 2 years ago 0.66
6 lacks consumer durables due to fi nancial defi cits 0.76
7 cannot afford basic expenditures 0.80
8 membership of (sports)club is too expensive 0.67
9 has diffi culties in obtaining a loan 0.65

Dimension 2a: Social rights: access to institutions and

provisions 0.82
1 often treated badly by public agencies 0.86
2 often long waiting periods for appointments/treatments at 0.84
public agencies
3 often problems with public agencies 0.81
4 refused by commercial service organisations 0.58
(banks, insurance companies etc.)
5 benefi t (according to respondent) wrongfully refused or 0.71
terminated

Dimension 2b: Social rights: access to adequate housing and safe

environment 0.71
1 frequent disturbances in neighbourhood 0.64
2 wants to move house within 2 years 0.82
3 had/expects a long search period in fi nding a new house 0.78
4 little social cohesion in neighbourhood 0.62
5 unsafe feeling in neighbourhood 0.42
6 unsafe feeling if one is home alone 0.39
7 often a victim of crime over the last 5 years 0.38

38
Table A.1 (Continued)

Component loading Cronbach’s alpha

Dimension 3: Social participation 0.76
1 feels left out of society 0.53
2 does not/hardly go out for amusement 0.67
3 experiences lack of social contacts 0.45
4 has no/little people to discuss intimate matters 0.55
5 has little social support 0.72
6 no/little membership of clubs, societies 0.60
7 no/little diversity in social contacts 0.81
8 social contacts hampered by health 0.44

Dimension 4: Cultural/normative integration 0.61

1 a false testimony is allowed if a friend faces trial 0.72
2 tresspassing the law is no problem as long as one does not
get caught 0.73
3 people with a payed job may moonlight for up to 0.64
150 euro a month
4 people on social assistance may moonlight for up to
150 euro a month 0.62
a Categorical Principal Component Analysis (CatPCA)

Source: SCP (Social exclusion 2003)

Appendix A 39
References

Abrahamson, P. (1997). Combating poverty and social exclusion in Europe, in Beck,

W., Van der Maesen, L. and Walker, A. (eds). The Social Quality of Europe, The Hague:
Kluwer Law International.
Abrahamson, P. (2001). Comments: Poverty and income distribution: inequality, financial
poverty persistent poverty and deprivation indicators. Paper presented at the International
Conference on indicators for social exclusion: Making Common European Objectives
Work.
Alcock, P. (1991). Understanding poverty. London: MacMillan Press.
Arbuckle, J.L. and Wothke, W. (1999). Amos 4.0 User’s Guide. Chicago: spss/Smallwaters
Corporation.
Atkinson, A.B. (1998). Social exclusion, poverty and unemployment. In Atkinson, A.B.
and Hills, J. (eds.) Exclusion, Employment and Opportunity, case Paper No. 4. London:
Centre for Analysis of Social Exclusion, lse, pp. 1-24.
Atkinson, T., Cantillon, B., Marlier, E. and Nolan, B. (2002). Social Indicators: the eu and
Social Exclusion. Oxford: Oxford University Press.
Atkinson, A.B., Cantillon, B., Marlier, E. and Nolan, B. (2005). Taking forward the eu
Social Inclusion Process; An independent Report Commissioned by the Luxembourg
Presidency of the Council of the European Union, downloadable at: [Link].
lu/eu2005lu/inclusion.
Bailey, N. (2006). Does work pay? Employment, poverty and exclusion from social
relations. In Pantazis, C., D. Gordon & R. Levitas (eds.) Poverty and social exclusion in
Britain; The millennium survey. Bristol: The Policy Press, pp. 163-189.
Bhalla, A. and Lapeyre, F. (1997). Social exclusion: Towards an Analytical and
Operational Framework. In: Development and Change 28: pp. 413-433.
Berghmann, J. (1995). Social exclusion in Europe: policy context and analytical
framework. In G. Room (ed.). Beyond the Threshold: the Measurement and Analysis of Social
Exclusion. Bristol: The Policy Press, pp. 10-28.
Burchardt, T., le Grand, J. and Piachaud, D. (2002). Introduction. In: Hills, J., le Grand, J.
and Piachaud, D. (eds.). Understanding Social Exclusion. Oxford: University Press, pp. 1-12.
Council of European Communities (1985). Council Decision of 19 December 1984 on specific
Community action to combat poverty. Luxembourg: Official Journal of the European
Communities, No L 2.
De Koning, J. and Mosley, H. (eds.) (2001). Labour Market Policy and Unemployment: Impact
and Process Evaluations in Selected European Countries. Cheltenham: Edward Elgar
Publishers.
Durkheim, E. (1897). Le Suicide; Étude de Sociologie. Paris: Félix Alcan.
Engbersen, G., K. Schuyt, J. Timmer & F. van Waarden (1993). Cultures of unemployment; a
comparative look at long-term unemployment and urban poverty. Boulder: Westview Press.
Engbersen, G. & R. Staring (2000). De morele economie van lage-inkomensgroepen:
armoede en informaliteit. In: G. Engbersen, J.C. Vrooman & E. Snel (eds.). Balans van
het armoedebeleid; vijfde jaarrapport armoede en sociale uitsluiting. Amsterdam: Amsterdam
University Press.
European Commission (2002). Joint Report on Social Inclusion 2001. Luxembourg: Office for
Official Publications of the European Communities.
European Commission (2004). Joint Report on Social Inclusion 2003. Luxembourg: Office for
Official Publications of the European Communities.

40
European Communities (ed.) (2004). Facing The Challenge; The Lisbon strategy for growth and
employment. Report from the High level group chaired by Wim Kok, November 2004.
Luxembourg: Official Publications of the European Communities.
Eurostat (2004a). Monetary Poverty in New Member States and Candidate Countries.
In: Statistics in Focus,12/2004. Luxembourg: Eurostat.
Eurostat (2004b). Poverty and social exclusion in the EU. In: Statistics in Focus,16/2004.
Luxembourg: Eurostat.
Foster J., Greer, J. and Thorbecke, E. (1984). A Class of Decomposable Poverty Measures.
In: Econometrica, 52: 761-766.
Gallie, D. and Paugam, S. (eds.) (2000). Welfare Regimes and the Experience of Unemployment
in Europe. Oxford: Oxford University.
Gifi, A. (1990). Nonlinear Multivariate Analysis. Chichester: Wiley.
Goodin, R.E., B. Headey, R. Muffels & H.-J. Dirven (1999). The real worlds of welfare
capitalism. Cambridge: Cambridge University Press.
Gough, I. (1997). Social aspects of the European model and its economic consequences.
In: Beck, W., van de Maesen, L. and Walker, A. (eds.). The Social Quality of Europe. The
Hague: Kluwer Law International.
Hills, J., Le Grand, J. and Piachaud, D. (2002). Understanding social exclusion. Oxford:
Oxford University Press.
Jäntti, M. and Danziger, S. (2000) ‘Income poverty in advanced countries’, in Atkinson,
A.B. and Bourguignon, F. (eds.) Handbook of Income Distribution (1), Amsterdam:
Elsevier.
Jehoel-Gijsbers, G. (2004). Sociale Uitsluiting in Nederland (Social Exclusion in the Netherlands).
The Hague: SCP (Social and Cultural Planning Office).
Jordan, B. (1996). A Theory of Poverty and Social Exclusion. Cambridge: Polity Press.
Levitas, R. (2006). The concept and measurement of social exclusion. In: Pantazis, C.,
D. Gordon and R. Levitas (eds.). Poverty and social exclusion in Britain; The millennium
survey. Bristol: The Policy Press, pp. 123-160.
Lewis, O. (1968). A Study of Slum Culture; Backgrounds for La Vida. New York: Random House.
Lewis, O. (1969). Cultures of poverty. In: P. Moynihan (ed.). On Understanding Poverty:
Perspectives from the Social Sciences. New York: Basic Books.
Katz, M.B. (1989). The undeserving poor: from the war on poverty to the war on welfare. New
York: Pantheon Books.
Kroft, H., G. Engbersen, K. Schuyt & F. van Waarden (1989). Een tijd zonder werk; een
onderzoek naar de levenswereld van langdurig werklozen. Leiden/Antwerpen: Stenfert
Kroese.
Murray, C. A. (1984). Losing Ground; American Social Policy, 1950-1980. New York: Basic
Books.
Murray, C. (1997). What it means to be a Libertarian; a Personal Interpretation. New York:
Broadway Books.
Nolan, B. and Whelan, C. (1996). Resources, Deprivation and Poverty. Oxford: Oxford
University Press.
Pantazis, C., D. Gordon and R. Levitas (eds.) (2006). Poverty and social exclusion in Britain;
The millennium survey. Bristol: The Policy Press.
Papadopoulos F. and Tsakloglou, P. (2001). Indicators of Social Exclusion in euromod.
euromod Working Paper em8/01. Athens: University of Economics and Business.
Paugam, S. (1997). La Disqualification Sociale. Paris: puf.
Room, G. (1995). Poverty in Europe. Competing paradigms of analysis. In: Policy and
Politics 23, 2: 103-113.

References 41
 Room, G. (1997). Social Quality in Europe: Perspectives on Social Exclusion. In Beck, W.,
van de Maesen, L. and Walker, A. (eds.). The Social Quality of Europe. The Hague: Kluwer
Law International.
Runciman, W.G. (1966). Relative Deprivation and Social justice. New York: Routledge.
Saraceno, C. (1997). Family change, family policies and the restructuring of welfare.
In: oecd Family, Market and Community, Social Policy Studies, no. 21: 81-100. Paris:
oecd publications.
Saraceno, C. (2001). Social exclusion, Cultural Roots and Diversities of a Popular Concept. Paper
presented at the conference ‘Social exclusion and children’, 3-4 may 2001at the
Institute for Child and Family Policy. Columbia: Columbia University.
Schuyt, C.J.M. and Voorham, C.A. (2000). Sociale Uitsluiting (Social Exclusion),
Stichting Leger des Heils welzijns- en gezondheidszorg. Amsterdam: Sociaal
Wetenschappelijke Pers.
Sen, A. (1985). A sociological approach to the measurement of poverty; a reply to
Professor Peter Townsend. In: Oxford Economic Papers, 37: 669-676.
Sen, A. (1992). Inequality re-examined. Cambridge (Mass.): Sage.
Short, K.S. (2005). Material and financial hardship and income-based poverty measures
in the usa. In: Journal of Social Policy 34:1, 21-38.
Silver, H. (1994). Social exclusion and social solidarity: three paradigms. In: International
Labour Review, Volume 133, Number 5-6, 1994: pp. 531-578.
Social exclusion Unit (2001). Preventing Social Exclusion: Report by the Social Exclusion Unit.
[Link]-offi[Link]/seu/[Link], consulted in march 2002.
Somerville, P. (1998). Explanations of Social Exclusion: Where does Housing fit in?.
In: Housing Studies, Volume 13, Number 6, 1998: pp. 761-780.
Todman, L.C. (2004). Reflections on Social Exclusion. Paper. Italy: University of Milan,
Department of Sociology and Social Research.
Townsend. P. (1979). Poverty in the United Kingdom. Harmondsworth: Penguin.
Uunk, W.J.G. and Vrooman, J.C. (2001). Sociale uitsluiting (‘Social exclusion’).
In: Armoedemonitor 2001. The Hague: scp/cbs (Social and Cultural Planning Office/
Statistics Netherlands).
Vrooman, C. and Snel, E. (1999). Op zoek naar de “echte armen” (‘Looking for the truly
poor’). In: G. Engbersen et al. (eds.). Armoede en Verzorgingsstaat: vierde jaarrapport
armoede en sociale uitsluiting. Amsterdam: Amsterdam University Press.
Wilson, J. (1994). Culture, incentives and the underclass. In: Aaron, H., T. Mann and
T. Taylor (eds.). Values and Public Policy. Washington: The Brookings Institution.
Wilson, W.J. (1987). The Truly Disadvantaged: the Inner City, the Underclass, and Public Policy.
Chicago: University of Chicago Press.
Wilson, W.J. (1997). When Work disappears; the World of the New Urban Poor. New York: Alfred
A. Knopf.

42 References

View publication stats