International benchmarking and key challenges for each Member State . 107

children seem to be relatively disadvantaged (compared to local children) in all countries but Ireland (Table A1.1.29).

The share of non-EU migrant children who live in overcrowded households reaches two-thirds in Estonia, Greece, Latvia, Lithuania and Slovenia.

Children with parents born in the EU appear to be in a better position than those with parents born outside the EU in Belgium, France, Luxembourg, Austria and Sweden:

overcrowding among EU migrant children varies between 10% and 38%, while in the non-EU migrant group the range is 34–63% in the same group of countries. Ireland is an exception in this respect (7% versus 4%), although it should be noted that the overcrowding rate is relatively low among all migrant groups.

1.5 International benchmarking and key challenges for

The analytical framework set up in the EU Task-Force report (2008), based on data from 2005, was further validated using 2006 data (EC 2008). According to both papers, Member States form four groups within the space defined by these dimensions (see Tables A1.5.1 and A1.5.2 for these earlier results).

Table 1.24 includes the validation of this analytical framework based on the 2007 data.⁵⁸ While in many respects the country clusters stay stable, there are some important changes as well. As a result of the 2007 validation process, again four country clusters can be identified (the labels follow those used by the EU Task-Force report).

Group A includes countries with good child poverty outcomes and that are also good performers in all determinant-side dimensions: the Nordic countries (Finland, Denmark, Sweden), Cyprus, the Netherlands, Austria and Slovenia all appeared in this cluster on the basis of the 2005 and 2006 data. Based on 2007 estimates, France also joins this group, having earlier been part of group B (see Table A1.5.1).

High labour-market participation of both parents is the key factor behind good outcomes in most of these countries. In Denmark, Finland, Sweden, Cyprus and Slovenia, children live predominantly with two-earner couples. In Denmark and Finland, the share of children with one parent working full time is also considerable; in Cyprus and Slovenia, the ‘single-breadwinner’ arrangement is still widespread. In the Netherlands, the role of the second earner in a part-time job is dominant, and it is not common to have both parents in full-time employment. As an outlier, in Austria the ‘single-breadwinner’ model is dominant, high earnings and income support compensating for the lack of a second earner, though the role of the model featuring one full-time earner and a part-time earner is also considerable.

In the Nordic countries, France and Slovenia, childcare provisions are a great help to parents participating in the labour market. Social transfers in Group A countries are not specifically targeted at children – only in France and Austria are they preferred by the benefit system; however, their effectiveness is generally high, with the exception of Cyprus (well below average) and the Netherlands (average).

Only in the Nordic countries (Denmark, Finland and Sweden) is the share of children living within single-parent families considerably higher than the EU average; most of these children have their parent (mostly the mother) in full-time employment and experience the lowest risk of poverty in the European Union. Nor are children in large families at high risk of poverty in these countries, except for in the Netherlands, where they not only experience close to the EU average risk of poverty, but also account for almost half of all children with an income below the poverty line.

Group B includes countries with high numbers of children in jobless households and low in-work poverty: Belgium and Germany, the Czech Republic, Estonia, Ireland, Hungary, Slovakia and the UK. The last three were part of Group C (below-average performance on both the joblessness and the in-work poverty dimension) on the basis of the 2005 data, and Slovakia had already shifted from Group C to Group B according to the 2006 data.⁵⁹

Within this group, Germany and Belgium have above-average child poverty outcomes, though no country performs really badly in this respect. One explanation could be that relatively effective income supports in almost all these Member States (Czech Republic, Ireland, Hungary and the UK, besides Belgium and Germany) result in narrower-than-average relative median poverty gaps (except in Estonia). Also, some countries perform well in the field of in-work poverty, resulting in levels of poverty incidence that are lower than the EU average (Germany, Belgium, Czech Republic). In the case of Slovakia and

58 The validation closely followed the methodology described in detail in EU Task-Force (2008). However, we cannot exclude the possibility of minor biases. We are aware of two deviations. Not being part of the publicly available EU-SILC dataset, Malta has been excluded from the analysis. Data on joblessness in Sweden were taken from the EU-SILC dataset, the jobless status of household members being estimated according to the ILO definition. Supporting tables and figures are included in Annex 1.5.

59 Also, according to the 2006 data, Hungary was characterised by very low performance in terms of child poverty outcomes, but only just below-average performance based on the 2005 data.

Estonia, slightly lower than average at-risk-of-poverty rates can be explained by considering that joblessness is in fact more related to weak labour-market attachment than to persistent joblessness.

In most of these countries, joblessness is strongly related to living in single-parent families.

In Belgium, Estonia, Ireland and the UK, the share of children with lone parents is high;

furthermore, in Belgium, Ireland and the UK those parents are also highly likely to be jobless. On the other hand, neither in Hungary nor in Slovakia does having a single parent result in a high risk of living in a jobless household at the same time. In Hungary, joblessness and weak labour-market attachment mainly affects children in large families, and is compensated for by generous income supports (mostly cash family benefits). In Slovakia, children in complex households are strongly affected: such children account for a considerable proportion of children and they experience high levels of poverty.

Group C consists of Member States with below-average performance in all dimensions:

Latvia and Lithuania (both part of Group D in previous analyses). Group C is the group most affected by changes from the previous wave of validation (2006). In the previous versions, this group contained Hungary, the UK and Slovakia (as well as Malta, which is not included in the present exercise). However, it is now Latvia and Lithuania that show Group C characteristics, while the other countries have entered the cluster (Group B) for which joblessness represents the key challenge in the field of child poverty (as we have seen, Slovakia had already shifted from Group C to Group B).

Group D includes countries with poor poverty outcomes and where children experience high levels of in-work poverty, but where their share in jobless households is low: all the Southern countries (Greece, Spain, Italy and Portugal), Luxembourg and Poland, all these countries staying in the same cluster across all validation waves. In earlier validation waves, Latvia and Lithuania were in this group, but they have shifted to Group C this time.

None of the countries in this group have child poverty outcomes near (or above) the average. In fact, these Member States have the worst performance in this respect. Not only is the extent of poverty high, but the poverty gap is likewise wide.

High levels of in-work poverty can be attributed to the high share of children in single-breadwinner households, and to the high risk of poverty among them. In all Member States concerned, the share of children in families where one parent works full time while the other is not in employment exceeds 40% of all children at risk of poverty, and they even account for the majority of such children in Italy, Spain and Greece. In Portugal, the high presence of mothers in the labour market (partly facilitated by the childcare services available) results in a relatively low share of all children with only one parent in employment – but also in the highest risk of poverty for those children in any country. In Poland, the relatively high level of full-time employment among mothers is associated with the highest risk of poverty of children in two-breadwinner households anywhere in the EU.

In countries where joblessness is defined as a key challenge, the problem is mostly associated with single-parent households. By contrast, in Group D Member States, in-work poverty is mostly related to couples with at least two dependent children; the share of children in single-parent families among those at risk of poverty is low, despite the high risk of poverty among them. In most of these countries, it is the children in households of couples with three or more dependent children that are most severely affected – except for in Greece, where those in households with two dependent children face a similar risk of poverty to children in large families, accounting for more than half of all children in poverty.

In Luxembourg, children in single-breadwinner households form not only the largest group of all children, but also the largest among those at risk of poverty. There, the main difference (compared to other countries in the group) is the extremely high risk of poverty (the highest in any Member State) faced by children in single-parent families and the relatively high proportion of them among those at risk of poverty, despite the lower than EU average share of all children in such households and the high participation of single parents in the labour market.

***

The country clusters gather together Member States that show similarities in some respects in the labour market and in the impact of income supports on child poverty outcomes.

Beyond similarities within clusters, these countries might differ in other respects that are based on the underlying differences in demographics, labour-market institutions and policy.

Chapter 2 digs deeper in the policy direction, by providing detailed country-level information on policies in place.

Table 1.24: Relative outcomes of countries related to child poverty risk and main determinants of child poverty risk

Child poverty risk outcomes

Joblessness:

children living in jobless households

In-work poverty:

children living in households confronted with such

poverty

Impact of social transfers (cash benefits excl.

pensions) on child poverty

FI + + + + + + + + + + +

CY + + + + + + + –

DK + + + + + + + + +

SI + + + + + + + + +

SE + + + + + + + + + +

FR + + + + + + +

NL + + + +

Group A

AT + + + + + +

DE + + – + + + + +

BE + – + + + +

SK – – + –

EE – – + –

CZ – – + + +

IE – – – + + +

HU – – – + + +

Group B

UK – – – – + +

LV – – – –

Group C

LT – – – – – –

PT – + – – –

LU – + + + – – –

EL – – + + + – – – – –

PL – – – + – – –

ES – – + – – – – – –

Group D

IT – – – + + – – – – –

Source: Tables A1.1.3, A1.1.4, Figure 1.12, Table 1.7, Table 1.11.

Notes: All calculations follow the one established by EU Task-Force (2008). Supporting tables and figures are in Annex 1.5.

111

C h ild p o ve rt y o u tc o m es a n d k ey c h al le n g es in e ac h M em b er S ta te – a s u m m ar y

The impact of policies GroupChild poverty outcomesKey challenges or strengths in the labour market Household composition

Other factors related to the key challenge Income supports ChildcareOther factors shaping child poverty EB

- relatively good - ARPR: higher than the NA, but lower than EU average - RMPG: same as the NA, and lower than EU average - PARP: medium - the youngest (0–5) are in the worst position - high numbers of children in JLHs - low work intensity of parents also affects children, but persistent unemployment is decisive - half of children in JLHs live with their single parent (mostly the mother), while a third of them in large families for children in JLHs and LWIHs: -young mother - low educated parents - 3+ children - migrant background - the level of redistribution is high towards children at risk of poverty - but income supports target children in JLHs and reduce their poverty to an extent that is lower than the EU average - the use of childcare is high in a cross-EU comparison, but remains lower among low-income families and migrant households

- regional disparities. children living in Brussels and Wallonia are at higher risk of poverty - children with anon-EU migrant background are among the most vulnerable ZB

- below average - ARPR: much higher than the NA, but lower than EU average - RMPG: same as the NA, narrower than the EU average - PARP: medium - high numbers of children in JLHs

- half of children in JLHs live with their single parent - children in large families are also affected by joblessness for children in JLHs and LWIHs: -parents with primary and secondary education -single parent - 3+ children - the level of redistribution is high towards children in JLHs - but arenot effective, since benefits are not particularly concentrated on those at risk of poverty

- verypoor availability of childcare facilities, together with relatively high levels of parental benefits - informal childcare is relatively widespread KA

- good - ARPR: lower than the NA, and much lower than the EU average - RMPG: wider than the NA and similar to the EU average - low number of children in JLHs and very low levels of in- work poverty - high participation rates among parents, the ‘2 breadwinner’ model is dominant - single parents are likely to be in full- time employment - children with jobless lone parent and with jobless couple are most exposed to poverty - low levels of redistribution towards children, but the transfers going to children are heavilyconcentrated on those at risk of poverty - transfers are highly effective: reduce the risk of poverty by more than half

- extensive childcare provisions are available - children with anon- EU migrant background are among the most vulnerable EB

- good - ARPR: lower than the NA, and much lower than the EU average - RMPG: narrower than the NA and similar to the EU average - relatively high numbers of children in JLHs - the ‘1 breadwinner’ model is widespread, but these children have a relatively low risk of poverty - three-fifths of children in JLHs live insingle-parent families

for children in JLHs and LWIHs: - low-educated parents - single parent -migrant background - high levels of redistribution towards children in JLHs - high level of poverty reduction among them, even in a cross-EU comparison - poor availability of childcare facilities in the western part of the country

- children withnon-EU migrant background are among the most vulnerable EB

- below average - ARPR: slightly lower than both the NA and the EU average - RMPG: wider than the NA and the EU average - PARP: medium - the risk of poverty increases with the age of child - relatively high numbers of children in JLHs - a fifth of children at risk of poverty have both parents in full- time employment - joblessness among children is associated with the presence of single parents - large families are also affected for children in JLHs and LWIHs: -parents with primary and secondary education -single parent - higher than EU average level ofredistribution towards children - but the benefitsystemis poorlytargetedatchildren at risk of poverty - the poverty-reduction impact of income supports for themis close to the EU average - there is asevere lack of childcare especially in urban areas

112

The impact of policies GroupChild poverty outcomesKey challenges or strengths in the labour market

Household composition

Other factors related to the key challenge Income supports ChildcareOther factors shaping child poverty IEB

- below average - ARPR: higher than the NA and same as the EU average - RMPG: does not differ from the NA and is narrower than the EU average - high numbers of children in JLHs - low work intensity of parents also affects children, but persistent unemployment is decisive - three-fifths of children in JLHs live insingle-parent households - half of children in JLHs live in households with 3 or more dependent children for children in JLHs and LWIHs: - low-educated parents -single parent

- the level of redistribution towards children is high - but that towards children at risk of poverty and in JLHs is near the EU average - the poverty-reduction effect of income supportsis somewhat higher than the EU average for all children and also for those in JLHs

- the limited availability and high cost of childcare is the biggest barrier to parents taking up employment ELD

- well below average - ARPR: higher than both the NA and the EU average - RMPG: larger than the NA and much wider than the EU average - children aged 12–17 are much more likely to be at risk of poverty than the youngest - high levels of in- work poverty - low work intensity of parents also affects children - the ‘1 breadwinner’ model is dominant - low work intensity and in-work poverty are strongly related to self-employment - households with 2 adults and 2 dependent children are affected most by both low work intensity and in-work poverty, large families are to some extent protected by the benefit system for children in ‘1 breadwinner’ households: - young mother - low educated parents - single parent - 2 children -migrant background - low levels of expenditure - social transfers have a negligible effect in reducing the risk of poverty among children with at least 1 parent in work - while strongly targeted by the social system, as a consequence of low levels of expenditure, even jobless families need torely heavily on other sources (inter- household transfers, pension) - low use of childcare facilities for children both under age 3 and those aged 3–5 - informal childcare is widespread

- regional disparities: children in the northern part of the country are affected most - children with anon-EU migrant background are among the most vulnerable ESD

- well below average - ARPR: higher than both the NA and the EU average - RMPG: slightly larger than the NA and much wider than the EU average - PARP: medium - children aged 12–17 are much more likely to be at risk of poverty than the youngest - high levels of in- work poverty - low work intensity of parents also affects children - the ‘1 breadwinner’ model is dominant

- children in households with 2 adults and 2 dependent children contribute most to child poverty, but large families are also strongly affected by both low work intensity and in-work poverty for children in ‘1 breadwinner’ households: -young mother - parents with primary and secondary education - 3+ children

- low levels of redistribution towards all children and also towards children at risk of poverty - low levels of spending and redistribution towards low- income children results in a low poverty-reduction effect of transfers - children in ‘1 breadwinner’ households receive only a small share of social transfers and these serve to reduce the risk of poverty among them only slightly - the use of formal childcare is near the EU average

- children with anon- EU migrant background are among the most vulnerable FRA

- above average - ARPR: slightly higher than the NA, but lower than the EU average - RMPG: narrower than both the NA and the EU average - PARP: low - relatively low number of children in JLHs and fairly low levels of in- work poverty - children injobless and low work- intensity households contribute most to child poverty - children with a coupleeither jobless or living in a ‘1 breadwinner’ arrangement are most affected - children with jobless lone parent are also affected, but to a smaller for children in JLHs and LWIHs: - parents with primary and secondary education

- the level of redistribution towards children is near the EU average, while the poverty-reduction effect is well above the EU average - the use of childcare for the youngest isabove the EU average, but the take-up is low among parents at risk of poverty - children with anon-EU migrant background are among the most vulnerable

113

The impact of policies GroupChild poverty outcomesKey challenges or strengths in the labour market

Household composition

Other factors related to the key challenge Income supports ChildcareOther factors shaping child poverty extent D

- well below average - ARPR: considerably higher than both the NA and the EU average - RMPG: considerably higher than both the NA and the EU average - PARP: high - high level of in-work poverty, affecting almost exclusively children in‘1 breadwinner’ households - fathers are often in self-employment - children in ‘1 breadwinner’ households at risk of poverty mostly live incouple with 2 or more dependent children, those large families being strongly over- represented for children in ‘1 breadwinner’ households: -young mother - parents with primary and secondary education - 2+ children

- low levels of expenditure - the level of redistribution towards children in general and children at risk of poverty islow, as is the poverty- reduction impact of transfers - the same holds for children in ‘1 breadwinner’ households - childcare use for the youngest islow, especially among those at risk of poverty

- children with anon-EU migrant background are among the most vulnerable - living in southern regions and on the main islands (Sicily and Sardinia) YA

- good - ARPR: lower than both the NA and the EU average - RMPG: lower than both the NA and the EU average - PARP: medium - very high ARPR of elderly also contributes to good outcomes for children - low number of children in JLHs and very low levels of in- work poverty - high participation rates among parents, the ‘2 breadwinner’ model is dominant - children in ‘1 breadwinner’ households are affected most by poverty - children in families ofcouples with 2 dependent children provide two-fifths of children at risk of poverty

- the level of redistribution towards children at risk of poverty isclose to the EU average - because of the relatively low level of transfers, they serve to reduce the relative number of children at risk of poverty at only just over half the EU average reduction; this effect, being, however, larger in single-parent and large households

- formalchildcare use for the youngestis low - mothers work full time, relying oninformal childcare provided by the extended family VC

- below average - ARPR: the same as the NA and slightly higher than the EU average - RMPG: much higher than both the NA and the EU average - PARP: medium - children in jobless and low work-intensity households are at high risk, but account for small share of all children - two-thirds of all children at risk of poverty live in in- work households - the ARPR is high among children living inlone-parent households and especially high among children in large families - the level of redistribution towards children is high, but transfers are not targeted at children at risk of poverty - children in in-work households receive only a relatively small share of social transfers, which go more to workless households and those with a low work intensity

- childcare places are very limited and those that exist are beyond the means of low-income families TC

- well below average - ARPR: higher than both the NA and the EU average - RMPG: much higher than both the NA and the EU average - PARP: high - high risk of poverty of children in jobless and low work- intensity households and relatively high levels of in-work poverty - the risk of poverty is especially high among children living inlone-parent households as well as inlarge families of three of more children - the level of redistribution towards all children and towards those at risk of poverty is low - the poverty-reduction effect of transfers is low, also due to low levels of transfers - transfers are redistributed to children in low work-intensity households, reducing the risk of poverty among them much