This chapter examines recent developments in social policies that are particularly relevant for children and families’ outcomes. It begins by situating the evolution of social policies within the broader context of demographic change and the successive economic recessions that welfare systems have had to address over the past two decades. The chapter also introduces the social investment perspective that has increasingly informed policy reforms in many OECD economies, highlighting the key policy dimensions that are especially important for children and their families. It then describes the expansion of in-kind social expenditures, the strengthening of support directed towards families with children, the increased policy efforts to support working parents, and the declining redistributive impact of income support.
Spending Better for Children through Social Policy
Abstract
Social policies today face a growing dilemma. On the one hand, they must respond to rising demands generated by population ageing and the diversification and expansion of social risks across the life course. Ageing populations increase the need for pensions, healthcare and long-term care, while changes in labour markets and family structures create new forms of economic insecurity and inequalities that call for policy responses. On the other hand, demographic change and fiscal constraints make it increasingly difficult to expand social spending. Slower labour force growth, pressures on public finances and competing budgetary priorities limit governments’ capacity to finance further increases in social expenditure. This tension places social policy at a crossroads, where the key challenge is not only how much to spend but how to spend better. Ensuring that social policies deliver “value for money”– that is, achieving their intended economic and social outcomes while remaining financially sustainable – has therefore become a central objective for policymakers.
To respond to these challenges, many countries are reconsidering how social spending is prioritised. One perspective that has gained prominence in this context is to place greater emphasis on programmes that strengthen individuals’ capacity to become economically self-sufficient along their lifetimes while providing fair opportunities to reach positive outcomes even when coming from disadvantaged backgrounds. This includes measures that promote early childhood development, support parents’ participation in the labour market, enhance educational attainment, and foster the accumulation of human and social capital. By improving people’s prospects for employment and economic participation, these policies seek to reduce long-term dependence on social assistance while strengthening productivity, social inclusion and social cohesion.
This approach, commonly referred to as social investment, emphasises preventive and enabling policies that help individuals build resilience and participate fully in society, alongside more traditional forms of income protection that buffer the consequences of economic shocks and adverse life events. The underlying premise is that well-targeted investments in people’s capabilities generate long-term returns through higher employment, stronger productivity, better well-being and lower future demands on public expenditure.
Whether families with children – and children themselves – benefit from these evolving policy priorities remains an open question. On the one hand, stronger emphasis on employment-oriented policies may benefit families if parents are able to secure stable jobs and reliable incomes, particularly when they are supported by measures that help reconcile work and family responsibilities. Children may also benefit directly from policies specifically targeted at families, such as early childhood education and care (ECEC), family benefits and parenting support. More broadly, investments in health, housing, education and other public services, although not exclusively designed for families with children, can substantially improve children’s living conditions, development opportunities and long-term life chances. On the other hand, these shifts in policy priorities may not benefit all children equally. Employment-oriented policies alone may be insufficient where parents face persistent barriers to work, labour markets fail to provide stable and adequately paid jobs, or employment is not accompanied by accessible childcare and other family supports. Likewise, inadequate access to affordable, high-quality services may prevent improvements in income from translating into better living conditions and reduced material deprivation. Fiscal pressures may also lead governments to concentrate resources on specific population groups or policy areas, potentially reducing support for others. Similarly, tighter targeting of benefits, often introduced to contain public spending, may exclude families experiencing financial hardship who do not meet strict eligibility thresholds. As a result, the net impact of these policy trends on child poverty and children’s well-being may be uncertain.
This chapter provides insights into recent developments in social policies that are particularly relevant for understanding how they may affect outcomes for children and families. The first section places the broad evolution of social policies in the context of demographic change and the series of crises to which welfare systems have had to respond over the past two decades. It also highlights the need to demonstrate value for money, as illustrated by the social investment perspective that underpins many of these policy developments and highlights the key dimensions of social policy that are especially important for children and their families. The following sections then describe how these interventions have evolved across OECD countries, focussing in particular on the expansion of in-kind social expenditures, the strengthening of support targeted at families with children, increased support for working parents, and the declining redistributive impact of income support.
Social policies pursue multiple, interconnected objectives. Their primary function is to protect individuals and families from social risks and economic hardship over the life course, including poverty, unemployment, sickness, disability and old age. Beyond this protective role, social expenditures also serve a redistributive function, helping to reduce inequalities in income and living standards, both vertically and horizontally (for example, between households with and without children).
Some social policies also play a promotional role by strengthening labour market attachment among working-age adults, particularly for disadvantaged groups. They also support families with children by helping to offset the costs of raising children and by facilitating the reconciliation of work and family life. Investments in childcare, education and active labour market programmes support employment and social mobility, while other in-kind benefits – such as healthcare, housing and family or social services – improve living conditions, promote good health and help reconnect vulnerable populations with work.
In addition, social policies contribute to macroeconomic stabilisation by sustaining demand during economic downturns and by supporting human capital development over the longer term. Taken together, social policies combine protection, investment in skills and capabilities, and measures that empower individuals to participate fully in economic and social life. In doing so, social policies aim to enhance the well-being of both children and adults, strengthen equality of opportunity, and promote greater social cohesion.
These multiple objectives are reflected in the way social expenditures are organised into distinct policy or risk areas, although their structure also varies across countries depending on historical and institutional factors, specific needs and the relative weight given to each objective. In the OECD Social Expenditure Database (SOCX) (OECD, n.d.[1]), which is used as the main source of comparative information in this report, social expenditures include health spending, which covers medical care and public health services; incapacity-related benefits, which provide income support in cases of sickness, disability, or work injury; and family benefits, which support households with children through cash transfers, parental leave, and childcare services. Other key areas include active labour market programmes aimed at improving employment prospects through training, job placement, and wage subsidies; unemployment benefits that protect income during joblessness; housing support that helps reduce accommodation costs; and social assistance and other income support programmes targeting groups at high risk of poverty. Old-age and survivors’ benefits form a final category, providing income security in retirement or following the death of a breadwinner, and represent one of the largest components of overall social spending.
Families with children benefit directly from expenditures explicitly targeted at them, but they may also gain – often indirectly – from other areas of social spending. For example, unemployment benefits and active labour market programmes help stabilise the incomes of jobless parents and support their return to work. Health expenditures play a crucial role in improving children’s health outcomes, safeguarding the health of prospective parents and pregnant women, as well as in supporting parental health and employability. Survivor pensions are likewise important for protecting families that become single‑parent households following the death of a partner from poverty. By contrast, families with children benefit far less from old-age pensions – except in countries where multigenerational households are common – and they can instead shift social expenditures towards childless households.
Growth in social expenditures in the context of economic and demographic changes
The share of national income that OECD countries devote to social expenditures has risen over the two past decades, increasing from an average of 20% of GDP in 2001 to around 25% in 2021 (Figure 2.1, Panel A). The two major crises over this period – the 2008 financial crisis and the COVID‑19 pandemic – both produced permanent upward shifts in the level of social spending, in gross and real terms. The share of social expenditures in GDP rose sharply between 2008 and 2010, reflecting combined effect of falling GDP per capita and the expansion of social programmes introduced in response to the financial crisis. Spending remained at these higher levels in the years that followed.
Social expenditures increased again during the COVID‑19 pandemic, peaking at 26% of GDP on average in 2020, driven by higher outlays on health, unemployment and active labour market programmes, as well as expanded income‑tested cash supports (OECD, 2023[2]) (Box 2.1). In per-capita real terms, social spending “stepped up” in 2020 due to rising expenditures and declining GDP per capita, and remained elevated in 2021 even as GDP per capita began to recover.
Important shifts in the composition of social expenditures have taken place together with increases in total levels. Total social expenditure per capita, measured in constant dollars, increased by over 50% between 2001 and 2021, from around USD 7 500 to nearly USD 12 000 (Figure 2.1, Panel B). This growth was driven primarily by rising expenditure on older people, alongside strong increases in per-capita health spending. Over this period, per-capita health expenditure grew by 72%, while expenditure on the elderly rose by 61%. Together, these two categories make up nearly two‑thirds of all social expenditure in 2021. Owing largely to the COVID‑19 pandemic, spending on active labour market policies and unemployment benefits also increased strongly since 2001 (+79%) but at less than 8% in 2021, they cover a much smaller share of per-capita expenditure. Similarly, per-capita spending on families with children remains relatively low – accounting for less than 10% of total social expenditure in 2021 on average. It grew in line with total social expenditure (+53% over 20 years), which represents a significant increase in spending per child given that the share of children in the population has declined over the same time period (‑16% for individuals under age 20).
Note: Social expenditure refers to in-cash and in-kind spending in public, mandatory private and voluntary private programmes in the policy areas old age, survivors, incapacity-related benefits, health, family, active labour market programmes, unemployment, housing and other social policy areas. Panel A: The right scale is indexed for 2015 = 100, based on data series measured in USD per person, PPP converted and at constant prices. Panel B: Data refer to USD per person, PPP converted at constant prices (base year = 2015).
Source: OECD (n.d.[3]), OECD Social Expenditure database, https://data-explorer.oecd.org/s/46u and OECD (n.d.[4]), OECD National Accounts at a Glance, https://data-explorer.oecd.org/s/46v.
The COVID‑19 pandemic severely disrupted labour markets across the OECD, prompting governments to rapidly expand social protection through sickness benefits, job retention schemes (JRS), unemployment benefits (UB), and support for self-employed workers (Dely, Hyee and Prinz, 2025[5]). Overall, these emergency measures were largely effective: they cushioned income losses, supported a swift labour-market recovery despite the initial economic shock, and helped prevent a lasting rise in labour-market inactivity. Structural impacts remained limited in the medium term, with no widespread increases in disability benefit claims and long-term unemployment largely returning to pre‑pandemic levels in most countries.
Countries differed in how they absorbed the shock. Most OECD countries introduced or expanded JRS, with European countries relying mainly on subsidised reductions in working hours, while countries such as the United States leaned more heavily on extended unemployment benefits, leading to greater job losses but rapid re‑employment during the recovery. Both approaches proved broadly successful: JRS were effective in preserving jobs, particularly where schemes already existed, while temporary UB extensions had limited adverse effects on job search given the exceptional circumstances. The pandemic also exposed gaps in protection for self-employed workers, prompting several countries to extend their social insurance coverage after the crisis. Despite temporary increases in long-term unemployment and sickness absence, there is little evidence so far of lasting spillovers into disability or permanent labour-market exit.
The pandemic also disrupted the usual balance between the state, market and family, bringing the relationship between families and public policy into sharper focus (Daly and Ryu, 2025[6]). Lockdowns and service closures led to a partial withdrawal of early childhood education and care and primary schooling, effectively “returning” children to the family sphere. Policy responses largely prioritised resourcing parental and family-based care rather than maintaining child-centred services, with access to education and care often conditioned on parental employment status. While some countries softened closures through limited access or distance learning, children’s needs were mainly interpreted through the lens of parental caregiving and labour-market participation. Overall, traditional family- or parent-centred approaches prevailed, with little progress in advancing children’s rights, autonomy, or participation, and with children often less visible than adults in crisis responses.
Cross-country variation in Europe was substantial but patterned (Daly and Ryu, 2025[6]). Countries with well-developed pre‑pandemic systems of income support and services for children and families – particularly the Nordic countries, as well as Austria, France and Luxembourg – were better able to protect children and sustain support during the crisis. Targeted welfare systems, such as those in Ireland and the United Kingdom, also performed relatively well in prioritising vulnerable children. In contrast, countries with weaker pre‑existing provision struggled to respond effectively. While these patterns reflect strong policy path dependency, the pandemic also disrupted prior child-centred policy trends, including service guarantees, early-years investment, and child poverty targets. This reversal highlights a tension between parent-centred and child-centred policy logics: although supporting parents is crucial, evidence from child rights and well-being frameworks suggests that a focus on families alone may be insufficient to safeguard children’s development and rights during times of crisis.
During the COVID‑19 crisis, the United States temporarily expanded the federal Child Tax Credit (CTC) as part of the American Rescue Plan Act of 2021, raising the maximum credit from USD 2 000 to as much as USD 3 600 per child under age 6 and USD 3 000 for older children, extending eligibility to 17‑year‑olds, making the credit fully refundable, and providing half of the benefit in monthly advance payments – changes that together helped reduce child poverty significantly in 2021. However, these enhancements were only in effect for tax year 2021 and expired at the end of the year, after which the CTC reverted to its earlier, less generous design, resulting in reduced support for low-income families and contributing to increases in child poverty once the expanded payments ended.
In Canada, the expansion of the Canada Child Benefit (CCB) and the introduction of the Canada Emergency Response Benefit (CERB) played a key role in cushioning the impact of the COVID‑19 pandemic on families with children. Child poverty fell markedly between 2019 and 2020, driven largely by the temporary emergency income support provided through CERB. However, following the expiry of the programme, child relative poverty increased by more than two percentage points (p.p.) in 2021, while the anchored child poverty rate rose even more sharply. This reversal highlights the extent to which the earlier reductions in child poverty were sustained by temporary income support and raises concerns about the vulnerability of low-income families once emergency measures were withdrawn (Thévenon, Clarke and Simard-Duplain, 2024[7]).
The drivers of rising social spending are multiple, and fully disentangling their respective contributions lies beyond the scope of this chapter. Nonetheless, the literature points to a range of economic, demographic, and political-institutional factors that systematically shape spending dynamics, exerting both upward and downward pressures on expenditure levels (Haelg, Potrafke and Sturm, 2022[8]).
Population ageing is a key driver of increasing social protection needs. Rising life expectancy and the growing share of older people are pushing up spending on pensions and healthcare, which rise sharply with age (Koutsogeorgopoulou and Morgavi, 2025[9]; Crowe et al., 2022[10]; OECD, 2017[11]). Declining fertility rates further accelerate ageing dynamics. While lower fertility can ease short-term pressures on family-related spending – including early childhood education and care (ECEC) and family benefits – potentially allowing for higher resources per child, it intensifies long-term fiscal pressures by shrinking the working-age population and eroding the tax base (Guillemette and Turner, 2021[12]; OECD, 2023[13]). This shift may increasingly constrain governments’ capacity to sustain investment in younger generations.
Other factors also contribute to upward pressures on spending. Economic downturns, rising unemployment, and financial crises typically trigger counter-cyclical expansions in welfare provision. In addition, higher public debt can reinforce political pressures to maintain or expand support, as crisis-related measures often prove difficult to reverse.
By contrast, several forces tend to restrain social expenditure. Trade globalisation exposes governments to greater international competition, strengthening incentives to contain public costs. Political fragmentation can further limit expansion by complicating coalition-building and decision making, while concerns about fiscal sustainability often prompt consolidation efforts. Consistent with this, Haelg et al. (2022[8]) found that, across OECD countries between 1980 and 2016, trade globalisation, political fragmentation, and fiscal balances are associated with lower social spending, whereas unemployment, population ageing, banking crises, public debt, and social globalisation are linked to higher expenditure levels.
The rising labour market participation of parents – particularly mothers – is another key trend increasing the demand for policies aimed at reconciling work and family life, including more generous parental leave provisions, greater investment in childcare services, and adjustments to tax – benefit systems to better reflect the growing prevalence of part-time and non-standard forms of employment (Adema, Clarke and Thévenon, 2020[14]). At the same time, fertility patterns and parenting behaviours have evolved, and family structures have become more diverse. These developments have required adaptations in social protection, income support, and family policies to better cover non-traditional family forms (Miho and Thévenon, 2020[15]; Rapp and Thévenon, 2025[16]). They have also prompted changes in the design and delivery of family and social services to respond more effectively to increasingly complex and varied needs (Riding et al., 2021[17]).
Taken together, these trends are driving significant adjustments in social policy strategies. They affect not only the overall level of spending, but also how resources are allocated across policy domains and the ways in which families with children are supported – whether through targeted transfers or more indirectly through broader policies that promote employment, protect incomes, and improve health and living conditions via in-kind services. While these shifts are rooted in structural demographic and economic changes, they also reflect growing expectations that social policies demonstrate their effectiveness and deliver measurable “returns”. This has contributed to the increasing prominence of a social investment perspective in the design and orientation of social policy.
Towards a social investment approach
Rising pressures on social expenditure have led many countries to reassess its impact and, in the context of fiscal constraints, to consider how best to prioritise spending. To inform these decisions – and to emphasise that social policies can generate broader benefits – many OECD countries have increasingly framed social policy as (social) investment as opposed to a mere cost, particularly when resources are directed towards interventions expected to yield high long-term returns (Bouget et al., 2015[18]). This approach seeks to move welfare states beyond a predominantly compensatory logic and toward a more preventive, forward-looking model. Rather than focussing mainly on income replacement once social risks materialise, social investment prioritises strengthening people’s capabilities over the life course through early intervention, education, and employment support (Hemerijck, Ronchi and Plavgo, 2023[19]; Garritzmann, Häusermann and Palier, 2022[20]; Esping-Andersen, 2002[21]). The overarching idea is that well-designed social policies are not merely redistributive transfers but productive investments that strengthen human capital, boost labour-market participation, reduce poverty and create more resilient and inclusive economies (OECD, 2018[22]).
As illustrated in Figure 2.2, the social investment approach rests on the idea of a virtuous cycle whereby well-designed policies reinforce one another across the life course. It emphasises both the cumulative effects of interventions whose benefits build over time and the complementarities between policies that address interconnected disadvantages across different domains of life. In practice, this approach prioritises spending that strengthens human and social capital, reduces inequalities from an early age, and enables children to realise their full potential. It therefore places strong emphasis on investment in early childhood education and care, schooling, and education systems more broadly. It also highlights policies that enhance households’ self-sufficiency, including active labour market programmes, lifelong learning, and measures that support the reconciliation of work and family life – which are also key to expanding the tax base and sustaining investment in poverty reduction and social inclusion, thereby completing the social investment cycle. At the same time, poverty prevention and reduction remain central, as they help create the conditions for equal opportunities and reinforce social cohesion.
More broadly, the social investment perspective underscores the importance of early intervention combined with sustained support over time. However, it has also been criticised for prioritising policies that promise the highest long-term economic returns, potentially at the expense of adequate income protection for those with immediate needs. Critics further argue that some social investment policies may disproportionately benefit better-off households (e.g. see later discussion on the expansion of early childhood care and education services). In doing so, it may underplay the needs of those facing severe and persistent disadvantages, who are often the hardest to support through activation-oriented measures or to reintegrate into the labour market (Box 2.2).
Source: Hemerijck, Ronchi and Plavgo (2023[19]), “Social investment as a conceptual framework for analysing well-being returns and reforms in 21st century welfare states”, Socio-Economic Review, https://doi.org/10.1093/ser/mwac035.
While the social investment approach has reshaped welfare thinking by emphasising early intervention, human capital formation, and long-term outcomes, several limitations have been identified in the literature.
First, the distinction between policies that mitigate social risks and those that prevent them is analytically helpful, but it can be difficult to make such distinction in practice, because social risks evolve over time and many policies serve multiple purposes. In reality, welfare systems rely on layered and interacting instruments that blur this boundary. Moreover, many policies simultaneously perform compensatory and preventive functions. For example, unemployment benefits mainly compensate for income loss after job loss, but they can also prevent longer-term harm by stabilising consumption, preserving skills, and enabling better job matching. Similarly, disability benefits mitigate income loss while helping prevent social exclusion and persistent poverty.
Whether a policy is preventive or compensatory also often depends on timing. Early childhood education and care (ECEC) can prevent the accumulation of disadvantage when provided early and universally, but when offered only after developmental problems emerge, it functions more as remedial support. Policy design and context also matter: while cash transfers are often seen as compensatory, reliable income support can prevent housing instability, poor health, or setbacks in child development; conversely, ostensibly preventive measures such as education or activation policies may have limited impact if access or quality is uneven.
Social risks are also interdependent and cumulative over the life course, further weakening a binary distinction. For instance, childcare services both mitigate work-family constraints in the short term and prevent long-term income penalties for mothers. Finally, political and institutional constraints mean that governments often combine preventive elements with existing compensatory schemes, creating hybrid policies. Overall, mitigation and prevention are best understood as points on a continuum, with effective social policies typically combining both logics (Hemerijck, 2017[23]).
Another concern is the risk of underemphasising income protection and adequacy, especially for individuals who cannot easily benefit from activation-oriented or skill-enhancement measures. Critics argue that a policy model centred predominantly on labour-market participation may leave those facing structural barriers – such as people with disabilities, long-term unemployed adults, or carers – insufficiently protected (Cantillon, 2011[24]; Morel, Palier and Palme, 2012[25]). This may result in a “Matthew effect”, where better-off groups benefit more from social investment policies than lower-income or precarious populations because they are linked to employment, tied to income or delivered as services. Working-income supplements, for instance, benefit those who work and do little for workless households. Likewise, persons with higher incomes tend to better avail themselves of public services on offer. Formal childcare services, for instance, are everywhere more attended by the children of double‑income households, and less so by those with a low-educated, low-income mother (Bonoli, Cantillon and Lancker, 2017[26]; Gingrich and Ansell, 2015[27]). Knise et al. (2021[28]), for instance, found that young adults with the lowest education and employment experience participate less often in the most advantageous measures; particularly in firm-based upskilling and employment assistance, and more often in measures that proved to be not as beneficial, such as workfare programmes. Regional contextual traits, which can present a variety of configurations, play a significant role in moderating the effects of ALMPs. Divergent outcomes per type and level of education also highlight the complexity of the landscape for ALMPs’ design and implementation (Cefalo and Scandurra, 2023[29]).
Social investment strategies can exhibit implementation and timing challenges. Because many of their returns – particularly those from early childhood investment – materialise only in the medium to long term, governments may find it difficult to sustain political and fiscal support. Short electoral cycles tend to favour visible, immediate spending rather than multi-decade commitments (Hemerijck, 2017[30]). Moreover, a strong social investment agenda requires well-co‑ordinated, high-quality services across sectors such as education, childcare, health and employment; in practice, fragmentation, limited administrative capacity, and siloed governance often weaken the intended “multiplier” effect (Vandenbroucke and Vleminckx, 2011[31]).
A social investment approach risks overstating the role of activation and human capital in securing social inclusion, particularly during periods of weak labour demand or structural economic change. It encourages governments to spend more on measures that support employment, and less on passive benefits (Cantillon, 2011[24]). Moreover, even well-designed social investment policies cannot fully compensate for precarious employment, declining job quality, or widening labour-market inequalities (Taylor-Gooby, Gumy and Otto, 2015[32]). Without complementary regulation and robust minimum-income protection, social investment can fall short of reducing poverty or cushioning new social risks (Cantillon and Vandenbroucke, 2014[33]).
Finally, concerns have been raised about the narrow economistic framing sometimes associated with social investment. When policies are primarily justified in terms of future fiscal returns or productivity gains, broader welfare objectives – such as social rights, equality, dignity, mental health, and overall well-being – may receive insufficient attention (Jenson, 2010[34]; Deeming and Smyth, 2015[35]). In response, increasing efforts are being made to broaden the assessment of social investment returns to better capture these non-economic dimensions. Such approaches include contingent valuation methods, which estimate individuals’ willingness to pay for specific social outcomes, and the use of shadow prices to monetise non-market dimensions of welfare (Fougère and Heim, 2019[36]). For example, the welfare effects of economic shocks or policy reforms can be evaluated by deriving income‑equivalent values for non-income dimensions of well-being, based on estimated shadow prices obtained from their relationship with life satisfaction (Boarini et al., 2022[37]).
Policies affecting children: what a social investment perspective suggests
Within the social investment framework, investment in children occupies a central place, reflecting the recognition that inequalities emerge early in life as differences in household resources, parental time, and developmental environments accumulate and can translate into long-term disadvantage (Esping-Andersen, 2005[38]; De Schutter et al., 2023[39]; OECD, 2009[40]). Some policies are explicitly designed to support children directly – for example, early childhood education and care, parenting support and family benefits – while others are primarily targeted at adults, such as employment and skills policies, but nevertheless influence children’s living standards by improving household resources and economic security. Together, these policies can help prevent or mitigate childhood disadvantage through complementary pathways.
Addressing childhood disadvantage requires continuous and multifaceted support throughout childhood, combining targeted measures for children and their families with broader policies that strengthen and stabilise household resources and living standards. Key dimensions highlighted in the literature include:
Early childhood policies. Early childhood is a critical window for intervention, when well-timed policies can generate particularly high social and economic returns. Early investments support cognitive and socio‑emotional development and lay the foundations for future skill acquisition (Heckman, Pinto and Savelyev, 2013[41]; Cunha and Heckman, 2007[42]). High-quality early childhood policies are therefore a cornerstone of a sustainable welfare state: they help prevent the intergenerational transmission of poverty while enabling parents – particularly mothers – to remain in employment, contributing to both gender equality and long-term economic performance.
Sustained support for material and social needs. Children and their families often require continued support into late childhood and adolescence – stages marked by new developmental and well-being challenges (OECD, 2021[43]). Evidence shows that interventions during these periods can have lasting effects on well-being, educational attainment, and labour market outcomes. Crucially, support must address not only income needs but also material and social conditions and adapt to children’s changing needs as they grow.
Pro‑employment policies and work-care reconciliation. Effective active labour market policies (ALMPs) and measures supporting the reconciliation of work and family life are essential to reducing family poverty (Cantillon and Vandenbroucke, 2014[33]; Sakamoto, 2023[44]). They help secure stable incomes while preserving parents’ capacity to provide care (Bakker and Van Vliet, 2022[45]; Sakamoto, 2024[46]; Nieuwenhuis, 2022[47]; Gingrich and Ansell, 2015[27]).
Family support and income protection. Policies that strengthen families’ resources and resilience are also critical. This includes income protection in the event of economic shocks or changes in family circumstances, as well as access to family and social services when families face more complex needs or difficulties affecting family functioning (Daly, 2025[48]; Riding et al., 2021[17]).
Social investment in practice
Although increasingly prominent in social policy debates, the extent to which the social investment approach is recognised and implemented in practice varies widely across countries. In Europe, the European Union (EU) has played a particularly important role in advancing this agenda since the early 2010s (European Commission, 2013[49]). The EU social investment vision closely aligns with the virtuous circle described earlier, which prioritises early intervention over later remediation and places strong emphasis on policies that support children and families, expand access to high-quality early childhood education and care, and prevent disadvantages from accumulating over the life course (European Commission, 2013[49]). A second pillar focusses on lifelong human capital development through continuous learning, skills upgrading, and active labour market policies, with the aim of helping individuals adapt to economic change and maintain attachment to employment. The third pillar views social protection as an enabling force: income‑support systems, social services, and work-family reconciliation policies are designed not only to cushion social risks but also to empower individuals to participate fully in the labour market and contribute to economic growth and broader societal functioning (European Commission, 2021[50]).
In practice, social investment has materialised to varying degrees and in different forms across European countries (Bouget et al., 2015[18]). An analysis of social investment strategies across European countries over 2004-2018 identifies three main clusters.1 balanced, basic, and bent (Baiocco et al., 2022[51]).
Countries pursuing balanced strategies combine relatively high spending across all major social investment domains throughout the life course – family and childcare, education, and labour market activation – with strong participation of children and young individuals in both pre‑primary and tertiary education and comparatively high levels of spending on early childhood services.
Basic strategies are characterised by low to medium overall spending and similarly modest enrolment rates in pre‑primary education. These countries focus primarily on more traditional investment areas, such as compulsory schooling and, to a lesser extent, tertiary education, while devoting comparatively limited resources to early childhood supports.
Bent strategies also involve low to medium levels of overall spending but differ in their emphasis on early-life interventions. These countries feature medium to high participation in pre‑primary and tertiary education and allocate a relatively larger share of resources for services benefiting very young children (and the first years of life), often through cash-based family supports such as parental leave.
A key driver behind the renewed emphasis on social investment in Europe is the need to address the EU’s competitiveness challenge, including stagnating productivity growth and persistent labour and skills shortages, in the context of shrinking working-age populations. Accordingly, many social investment policies and reforms are framed not only as tools to achieve social objectives, but also as public investments expected to generate economic returns through strengthened human capital and higher productivity (Council of the European Union, 2024[52]).
In New Zealand, the social investment approach emerged during the 2011-2012 welfare reforms as an effort to shift social policy from a reactive model to one focussed on long-term outcomes. Central to this shift was the use of actuarial assessments and integrated administrative data to estimate individuals’ future social and fiscal trajectories. Its first major application was in the welfare system, where long-term liability estimates informed the redesign of employment and income‑support programmes (Government of New Zealand, 2013[53]; Chapple, 2013[54]). Over subsequent years, the framework expanded to child protection, justice, and health, supported by the growing analytical capacity of Statistics New Zealand’s Integrated Data Infrastructure (OECD, 2024[55]; Gluckman, 2017[56]).
After 2017, part of this approach was reframed from a broader “well-being” perspective, which reduced the emphasis on actuarial liability while retaining core features such as data-driven targeting, cross-agency co‑ordination, and a stronger focus on addressing complex needs (Government of New Zealand, 2019[57]). This approach recognises that inclusive activation policies require not only labour-market services and income support (e.g. childcare assistance, job-search support) but also a wide range of social services, including parenting programmes, housing provision, and financial counselling for the most disadvantaged groups. As part of this shift, the government enacted landmark legislation in 2018 to make reducing child poverty and improving child well-being an enduring national priority (Government of New Zealand, 2025[58]; OECD, 2023[59]). The legislation requires successive governments to adopt and regularly update a national child well-being strategy, first introduced in 2019 as an open-ended, integrated policy framework. As a result, New Zealand’s social investment model has evolved from a fiscally oriented tool for welfare reform into a whole‑of-government strategy aimed at improving long-term social outcomes, and more recently it has placed a particular focus on supporting the most vulnerable populations facing entrenched and complex disadvantages.
Other OECD countries have also implemented social investment strategies. Comparative studies of welfare state reforms – most notably Garritzmann, Häusermann and Palier (2022[20]) – show that a wide range of countries have introduced policies aimed at creating, mobilising, and preserving human skills and capabilities. These strategies differ not only in their core functions but also in their financing models and distributive profiles. In some cases, social investment programmes are conceived as repayable or short-term instruments expected to deliver measurable fiscal returns; in others, they are framed as long-term public investments generating diffuse social benefits, often embedded within broader “well-being” policy frameworks.
The way distributional considerations are integrated into social investment strategies also varies. Some approaches are inclusive, providing broad-based and relatively egalitarian support that fosters large and cohesive constituencies. Others are stratified, offering differentiated levels of support across social groups, thereby potentially reinforcing existing inequalities while maintaining political support. A third approach is more targeted, directing resources primarily towards lower-income or otherwise disadvantaged groups based on assessed need. These differences in design and financing give rise to diverse models of social investment, each associated with distinct distributive outcomes (Garritzmann, Häusermann and Palier, 2022[20]).
The remainder of this chapter shows that key aspects of the evolution of social expenditure are consistent with a social investment approach, while also documenting substantial cross-country variation in dimensions central to supporting children and their families. It highlights, first, the overall expansion of social spending, including a growing share devoted to in-kind benefits (Section 2.2). It then documents the strengthening of support for families with children – particularly in the early years – reflecting a greater emphasis on early intervention and prevention (Section 2.2).
The chapter also examines the reinforcement of labour market activation policies, including expanded employment support and efforts to reduce barriers to work for parents, notably through increased provision of early childhood education and care services and tax-benefit reforms aimed at improving work incentives for low-income families (Section 2.3). Finally, it points to a decline in the redistributive impact of income transfers in some countries, consistent with a shift away from predominantly compensatory welfare models (Section 2.4).
Expenditure on social benefits in kind is central to a social investment perspective, as it directly builds human capital and expands access to essential services – such as childcare, health, and housing – thereby strengthening long-term opportunities and reducing inequalities more effectively than income transfers alone. In OECD social expenditure data, in-kind benefits encompass publicly or privately provided services, including healthcare (in- and out-patient care, medical goods, and preventive services), early childhood education and care, housing support, and a broad range of social services (such as care for persons with disabilities, and families).
While not formally classified as in-kind benefits, many elements of active labour market policies share similar characteristics, as they provide service‑based support – including public employment services (placement, counselling, job-search assistance), training programmes, hiring incentives, integration support for people with disabilities, and direct job-creation subsidies.
In most OECD countries, the largest share of social spending is devoted to in-kind support, which accounted for about 64% of total social expenditures per capita in 2021 across the OECD, up from around 58% at the start of the 2000s (Figure 2.3, Panel A). Reliance on in-kind provision is particularly strong in Colombia, Costa Rica, Mexico and the United States, where 85% or more of social spending takes this form. By contrast, Greece, Italy and Poland allocate social expenditures almost evenly between cash and in-kind support.
Levels of in-kind spending – and how this spending is distributed across social risk areas – vary widely across countries (Figure 2.3, Panel B). In 2021, OECD countries spent on average around a total of USD 4 700 (in 2015 prices) per capita on in-kind support, but this amount is over twice as high in the United States and almost six times lower in Mexico. Looking at the composition of in-kind provision, health-related services account for by far the largest share: on average, nearly three‑quarters of per-capita in-kind spending goes to health services. In-kind support for families is the second most important area, although spending levels are much lower – on average, about seven times lower than in health. Nordic countries (notably Denmark, Iceland, Norway and Sweden) and Luxembourg stand out, with spending on in-kind family support roughly twice the OECD average.
Active labour market policies also show substantial cross-country variation. Per-capita spending is highest in Ireland – around USD 1 260 (in 2015 prices), which is nearly four times the OECD average – and is also comparatively high in Denmark, Luxembourg, the Netherlands and the United States. In Denmark, this reflects substantial investments in training and upskilling programmes, vocational rehabilitation and supported employment for people with reduced work capacity, as well as wage‑subsidised “flex jobs” for individuals with disabilities or health limitations. In Luxembourg, high spending is driven mainly by extensive training programmes and employment incentives. In the United States, elevated spending reflects strong investment in training programmes, youth-focussed initiatives such as Job Corps – which supports young people transitioning from school to work – and measures promoting job rotation and job-sharing schemes.
In-kind support related to incapacity is likewise particularly elevated in Denmark and Luxembourg, where per-capita spending exceeds the OECD average by a factor of four or more. Most of this spending goes to residential care for people with disabilities or incapacity, rehabilitation services, and subsidies for the care of children and adults with disabilities, as well as specialised day-care facilities for children requiring special support.
Regarding in-cash support, total per capita expenditures have risen markedly since the early 2000s in many countries, increasing by about 30% on average in constant USD compared with 2001 (Figure 2.3, Panel C). However, this type of spending has declined in Luxembourg, Norway, Sweden, Finland, Australia and Mexico during the same period. Levels vary even more widely across countries, and in-cash support remains particularly low in Chile, Colombia, Costa Rica, Mexico and Türkiye.
Among the different categories of in-cash support outside pensions, incapacity-related benefits account for the highest average spending, though with substantial cross-country variation. Spending is especially high in Luxembourg, Iceland, Norway and Switzerland, where a large share is devoted to disability pensions – and, in Norway, also to paid sick leave and care allowances. In all countries, unemployment benefits and family-related cash benefits also represent important components of in-cash expenditure; the scope and design of family benefits are discussed in the next section.
Note: The data show per capita spending across public, mandatory private and voluntary private programmes. There is no in-cash spending on housing, and no in-kind spending on unemployment. Spending on active labour market programmes is included in the in-kind expenditure despite not being classified as in-kind or in cash spending in the OECD Social Expenditure Database; in-kind expenditure on survivors is excluded as it mostly covers funeral-related expenses, which are less relevant for this analysis; in-cash expenditure on health is excluded as it is only relevant for Chile. Similarly, old age spending is excluded from all figures.
OECD and EU averages are computed for all respective Member countries with available data for both 2001 and 2021. Data refer to 2021 unless specified otherwise. In each panel, “Total 2001” refers to the 2001 total spending amount across all categories shown in the figure. Panels A and C: Data for Chile refer to 2022 instead of 2021. In Chile, cash support for families expanded during COVID‑19 through the “Programa Ingreso Familiar de Emergencia” (Emergency Family Income Programme) which in 2021 amounted to 7.3% of GDP. The programme was phased out in 2022. Panels B and C: Data are reported in per capita USD PPP in constant prices.
Source: OECD (OECD, n.d.[1]), OECD Social Expenditure Database (SOCX), https://www.oecd.org/en/data/datasets/social-expenditure-database-socx.html.
Families with children are the direct target of a broad set of supports that pursue multiple, interconnected goals: (1) supporting families in raising children and reducing the economic costs of childrearing; (2) promoting gender equality and helping parents – especially mothers – reconcile work and family life; (3) reducing child poverty and social inequalities; and (4) fostering children’s development and long-term opportunities through investments in early education, childcare, and parenting support. (Adema, Clarke and Thévenon, 2020[14]; Thévenon, 2011[60]).
Public spending on family benefits can be grouped into three broad categories:
Child-related cash transfers, including child allowances (sometimes income‑tested and often varying with the age or number of children), income support during parental leave, and – in some countries – specific support for single‑parent families.
Public spending on services for families, such as the direct financing or subsidisation of childcare and early childhood education, earmarked childcare payments to parents, support services for young people and residential care facilities, and broader family services, including centre‑based support and home‑help services for families in need.
Tax-based financial support, including tax exemptions (e.g. excluding child benefits from taxable income), child tax allowances (amounts deducted from gross income), and child tax credits (amounts deducted from the tax liability).
An increasing allocation of national income to families with children
Most OECD countries have strengthened their support for families since the early 2000s, as reflected in the rise in family spending as a share of GDP – from about 2% in 2001 to 2.3% on average in 2021 (Figure 2.4, Panel A). Increases were particularly pronounced in Iceland (+1.4 p.p.), Poland (+2.4 p.p.), Japan (+1.4 p.p.), Korea (+1.5 p.p.) and Spain (+1 p.p.). By contrast, family spending declined in a few countries, most notably in Australia (‑1 p.p.) and Ireland (‑0.9 p.p.), where GDP grew more rapidly than family spending.
Trends are broadly similar in per-capita terms.2 Most countries increased per-capita family support in cash, though there are notable exceptions – including Luxembourg, Norway, Australia, Finland, the United Kingdom and the United States – where per-capita cash support declined (Figure 2.4, Panel B). In many countries, family allowances account for the largest share of expenditures to raise families’ income, although income replacement during maternity and/or parental leave also represents a substantial expenditure in several cases. Growth in in-kind support for families has been even more pronounced: on average across the OECD, per-capita in-kind family spending in 2021 was over twice its 2001 level (Figure 2.4, Panel C). The bulk of this in-kind expenditure is devoted to early childhood education and care (ECEC) services for preschool children, which account for around two‑thirds of in-kind family spending on average.
Note: OECD and EU averages are computed for all respective Member countries with available data for both 2001 and 2021. Data refer to 2021 unless specified otherwise. In each panel, “Total 2001” refers to the 2001 total spending amount across all categories shown in the figure. * Data refer to 2002 for lack of available data in 2001. ** Data for 2021 refer to 2022. In Chile, cash support for families expanded during COVID‑19 through the “Programa Ingreso Familiar de Emergencia” (Emergency Family Income Programme) which in 2021 amounted to 7.3% of GDP. The programme was phased out in 2022. Panels B and C: Spending is denoted in per capita expenditure in USD PPP in constant prices. Per capita refers to the total population size rather than per-child expenditure.
Public spending accounted for here concerns public support that is exclusively for families (e.g. child payments and allowances, parental leave benefits and childcare support), only. Spending in other social policy areas such as health and housing support also assists families, but not exclusively, and is not included here. Coverage of spending on family and community services in the OECD Social Expenditure data may be limited as such services are often provided and/or co-financed by local governments. The latter may receive general block grants to finance their activities, and reporting requirements may not be sufficiently detailed for central statistical agencies to have a detailed view of the nature of local spending. In Nordic countries (where local government is heavily involved in service delivery) this does not lead to large gaps in the measurement of spending, but it does for some countries with a federal structure, for example, Canada and Switzerland. National authorities provided estimates on the value of tax breaks for Switzerland. Spending for the United Kingdom is likely to be underestimated, as information on the tax part of the WTC and CTC are no longer available they have been estimated.
Source: OECD (n.d.[61]), OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html and OECD (OECD, n.d.[1]), OECD Social Expenditure Database (SOCX), https://www.oecd.org/en/data/datasets/social-expenditure-database-socx.html.
A shift towards greater investment across childhood, especially in early years
A closer examination of how public social and education expenditures are distributed across different stages of childhood shows that per-child spending has increased substantially since the early 2000s at all ages, with particularly strong growth for children below compulsory school age. As shown in Figure 2.5 for the OECD average, the sharp rise in support for families with preschool-aged children reflects significant increases in both cash benefits around childbirth – linked to the expansion of maternity, paternity and parental leave income support (Box 2.3) – and in the provision of childcare and early education services.
Nevertheless, despite this significant expansion, the early years (children under age 6) continue to account for a smaller share of total public expenditure on children than later stages of childhood. On average across the OECD, spending on the early years represented 29% of total child-related expenditure in 2021, compared with 35% for middle childhood (children between age 6 and 11) and 36% for adolescence (ages 12‑17) (OECD, n.d.[62]). Only a limited number of countries – namely Iceland, Lithuania, Estonia, Hungary, Latvia and Japan – allocate sufficient resources to families with preschool-aged children for the early years to account for one‑third or more of total public spending on children. By contrast, the early-years share remains below 15% in Costa Rica, Türkiye and the United States (OECD, n.d.[62]).
Note: The graphs show OECD averages of per-child spending patterns by age and by type in USD PPP. Spending figures are central government spending amounts, and do not include spending at a local or regional level since these data are not readily available. The figures offer approximate spending patterns, developed from aggregate spending data and spending rules for each benefit (e.g. eligibility by age or enrolment in education, payment amounts). The indicators are calculated using the Age‑Spending Profiles methodology used in OECD (2011[63]), Doing Better for Families, https://doi.org/10.1787/9789264098732-en.
Panel A: Data refer to 2021 unless specified otherwise. “Total 2003” comprises all spending categories shown in the figure for 2003.
Panel B: The age categories represent the sum of per-child spending for each year within the age range, providing an estimation of what would be the cumulative spending on one child during that age period if the spending rules for the given year remained unchanged.
For more information on the Age‑Spending Profiles methodology, see Annex 2.A1 in OECD (2011[63]), Doing Better for Families, https://doi.org/10.1787/9789264098732-en.
Source: OECD (n.d.[61]), OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html based on OECD (OECD, n.d.[1]), OECD Social Expenditure Database (SOCX), https://www.oecd.org/en/data/datasets/social-expenditure-database-socx.html and OECD (n.d.[64]), OECD Education Database, https://data-explorer.oecd.org/s/5ca.
Paid leave entitlements have expanded considerably across OECD countries over the past five decades. In 1970, mothers had access to an average of just 16 weeks of paid leave, rising to 43 weeks by 1995 and roughly one year by 2024. The largest increases occurred in Finland and the Slovak Republic, where mothers can now receive more than three years of paid leave, compared with just 9 and 26 weeks respectively in 1970. Reductions in total paid leave are rare and mostly due to reforms introducing flexible payment-rate options, as seen in Czechia and Germany.
Countries that offer the longest leave entitlements today are generally those that introduced parental leave earliest (OECD, n.d.[65]; Thévenon, 2018[66]). With the exception of Italy and Spain, all countries that had parental leave policies in place by 1980 provide mothers with at least one year of employment-protected leave, and many of these early adopters offer two years or more. By contrast, countries that introduced parental leave after 1980 – especially after 1990 – tend to offer shorter durations, though Germany and Portugal stand out as later adopters that now provide mothers with more than two‑and-a-half years of protected leave.
Paid father-specific leave is a much more recent development. In 1970 it existed in only three OECD countries, and by 1995 the OECD average remained under one week. Since then, however, most countries have introduced dedicated entitlements for fathers: by 2024, 35 of 38 OECD countries offered paid father-specific leave, averaging 12.7 weeks. Major reforms in Korea and Japan have extended fathers’ individual entitlement to one year of paid leave, while several EU countries expanded their provisions to comply with the 2019 Work-Life Balance Directive, which requires at least 10 working days of paternity leave and a minimum of two non-transferable months of paid parental leave for each parent.
Source: OECD (n.d.[65]), OECD Family Database, PF2.5 Trends in leave entitlements around childbirth since 1970, https://www.oecd.org/en/data/datasets/oecd-family-database.html.
A broad portfolio of cash-based family support
Expenditure on cash transfers that support family incomes is particularly sizeable during the early years of a child’s life. In most countries, however, cash benefits continue throughout childhood until the child reaches adulthood – and in some cases beyond, when young adults remain financially dependent on their parents’ household. Across OECD countries, cash support for families typically takes the form of a combination of benefits, either broadly available to families or targeted at specific groups defined by income level, family circumstances (such as single‑parent households), and in some cases conditional on certain behaviour.
Among the 35 OECD countries with available data, all provided at least one type of family benefit in 2024, and the large majority (30 countries) offered two or more. Greece, Spain, Switzerland, Türkiye and the United Kingdom provided only one family benefit, while Norway offered six distinct benefits, each with its own eligibility criteria and not necessarily mutually exclusive (OECD, n.d.[67]). Around half of OECD countries apply means testing to their main family benefit, requiring household income – and, in Spain and Portugal, also assets – to fall below a specified threshold.
In addition, 22 OECD countries provide benefits specifically targeting single‑parent families or equivalent forms of support. Ten countries – Denmark, Estonia, Finland, Germany, Latvia, Lithuania, Norway, the Slovak Republic, Sweden and Switzerland – also operate alimony-advance schemes, which are often (though not always) activated when the non-resident parent fails to pay child support, as is the case in Denmark, Estonia, Germany (where the scheme is also activity-tested), Lithuania, Norway, the Slovak Republic and Sweden.
Several OECD countries also provide cash benefits linked to child characteristics, most commonly school-related allowances intended to help cover the costs of school supplies, clothing, and materials, or to support low-income students by reducing or eliminating voluntary or ancillary school fees. Three countries – France, Italy and Norway – offer means-tested benefits targeted specifically at families with younger children.
Finally, some OECD countries provide family benefits linked to parents’ labour market status, typically through in-work benefits or tax credits or childcare allowances, as observed in Belgium, Czechia, Finland, Korea and Norway (OECD, n.d.[67]). Additional financial support is often delivered through the tax system rather than direct cash transfers. This includes tax exemptions (e.g. excluding child benefits from taxable income), child tax allowances (amounts deducted from gross income before taxation), and child tax credits (amounts directly deducted from tax liability).
A recent comparative study of child benefit distribution by Nightingale et al. (2025[68]) shows that European countries cluster into distinct models of child-benefit spending based on three dimensions: overall investment levels, the balance between cash and in-kind support, and the extent of means-testing. Among high-investment countries, some combine generous spending with limited targeting and a strong reliance on cash transfers (Austria, Poland, Estonia, Lithuania, Luxembourg), while others prioritise in-kind services under similarly low levels of means-testing (Bulgaria, Denmark, Finland). Germany stands alone in pairing high investment with strongly targeted benefits delivered mainly in-kind. Notably, no country adopts a model that features both high overall investment and a strong emphasis on cash support with extensive means-testing.
Among lower-investment countries, three broad approaches emerge. A first group relies heavily on means-tested cash transfers despite limited overall spending (Ireland, Greece, Cyprus, Portugal, Slovenia). A second, larger group maintains low means-testing and favours cash benefits (Romania, the Slovak Republic, Czechia, Croatia, Italy, Latvia, Hungary, Belgium). A final set of countries combines low investment with low means-testing but places more weight on in-kind support, including Spain, France, Malta, the Netherlands and Sweden. No low-investment country combines heavy means-testing with a strong emphasis on in-kind benefits.
Cash support responds to immediate income needs and leaves parents free to decide how resources are used. In contrast, in-kind support is designed to respond more directly to specific needs. In-kind support to families with children consists of benefits provided as services or goods rather than as cash transfers. The most important forms include early childhood education and care, education and health services, housing support, school meals and transport, and a range of family, child protection and disability-related services (Box 2.4).
Compared with cash transfers, in-kind support has the advantage of ensuring that public resources are spent on specific goods and services which are considered as essential for child healthy development and well-being, thereby helping governments pursue objectives such as promoting health and human capital formation, equal access to essential services. In-kind provision can also reduce inequalities in access that would otherwise depend on family income, information or preferences, and therefore help prevent the formation of long-term disadvantage. Universal or heavily subsidised services limit the extent to which children’s opportunities are shaped by parental resources, while subsidised childcare in particular lowers barriers to employment for parents – especially mothers – making in-kind support an important complement to family cash benefits. Moreover, in-kind services can address market failures, such as under-provision or uneven quality of childcare and education, that cash transfers alone may not resolve.
For this reason, most OECD countries rely on a combination of cash and in-kind support. Cash benefits help stabilise household incomes and reduce child income poverty, and are expected to enable parents to invest more in the goods and services needed for children to grow and develop well. In-kind services, by contrast, are more directly linked to policy objectives such as promoting child development, ensuring equal access to essential services, and supporting parental employment. In addition, conditioning cash transfers on the take‑up of targeted services can help focus support on priority groups while ensuring that households have the resources needed to access those services (Box 2.4).
Family support services encompass a broad range of public and community-based interventions designed to strengthen parenting capabilities, improve family functioning, and help children thrive across childhood. In the OECD context, these services are typically organised along several key areas of support, each responding to different aspects of family and child well-being (Riding et al., 2021[17]).
Basic material needs
Support for basic material needs helps families meet essential everyday requirements, including food, clothing and housing. Programmes include school meal schemes, food and clothing vouchers, direct provision of items through community services, and access to subsidised or social housing. These supports play a crucial role in reducing short-term deprivation and removing barriers to equitable participation in school and community life.
Healthcare services
Health-related family support services promote maternal, child and adolescent health. This includes prenatal and postnatal care, regular check-ups and vaccinations, specialist referrals, prescription coverage and mental health support such as counselling or therapy. These services are essential for preventive care and early intervention in health risks that can affect long-term development.
Family functioning support
These services assist families with strengthening relationships and coping with everyday challenges. They include family counselling, casework and specialised social services, respite care during family crises, and in-home support tailored to individual family needs. Such services help families manage stress, improve caregiving practices, and address issues before they escalate to higher-risk situations.
Parenting support and early intervention
Parenting programmes and early intervention services aim to build parental skills and enhance the parent-child relationship. They can be delivered through home‑visiting practitioners, community centres or schools and take the form of goal-oriented casework, individual or group counselling, and evidence‑based parenting curricula. These interventions are particularly effective in supporting families before crises develop.
Specialised services for complex needs
Specialised support targets families facing multiple or deep-seated challenges. This includes services for children with disabilities, families affected by addiction, and supports tailored to socially marginalised groups such as migrants or refugees. These specialised programmes help bridge gaps that universal supports alone cannot fill.
A growing body of evidence highlights pregnancy and the first 1 000 days of life as a critical window for child development and long-term outcomes. In response, several countries have adopted a “first 1 000 days” approach that co‑ordinates policies to provide tailored, continuous support to pregnant women and families with infants. This approach emphasises early identification of needs through health and social services, timely referral to appropriate support, and preventive interventions to address emerging difficulties before they escalate, including coaching and guidance for new and expectant parents.
Families with school-age children can access a broad range of services delivered through schools, which are well placed to identify disadvantaged children, assess their needs, and maintain contact with parents or caregivers. By bringing children together in a single setting, schools facilitate regular follow-up and access to services that might otherwise depend on parental availability, including health check-ups provided by school health services, whose scope and frequency vary across countries. Many education systems also integrate psychological support, as in Copenhagen, where school counsellors can refer adolescents to targeted interventions with psychologists addressing issues such as stress, loneliness, substance use, depression or anxiety. In addition, schools play a key role in delivering school-meal programmes in many countries, often with income‑tested or targeted eligibility, both to support family budgets and to promote child nutrition and food security, sometimes in partnership with local authorities, NGOs and voluntary organisations.
Combining conditional cash transfers with support services can effectively increase service take‑up and improve family outcomes by linking financial assistance to participation in programmes such as health screenings, school meals or parenting classes. Evidence from experimental “Cash+” programmes shows that this integrated approach delivers better educational, health and nutrition outcomes than providing cash or services alone (Little et al., 2021[69]). Reflecting this, conditional cash transfers are increasingly used in OECD countries, with around half of capital cities supporting vulnerable families through such schemes, particularly when higher benefit levels, longer payment durations and well-timed transfers help sustain engagement with services.
An important dimension of support for families with children is aimed at promoting parents’ employment and helping them remain in work after the birth of a child, by facilitating the reconciliation of work and family life and ensuring that paid employment remains financially worthwhile for both partners in couple households as well as for single parents. This objective is pursued primarily through child-related leave entitlements for pregnant women and parents of newborn children – which, as discussed in the previous section, have expanded substantially over recent decades (see Box 2.3) – and through the provision of early childhood education and care (ECEC) services. Beyond enabling parents to balance employment and caregiving responsibilities, ECEC services also seek to provide children with high-quality, professionally delivered early learning and care, helping to foster child development and reduce inequalities in early life opportunities (OECD, 2025[70]; Duncan et al., 2022[71]).
Unequal development of early childhood childcare and education services
As illustrated in Figure 2.5, public investment in ECEC services has grown substantially across the OECD since the early 2000s. However, spending on ECEC services as a percentage of GDP varies widely across countries, ranging from less than 0.3% in Türkiye to more than 2% in Iceland – around seven times higher (Figure 2.5, Panel A). These differences reflect varying national priorities in expanding service availability, improving affordability for families, and enhancing the quality of early childhood education and care.
Childcare service coverage is particularly important for enabling parents – especially mothers, who still provide the bulk of childcare – to participate in the labour market and progress in their careers. As shown in Figure 2.5, Panel B enrolment in formal childcare for children under age 3 has risen markedly since the mid‑2000s, increasing from around one in five children to more than one in three across the OECD by 2024. These aggregate trends, however, conceal substantial inequalities by family income. In many European countries with available data, children from low-income families are significantly less likely to attend formal childcare services. On average across European countries in 2024, about 49% of children under age 3 from the highest income third were enrolled in formal childcare, compared with 33% among those from the lowest income third. This disparity reflects the so-called “Matthew effect”, whereby childcare provision disproportionately benefits more advantaged families, for a variety of structural, financial, and informational reasons (Box 2.5).
Note: Panel A: Enrolment rates for ages 0‑2 refer to 2019 for Canada and Japan and to 2018 for Costa Rica and the United Kingdom; for ages 3‑5, data refer to 2019 for Belgium, Canada and Greece and to 2018 for the United States. In some countries local governments play a key role in financing and providing childcare services. Such spending is comprehensively recorded in Nordic countries, but in some other (often federal) countries it may not be fully captured by the OECD social expenditure data. Panel B: Data from the 2024 series for Hungary refer to 2022; for Australia, Austria, Brazil, Chile, Colombia, Denmark, Estonia, Finland, Germany, Iceland, Israel, Japan, Korea, Lithuania, Mexico, New Zealand, Norway, Slovenia, Spain, Sweden, Switzerland and Türkiye to 2023; for Canada to 2025. Data from the 2005 series for Bulgaria and Romania refer to 2007; for Malta to 2008. Data generally include children enrolled in early childhood education services (ISCED 2011 level 0) and other registered ECEC services (ECEC services outside the scope of ISCED 0, because they are not in adherence with all ISCED-2011 criteria). See notes of Chart PF3.2.A in the OECD Family Database (n.d.[61]) for differences in data source and service definition. Panel C: Data for Switzerland refer to 2023; for Germany and Hungary to 2022. Data are OECD estimates based on information from EU-SILC. Data refer to children using centre‑based services (e.g. nurseries or day care centres and pre‑schools, both public and private), organised family day care, and care services provided by (paid) professional childminders, regardless of whether or not the service is registered or ISCED-recognised. Equivalised disposable income tertiles are calculated using the disposable (post tax and transfer) income of the household in which the child lives – equivalised using the square root scale, to account for the effect of family size on the household’s standard of living – and are based on the equivalised disposable incomes of children aged less than or equal to 12.
Source: Panel A: “PF3.1 Public spending on childcare and early education” in OECD (n.d.[61]), OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html; Panels B and C: “PF3.2 Enrolment in childcare and pre-school” in OECD (n.d.[61]), OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html.
The “Matthew effect”1 in early childhood education and care (ECEC) describes a pattern where policies or service expansions intended to help all children end up disproportionately benefiting better-off families – so the advantaged gain more and the disadvantaged fall further behind. In the context of childcare this typically shows up as higher uptake and better quality experiences among children from higher-income or more educated families, even where overall provision grows (Pavolini and Van Lancker, 2018[72]; Van Lancker, 2023[73]; OECD, 2025[70]).
Key determinants of the “Matthew effect” include:
Supply constraints and uneven availability. Limited places, long waiting lists and geographically uneven provision mean better-connected or higher-income parents (who can commute, pay higher fees or navigate admissions) benefit first (OECD, 2025[70]).
Cost and affordability barriers. Fee structures and poor targeting of subsidies leave low-income families facing higher effective prices or administrative obstacles to take‑up (OECD, 2020[74]).
Demand-side factors and parental resources. Differences in parents’ employment patterns, access to information, preferences, social networks, and capacity to navigate and use available services (e.g. because of irregular working hours, language barriers, administrative complexity or limited familiarity with the social protection system) can reduce service uptake among disadvantaged families. Recent work teases apart how much of the gap is about preferences versus structural constraints (Wood, Neels and Maes, 2023[75]).
Policy design and sequencing. Policies that expand high-quality places but leave fees, accessibility or complementary supports unchanged tend to favour those already best placed to benefit (Van Lancker, 2023[73]).
The Matthew effect in early childhood education and care (ECEC) has far-reaching consequences for both children and families. When better-off households are more likely to access high-quality services, early cognitive and socio‑emotional gaps widen and can persist throughout schooling and into adulthood, reinforcing existing patterns of inequality (OECD, 2025[70]). At the same time, the labour-market and anti-poverty benefits typically associated with ECEC expansion are weakened: if advantaged parents are the main beneficiaries, improvements in maternal employment, family income and poverty reduction fall short of expectations (OECD, 2018[76]). Finally, unequal take‑up can undermine the legitimacy and efficiency of social-investment strategies. Reforms that deliver disproportionate gains to higher-income families risk political backlash and fail to meet their redistributive goals, limiting their long-term impact and public support (Van Lancker, 2023[73]).
Reversing the Matthew effect in ECEC requires a comprehensive policy approach that tackles both supply- and demand-side barriers. Expanding the supply of high-quality places – particularly in underserved neighbourhoods-helps ensure that limited availability does not systematically favour better-off families (Abrassart and Bonoli, 2015[77]). Affordability is equally critical: progressive fee structures, simple subsidy rules and automatic or streamlined enrolment can substantially reduce financial barriers for low-income households (OECD, 2020[74]). At the same time, complementary policies such as well-aligned parental leave, extended opening hours, targeted outreach and accessible information, and strong staff training are essential to address the administrative, cultural and practical hurdles that often deter disadvantaged parents (OECD, 2011[63]; Adema, Clarke and Thévenon, 2020[14]). Finally, governments need to monitor not just overall coverage but who benefits: regularly collecting and publishing data on participation and quality by income, education and location allows early detection of emerging Matthew effects and supports timely policy correction (Van Lancker, 2023[73]).
1. In social policy, the “Matthew effect” is named after a passage in the Gospel of Matthew (“to those who have, more will be given”), reflecting the idea that social benefits often disproportionately advantage those who are already better off.
Increased investments in active labour market policies
Active labour market policies constitute another important pillar of support for working-age adults. While not specifically targeted at parents, families with children can benefit substantially from these measures, as care responsibilities remain a significant barrier to employment for many – particularly women (Fernandez et al., 2016[78]; Farchy and Immervoll, 2021[79]). Parents may rely on these programmes to re‑enter employment following career interruptions related to caregiving, to upgrade skills that facilitate a return to work, or to support longer-term career progression. Active labour market policies encompass a range of measures, including public employment services (such as job placement, counselling and job-search assistance) and training programmes designed to improve labour market integration for groups facing specific barriers. These groups often include people with disabilities, single parents, and parents who have been out of the workforce for extended periods following the birth of a child (Martin, 2015[80]; Nieuwenhuis, 2022[47]).
Available data do not allow for spending to be disaggregated by beneficiary group, making it difficult to determine to what extent families with children benefit from these programmes. Evidence in Figure 2.7 shows that per-capita spending on ALMPs has increased in many countries over the past two decades, with substantial cross-national variation. Ireland stands out as the country experiencing the largest increase, with per-capita spending of almost four times the OECD average in 2021. While a significant share of the 2021 spending patterns can be expected to reflect countries’ responses to the COVID‑19 pandemic, the overall trend of increased per-capita spending on ALMP across the OECD also holds for comparisons with pe‑pandemic years (see Figure 2.7, Panel B above). The same is true for the large cross-national variation in both trends and levels (not shown).
Per capita expenditures in constant USD PPP, 2021 and 2001
Note: OECD and EU averages are computed for all respective Member countries with available data for both 2001 and 2021. Spending is denoted in per capita expenditure in USD PPP in constant prices. Per capita expenditure refers to the total population size rather than expenditure per unemployed person or per working-age population. Active labour market programmes (ALMPs) include all social expenditure (other than education) which is aimed at the improvement of the beneficiaries’ prospect of finding gainful employment or to otherwise increase their earnings capacity. This category includes spending on public employment services and administration, labour market training, special programmes for youth in the transition from school to work, special labour market and rehabilitation programmes for persons with disabilities, and labour market programmes to provide or promote employment for all other groups of unemployed and inactive persons.
Source: OECD (n.d.[3]), Social Expenditure database, https://data-explorer.oecd.org/s/46u.
Making work pay for low-income families
The financial returns to employment – in other words, whether work pays – are another critical determinant of parental labour market participation and families’ ability to escape poverty. These returns depend not only on wage levels, but also, crucially, on the design of tax and benefit systems, which shape the effective tax rates faced by low-paid working parents. Figure 2.8 shows effective tax rates for parents with two children who move into full-time employment after five years of unemployment. The figure shows the share of additional earnings that is effectively “taxed away” through higher taxes and reduced benefits, reported separately for single parents earning 67% of the average wage and for partnered parents who are both employed full-time at 50% of the average wage. This earnings level represents a relatively modest income and, for a family with two children, is close to the poverty line in many OECD countries, highlighting the importance of tax-benefit design for making work financially worthwhile for low-income families.
Panel A of Figure 2.8 shows substantial cross-country variation in the share of earnings that single parents effectively “pay back” through higher taxes or reduced benefits when taking up such employment. In Estonia, only about 13% of earnings are lost – far below the OECD average of 56% – while Slovenia displays the highest effective tax rate, with around 87% of earnings offset by lower benefits or higher taxes, leaving households with a take‑home gain of just 13% of their earnings gained from work.
Similar cross-national disparities appear for couple households in which a second earner enters employment (Figure 2.8, Panel B). On average, effective tax rates for these couples are lower than for the selected category of single parents, partly because single parents are more likely to lose access to social assistance and other minimum-income benefits when they take up work. Figure 2.8, Panel C maps countries according to whether effective tax rates for single parents and for couple families fall above or below the OECD average. The lower-left quadrant identifies countries with below-average rates for both household types, whereas the upper-right quadrant highlights those with above‑average rates for both.
Note: This indicator measures the financial disincentives to participate in the labour market after five years of unemployment. It calculates the proportion of earnings in the new job that are lost to either higher taxes or lower benefit entitlements when a jobless person takes up full-time employment paid at 67% of average wage for single‑parents and at 50% for a second earner in couple families. The indicator assumes the presence of two children in the household, aged 4 and 6. Childcare benefits or costs are not accounted for and in couple families, the primary earner is assumed to work full time at 50% of the average wage. Cash housing benefits are calculated assuming private market rent, plus other charges, amounting to 20% of the full-time wage for all family types.
Source: OECD Secretariat calculations based on output from the OECD (n.d.[81]) Tax-Benefit Model (version 2.7.1), https://www.oecd.org/en/data/tools/oecd-calculator-of-taxes-and-benefits.html.
In couple families, having two partners creates the potential to draw on two sources of labour income, which can raise overall family income and provide greater income stability over time. This is particularly important for low-income households with children, where the first earner (often the father) is employed in a low-paid job and the second earner (often the mother) can enter the labour market to help lift the family out of poverty. Whether this second earner chooses to work, however, depends crucially on the effective tax rates they would face and on whether supplying labour as a second earner is financially more attractive than increasing the income of the first earner. Differences in the tax and benefit burden associated with a given level of household income – depending on how earnings are split between partners – therefore provide a useful indication of the financial incentives for second earners in low-income families to enter employment, compared with achieving the same income through a single earner.
Figure 2.9 shows the difference in net household income between single‑earner couples and couples where both partners earn the same gross income (assumed to equal the average earnings in total). A larger difference indicates a stronger financial gain from having two earners rather than one. The results suggest that the financial advantage of dual earner families is particularly strong in the Netherlands, Italy, Finland and Belgium, where the gain is at least twice the OECD average. In contrast, countries on the right-hand side of the figure have tax systems that are broadly neutral with respect to the distribution of earnings, and only in Japan do single‑earner couple families appear to have a financial advantage.
Additional net household income of an equal dual-earner couple compared to a single‑earner couple, as a proportion (%) of net household income for the single‑earner couple, 2024
Note: The estimates here relate to the situation for a couple household with two children aged 4 and 6 and gross household earnings equal to 100% of average earnings. In those situations where one member of the household is not working, it is assumed that they are not entitled to unemployment benefits (for example, because their entitlements have expired). The household is however assumed entitled to social assistance and other means-tested benefits, subject to the relevant income conditions. Where receipt of social assistance is subject to activity tests (such as a non-working household member engaging in an active job-search or being “available” for work), these requirements are assumed to be met. Cash housing benefits are calculated assuming private market rent, plus other charges, amounting to 20% of the full-time wage for all family types. Neither childcare benefits nor childcare costs are considered.
Source: OECD Secretariat calculations based on output from the OECD (n.d.[81]) Tax-Benefit Model (version 2.7.1), https://www.oecd.org/en/data/tools/oecd-calculator-of-taxes-and-benefits.html.
As noted earlier, labour market income can provide families with durable protection against poverty when employment is stable and earnings are sufficient once shared among household members. However, the rise in in-work poverty documented in Chapter 3 suggests that, for many households with children, labour income is increasingly insufficient to lift families above the poverty line.3 In such cases, policies that address the structural drivers of in-work poverty – including adequate minimum wages, skills development, secure employment contracts, and access to affordable childcare – are essential to prevent poverty among working households. At the same time, income support through cash benefits, tax credits, or tax reductions for low-income families can be also important. The need for these income transfers – and the extent to which they must complement low earnings – depends critically on working hours during a week, month and a year, and how wages compare to poverty threshold at the lower end of the income distribution.
Stronger protection from minimum wages, yet insufficient to curb in-work poverty
Figure 2.10 shows that, on average across the OECD, minimum wages have increased relative to median income over time – from just over 42% in the early 2000s to around 55% in 2023 – suggesting that minimum-wage employment now offers better protection against poverty than two decades ago. This upward trend is not universal, however. In a significant number of countries the increase has been modest, and in some cases the minimum wage relative to median income has declined, notably in the United States and, more recently, in Türkiye (Annex 2.A).
At the same time, the poverty line anchored in 2005 represents a declining share of median income, suggesting that minimum wages may have played an increasingly important role in mitigating child and family severe poverty during the 2010s. However, although earnings from full-time work at the statutory minimum wage have, on average, risen slightly above the relative poverty threshold, the growing incidence of in-work poverty indicates that low-paid employment – often combining low wages with insufficient working hours – frequently fails to provide adequate income to lift families with children out of poverty (OECD, 2023[82]). In principle, this gap could be compensated by income transfers. Yet Figure 2.10 shows that transfers received by the average poor household have, on average, declined, despite considerable cross-country variation (see Annex), and in many countries have remained broadly stable as a share of median income. Taken together, these trends suggest that while earnings from minimum-wage (full-time) employment have become relatively more important in reducing income poverty, the stagnation – or decline – of income support has made it increasingly difficult for households with low earnings to escape poverty.
OECD average evolution of poverty lines, minimum wage and income transfers, as a share of median income
Note: Data are expressed as a percentage share of the median equivalised disposable income in a given year. Transfers to the average poor household are computed by comparing the average equivalised household income of individuals below the relative poverty line before and after income transfers. The anchored poverty line displays the inflation-adjusted relative poverty line from 2005. The minimum wage represents the theoretical earnings of a full-time employee under the national minimum wage legislation, averaged across countries where such legislation exists. Across all series, the composition of the OECD average changes over time according to data availability (ranging from 14 to 36 countries across series and years). However, the displayed trends are not sensitive to the fluctuation of countries in the composition of the average.
Source: OECD (n.d.[83]), OECD Income Distribution Database, https://www.oecd.org/en/data/datasets/income-and-wealth-distribution-database.html.
The redistributive role of income transfers
Tax and benefit systems pursue multiple objectives, of which income redistribution and poverty protection are only one – albeit a central – component.4 Their redistributive role operates along two dimensions: vertical and horizontal redistribution. Vertical redistribution involves transfers from higher-income to lower-income households, ensuring that individuals and families have access to a minimum income sufficient to meet basic needs and are protected against key social risks such as unemployment, disability, old age, or family-related shocks. Horizontal redistribution, by contrast, reallocates resources between households with similar incomes but different needs – particularly due to differences in family size and composition. In this context, it typically involves transfers from childless households to those with children, helping to ensure that families have adequate resources to meet the costs of raising children.
Vertical redistribution
The vertical redistributive effect of taxes and transfers is typically assessed by comparing household income inequality before and after taxes and transfers, using indicators such as the Gini coefficient. Looking at trends since the early 1990s, Causa and Hermansen (2017[84]) show that redistribution through taxes and transfers has declined on average and in most countries with data extending back to the mid‑1990s. Most of this decline occurred between the mid‑1990s and mid‑2000s. The onset of the 2008-2009 crisis temporarily halted the trend, as automatic stabilisers and discretionary fiscal measures cushioned the labour market and social shock (Causa and Hermansen, 2017[84]). The decrease in redistribution was largely driven by reductions in cash transfers, while changes in personal income taxes played a smaller and more heterogeneous role across countries. In particular, the fall in transfer-based redistribution mainly reflected less generous social insurance benefits – especially unemployment-related transfers – though in some countries this was partly offset by more progressive social assistance in the form of means-tested benefits (Causa, Browne and Vindics, 2019[85]).
Another stylised fact highlighted by Causa and Hermansen (2017[84]) is that the redistributive effect of transfers is driven primarily by their overall size and less so by the degree of targeting. Social insurance transfers account for the largest reduction in inequality primarily because they represent the bulk of transfer spending. By contrast, social assistance transfers are more tightly targeted but smaller by design, which limits their aggregate impact on inequality. Nonetheless, they remain crucial for ensuring income adequacy and securing minimum living standards for vulnerable groups (Causa and Hermansen, 2017[84]).
Across OECD countries, cash transfers constitute the main source of redistribution. Personal income taxes play a comparatively larger role in countries where overall redistribution is low – such as Japan, Korea, Israel and the United States – but remain secondary to transfers in most systems. Social security contributions, while generally having a weak redistributive impact, can exert a modest disequalising effect in several countries (Causa and Hermansen, 2017[84]).
The degree of redistribution across income groups can be assessed by examining shares of taxes paid and benefits received by lower-income groups relative to their population shares. Figure 2.11 shows that, on average across OECD countries, roughly one‑third of all cash benefits from public social security accrue to households in the bottom 30% of the income distribution. In several countries – including Australia, Denmark, Finland, the Netherlands, New Zealand and the United Kingdom – this share reaches or exceeds one half, indicating comparatively strong targeting of cash benefits towards low-income households. By contrast, countries located on the right-hand side of the figure display much weaker vertical redistribution towards the 30% bottom income, with around 20% or less of total transfers accruing to this group, as seen in Costa Rica, Greece, Luxembourg, Spain and Portugal. In comparison, cross-country variation in the share of taxes paid by the bottom 30% is much more limited, averaging around 8% of total tax revenues. Overall, redistribution towards the bottom 30% is particularly pronounced in countries on the left-hand side of Figure 2.11, where this group receives more than half of all cash benefits while contributing around 10% or less of total taxes paid.
2022 or latest available year
Note: Data for Germany and the Netherlands refer to 2021, for France to 2020, for Belgium and Iceland to 2017, for Türkiye to 2015 and for Luxemburg to 2014. Benefits received include current transfers received from social security (including accident and disability benefits, old-age and survivor cash benefits, unemployment benefits, maternity allowances, child and/or family allowances, all income‑tested and means-tested benefits that are part of social assistance, including quasi-cash transfers given for a specific purpose such as food stamps) but exclude transfers received from employment-related social insurance, non-profit institutions and other private households (e.g. alimonies). Taxes paid comprise direct taxes on income and wealth paid by households (net of refunds), as well as contributions paid by households to public social security schemes. They exclude contributions paid by households to employment-related social insurance schemes, non-profit institutions and other households.
Source: OECD (n.d.[83]), OECD Income Distribution Database, https://www.oecd.org/en/data/datasets/income-and-wealth-distribution-database.html.
Although households in the bottom 30% of the income distribution remain a key target of income transfers, the share of net transfers to their disposable income has declined since the early 2000s. Figure 2.12 shows that net transfers to households in the bottom 30% of the income distribution are positive overall – meaning they receive more in benefits than they pay in taxes. However, in many countries, the share of income represented by these net transfers has declined over the past two decades, following a period of increase between the mid‑2000s and the mid‑2010s in the aftermath of the 2008 financial crisis.
This pattern is consistent with findings by Causa and Hermansen (2017[84]), who document a reduction in income support from social transfers to workless households in the bottom 40% of the income distribution. According to the authors, this decline has been largely driven by reductions in insurance‑based transfers and only partly offset by increases in means-tested assistance in some countries. By contrast, income support for working households in the bottom 40% has tended to increase, mainly reflecting declines in income taxes and social security contributions, which have only partially compensated for widespread falls in market incomes and reductions in social transfers (Causa and Hermansen, 2017[84]).
Income contribution of net public transfers among the bottom 30% of the income distribution
Note: OECD and EU averages are computed for all respective Member countries with available data across the three years. Data refer to mean equivalised net public transfers received by the bottom 30% divided by the mean equivalised disposable income of households in the bottom 30%. Net public transfers received are calculated as the mean equivalised current transfers received from social security minus the sum of mean equivalised direct taxes paid on income and wealth (net of refunds) as well as contributions paid to public social security schemes.
Source: OECD (n.d.[83]), OECD Income Distribution Database, https://www.oecd.org/en/data/datasets/income-and-wealth-distribution-database.html.
Redistribution levels can vary for various reasons, including family composition and employment across income groups that influence the respective tax and benefits rates that apply, as well as the take‑up level of benefits, and the tax-benefit design parameters themselves. The RED index provides a summary measure specifically of the redistributive design of tax-benefit policies5 for a set of hypothetical household types, in this case low- and middle‑income families in different employment situations (OECD, 2024[86]) (Box 2.5). Increases in the index indicate that changes in tax-benefit policies have strengthened redistribution towards household types with greater financial needs, whereas declines signal that designs have become less redistributive. Figure 2.13 shows that, in most OECD countries, and for families with children, the design of tax-benefit policies has in fact become less redistributive between the early 2000s and 2022. Notable decreases are observed in countries whose policy design is comparatively less redistributive (e.g. Hungary, the Slovak Republic), shows medium levels of redistribution (e.g. Norway, the United Kingdom), but also among countries with traditionally highly-redistributive tax and benefits systems (e.g. Denmark). Only in a small group of countries – including Japan, Ireland, Lithuania, the Netherlands, Latvia, Luxembourg, Spain, Estonia, Greece and Italy – has the redistributive design become more generous toward low-income families since the early 2000s.
Evolution between 2001 and 2022 of the RED index measuring the progressivity of net taxes with growing income
Note: Data in the 2001 series refer to 2005 for Cyprus, Estonia, Latvia, Lithuania, Malta, Slovenia and Türkiye; to 2008 for Bulgaria, Israel and Romania; and to 2013 for Croatia. OECD and EU averages are computed for all respective Member countries with available data for both 2001 and 2022. The RED index theoretically ranges from ‑1 (in the case of fully regressive transfers) to +1 (when net taxes and social contributions are all paid by the highest income groups). All countries display positive values, indicating that the tax and benefit systems for the income and family groups considered are progressive. A value of zero indicates that net transfers are strictly proportional to gross income, implying neither progressivity nor regressivity. Higher positive values reflect greater progressivity, meaning that higher-income households pay higher net taxes as a proportion of their gross income. For more details, see Box 2.6.
Source: OECD Secretariat calculations based on output from the OECD (n.d.[81]) Tax-Benefit Model (version 2.7.1), https://www.oecd.org/en/data/tools/oecd-calculator-of-taxes-and-benefits.html.
The RED index measures the progressivity of taxes and benefits weighted by the volume of redistribution for a hypothetical set of households, following the methodology outlined in the OECD tax-benefit simulation model (OECD, 2024[86]). Social contributions are counted as “taxes” in the computation of the RED index. For every pair of country and time , the formula is.
where
measures the size of redistribution, or the incidence, and is the total of net taxes paid as a share of total gross income of across households .
is the Kakwani index and the formula for the concentration index takes the form with the rank of household being calculated as where represents the total number of households
The RED index yields values between ‑1 and 1, with higher values indicating a more progressive tax schedule. A tax schedule that is exactly proportional to gross income has a RED index of 0 as the concentration of net tax liabilities would be exactly the same as the distribution of gross income, which results in a Kakwani index of 0. A tax schedule that generates larger tax liabilities as a percentage of gross income for higher-income groups will result in a positive RED index. The incidence – or size – weighs the redistributive effect by the volume of redistribution; as a consequence, highly progressive benefits will not change the RED index by a lot if the benefit size is small.
The set of hypothetical households is comprised of observations across 7 types of families, each with 0‑3 children (28 household types in total). The seven family types are:
a single earner working full-time,
a single earner working part time (50%),
a single earner who is unemployed,
a couple where the first earner works full-time and the second earner is employed full-time at 67% of the average wage,
a couple where the first earner works full-time and the second earner is employed part-time (50%) at 67% of the average wage,
a couple where the first earner works full-time and the second earner is employed does not work, and
a couple where both are unemployed
The earnings of the first earner, where they are employed, range from the national minimum wage (or the equivalent of the OECD average where there is none) to 1.5 times the national median equivalent disposable income. Every percentile of the national average wage within that range represents one observation of a hypothetical household. In addition, observations for the two household types where all adults are unemployed are replicated to represent 10% of the other five family types.
Source: OECD (2024[86]), “TaxBEN: The OECD tax-benefit simulation model Methodology, user guide and policy applications”, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/income-support-redistribution-and-work-incentives/OECD-TaxBEN-methodology-and-manual.pdf.
Horizontal redistribution
Horizontal redistribution refers to the reallocation of resources across households with different needs or circumstances, such as household size and the presence of children, with the aim of narrowing living standards gaps for a given level of market income. One way to assess how horizontal redistribution operates is to examine how transfers increase with the number of children, holding households’ gross income constant. Figure 2.14 shows the average additional net transfers received by households with children – relative to otherwise identical childless households – by number of children, expressed as a percentage of gross income. The level of support provided to families with children varies substantially across countries. For example, net transfers to families with children exceed the OECD average for all family sizes in Japan and Lithuania (Figure 2.14, Panel A).
As expected, support increases with family size. On average across countries, households with two children receive about 1.9 times more in net transfers than households with one child, while households with three children receive roughly 3.1 times more. However, cross-country differences are large. In some countries, support for two‑child families is significantly more than double that for one‑child families (notably in Hungary and Slovenia). An even larger group of countries provides substantially higher support to families with three children relative to one‑child families, particularly in Estonia, Latvia, Hungary, Slovenia and Spain, where the ratio exceeds four.
Looking at trends over time, additional support for families with children in most countries peaked in the early 2010s, following the financial crisis. By 2022, levels of support in many countries had returned close to those observed in the mid‑2000s. There are, however, notable exceptions. Estonia, Greece, Korea, Latvia, Lithuania, Japan and Poland have seen significant increases in transfers for families with three children. In contrast, support for two and three children families has declined substantially in Hungary, Ireland and the United Kingdom compared to the mid‑2000s, although in Hungary it remains well above the OECD average. One‑child households have also seen marked increases in Japan, Lithuania, Greece and Korea. For two‑child families, sizeable increases are visible in Lithuania, Japan, Ireland, Israel, Switzerland and France.
Note: This indicator measures the additional benefits (net of taxes and social contributions paid) that a hypothetical set of low- to middle‑income families with 1 child (or 2 or 3 children, respectively) receive compared to the same set of families without any children, as a share of their gross income. The hypothetical set of families comprises single parents who are unemployed, working part-time or full-time as well as couple families where both parents are unemployed and where one parent works full-time and the other parent either does not work, works part-time or full-time. For each of the employment situations, a range of different earnings are considered ranging from the minimum wage to 1.5 times the national median equivalent disposable income. Panel A: The stacked bars represent the additional equivalised transfers that the average hypothetical family receives when a child is added to the family. The sum of the stacked bars shows the additional equivalised transfers compared to a family without any children. The ratios show the additional equivalised benefits that families with 2 (or 3) children received compared to families without children, relative to the additional equivalised benefits that families with 1 child receive compared to families without children.
Source: OECD Secretariat calculations based on output from the OECD (n.d.[81]) Tax-Benefit Model (version 2.7.1), https://www.oecd.org/en/data/tools/oecd-calculator-of-taxes-and-benefits.html.
References
[77] Abrassart, A. and G. Bonoli (2015), “Availability, Cost or Culture? Obstacles to Childcare Services for Low-Income Families”, Journal of Social Policy, Vol. 44/4, pp. 787-806, https://doi.org/10.1017/S0047279415000288.
[14] Adema, W., C. Clarke and O. Thévenon (2020), “Family Policies and Family Outcomes in OECD Countries”, in Neuwenheuwiss, R. and W. Van Lancker (eds.), The Palgrave Handbook of Family Policy, Palgrave Macmillan, https://doi.org/10.1007/978-3-030-54618-2_9.
[51] Baiocco, S. et al. (2022), “Changing Social Investment Strategies in the EU”, JRC Working Papers Series on Labour, education and Technology, No. 1, European Commission - Joint Research Center, https://publications.jrc.ec.europa.eu/repository/handle/JRC127769.
[45] Bakker, V. and O. Van Vliet (2022), “Social Investment, Employment and Policy and Institutional Complementarities: A Comparative Analysis across 26 OECD Countries”, Journal of Social Policy, Vol. 51/4, pp. 728-750, https://doi.org/10.1017/S0047279421000386.
[37] Boarini, R. et al. (2022), “Well-being during the Great Recession: new evidence from a measure of multi-dimensional living standards with heterogeneous preferences*”, The Scandinavian Journal of Economics, Vol. 124/1, pp. 104-138, https://doi.org/10.1111/SJOE.12461.
[26] Bonoli, G., B. Cantillon and W. Lancker (2017), “Social investment and the Matthew effect: Limits to a strategy”, in Hemerijck, A. (ed.), The Uses of Social Investment, Oxford Academic, https://doi.org/10.1093/oso/9780198790488.003.0005.
[18] Bouget, D. et al. (2015), Social investment in Europe – A study of national policies 2015, European Commission: Directorate-General for Employment, Social Affairs and Inclusion, Liser, European Social Observatory, Applica, Publications Office, https://data.europa.eu/doi/10.2767/084978.
[24] Cantillon, B. (2011), “The paradox of the social investment state: Growth, employment and poverty in the Lisbon era”, Journal of European Social Policy, Vol. 21/5, pp. 432-449, https://doi.org/10.1177/0958928711418856.
[33] Cantillon, B. and F. Vandenbroucke (2014), Reconciling Work and Poverty Reduction. How Successful Are European Welfare States?, Oxford University Press, https://ssrn.com/abstract=2989529.
[85] Causa, O., J. Browne and A. Vindics (2019), “Income redistribution across OECD countries: Main findings and policy implications”, OECD Economic Policy Papers, No. 23, OECD Publishing, Paris, https://doi.org/10.1787/3b63e61c-en.
[84] Causa, O. and M. Hermansen (2017), “Income redistribution through taxes and transfers across OECD countries”, OECD Economics Department Working Papers, No. 1453, OECD Publishing, Paris, https://doi.org/10.1787/bc7569c6-en.
[29] Cefalo, R. and R. Scandurra (2023), “What, for whom, and under what circumstances: Do activation policies increase youth employment in the EU?”, Journal of European Social Policy, Vol. 33/4, pp. 391-406, https://doi.org/10.1177/09589287231199568.
[54] Chapple, S. (2013), “Forward liability and welfare reform in New Zealand”, Policy Quarterly, Vol. 9/2, https://doi.org/10.26686/PQ.V9I2.4449.
[52] Council of the European Union (2024), Council conclusions on the role of the labour market, skills and social policies for resilient economies, https://www.consilium.europa.eu/en/meetings/epsco/2024/06/20/ (accessed on 1 December 2025).
[10] Crowe, D. et al. (2022), “Population ageing and government revenue: Expected trends and policy considerations to boost revenue”, OECD Economics Department Working Papers, No. 1737, OECD Publishing, Paris, https://doi.org/10.1787/9ce9e8e3-en.
[42] Cunha, F. and J. Heckman (2007), “The technology of skill formation”, American Economic Review, Vol. 97/2, pp. 31-47, https://doi.org/10.1257/aer.97.2.31.
[48] Daly, M. (ed.) (2025), Families, Welfare States and Resilience – Low-Resource Families Navigating Care, Employment and Welfare in Europe, Elgar Online, https://doi.org/10.4337/9781035346769.
[6] Daly, M. and S. Ryu (2025), “Child-related Social Policies in Europe during the COVID-19 Pandemic”, Journal of Social Policy, Vol. 54/3, pp. 894-911, https://doi.org/10.1017/S0047279423000351.
[39] De Schutter, O. et al. (2023), The Escape from Poverty: Breaking the Vicious Cycles Perpetuating Disadvantage, Policy Press, Bristol, https://doi.org/10.51952/9781447370611.
[35] Deeming, C. and P. Smyth (2015), “Social Investment after Neoliberalism: Policy Paradigms and Political Platforms”, Journal of social policy, Vol. 44/2, pp. 297-318, https://doi.org/10.1017/S0047279414000828.
[5] Dely, B., R. Hyee and C. Prinz (2025), “What worked well in social protection during the COVID-19 pandemic?”, OECD Social, Employment and Migration Working Papers, No. 325, OECD Publishing, Paris, https://doi.org/10.1787/b31ac155-en.
[71] Duncan, G. et al. (2022), “Investing in Early Childhood Development in Preschool and at Home”, NBER Working Paper Series, No. 29985, https://doi.org/10.3386/w29985.
[38] Esping-Andersen, G. (2005), “Education and Equal Life-Chances: Investing in Children”, in Kangas, O. and J. Palme (eds.), Social Policy and Economic Development in the Nordic Countries. Social Policy in a Development Context, Palgrave Macmillan, London, https://doi.org/10.1057/9780230523500_6.
[21] Esping-Andersen, G. (ed.) (2002), Why We Need a New Welfare State, Oxford Academic, https://doi.org/10.1093/0199256438.001.0001.
[50] European Commission (2021), “European Pillar of Social Rights Action Plan”, https://employment-social-affairs.ec.europa.eu/policies-and-activities/european-pillar-social-rights-building-fairer-and-more-inclusive-european-union/european-pillar-social-rights-action-plan_en (accessed on 1 December 2025).
[49] European Commission (2013), Towards Social Investment for Growth and Cohesion – including implementing the European Social Fund 2014-2020, Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52013DC0083 (accessed on 7 September 2026).
[79] Farchy, E. and H. Immervoll (2021), “Faces of joblessness in Finland: A people-centred perspective on employment barriers and policies”, OECD Social, Employment and Migration Working Papers, No. 258, OECD Publishing, Paris, https://doi.org/10.1787/ca3bc4a4-en.
[78] Fernandez, R. et al. (2016), “Faces of joblessness: Characterising employment barriers to inform policy”, OECD Social, Employment and Migration Working Papers, No. 192, OECD Publishing, Paris, https://doi.org/10.1787/5jlwvz47xptj-en.
[36] Fougère, D. and A. Heim (2019), “L’évaluation socioéconomique de l’investissement social : Comment mettre en oeuvre des analyses coûts-bénéfices pour les politiques d’emploi, de santé et d’éducation”, Document de travail, No. 6, France Stratégie, https://www.strategie-plan.gouv.fr/files/files/Publications/2019/dt-investissement_social-fougere-heim-19novembre-final.pdf.
[20] Garritzmann, J., S. Häusermann and B. Palier (eds.) (2022), The World Politics of Social Investment : Volume I Knowledge Economy, Oxford University Press, London, https://global.oup.com/academic/product/the-world-politics-of-social-investment-volume-i-9780197585245.
[27] Gingrich, J. and B. Ansell (2015), “The Dynamics of Social Investment: Human Capital, Activation, and Care”, The Politics of Advanced Capitalism, pp. 282-304, https://doi.org/10.1017/CBO9781316163245.012.
[56] Gluckman, P. (2017), Enhancing evidence-informed policy making, Office of the Prime Minister’s Chief Science Advisor, New Zealand, https://www.dpmc.govt.nz/sites/default/files/2021-10/pmcsa-17-07-07-Enhancing-evidence-informed-policy-making.pdf (accessed on 1 December 2025).
[58] Government of New Zealand (2025), Putting the Strategy into action - Taking a social investment approach, Ministry of Social Development, https://www.msd.govt.nz/about-msd-and-our-work/child-wellbeing-and-poverty-reduction/child-youth-strategy/putting-the-strategy-into-action.html (accessed on 12 September 2025).
[57] Government of New Zealand (2019), The Wellbeing Budget, https://www.treasury.govt.nz/sites/default/files/2019-06/b19-wellbeing-budget.pdf (accessed on 29 January 2026).
[53] Government of New Zealand (2013), The Investment Approach - 2013 Benefit System Performance Report, Ministry of Social Development, https://www.msd.govt.nz/about-msd-and-our-work/publications-resources/evaluation/investment-approach/2013-index.html (accessed on 1 December 2025).
[12] Guillemette, Y. and D. Turner (2021), “The long game: Fiscal outlooks to 2060 underline need for structural reform”, OECD Economic Policy Papers, No. 29, OECD Publishing, Paris, https://doi.org/10.1787/a112307e-en.
[8] Haelg, F., N. Potrafke and J. Sturm (2022), “The determinants of social expenditures in OECD countries”, Public Choice, Vol. 193/3-4, pp. 233-261, https://doi.org/10.1007/S11127-022-00984-4/TABLES/5.
[41] Heckman, J., R. Pinto and P. Savelyev (2013), “Understanding the Mechanisms through Which an Influential Early Childhood Program Boosted Adult Outcomes”, American Economic Review, Vol. 103/6, pp. 2052-86, https://doi.org/10.1257/AER.103.6.2052.
[30] Hemerijck, A. (2017), “Social Investment and Its Critics”, in Hemerijck, A. (ed.), The Uses of Social Investment, Oxford University Press, https://doi.org/10.1093/oso/9780198790488.003.0001.
[23] Hemerijck, A. (ed.) (2017), The Uses of Social Investment, Oxford University Press, https://doi.org/10.1093/OSO/9780198790488.001.0001.
[19] Hemerijck, A., S. Ronchi and I. Plavgo (2023), “Social investment as a conceptual framework for analysing well-being returns and reforms in 21st century welfare states”, Socio-Economic Review, Vol. 21/1, pp. 479-500, https://doi.org/10.1093/ser/mwac035.
[34] Jenson, J. (2010), “Diffusing Ideas for After Neoliberalism: The Social Investment Perspective in Europe and Latin America”, Global Social Policy, Vol. 10/1, pp. 59-84, https://doi.org/10.1177/1468018109354813.
[28] Knize, V., M. Wolf and C. Zabel (2021), “Do Disadvantaged Young Adults Benefit Less from Social Investment? Studying Matthew Effects in Activation Measures”, Youth and Globalization, Vol. 3/1, pp. 13-55, https://doi.org/10.1163/25895745-03010002.
[9] Koutsogeorgopoulou, V. and H. Morgavi (2025), “Ageing populations, their fiscal implications and policy responses”, OECD Economics Department Working Papers, No. 1844, OECD Publishing, Paris, https://doi.org/10.1787/6aec03b3-en.
[69] Little, M. et al. (2021), “Effectiveness of cash-plus programmes on early childhood outcomes compared to cash transfers alone: A systematic review and meta-analysis in low- and middle-income countries”, PLOS Medicine, Vol. 18/9, p. e1003698, https://doi.org/10.1371/JOURNAL.PMED.1003698.
[80] Martin, J. (2015), “Activation and active labour market policies in OECD countries: stylised facts and evidence on their effectiveness”, IZA Journal of Labor Policy 2015 4:1, Vol. 4/1, pp. 4-, https://doi.org/10.1186/S40173-015-0032-Y.
[15] Miho, A. and O. Thévenon (2020), “Treating all children equally?: Why policies should adapt to evolving family living arrangements”, OECD Social, Employment and Migration Working Papers, No. 240, OECD Publishing, Paris, https://doi.org/10.1787/83307d97-en.
[25] Morel, N., B. Palier and J. Palme (eds.) (2012), Towards a social investment welfare state? Ideas, policies and challenges, Bristol University Press, https://doi.org/10.2307/J.CTT9QGQFG.
[47] Nieuwenhuis, R. (2022), “No activation without reconciliation? The interplay between ALMP and ECEC in relation to women’s employment, unemployment and inactivity in 30 OECD countries, 1985-2018”, Social Policy and Administration, Vol. 56/5, pp. 808-826, https://doi.org/10.1111/spol.12806.
[68] Nightingale, M. (2025), Filling in the knowledge gaps and identifying strengths and challenges in the effectiveness of the EU Member States’ child benefits, RAND Europe.
[70] OECD (2025), Reducing Inequalities by Investing in Early Childhood Education and Care, Starting Strong, OECD Publishing, Paris, https://doi.org/10.1787/b78f8b25-en.
[55] OECD (2024), New Zealand’s Integrated Data Infrastructure: Using data to identify need and improve the well-being of children and youth, OECD Publishing, Paris, https://www.oecd.org/en/publications/well-being-knowledge-exchange-platform-kep_93d45d63-en/new-zealand-s-integrated-data-infrastructure-using-data-to-identify-need-and-improve-the-well-being-of-children-and-youth_5088d51e-en.html.
[86] OECD (2024), TaxBEN: The OECD tax-benefit simulation model Methodology, user guide and policy applications, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/income-support-redistribution-and-work-incentives/OECD-TaxBEN-methodology-and-manual.pdf (accessed on 11 December 2025).
[13] OECD (2023), Exploring Norway’s Fertility, Work, and Family Policy Trends, OECD Publishing, Paris, https://doi.org/10.1787/f0c7bddf-en.
[82] OECD (2023), “Living wages in practice”, OECD Policy Insights on Well-being, Inclusion and Equal Opportunity, No. 8, OECD Publishing, Paris, https://doi.org/10.1787/699b3f9b-en.
[59] OECD (2023), New Zealand’s Child and Youth Wellbeing Strategy, OECD Publishing, Paris, https://www.oecd.org/en/publications/well-being-knowledge-exchange-platform-kep_93d45d63-en/new-zealand-s-child-and-youth-wellbeing-strategy_536ab28d-en.html.
[87] OECD (2023), On Shaky Ground? Income Instability and Economic Insecurity in Europe, OECD Publishing, Paris, https://doi.org/10.1787/9bffeba6-en.
[2] OECD (2023), The rise and fall of public social spending with the COVID-19 pandemic, OECD Publishing, Paris, https://doi.org/10.1787/12563432-en.
[43] OECD (2021), Measuring What Matters for Child Well-being and Policies, OECD Publishing, Paris, https://doi.org/10.1787/e82fded1-en.
[74] OECD (2020), “Is Childcare Affordable?”, Policy Brief on Employment, Labour and Social Affairs, OECD, Paris, https://web-archive.oecd.org/pdfViewer?path=/2020-06-05/554683-OECD-Is-Childcare-Affordable.pdf (accessed on 6 January 2021).
[22] OECD (2018), Opportunities for All: A Framework for Policy Action on Inclusive Growth, OECD Publishing, Paris, https://doi.org/10.1787/9789264301665-en.
[76] OECD (2018), Poor children in rich countries: why we need policy action, OECD, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/child-and-family-well-being/poor-children-in-rich-countries-policy-brief-2018.pdf (accessed on 3 September 2026).
[11] OECD (2017), Preventing Ageing Unequally, OECD Publishing, Paris, https://doi.org/10.1787/9789264279087-en.
[63] OECD (2011), Doing Better for Families, OECD Publishing, Paris, https://doi.org/10.1787/9789264098732-en.
[40] OECD (2009), Doing Better for Children, OECD Publishing, Paris, https://doi.org/10.1787/9789264059344-en.
[81] OECD (n.d.), Calculator of taxes and benefits, https://www.oecd.org/en/data/tools/oecd-calculator-of-taxes-and-benefits.html.
[64] OECD (n.d.), OECD Education Database, https://data-explorer.oecd.org/s/5ca (accessed on 14 September 2026).
[61] OECD (n.d.), OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html (accessed on 14 September 2026).
[83] OECD (n.d.), OECD Income Distribution Database, https://www.oecd.org/en/data/datasets/income-and-wealth-distribution-database.htm (accessed on 14 September 2026).
[4] OECD (n.d.), OECD National Accounts at a Glance, https://data-explorer.oecd.org/s/46v (accessed on 14 September 2026).
[3] OECD (n.d.), OECD Social Expenditure (database), https://data-explorer.oecd.org/s/46u (accessed on 14 September 2026).
[1] OECD (n.d.), OECD Social Expenditure Database (SOCX), https://www.oecd.org/en/data/datasets/social-expenditure-database-socx.html (accessed on 14 September 2026).
[67] OECD (n.d.), “PF1.3 Family cash benefits”, OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html (accessed on 21 November 2025).
[62] OECD (n.d.), “PF1.6 Public spending by age of children”, OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html (accessed on 24 November 2020).
[65] OECD (n.d.), “PF2.5 Trends in leave entitlements around childbirth since 1970”, OECD Family Database, https://www.oecd.org/en/data/datasets/oecd-family-database.html (accessed on 7 September 2026).
[72] Pavolini, E. and W. Van Lancker (2018), “The Matthew effect in childcare use: a matter of policies or preferences?”, Journal of European Public Policy, Vol. 25/6, pp. 878-93, https://doi.org/10.1080/13501763.2017.1401108.
[16] Rapp, S. and O. Thévenon (2025), “Parenting on a Budget: The Cost of Raising Children in changing family configurations and Its Policy Implications”, OECD Papers on Well-being and Inequalities, No. 36, OECD Publishing, Paris, https://doi.org/10.1787/bd0b594e-en.
[17] Riding, S. et al. (2021), “Looking beyond COVID-19: Strengthening family support services across the OECD”, OECD Social, Employment and Migration Working Papers, No. 260, OECD Publishing, Paris, https://doi.org/10.1787/86738ab2-en.
[46] Sakamoto, T. (2024), “Do social investment policies promote employment among the vulnerable? A case of single mothers”, International journal of sociology and social policy, Vol. 44/5-6, pp. 607-627, https://doi.org/10.1108/IJSSP-08-2023-0194.
[44] Sakamoto, T. (2023), “Poverty, inequality, and redistribution: An analysis of the equalizing effects of social investment policy”, International Journal of Comparative Sociology, Vol. 65/3, pp. 310-334, https://doi.org/10.1177/00207152231185282.
[32] Taylor-Gooby, P., J. Gumy and A. Otto (2015), “Can ‘New Welfare’ Address Poverty through More and Better Jobs?”, Journal of Social Policy, Vol. 44/1, pp. 83-104, https://doi.org/10.1017/S0047279414000403.
[66] Thévenon, O. (2018), “Leave policies for parents in a cross-national perspective: various paths along the same course?”, in Eydal, G. and T. Rostgaard (eds.), Handbook of Family Policy, Edward Elgar Publishing, https://doi.org/10.4337/9781784719340.00018.
[60] Thévenon, O. (2011), “Family Policies in OECD Countries: A Comparative Analysis”, Population and Development Review, Vol. 37/1, https://doi.org/10.1111/j.1728-4457.2011.00390.x.
[7] Thévenon, O., C. Clarke and G. Simard-Duplain (2024), “The economic cost of childhood socio-economic disadvantage in Canada”, OECD Papers on Well-being and Inequalities, No. 25, OECD Publishing, Paris, https://doi.org/10.1787/7abf1837-en.
[73] Van Lancker, W. (2023), “The Matthew Effect in Early Childhood Education and Care: How family policies may amplify inequalities”, in The Oxford Handbook of Family Policy: A Life-Course Perspective, Oxford University Press, Oxford, https://doi.org/10.1093/oxfordhb/9780197518151.013.36.
[31] Vandenbroucke, F. and K. Vleminckx (2011), “Disappointing poverty trends: Is the social investment state to blame?”, Journal of European Social Policy, Vol. 21/5, pp. 450-471, https://doi.org/10.1177/0958928711418857.
[75] Wood, J., K. Neels and J. Maes (2023), “A closer look at demand-side explanations for the Matthew effect in formal childcare uptake in Europe and Australia”, Journal of European Social Policy, Vol. 33/4, pp. 451-468, https://doi.org/10.1177/09589287231186068.
Note: Data are expressed as a percentage share of the median equivalised disposable income in a given year. Transfers to the average poor household are computed by comparing the average equivalised household income of individuals below the relative poverty line before and after income transfers. The anchored poverty line displays the inflation-adjusted relative poverty line from 2005. The minimum wage represents the theoretical earnings of a full-time employee under the national minimum wage legislation, averaged across countries where such legislation exists.
Source: OECD Secretariat calculations based on OECD (n.d.[83]), OECD Income Distribution Database, https://www.oecd.org/en/data/datasets/income-and-wealth-distribution-database.html.
← 1. At the same time, substantial variation remains within each cluster, both in terms of spending levels and policy design, suggesting that fully developed and coherent social investment strategies have not yet consolidated into clearly defined national models.
← 2. Per capita refers to the population as a whole. The amounts are not average per-child expenditures.
← 3. Income instability is another important dimension of in-work poverty (OECD, 2023[87]). Across European OECD countries, for example, only around 20% of individuals in working-age households experienced sustained income growth of at least 25% over a four‑year period. For most people, income trajectories were instead characterised by either volatility or decline. Such instability is concentrated among groups already at greater risk of poverty, including unemployed individuals, workers in temporary or insecure employment, and members of single‑earner or younger households. Financial resilience is often limited: around two‑thirds of people in lower-income working-age households lack sufficient liquid assets to remain above the poverty line for more than three months in the event of an income shock.
← 4. For instance, fiscal instruments and benefit payments are also designed to support economic growth by creating incentives for education, employment, innovation and risk-taking, as well as by financing public investment. In addition, they play a central role in promoting macroeconomic stability by sustaining demand over the business cycle and cushioning the impact of economic shocks.
← 5. In addition to taxes and benefits, social contributions are also considered in the computation of the RED index. Although not always mentioned explicitly, they enter as part of households’ “taxes”.