Material deprivation in the EU: A multi-level analysis on the influence of decommodification and defamilisation policies

Abstract

The central aim of this paper is identifying the existing political leeway for the reduction of deprivation levels in Europe. The links between household and individual characteristics and risks of material deprivation have been abundantly researched, but what are the political institutions that modify and possibly buffer these odds? Welfare state differences have been portrayed in depth but their association with social outcomes such as deprivation is less clear. By identifying the impact of decommodification and defamilisation policies on deprivation exposure, we seek to fill this gap. Our results, based on European Union Statistics on Income and Living Conditions 2012 to 2013 cross-sectional micro data, reveal that social programmes that cover large segments of the population and follow a needs-based approach are linked to lower odds of being materially deprived. A significant number of European cross-country differences in deprivation rates can thus be traced back to varying levels of social assistance, as well as to differences in the provision of public healthcare. Defamilisation policies are, despite increased risks for (single parent) families, not yet addressing the problem of low living standards successfully. An extension of childcare services, however, seems promising for lowering deprivation among families, particularly in countries with high levels of involuntary part-time work.

Keywords

Material deprivation poverty welfare states decommodification defamilisation EU-SILC multi-level analysis

Introduction

Combating material deprivation is one of the core objectives of the European Union’s social inclusion strategy. As part of the 2020 strategy, the Member States agreed on a social inclusion target, the aim being to lift 20 million people out of poverty and social exclusion by the year 2020. Among the three social inclusion indicators, material deprivation covers the individual’s actual standard of living. In contrast to the other two indicators (income poverty and low work intensity), it directly depicts material hardship, understood as a lack of access to goods and services that are common in a society. Combating material deprivation is therefore crucial towards fostering the individual’s ability to fully participate in all societal aspects.

Although the social inclusion target was declared some years ago, we can see in Figure 1 that material deprivation is still widely prevalent throughout the EU, especially in southern and eastern Member States. Against this backdrop, it is surprising that in the past years the academic debate on how to successfully combat material deprivation has not been a priority. Research on deprivation has concentrated on the impact of individual and household characteristics (Layte et al., 2001; Nolan and Whelan, 2011; Whelan et al., 2004) and welfare regime patterns (Muffels and Fouarge, 2003; Whelan and Maitre, 2012). However, comparative research on the impact of different policy instruments on deprivation levels is only just emerging (Nelson, 2012; Visser et al., 2013; Saltkjel and Malmberg-Heimonen, 2017). Some insights into the effect of social policies on deprivation might be gained from studies on income poverty (Lelkes and Gasior, 2012). Nonetheless, as groups do not overlap completely, policies that lower poverty do not necessarily lower deprivation. The question of how a country’s social policies are linked to the prevalence of material deprivation remains therefore unanswered (Saltkjel and Malmberg-Heimonen, 2017).

Figure 1.

Levels of deprivation in Europe, 2012–2013.

In this paper, we take a closer look at the institutional determinants that underlie the welfare regime’s significant influence on the prevalence of material deprivation. We aim at bringing together the academic debate on deprivation with the one on welfare states, by studying the role of welfare dimensions in producing social outcomes. The research question is: In what ways are decommodification and defamilisation policies associated with deprivation levels for the working-age population and their specific subgroups? Applying multi-level models to cross-sectional data from European Union Statistics on Income and Living Conditions 2012 to 2013 (EU-SILC 2012–2013) waves, we do not only take the benefit generosity into account, but also look at the effective coverage of the policies. Our results point towards the importance of decommodification, in contrast defamilisation policies are not significantly related to deprivation risks. The social assistance index and the public healthcare index are significantly associated with lower deprivation, presumably because they form a long-term strategy to directly cover individual needs. Childcare services, however, seem to also have the potential to address deprivation risks for families. The article is organised as follows: First, the theoretical concept of deprivation is discussed and household level risk factors are explained. The second section deals with the welfare dimensions’ influence on these risk factors. After describing the methods, we proceed to portray the welfare index and show the results from the multilevel model. Finally, our findings are synthesised and discussed critically against their political implications.

The theoretical concept of material deprivation

Deprivation is a directly observable lack of material goods and services, measured relative to what is regarded as common in society (Eurostat, 2010; Guio et al., 2012; Whelan and Whelan, 1995). Two aspects are of special relevance: First, to be deprived, there needs to be an ‘enforced lack’, meaning that individuals do not possess these goods, not because they do not want them, but because they cannot afford them (Mack and Lansley, 1985). Second, this approach emphasises the relativity of being deprived, as it divides the deprived from the non-deprived using a list of common living conditions. These common living conditions have to be adapted to what is customarily perceived as a necessity within society.

In the case of the EU, an EU-wide common living standard was defined, including among others the ability to afford a protein-rich meal every second day, an adequately warm house, and a washing machine, but also consumer products less closely related to economic strain such as a television, a car, and a one-week holiday a year. The list has remained unchanged since it was adopted in 2009. However, a renewed index was proposed in 2016 (Guio et al., 2016). The main drivers of deprivation in 2012/3 were the inability to afford a holiday and to face unexpected expenses (with approximately 40% of deprived people); least problematic were the affordability of a washing machine, a television, and a telephone (with 1% of deprived Europeans).

Income poverty describes the absence of sufficient financial resources, whereas material deprivation has been called ‘an outcome of poverty’ (Townsend, 1979): Households often become material deprived after suffering from income poverty. However, poverty does not automatically lead into material deprivation. There are households that even after a prolonged sequence of poverty are still able to afford basic goods and therefore are not materially deprived. And of course there are also households that are materially deprived without being income poor. Nonetheless, material deprivation as an enforced absence of goods is in most cases linked to reduced monetary resources.

As a so-called ‘direct’ poverty measure (Ringen, 1988), material deprivation does not only portray the resources of a household. It also reveals the extent to which the resources available to a household match the needs of its members (Notten and Guio, 2017). Whereas income poverty measures cannot live up to the challenge the challenge of adequately portraying a household’s needs (such as a situation where chronic illnesses or bad health affects household members), the direct measurement of deprivation does not face such a challenge as products would be bought according to the perceived need in a household and not according to the material deprivation list. Moreover, deprivation will reflect upon additional assets that a household owns. Savings, gifts or services can be used to finance a living standard over a certain period of time (Israel, 2015; Nelson, 2012; Till and Eiffe, 2010).

During the 2008–2012 European economic crisis, it also proved that material deprivation responds more directly to institutional and economic changes than the indirect at-risk-of-poverty (AROP) rate. Despite a certain time lag in the effect of economic changes on deprivation, because of the durable nature of many commodities on the deprivation list, it showed the extension of poverty among the middle class better than traditional poverty measures.

The social distribution of risks and resources

Although the interaction between a household’s income and needs is important for estimating its risk of deprivation, low-earners and people with higher needs are not inevitably exposed to deprivation. A household’s endowment with goods and services is fundamentally a question of the national institutions that can buffer or increase the risk of deprivation. Welfare states shape the distribution of economic outcomes such as income and living conditions by intervening in the market allocation of important resources (Kammer et al., 2012). Esping-Andersen termed this mechanism decommodification in his cross-country welfare analysis (Esping-Andersen, 1990; 1993) and defined it as the extent to which individuals are sufficiently emancipated to sustain their livelihood independent from having to participate in the labour market. He argued that a minimum standard of decommodification includes the aspect that ‘citizens can freely and without potential loss of job, income or general welfare, opt out of work when they themselves consider it necessary’ (Esping-Andersen, 1990, p. 23). The higher the degree of decommodification in a country, the easier it is for persons outside the labour market, be it temporarily or constantly, to secure a socially acceptable standard of living. Core decommodification policies for working-aged individuals are unemployment insurance, social assistance, sickness leave and disability benefits. Critics urge, however, that decommodification should not only be restricted to monetary benefits, but also include services as a second dimension (Bambra, 2004; Kautto, 2002). The decommodifying role of services is particularly applicable to public health services¹ (Bambra, 2005).

Countries do not only intervene in the market allocation, but also in the division of work within families. The choice for dual-earner or traditional family policies influences the accessibility of work and income for parents and carers. This is a second, indirect pathway through which institutions can influence a household’s resource base. These policies are usually termed defamilisation policies. Although this concept does not stand at the core of Esping-Andersen’s typology (1990), other authors, such as McLaughlin and Glendinning (1994) or Lister (1994) have dealt with it more explicitly. Along the lines of decommodification, defamilisation can be defined as the extent to which adults are emancipated in the ability to sustain their livelihoods independent of financial and non-financial support from other family members (Bambra, 2007). Following Lohmann and Zagel (2016), we regard those welfare state policies as defamilising that aim at reducing financial dependencies and/or care obligations between family members. Core defamilisation instruments are: free publicly available childcare facilities, child benefits paid to families to cover some of the children’s material needs, and parental leave or policies that allow individuals to take periods of leave to care for older or sick family members (Lohmann and Zagel, 2016).

The influence of decommodification and defamilisation on deprivation

van Kersbergen and Vis (2015) emphasise the importance of an academic deviation from a sole application of welfare dimensions towards studying their role in producing social outcomes, which fundamentally affect the life chances and capabilities of individuals. Two questions arise: How exactly is deprivation linked to these welfare dimensions? And which policies that can be summed up under the umbrella terms ‘decommodification’ and ‘defamilisation’ are likely to be effective in buffering deprivation risks? In the following section, we will review the links between welfare institutions and material deprivation, and present our hypotheses.

Decommodification. In general, the European welfare states aim at supplementing the gross-market income to allow citizens to meet their basic needs. By this means welfare states reduce the extent and the depth of deprivation (Notten and Guio, 2016). Decommodification can be expected to buffer deprivation risks through monetary benefits, such as unemployment benefits or social assistance, which are bestowed upon an individual as part of his or her citizenship rights. Such benefits have a direct anti-deprivation impact. First, income-maintenance programmes (unemployment benefits) smooth consumption in the early stage of inactivity. Second, when employment is not acquired after a certain period, social assistance provides minimum levels of consumption to households. Social assistance has already been proven to be associated with lower deprivation levels (Nelson, 2012), whereas the influence of unemployment benefit provision is as yet unexplored.

Furthermore, universally provided governmental services have been shown to lower material deprivation by increasing the purchasing power of a household (Israel, 2015). Among all governmental services that decommodify risks, healthcare provision is most likely to be a determinant of deprivation (Bambra, 2005), as the cost of a long-lasting or chronic illness is a significant burden on a household’s financial resources. The higher the number of people exempted from healthcare user fees (best through a universal provision of services), and the more extensive the free basic bundle of treatments and services, the lower the cost burden will be for individuals, making the occurrence of deprivation less likely.

Thus, decommodification measures improve the purchasing power of households in a twofold manner: directly, by supplementing the household income; and indirectly, for example, in the case of free access to health care services, by decreasing household expenses. Based on this argument, we develop our first hypothesis: The higher the extent of decommodification in a given country, in terms of effective coverage and size, the lower the deprivation rate.

Defamilisation. This measures aim at increasing the labour market participation of a household by providing benefits for households with children and ageing dependent parents. As we know from the micro-level picture of deprivation, a low work intensity of a household is highly correlated with the risk of suffering from deprivation. Dual-earner household structures, on the contrary, are associated with higher household incomes that protect from poverty and deprivation. Affordable childcare provision and provision of elderly care is a means of enabling dual-earner households by predefining the degree to which women – who traditionally tend to take the role of the carer – are included in the labour market.

In addition to services, family benefits are a way of addressing a group at high risk of deprivation and poverty, as dependent children significantly increase the material need of a household. Especially single parent households are prone to deprivation, as they rely largely on the income of only one earner. Therefore, defamilisation measures can be expected to lower the extent of deprivation by socialising the costs of family and care obligations. However, they could also have a contrary effect in lowering maternal employment by providing an alternative income that, despite being low, reduces a household’s need to work and allows women in particular to be inactive on the labour market (Thévenon, 2011; Jaumotte, 2003; Nieuwenhuis et al., 2012). In households that are at risk of deprivation, family benefits are expected to be less likely to replace in-work income but rather to supplement it, given that their reservation wage will be lower (if the charges for childcare services are no disincentive). Therefore, high family benefits and childcare services are expected to have a positive effect on material endowment. Accordingly, the hypothesis on the influence of defamilisation is: The higher the extent of defamilisation in a given country, in terms of effective coverage and size, the lower the deprivation rate.

Methodology

An examination of the data, the EU-approach to material deprivation, and the operationalisation of the welfare state effort will now be introduced.

Data

The assessment of the pattern of material deprivation is carried out using the EU-SILC database from the March 2015 revision. The dataset provides detailed individual and household-level information on all 28 EU Member States plus the European Free Trade Association countries Norway, Switzerland and Iceland. Due to limited access to macro data this study will focus on 29 European countries, excluding Cyprus and Croatia. Cross-sectional data from 2012 to 2013, which covers a time of high demand on the welfare system, will be merged to provide a robust picture of material deprivation within European countries and not just a one-year snapshot. In total, our data comprises 832,900 households with the head of the household of working age (18–59 years). Although the EU-SILC is a unique dataset that gives us the opportunity to analyse social phenomena on the supranational EU level, its output-harmonisation compromises the coherence and accuracy of comparative analysis (Iacovou et al., 2012). Some variation in the pattern of deprivation might therefore stem from diverging survey methods and designs.

Macro-level data to analyse the differences in welfare state decommodification and defamilisation were obtained from the Organisation for Economic Co-operation and Development (OECD) or aggregated based on EU micro data. For the macro-level, only the year 2012 will be portrayed. A cross-sectional time series approach is avoided due to the issues of autocorrelation and non-stationarity that arise from error terms and values that do not change to meaningful extents over the examined time period (Vis, 2013).

Operationalisation

We analyse deprivation at the household level and define the materially deprived based on their scoring on the EU deprivation indicator (Fusco et al., 2011). This indicator comprises the following nine items: the ability to afford or to pay for (a) unexpected expenses, (b) a one-week annual holiday away from home, (c) arrears such as rent, (d) a meal with meat, chicken, fish or a vegetarian equivalent every second day, (e) an adequately warm house, (f) a washing machine, (g) a colour TV, (h) telephone, and (i) a personal car (Guio et al., 2009). We use the deprivation rate as our key indicator, and apply the original 3+-threshold. This means that we consider a household to be deprived that lacks a minimum of three out of these nine items because it cannot afford them. When estimating deprivation, we control for certain household characteristics. Employment status, bad health,² and migrant background are calculated as a share of the household members with these characteristics, whereas education is based on information on the head of the household.

Different approaches exist to measure the welfare state’s dimensions of decommodification and defamilisation (see Siegel, 2005; Scruggs and Allan, 2006; Marx et al., 2014). A valid welfare state index should reflect on policy intent and outcome (Marx et al., 2014) and on the programmes’ progressivity and eligibility (Ferrarini et al., 2013). The analysis in this paper will combine the social citizenship approach (Ferrarini et al., 2013), which uses simulations of model families’ claims to benefits, with an approach to effective coverage as developed by Maquet et al. (2016). National-level social policy data will be combined into a two-dimensional index that includes first, the levels of benefits and services (the entitlements) and second, the effective population coverage (addressing issues of eligibility).

The benefit level is based on a model family with one person earning 50% of the average national wage, a second inactive adult, and two children. Simulations of the level of unemployment benefits within the first month of unemployment were retrieved from the OECD’s net income database, as were family benefits (in the case of employment). Both values were adjusted for national income differences by dividing the amount by the median national income. The figures for social assistance benefits stem from the OECD’s cash minimum income database and are also adjusted by median national income. In contrast to the replacement rate of individual income, the benefit level as a proportion of the median national income (which maintains a strong relation to the acceptable living conditions within a country) offers a tool to assess how appropriate the benefits are to address issues of poverty and deprivation.

Coverage is a crucial dimension to the effectiveness of social protection apart from adequacy (see also Bargain et al., 2012; Eurofound, 2015; Matsaganis et al., 2014). It is important to look at it as a separate dimension, as some countries offer relatively high benefits but restrict them to some core groups, thereby making them unattainable to a large segment of their population. We chose the approach of effective coverage (Maquet et al., 2016), to not only include all those eligible to a certain benefit or service at a legal level, but also to show the actual take-up rates. These reveal not only national differences in the strictness of eligibility rules and the duration of a benefit, but also non-take up rates and additional factors (e.g. bureaucratic hurdles or activation and contribution requirements in the case of social insurance schemes) that limit the generosity of a service in a qualitative way (Maquet et al., 2016). The definition of coverage is therewith one of actual receipt in contrast to one of social rights as adopted in most current approaches. This implies that when talking about effective coverage, it is also the demand for the services or benefits offered that play a role in national rates,³ as the decision to take up a service or a benefit remains at the discretion of a household or person.

Data for the second dimension of the effective coverage of income support schemes were obtained from the EU-SILC and EU Labour Force Survey following the methodology of Maestri (2016). Effective unemployment benefit coverage is thus measured as the share of the recipients of unemployment benefit among the unemployed of working-age (18–59 years).⁴ Effective social assistance coverage is defined as the share of working-aged individuals living in poor households receiving more than 10% of their household gross income from the state. With this broader definition, we do not just focus on social assistance per se, but on state support to households in the event of insufficient resources (for a longer discussion, see Maquet et al., 2016). Finally, effective family benefit coverage is shown according to the share among families with dependent children.

To measure services, the following approach is chosen: Healthcare decommodification will be defined according to Bambra (2005) as public health expenses (as a percentage of overall costs in the health sector), whereas the effective coverage dimension will be shown by the number of persons reporting unmet medical needs for financial reasons. Defamilisation through the provision of childcare is measured by the amount of hours that children over 3 years to pre-school age spend in formal childcare institutions relative to the median national working hours. This measure enables us to effectively capture defamilisation by showing how easy it is to combine childcare with a regular job. Simply assuming a 40-hour working week would not capture the variety in work-time arrangements that exist in Europe. The effective coverage dimension is measured by the proportion of children between 3 and 5 years enrolled in public and private childcare facilities. This presents one of the limitations of the study because, ideally, unmet childcare need should have been portrayed in the coverage section, as pure levels of enrolment do not reflect the preferences of the parents (which can also vary depending on the national and societal context of cultural preferences, presence and closeness to further family members, and so on). Moreover, services provided to the elderly would ideally also be included in the defamilisation index; however, the variety of need for support based on different degrees of disability makes it difficult to capture elderly care provision.

A summary of the operationalisation of our two dimensions of welfare regimes is shown in Table 1.

Table 1.

Conceptualisation of defamilisation and decommodification policies.

Benefit/service	Level	Measurement	Coverage	Measurement
Decommodification policies
Social Assistance Provision
Social assistance benefits	Income levels provided by cash minimum-income benefits	Minimum income provided. Net income value in % of median household incomes. Source: Organisation for Economic Co-operation and Development (OECD) Benefits and Wages statistics	Coverage rate of social assistance benefits	Share of individuals aged 18–59 living in poor households receiving less than 10% of their household gross income from social transfers. Source: European Union Statistics on Income and Living Conditions (EU-SILC)
Unemployment Benefit Provision
Unemployment benefits	Income level provided by unemployment benefits	Net income value in % of median household incomes for a household with an unemployed, former 50% Average Worker, inactive spouse, 2 children, 1st month of unemployment. Source: OECD Benefits and Wages statistics	Coverage rate of public unemployment benefits	Share of recipients of unemployment benefit among the working-aged unemployed (18–59 years old). Source: EU-Labour Force Survey
Public Healthcare Provision
Public healthcare provision	Public health expenses	In % of total (public and private) healthcare costs. Source: OECD	Coverage rate of public healthcare services	Proportion of population with financial inaccessibility of healthcare due to reasons of costs, waiting lists or transport costs. Source: EU-SILC
Defamilisation policies
Family Benefit Provision
Family benefits	Income level provided by child allowances	Net income value in % of median household incomes for a household with a 50% average worker, inactive spouse, 2 children. Source: OECD Benefits and Wages statistics	Coverage rate of family benefits	Share of recipients of family benefits among all families with dependent children. Source: EU-SILC
Childcare Provision
Childcare provision	Average number of weekly hours of formal care	Formal care for children >3 until pre-school age Including children with at least 1 hour of formal care, measured in weekly hours in % of average weekly hours worked. Source: Eurostat (ilc_camnforg0)	Coverage rate of childcare services	Proportion (%) of children aged 3–5 enrolled in public or private pre-primary education. Source: OECD Family database

Decommodification and defamilisation: Effective coverage and size in Europe

Two indices were calculated to account for the degree of decommodification in a country: The first one comprises social assistance and unemployment benefits, the second additionally includes healthcare provision. Both indices are highly correlated and show a similar ranking of countries, therefore only the full index will be described in detail. The defamilisation index includes family benefits and childcare provision, which appear to be highly compensatory: Countries providing low childcare services are characterised by higher family benefits. In the following sections, we briefly sketch out how our decommodification and defamilisation indicators vary across countries. The data is displayed in the Appendix and in Tables 2 and 3.

Table 2.

Countries ranked by decommodification index.

Country	Unemployment benefit index	Social assistance index	Health provision index	3-item decommodification index	Rank
Denmark	45.03	100.00	96.83	80.62	1
Republic of Ireland	100.00	94.37	31.86	75.41	2
Netherlands	34.15	76.71	100.00	70.29	3
Iceland	46.68	79.99	70.45	65.71	4
Luxembourg	22.72	80.32	93.24	65.43	5
Germany	63.56	56.99	69.05	63.20	6
Belgium	41.37	66.38	66.02	57.92	7
Austria	26.26	73.60	68.82	56.23	8
Finland	41.01	56.31	68.73	55.35	9
United Kingdom	10.49	66.48	88.90	55.29	10
Czech Republic	19.02	47.52	94.91	53.82	11
Norway	16.80	40.20	94.90	50.63	12
France	31.47	50.43	68.39	50.10	13
Latvia	12.40	74.05	54.93	47.13	14
Sweden	14.20	41.00	84.69	46.63	15
Slovenia	14.30	60.61	62.80	45.90	16
Spain	32.77	35.46	62.88	43.70	17
Estonia	8.76	32.41	79.09	40.09	18
Switzerland	38.92	37.75	27.73	34.80	19
Portugal	27.32	48.74	28.10	34.72	20
Poland	2.35	53.11	47.79	34.41	21
Malta	5.45	55.64	38.93	33.34	22
Romania	6.97	30.51	61.67	33.05	23
Hungary	24.63	33.10	30.97	29.57	24
Slovak Republic	0.00	27.04	53.36	26.80	25
Italy	0.89	4.05	66.14	23.69	26
Greece	25.26	0.00	36.49	20.58	27
Latvia	4.62	44.13	0.00	16.25	28
Bulgaria	7.91	30.76	5.95	14.88	29

Displayed are the indices constituted from size and coverage rate for unemployment, social assistance and healthcare provision. They were added into a simple multiplicative index and then the values were rescaled to portray the scale from real lowest value (0) to the highest real value (100) among the analysed states. For detailed data, see Annex 1.

Table 3.

Countries ranked by defamilisation index.

Country	Family benefit index	Childcare provision index	Defamilisation index	Rank
Germany	82.36	72.96	77.66	1
United Kingdom	100.00	42.32	71.16	2
Denmark	22.59	100.00	61.29	3
Slovenia	45.99	75.87	60.93	4
Hungary	45.57	75.85	60.71	5
Belgium	27.41	85.05	56.23	6
Iceland	17.30	92.42	54.86	7
Norway	11.74	96.69	54.21	8
Sweden	20.42	85.85	53.13	9
Italy	23.74	82.53	53.13	10
Luxembourg	41.36	58.85	50.11	11
Portugal	9.70	88.67	49.19	12
Estonia	10.06	86.98	48.52	13
Malta	29.60	64.88	47.24	14
Latvia	5.47	83.08	44.28	15
France	14.01	73.58	43.79	16
Netherlands	27.71	56.98	42.35	17
Bulgaria	21.59	61.88	41.74	18
Austria	36.05	41.66	38.86	19
Finland	19.42	57.84	38.63	20
Lithuania	6.82	68.80	37.81	21
Republic of Ireland	31.95	39.58	35.77	22
Spain	0.00	62.76	31.38	23
Slovak Republic	11.80	50.20	31.00	24
Romania	32.59	29.31	30.95	25
Poland	5.90	45.33	25.61	26
Czech Republic	4.51	46.40	25.45	27
Greece	20.43	14.40	17.41	28
Switzerland	16.90	0.00	8.45	29

Displayed are the indices constituted from size and coverage rate for family benefit and childcare provision. They were added into a simple multiplicative index and then the values were rescaled to portray the scale from real lowest value (0) to the highest real value (100) among the analysed states. For detailed data, see Annex 1.

Decommodification policies

Unemployment benefits. The generosity of unemployment benefits within the first month of unemployment is only in very few countries sufficient to prevent a worker with an income of 50% of the national average and his or her family (a non-working spouse and two children) from falling below the national poverty line. Only Ireland, the Netherlands, Denmark, Switzerland, Sweden, Greece and Romania provide benefits above 60% of the median income. In 13 other countries the benefits are between 50 and 60% of the median income and in the remaining nine countries even lower, falling to merely 31% of median income in Slovakia. The rate to which the unemployed are covered by benefits is low in most European countries, due to strict entitlement criteria, which, for example, do not cover the long-term unemployed or young adults having only a short period of social insurance contributions. Only in six countries are more than half of the unemployed covered by benefits. In 10 more countries, at least one-third of the population is covered, whereas in all other countries effective coverage levels range between 13% and 32%.

Social assistance benefits. The size of social assistance benefits provided to a married couple with two children lies between 40% and 60% of the average income in seven countries. In 19 countries it lies between 20% and 40% and in Greece, Italy and Slovakia it stands at less than 20%. The effective coverage of social assistance benefits is however quite high, when compared to the low reach of the unemployment programmes. On average, 90% of the poor households receive at least 10% of their household income from national benefits. The lowest effective coverage rates are to be found in Eastern and Mediterranean countries.

Healthcare provision. Healthcare decommodification in terms of the public share of total health expenses is particularly high in the Netherlands, Denmark, Norway, Luxembourg and the Czech Republic (approximately 85%). In 16 other countries the state takes over more than 70% of the costs and in all the others (with the exemption of Latvia and Bulgaria) more than 60% of the costs. Effective healthcare coverage (in terms of met medical needs) is very high at an average of 98% in Europe. Nonetheless, differences exist, with 10% of the population in Latvia recording unmet medical needs, 6% in Romania and 5% in Bulgaria, Greece and Ireland.

Decommodification index. When combining the unemployment, social assistance and health provision indices (which include the factor of effective coverage times benefit size), we find an interesting picture, with Denmark showing the highest levels of decommodification, followed by Ireland, the Netherlands, Ireland, and Luxembourg. At the lower end of the scale are Italy, Greece and Latvia and finally, Bulgaria.

Defamilisation policies

Family benefits. In most countries, benefits for the 50%-average-wage worker family with two children are at approximately 15% of the median income and therefore fairly low. They are highest in the United Kingdom (48%), followed by Germany (38%), Greece, Slovenia and Hungary (23–24%), and are lowest in Spain, Latvia, Slovakia, Norway, Estonia and the Czech Republic (4–6%). Effective coverage also varies highly but can be considered universal (above 90% of population coverage) in the majority of the countries. In five countries, effective coverage is for no more than 70% of families with dependent children, six more countries provide effective coverage of more than 50%, and Greece and Spain cover approximately 40% of all families.

Childcare services. The proportion of children aged 3 years to pre-school age enrolled in private or public childcare stands on average at 85%. In 12 countries, more than 90% of children in this age group are in childcare, in 13 countries more than 70%, and least of all in Switzerland (46%) and Greece (47%). In addition to the degree of usage of care facilities, the care services’ adaptation to the working schedule (average weekly hours worked) within a country tells a lot about their capacity to enable the labour market participation of both partners. Only in Lithuania and Denmark do the childcare hours offered meet the average number of working hours and a further 12 countries are able to approximately meet the average working hours, supplying childcare at a rate of 85%. Particularly low rates are shown in Greece and Spain (42 and 37%, respectively).

Defamilisation index. Our data show that some countries that are particularly low on the childcare index show above-average levels on the family benefit index and vice versa. In total, the ranking reveals that defamilisation (when combining family benefit and childcare dimensions) is especially high in Germany, followed by the UK, Denmark, Slovenia, and Hungary. At the lower end are Romania, Poland, the Czech Republic, Greece and finally, Switzerland.

Multi-level analysis of material deprivation

We conduct an analysis of the determinants of material deprivation using binary multi-level models, which enables us to take multiple levels of aggregation into consideration. We apply a two-level variance component model. By this means, we can differentiate between influences from the micro-level and influences from country-specific factors (Cameron and Trivedi, 2010). In each of the models we analyse how certain welfare state policies are associated with deprivation. We control for the median net income of the lowest income quintile (p20) to avoid confusing effects stemming from the affluence of a country and the relatively high net income of its working population with specific policy effects. The coefficients will be displayed as marginal effects that are comparable throughout models (Mood, 2010; Best and Wolf, 2012). The subsequent steps for the analysis are as follows: First, the individual characteristics of households exposed to material deprivation will be shown. Next, we analyse the differences in material deprivation between welfare institutions. Finally, the multiplicative indices of benefit eligibility and entitlements to the different welfare state policies will be portrayed and tested using separate models, first for the whole population and second for the targeted subgroup.

The effect of household characteristics on deprivation

Based on earlier studies (in particular Fusco et al., 2010; Muffels and Fouarge, 2003; Nolan and Whelan, 2011), we include variables reflecting the households’ resources and needs in our micro-level model. Looking at the household level (Table 4, M1), we find that all included variables have the expected effect. Education forms the most important resource to lowering the risk of material deprivation. The risk of material deprivation is 19 percentage points higher amongst those poorly educated than amongst the highly educated. Being in a single parent household increases the risk of material deprivation to the next highest degree of all micro-level variables controlled for. These households are 16 percentage points more likely to be exposed to deprivation than households with two adults and no children. Receiving an income below the national pre-transfer poverty line,⁵ set at 60% of the median gross income of a country’s population, is another important variable.

Table 4.

Multi-level (xtmelogit) regression of material deprivation (binary) 2012–2013 data with welfare states’ characteristics.

Multi-Level Regression
	M0	M1	M2	M3	M4	M5
	Margins_b/SE	Margins_b/SE	Margins_b/SE	Margins_b/SE	Margins_b/SE	Margins_b/SE
Household Composition
Adults, no children		Reference group	Reference group	Reference group	Reference group	Reference group
One-person household		0.075** (0.01)	0.078** (0.00)	0.078** (0.00)	0.077** (0.00)	0.078** (0.00)
Single parent household		0.162** (0.02)	0.162** (0.00)	0.161** (0.00)	0.160** (0.00)	0.161** (0.00)
Adults with children		0.031** (0.00)	0.033** (0.00)	0.033** (0.00)	0.033** (0.00)	0.033** (0.00)
Share of migrants in household		0.001** (0.00)	0.001** (0.00)	0.001** (0.00)	0.001** (0.00)	0.001** (0.00)
Share of members with bad health		0.002** (0.00)	0.002** (0.00)	0.002** (0.00)	0.002** (0.00)	0.002** (0.00)
Education
Higher education (ISCED 5–6)		—	—	—	—	—
Low education (ISCED 1–2)		0.184** (0.02)	0.194** (0.01)	0.193** (0.01)	0.191** (0.01)	0.193** (0.01)
Medium education (ISCED 3–4)		0.090** (0.01)	0.100** (0.00)	0.100** (0.00)	0.098** (0.00)	0.100** (0.00)
Market Income
Gross income below 60% of median		0.001** (0.00)	0.002** (0.00)	0.002** (0.00)	0.001** (0.00)	0.002** (0.00)
Macro Level Variables
Q1 average market income		—	−0.005** (0.00)	−0.005** (0.00)	−0.005** (0.00)	−0.005** (0.00)
		—	−0.005** (0.00)	−0.005** (0.00)	−0.005** (0.00)	−0.005** (0.00)
Two-item decommodification index		—	−0.001* (0.00)	—	—	—
		—	−0.001* (0.00)	—	—	—
Three-item decommodification index		—	—	−0.002** (0.00)	—	−0.002** (0.00)
		—	—	−0.002** (0.00)	—	−0.002** (0.00)
Defamilisation index		—	—	—	−0.001 (0.00)	−0.000 (0.00)
		—	—	—	−0.001 (0.00)	−0.000 (0.00)
Year (Reference 2012)		−0.002* (0.00)	−0.002* (0.00)	−0.002⁺ (0.00)	−0.002* (0.00)	−0.002* (0.00)
N	832,900	832,900	832,900	832,900	832,900	832,900
χ ²	—	87,495	87,617	87,668	87,598	87,670
var_u1	1.122	1.306	0 .16	0.12	0.19	0.12
var_sum	4.411	4.596	3.45	3.41	3.48	3.41
rho1	0.254	0.2843	0.046	0.034	0.054	0.034
r ²	—	0.143	0.123	0.143	0.143	0.143
aic	721,000	617,275	617,218	617,209	617,223	617,211
bic	721,000	617,415	617,381	617,372	617,386	617,386
ll	−360,000	−308,625	−315,810	−308,590	−308,597	−308,590

Displayed are average marginal effects. SEs are in parentheses. R ² is calculated based on improvement of log-likelihood compared to M0 (ll_c). Dependent variable = material deprivation. 0: not deprived; 1: deprived in three or more areas.

Source: EU-SILC; own calculations.

EU-SILC: European Union Statistics on Income and Living Conditions; ISCED: International Standard Classification of Education; M: multi-level; N: number; χ ²: chi-squared; var_u1: random variable; var_sum: sum of variables; rho1: Spearman’s rank coefficient; r ²: coefficient of determination; aic: Akaike information criterion; bic: Bayesian information criterion; II: Type 2 errors.

⁺ p/z < 0.1.

*p/z < 0.05.

**p/z < 0.001.

***p/z < 0.0001.

If we want to disentangle the welfare regime differences stemming from the varying composition of the reference population from concrete country-specific effects, we need to include further explanatory factors in the analysis. First, we consider our control variable in M2 of Table 4. The pre-governmental income level of the lowest income quintile is very strong and is significantly linked to lower deprivation. The wages within a country are thus strongly related to the level of deprivation. The question that we now aim to answer is: Exactly which welfare state policies drive the differences between the Member States in deprivation levels? We will therefore supplement the picture described so far by looking at the main dimensions of welfare state systems, which buffer the repercussions of market deficiencies.

The influence of decommodification on deprivation

When measuring the two-item decommodification index, it shows a significant effect at the 95% significance level. A smart combination of income maintenance measures is therefore apt to lower deprivation. The effect of decommodification becomes even stronger, when the three-item index including healthcare decommodification is used (Table 4, M3). Next to income maintenance, the provision of public services is therefore central to addressing deprivation risks.

When including the single indicators separately, we find that higher social assistance and healthcare provision are significantly linked to lower deprivation risks, but unemployment benefits are not (Table 5, M2). Unemployment benefits have no significant influence on deprivation when the whole population is analysed, and interestingly neither when looking only at the unemployed. A certain consumption smoothing is certainly occurring, but the effect of the unemployment benefit index is low as it has a low effective coverage rate and it provides support only for the first months of unemployment, during which the risk of material deprivation is still low. The social assistance level, however, is highly significant even at the level of the whole population and has a strong deprivation-lowering effect. In contrast to unemployment support, which is being increasingly restricted (Cantillon, 2011), social assistance proves as an important and long-term source for a household’s access to non-market income. The effect for the subgroup of low-income households is even stronger. This confirms the finding of Nelson (2012), who already pointed towards the high importance of minimum income provision for deprivation. Even more than social assistance, public healthcare is linked to lower deprivation rates. It proves to be a key resource in lowering market dependence for the whole population, but in particular for individuals and households with an ill household member. Given the social gradient in health (Siegrist and Marmot, 2006), showing that individuals from lower-income groups suffer more often from illness, public health expenditure has a clear pro-poor impact.⁶ The first hypothesis can therefore be affirmed strongly: A high focus on decommodification by welfare states is significantly related to lower risks of material deprivation.

Table 5.

Multi-level (xtmelogit) regression of material deprivation and index parts.

	M1a	M1b	M2a	M2b	M3a	M3b	M4a	M4b	M5a	M5b
Group	All house- holds	Target group	All house-holds	Target group	All house-holds	Target group	All house-holds	Target group	All house-holds	Target group
	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE	Margins b/SE
Q1 average market income	−0.005**	−0.007**	−0.005**	−0.008**	−0.005**	−0.006**	−0.005**	−0.006**	−0.005**	−0.006**
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)
Social assistance index	−0.001**	−0.003**
	(0.00)	(−0.00)
Unemployment index			−0.001	−0.001
			(0.00)	(0.00)
Health provision index					−0.001**	−0.002**
					(0.00)	(0.00)
Family benefit index							−0.000	−0.000
							(−0.00)	(−0.00)
Childcare provision index									−0.000	−0.003⁺
									(0.00)	(0.00)
N	832,900	113,026	832,900	74,090	832,900	53,958	832,900	500,118	832,900	153,947
χ²	84,095	12,137	87629	4317	87,599	6226	87,591	58,709	87,597	19,598
var_u1	0.1339	0.1017	0.1476	0.1986	0.18362	0.2156	0.1951	0.1839	0.1860	0.1703
var_sum	3.423	3.391	3.437	3.488	3.473	3.505	3.485	3.473	3.475	3.460
rho1	0.039	0.03000	0.0429	0.0569	0.0528	0.0615	0.0559	0.0529	0.0535	0.049
r²	0.135	0.208	0.143	0.148	0.143	0.231	0.143	0.270	0.143	0.264
aic	622,662	125,097	617,216	86,800	617,222	60,254	617,224	368,021	617,223	113,438
bic	622,814	125,222	617,379	86,920	617,385	60,378	617,387	368,143	617,386	113,547
ll	−311,318	−62,535	−308,594	−44,846	−308,597	−30,113	−308,598	−183,999	−308,597	−56,708

Target groups: Households with low income for social assistance benefits; households with unemployed members for unemployment benefits; households including individuals with bad health for healthcare provision; households with dependent children for child benefit provision; and households with children aged 3–6 years for childcare provision. Controlled for but not displayed: individual-level variables and year. In M4 and M5, the household composition was excluded and controlled only for single parent households. Displayed are average marginal effects. Standard errors (SEs) are in parentheses.

Source: EU-SILC; own calculations.

⁺ p/z < 0.1; *p/z < 0.05; **p/z < 0.001; ***p/z < 0.0001.

EU-SILC: European Union Statistics on Income and Living Conditions; M: multi-level; N: number; χ ²: chi-squared; var_u1: random variable; var_sum: sum of variables; rho1: Spearman’s rank coefficient; r ²: coefficient of determination; aic: Akaike information criterion; bic: Bayesian information criterion; II: Type 2 errors.

The influence of defamilisation on deprivation

Our multi-level model testing the defamilisation hypothesis shows that the defamilisation index is not significantly related to deprivation (M4). In addition, when breaking the index down into its dimensions (Table 5), it shows that neither the provision of family benefits nor the provision of childcare is significantly related to lower material deprivation risks. However, when restricting the analysis for childcare to the target group of families with children between 3 and 6 years of age, the childcare index becomes significant at the 10% significance level. Childcare services therefore seem to also have the potential to address deprivation risks for families. Instead of providing ambiguous incentives such as family benefits, the defamilising strategy of childcare – enabling a dual-earner household structure and a higher number of hours worked per household – has the capacity to increase household income and lower deprivation. Future studies should investigate further in this direction and address the limitation that we could not measure unmet childcare needs and employed simple enrolment rates to measure coverage (an indicator that also varies with the start of the obligatory pre-school age in European countries).

Moreover, the lack of effect of the family benefit index does not imply that such benefits are irrelevant to the risk of deprivation – after all single parent families are a key risk group whose deprivation risk needs to be addressed by more than childcare facilities – but that the policy design is not targeted enough to generate a significant difference in the overall extent of deprivation in any given country. Moreover, the shift towards a fiscalisation of child benefits might obscure our results, as we did not portray the role played by the tax system in providing income support to families (Marchal and Marx, 2018).⁷ Accordingly, defamilisation, in terms of cash and in-kind benefits, remains secondary for explaining a country’s overall deprivation levels.⁸ We therefore refute the second hypothesis: Defamilisation is currently not significantly related to lower material deprivation levels; however, it shows certain potential.

Discussion

Healthcare decommodification clearly stands out as one of the most effective anti-deprivation policy. Whole households seem to be protected from deteriorated living standards when medical treatment is covered by the government, which prevents these costs from becoming a drain on a family’s budget. This result points to one of the key differences between poverty and deprivation: Material deprivation shifts the focus further to the ‘demand’ side of a household, whereas income poverty focuses solely on the ‘supply’ side.

Even though income poverty measures try to accommodate for a household’s needs by using equivalence scales, they are not able to capture what the actual standard of living is, as further needs that may be due to bad health or disability are not reflected. Deprivation as an ‘outcome measure’ reveals to what extent bad health and healthcare are fundamental for a household’s living conditions. Services such as healthcare, which support households not through direct cash transfers but by reimbursing the costs of medical treatment, have been underestimated in the current debate.

Moreover, we find that the duration of decommodification measures matters for preventing households from being deprived. Although the unemployment replacement rate is not associated with the deprivation rate on the overall population level, long-term payments such as social assistance, which operate based on a needs-based approach and actually manage to cover great parts of the population at risk, play an important role. Thereby we reinforce the view that deprivation often occurs after persistent periods of poverty, when initial economic and social resources are depleted.

Furthermore, formal childcare does seem to have the potential to give a financial advantage to deprived households by reducing the extent of involuntary part-time work (a strong determinant of low wages). Nonetheless, we know that families from the middle class take advantage of these services more frequently than poorer families. Such policies therefore do not yet seem to be targeted enough at addressing the increased needs of households that are especially prone to being deprived.

Conclusions

In this article, we set out to analyse how national welfare institutions influence the degree of material deprivation to contribute to the debate on Europe 2020 goals. Departing from Esping-Andersen’s (1990) concept of the welfare state, and its important critiques (among them Lister, 1994; McLaughlin and Glendinning, 1994), we distinguished two important strands of welfare policies – decommodification and defamilisation measures – and assessed their association with deprivation. Our analyses clearly revealed that there are salient differences in deprivation incidences by welfare policies. The article amplifies the current knowledge on deprivation by revealing the importance of the provision of public healthcare to the population, next to the provision of social assistance benefits. Both decommodifying policies are key to addressing households that are at risk of being materially deprived.

On the contrary, policies targeted at countering the high exposure to deprivation experienced among families, and in particular single parent households, are still lacking in most welfare states in Europe. Childcare services seem to offer a promising policy for increasing household resources, which has to be explored further.

Our analysis of material deprivation is exposed to several limitations due to the way deprivation is conceptualised and measured in EU-SILC (Guio et al., 2016) but also due to the measurement of defamilisation and decommodification. Measuring the size of benefits is complicated by the fact that benefits are dependent on different factors (e.g. past income, duration of contribution, duration of receipt, number and age of children). This is best exemplified by the case of unemployment benefits where long-term unemployed are treated in most Member States very differently from recently unemployed. Nonetheless, for simplification, only one value (the initial period) was portrayed in this paper.

Another important point concerns the measurement of defamilisation. Here, data availability restricted us to analysing childcare and leaving elderly care aside. But even in the case of childcare, difficulties, such as the unavailability of information on ‘unmet childcare need’, which is a strong determinant of involuntary part-time and therefore likely to influence income options of households, was not available and had to be replaced by pure childcare enrolment rates. The results of defamilisation on deprivation portrayed in this paper can therefore only form a starting point for research in the area of ‘unmet childcare needs’ and their influence on income and living conditions.

Supplemental material

Supplemental Material, Annex - Material deprivation in the EU: A multi-level analysis on the influence of decommodification and defamilisation policies

Supplemental Material, Annex for Material deprivation in the EU: A multi-level analysis on the influence of decommodification and defamilisation policies by Sabine Israel, and Dorothee Spannagel in Acta Sociologica

Footnotes

Author’s note

Sabine Israel is now affiliated to GESIS Leibniz Institute for the Social Sciences, Germany

Acknowledgements

We would like to very much thank the anonymous reviewers and Eric Seils for their excellent comments and questions.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research benefited from funding from the Combating Poverty in Europe (COPE)-project, financed under the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no. 290488.

Supplemental material

Supplemental material for this article is available online.

Notes

Appendix A

average market income for working-aged population (18-59 years of age).

Country characteristics		Unemployment assistance benefit		Social assistance benefit		Public healthcare		Childcare provision		Family benefits
code	Q1 average market in-come (€ PPP)	Size (in Euro PPP)	Coverage (in % of eligible)	Size (in Euro PPP)	Coverage (in % of eligible)	Size (in % of total expenditure)	Coverage(of medical need)	Size (in % of working hours)	Coverage (in % of eligible)	Size (in Euro PPP)	Coverage (in % of eligible)
AT	6951	44,13	50	42,39	91,50	75,55	99,76	64,32	84,13	17,67	96,23
BE	5183	56,71	56	38,08	92,50	75,91	97,99	87,71	98,67	13,58	97,77
BG	1783	48,15	22	20,77	85,27	56,29	94,67	86,96	79,61	15,02	71,69
CH	9575	65,60	46	23,27	90,89	61,67	98,81	49,80	46,10	8,81	99,23
CZ	4808	47,87	37	27,83	93,26	84,81	99,67	75,42	76,44	6,08	55,88
DE	5798	55,04	83	33,67	90,92	76,28	98,91	81,93	94,61	37,76	97,95
DK	6124	73,14	46	58,33	88,75	85,54	99,61	100	97,73	11,76	95,21
EE	2712	37,08	31	21,83	84,83	79,91	98,83	98,17	89,63	6,20	93,38
EL	3457	61,04	35	3,26	79,10	67,51	94,82	70,29	47,98	24,17	42,46
ES	3968	59,68	44	23,06	86,83	73,56	99,62	72,33	96,62	3,83	37,82
FI	5933	59,00	53	31,45	96,27	75,42	99,89	89,46	74,00	10,38	94,67
FR	6620	53,77	47	29,08	94,16	76,95	97,75	78,32	99,55	8,80	85,22
HU	2303	57,49	37	20,74	90,94	63,64	97,54	90,86	87,68	23,10	91,38
IE	2390	91,12	75	53,35	91,85	64,45	96,8	66,50	79,03	15,81	96,38
IS	7180	55,59	63	44,19	94,88	80,65	94,16	95,11	96,80	11,83	75,33
IT	5443	50,38	13	5,37	85,08	78,17	95,21	89,00	95,12	17,68	66,12
LT	2168	52,09	26	42,85	91,03	70,81	99,55	100,00	74,40	8,33	52,69
LU	8768	59,40	34	47,01	89,52	84,46	99,39	74,75	89,58	20,23	95,38
LV	1987	42,61	21	28,38	85,57	56,73	90,25	97,42	87,34	4,31	88,31
MT	5293	57,31	16	32,94	90,93	65,6	98,89	71,47	100,00	15,46	91,99
NL	6886	84,98	32	43,09	93,56	86,63	99,64	69,64	94,15	14,30	93,75
NO	8857	48,09	34	24,55	91,04	85,08	99,35	98,66	96,54	6,74	96,70
PL	3247	45,01	16	33,71	85,14	70,07	97,02	82,12	69,23	7,51	53,19
PT	3494	58,75	39	30,90	85,93	62,64	97,49	99,80	89,44	9,34	60,37
RO	1093	61,88	16	21,87	80,43	77,73	93,73	57,80	77,63	17,93	86,51
SE	5978	62,71	23	24,94	91,23	81,7	99,07	92,68	94,02	10,91	94,04
SI	4998	51,81	29	35,30	91,77	73,29	99,95	89,77	88,77	23,50	90,62
SK	3802	30,81	19	17,53	90,60	70,51	99,19	83,97	72,05	6,80	96,28
UK	4527	35,94	35	38,40	91,87	82,51	99,89	56,70	96,30	47,62	93,66

Note: Income displayed in Euro (PPP). Q1 average market income and poverty gap for working-aged population (18-59 years of age). Size of benefit displayed in % of median income, coverage of benefits in % of eligible population receiving the benefit. Public healthcare size displayed in % of total health expenditures, coverage as % of persons not reporting unmet medical needs for cost reasons. Size of childcare provision displayed as the proportion of childcare hours claimed in % of average working hours in a country, coverage as enrolment rate (3-5 year olds). For more detail see Table 1. Data retrieved from OECD database, except for data in bold (EU-SILC, own calculation) and the unemployment assistance coverage (EU-LFS, provided by V. Maestri).

References

Bambra

(2004) The worlds of welfare: Illusory and gender blind? Social Policy & Society 3(3): 201–211.

Bambra

(2005) Worlds of welfare and the health care discrepancy. Social Policy & Society 4(1): 31–41.

Bambra

(2007) Defamilisation and welfare state regimes: A cluster analysis. International Journal of Social Welfare 16(4): 326–338.

Bargain

Himmervol

Viitamäki

(2012) No Claim, No Pain: Measuring the Non-Take-up of Social Assistance Using Register Data. Journal of Economic Inequality 10(3): 375–395.

Best

Wolf

(2012) Model comparison and result interpretation in logit and probit regressions [Modellvergleich und Ergebnisinterpretation in Logit und Probitregressionen]. Cologne Journal of Sociology and Social Psychology [Kölner Zeitschrift für Soziologie und Sozialpsychologie] 64: 377–396.

Cameron

Trivedi

(2010) Microeconomics Using Stata. College Station: Stata Press.

Cantillon

(2011) The paradox of the social investment state: Growth, employment and poverty in the Lisbon era. Journal of European Social Policy 21(5): 432–449.

Esping-Andersen

(1990) The Three Worlds of Welfare Capitalism. Cambridge: Polity Press.

Esping-Andersen

(1993) Changing Classes. Stratification and Mobility in Post-Industrial Societies. London: SAGE.

10.

Eurofound (2015) Access to Social Benefits: Reducing Non-Take-up. Luxembourg: Publications Office of the European Union.

11.

Eurostat (2010) Income Poverty and Material Deprivation in European Countries. Luxembourg: Eurostat.

12.

Ferrarini

Nelson

Korpi

. (2013) Social citizenship rights and social insurance replacement rate validity: Pitfalls and possibilities. Journal of European Public Policy 20(9): 1251–1266.

13.

Fusco

Guio

Marlier

(2010) Characterising the income poor and the materially deprived in European countries. In: Atkinson

Marlier

(eds) Income and Living Conditions in Europe. Luxemburg: EU, pp. 133–154.

14.

Fusco

Guio

Marlier

(2011) Income Poverty and Material Deprivation in European Countries. CEPS/INSTEAD Working Paper Series 04. https://ideas.repec.org/p/irs/cepswp/2011-04.html (accessed 24 May 2018).

15.

Guio

Fusco

Marlier

(2009) A European Union Approach to Material Deprivation using EU-SILC and Eurobarometer data. IRISS Working Paper Series (19). https://ideas.repec.org/p/irs/iriswp/2009-19.html (accessed 24 May 2018).

16.

Guio

Gordon

Marlier

(2012) Measuring Material Deprivation in the EU. Indicators for the Whole Population and Child-Specific Indicators. Luxembourg: Eurostat.

17.

Guio

Marlier

Gordon

. (2016) Improving the measurement of material deprivation at the European Union level. Journal of European Social Policy 26(3): 219–233.

18.

Iacovou

Kaminska

Levy

(2012) Using EU-SILC data for cross-national analysis: Strengths, problems and recommendations. ISER Working Paper 2012-02. Essex: ISER. Available at: https://econpapers.repec.org/paper/eseiserwp/2012-03.htm (accessed 24 May 2018).

19.

Israel

(2015) More than cash: Societal influences on the risk of material deprivation. Social Indicators Research 129(2): 619–637.

20.

Jaumotte

(2003) Female labour force participation: Past trends and main determinants in OECD countries. SSRN Electronic Journal. Available at: https://dx-doi-org.web.bisu.edu.cn/10.2139/ssrn.2344556 (accessed 24 May 2018).

21.

Kammer

Niehues

Peichl

(2012) Welfare regimes and welfare state outcomes in Europe. Journal of European Social Policy 22(5): 455–471.

22.

Kautto

(2002) Investing in services in West European welfare states. Journal of European Social Policy 12(1): 53–65.

23.

Layte

Whelan

Maitre

. (2001) Explaining levels of deprivation in the European Union. Acta Sociologica 44(2): 105–122.

24.

Lelkes

Gasior

(2012) Income poverty and social exclusion in the EU. Situation in 2008 and trends (based on EU-SILC 2005–2009). European Centre Policy Briefs. January 2012. Available at: https://www.euro.centre.org/publications/detail/351 (accessed 24 May 2018).

25.

Lister

(1994) She has other duties: Women citizenship and social security. In: Baldwin

Falkingham

(eds) Social Security and Social Change: New Challenges to the Beveridge Model. New York, London: Harvester Wheatsheaf, pp. 31–44.

26.

Lohmann

Zagel

(2016). Family policy in comparative perspective: The concepts and measurement of familisation and defamilisation. Journal of European Social Policy 26(1): 48–65.

27.

Mack

Lansley

(1985) Poor Britain. London: Allen & Unwin.

28.

Maquet

Maestri

Thévenot

(2016) The coverage rate of income support measures in the EU: measurement and challenges. Directorate-General for Employment, Social Affairs and Inclusion Working Paper 2/2016. Luxembourg: Publications Office of the European Union. Available at: http://ec.europa.eu/social/BlobServlet?docId=16672&langId=en (accessed 24 May 2018).

29.

Marchal

Marx

(2018) Stemming the tide: What have European Union countries done to support low-wage workers in an era of downward wage pressures? Journal of European Social Policy 28(1): 18–33.

30.

Marx

Nolan

Olivera

(2014) The welfare state and anti-poverty policy in rich countries. IZA Discussion Paper, No. 8154. Bonn: Institute for the Study of Labour. Available at: http://ftp.iza.org/dp8154.pdf (accessed 24 May 2018).

31.

Matsaganis

Ozdemir

Ward

(2014) The coverage rate of social benefits. Social Situation Observatory. Income Distribution and Living Conditions Research note 9/2013. Luxembourg: Publications Office of the European Union. Available at: http://ec.europa.eu/social/BlobServlet?docId=11523&langId=en (accessed 24 May 2018).

32.

McLaughlin

Glendinning

(1994) Paying the care in Europe: Is there a feminist approach? In: Hantrais

Mangen

(eds) Family Policy and the Welfare of Women. Loughborough: European Research Centre, pp. 67–82.

33.

Mood

(2010) Logistic regression: Why we cannot do what we think we can do, and what we can do about it. European Sociological Review 26(1): 67–82.

34.

Muffels

Fouarge

(2003) The role of European welfare states in explaining resources deprivation. Social Indicators Research 68(3): 299–330.

35.

Nelson

(2012) Counteracting material deprivation: The role of social assistance in Europe. Journal of European Social Policy 22(2): 148–163.

36.

Nieuwenhuis

Need

van der Kolk

(2012). Institutional and demographic explanations of women’s employment in 18 OECD countries, 1975–1999. Journal of Marriage and the Family 74(3): 614–630.

37.

Nolan

Whelan

(2011) Poverty and Deprivation in Europe. Oxford: Oxford University Press.

38.

Notten

Guio

(2017) The impact of social transfers on income poverty and material deprivation. ImPRovE Working Paper No. 16/17. Antwerp: Herman Deleeck Centre for Social Policy. pp. 1–31. http://improve-research.eu/?wpdmact=process&did=MTQ4LmhvdGxpbms= (accessed 24 May 2018).

39.

Ringen

(1988) Direct and indirect measures of poverty. Journal of Social Policy 17(3): 351–365.

40.

Saltkjel

Malmberg-Heimonen

(2017) Welfare generosity in Europe: A multi-level study of material deprivation and income poverty among disadvantaged groups. Social Policy and Administration 51(7): 1287–1310.

41.

Scruggs

Allan

(2006). Welfare-state decommodification in 18 OECD-countries: A replication and revision. Journal of European Social Policy 16(1): 55–72.

42.

Siegel

(2005) When (only) money matters: the pros and cons of expenditure analysis. In: Clasen

Siegel

(eds) Investigating Welfare State Change: The ‘Dependent Variable Problem’ in Comparative Analysis. Cheltenham: Edward Elgar, pp. 43–71.

43.

Siegrist

Marmot

(2006) Introduction. In: Siegrist

Marmot

(eds) Social Inequalities in Health: New Evidences and Policy Implications. Oxford: Oxford University Press, pp. 1–26.

44.

Thévenon

(2011) Family policies in OECD countries: A comparative analysis. Population and Development Review 37(1): 57–87.

45.

Till

Eiffe

(2010) Progress of living conditions: A dynamic model of material deprivation for a European Society. In: Atkinson

Marlier

(eds) Income and Living Conditions in Europe. Luxembourg: EU, pp. 241–264.

46.

Townsend

(1979) Poverty in the UK. A Survey of Household Resources and Standards of Living. London: Lane.

47.

van Kersbergen

Vis

(2015) Three worlds’ typology: Moving beyond normal science? Journal of European Social Policy 25(1): 111–123.

48.

Vis

(2013) How to analyse welfare states and their development. In: Greve

(eds) The Routledge Handbook of the Welfare State. London and New York: Routledge, pp. 274–282.

49.

Visser

Gesthuizen

Scheepers

(2013) The impact of macro-economic circumstances and social protection expenditure on economic deprivation in 25 European Countries, 2007–2011. Social Indicators Research 115: 1179–1203.

50.

Whelan

(1995) In what sense is poverty multidimensional? In: Room

(eds) Beyond the Threshold. The Measurement and Analysis of Social Exclusion. Bristol: Policy Press, pp. 29–48.

51.

Whelan

Layte

Maitre

(2004) Understanding the mismatch between poverty and deprivation. European Sociological Review 20(4): 287–302.

52.

Whelan

Maitre

(2012) Understanding material deprivation in Europe: A multi-level analysis. UCD Geary Institute Discussion Paper Series (05). Available at: http://www.ucd.ie/geary/static/publications/workingpapers/gearywp201205.pdf (accessed 24 May 2018).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB