Report No: ACS13024 WHY SHOULD WE CARE ABOUT CARE? THE ROLE OF INFORMAL CHILDCARE AND ELDERCARE IN AGING SOCIETIES . The World Bank Group 1 Why should we care about care? The Role of Informal Childcare and Eldercare in Aging Societies in the ECA Region . Standard Disclaimer: . This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. . Copyright Statement: . The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978- 750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. 2 ACKNOWLEDMENTS This report was authored by a World Bank team including Victoria Levin, Ana Maria Munoz-Boudet and Beth Zikronah Rosen, with substantive inputs from Julianna Flanagan and Tami Aritomi. The report draws on background work produced by Tami Aritomi (care projections), Isabella Buber-Ennser (Grandparental childcare in Europe) and Marco Albertini (Intergenerational support in Europe). The team received valuable inputs and comments from Elena Glinskaya, Alessandra Marini, Veronica Silva and Emily Sinnott. 3 Contents I. Motivation: Why should we care about care? ................................................................................... 7 II. How is care organized in ECA? Formal and informal supply ........................................................ 15 Social norms and views....................................................................................................................... 21 Social policies and supply of formal care ........................................................................................... 25 III. Women are the main informal care providers in ECA ................................................................ 35 The role of grandparents as childcare providers ................................................................................. 39 Eldercare and upward generational flows of care ............................................................................... 42 Caught in the middle: Simultaneous intergenerational flows of care ................................................. 47 IV. How does caregiving affect women at different stages of the life cycle? ................................... 49 Caregiving and labor force attachment: childcare .............................................................................. 49 Caregiving and labor force attachment: eldercare .............................................................................. 56 Caregiving and income ....................................................................................................................... 60 V. Policy discussion ............................................................................................................................. 62 Providing accessible, affordable, quality formal care services ........................................................... 62 Acknowledging and improving informal care .................................................................................... 67 Flexible work arrangements ................................................................................................................ 69 Allowances and tax incentives ............................................................................................................ 70 Conclusion .......................................................................................................................................... 73 References ............................................................................................................................................... 75 4 Figures Figure 1: Projections of demand for childcare and eldercare in ECA, 2010–2060 .................................... 10 Figure 2: School Enrolment, pre-primary (gross % of relevant age groups) .............................................. 18 Figure 3: Children in formal childcare by age group (children with at least 1 hour of formal care per week as % of the population in the age group) .................................................................................................... 18 Figure 4: Percent of households with at least one child under 7 years who use institutional, paid at home, and unpaid childcare ................................................................................................................................... 19 Figure 5: Share of adults 50+ years with self-reported limitation in activities of daily living that receive care .............................................................................................................................................................. 20 Figure 6: Percent of households with at least one elder 65+ years who use institutional, paid at home, and unpaid eldercare .......................................................................................................................................... 20 Figure 7: Agreement with the following statements by respondents’ age and country .............................. 22 Figure 8: Percentage of respondents who think that the state (as opposed to the family) should bear responsibility for older persons who are in need, in each of the following categories ............................... 24 Figure 9: Opinions about state involvement in provision of childcare and eldercare ................................. 24 Figure 10: Assessment of childcare policies in selected ECA countries, 2007 .......................................... 26 Figure 11: Availability of places at care facilities in selected ECA countries, 2014 .................................. 27 Figure 12: Length of parental leave and effective parental leave in selected European countries ............. 29 Figure 13: Affordability of care facilities in selected ECA countries, 2014 ............................................... 29 Figure 14: Quality of care services in selected ECA countries, 2014 ......................................................... 31 Figure 15: Distribution of source of care for those with severe and moderate limitations by living arrangement (self-reported health) .............................................................................................................. 32 Figure 16: Correlates of using institutional/paid childcare arrangements in selected ECA and EU- 15/EFTA countries ...................................................................................................................................... 33 Figure 17: Proportion of working age population (aged 20–64 years) undertaking activities by gender and time of day .................................................................................................................................................. 36 Figure 18: Proportion of working age population (aged 20–64 years) engaging in work by gender and time of day .................................................................................................................................................. 36 Figure 19: Average daily minutes spent on childcare, by sex, country, and age of the youngest child ...... 37 Figure 20: Care flows throughout women’s life cycles .............................................................................. 37 Figure 21: Participation and intensity of social support.............................................................................. 38 Figure 22: Co-residence patterns for adults over 50 years of age ............................................................... 39 Figure 23: Grandmothers and grandfathers who provided any childcare by country (in percent) ............. 40 Figure 24: Estimated country coefficients for grandmothers’ care provision (any and regular grandchild care) ............................................................................................................................................................ 42 Figure 25: Care provided to parents and/or parents-in-law among women who report providing help to someone in the previous year. Female respondents, by age ....................................................................... 43 Figure 26: Difference of demographic and household characteristics between respondents who are non- elder caregivers and elder caregivers .......................................................................................................... 44 Figure 27: Participation of children in social support to SHARE respondents ........................................... 46 Figure 28: Intensity of social support provided to SHARE respondents by participating children (hours) 46 5 Figure 29: Multiple tasks of grandparents by age and gender (in percent) ................................................. 48 Figure 30: Multiple tasks of women aged 50–59 years with young grandchildren by country (in percent) .................................................................................................................................................................... 48 Figure 31: Employment rate of mothers by age of their youngest child ..................................................... 50 Figure 32: Career interruptions of over 12 months due to childcare, by education .................................... 51 Figure 33: Working mothers as clients of childcare facilities..................................................................... 53 Figure 34: Childcare-related reasons for reduced labor supply .................................................................. 54 Figure 35: Labor force status for mothers with and without informal care ................................................ 55 Figure 36: Employment rate of older individuals with/without grandchildren, by sex and age ................. 56 Figure 37: Formal eldercare utilization and middle-aged female labor force participation ........................ 58 Figure 38: Working daughters as clients of eldercare facilities (% of all clients) ...................................... 59 Figure 39: Early childcare gap in the EU, 2003–2007 ................................................................................ 64 Figure 40: Free ECEC provision in the EU, by age and weekly hours, 2012/13 ........................................ 65 Figure 41: Range of monthly allowances to care recipients or providers ................................................... 71 Tables Table 1: Summary of data sources by ECA countries ................................................................................ 13 Table 2: Typologies of care ........................................................................................................................ 15 Table 3: Typology based on use and financing of care ............................................................................... 16 Table 4: Number of care facilities in the field data sample ........................................................................ 27 Table 5: Summary of regulatory environment around childcare in selected countries ............................... 66 Table 6: Percent of facilities that are formally accredited .......................................................................... 67 Table 7: Availability of cash benefits ......................................................................................................... 72 6 I. Motivation: Why should we care about care? 1. Attainment of greater longevity by many European societies implies a longer period of multi- generational coexistence. The combined forces of lower fertility and higher life expectancy drive population aging in European societies. The increase in longevity implies more years of shared living between generations. This phenomenon provides an opportunity for people of different generations to interact with and learn from each other, but it also presupposes a rising importance of multi-generational bonds in the provision of social support and care for the old and the young (Bengston 2001; Hank 2007; Brandt et al. 2009). 2. The exchange of social and economic support between the generations is important from both micro and macro perspectives. Within the family, parents generally support their children not only while they grow up but also when they have become independent and left the parental home, whereas grown-up children often support their parents when they have become disabled and need help. Within the welfare system, those of working age provide the economic resources both for the young, in terms of family support and schooling, and for the old, in terms of pension and health care financing. This may be conceptualized not only as redistribution among age groups, but it also refers to the succession of generations. Today’s children will provide for the welfare of their parents’ generation when the latter has become old, and in exchange, their parents will provide care for their grandchildren (Hank and Buber 2009; Silverstein et al. 2006; Grundy and Henretta 2006). 3. The demand for informal elder care in most countries in Eastern Europe and Central Asia (ECA) is expected to increase due to larger older cohorts and longer periods of care needs. As the age profile of many ECA countries changes from the traditional pyramid to a ‘mushroom’ with a distinct bulge at older ages, there will be more people in each of the older age groups who may require some help with activities of daily living. As can be observed in the projections of care demand in Figure 1a, the number of adults with limitations in activities of daily living (ADL) per adult who is healthy (that is, without ADL limitations) is projected to rise significantly from 2010 to 2060. Combined with a slightly decreasing trend in the demand for childcare (based purely on the United Nations [UN] demographic projections for children aged 0 to 5 years), Figure 1b demonstrates that the increase in the aggregate demand for care in an average ECA country stems largely from the increase in moderate rather than severe disability. This implies a rising demand for informal care, as the existing systems of formal care provision, which are largely focused on health care for the severely disabled, will see less of an increase in its already oversubscribed capacity. In most countries, the formal long-term care system for elderly with severe ADL limitations are already experiencing excess demand, resulting in queuing and inefficient use of hospital stays for long-term care needs among individuals that do not require such intensive and intrusive care and could be cared for either in a special eldercare facility (non-medical), or at home, given adequate support. In addition, if growth in life expectancy outpaces growth in health outcomes, individuals may spend more years of life in frail health and may need longer spans of non-medical support with daily activities, such as shopping, housekeeping, food preparation, and/or handling medications and finances, even if they do not require specialized medical help.1 Such formal help systems do not exist in the majority of ECA countries, where the next of kin of the 1 This includes elderly experiencing age-related mental health and emotional well-being problems, such as dementia, Alzheimer’s disease, and other type of disorders affecting memory, thinking, behavior and ability to perform everyday activities (WHO 2013). 7 elderly largely absorbs the responsibility. Thus, without any policy adjustment, we can expect to see increasing demand for informal care, especially in some of the ECA countries where the age pyramid dynamics combine with a high rate of disability in certain age groups, such as countries in the Western Balkans (Serbia, the former Yugoslav Republic [FYR] of Macedonia, and Bosnia and Herzegovina) as well as Armenia and Slovakia (Figure 1c). 4. The extent to which eldercare and childcare needs are met by formal providers is shaped by normative, institutional, and labor market factors. Caregiving decisions are affected by normative views about filial obligations and intergenerational solidarity (Finley 1988; Gans and Silverstein 2006). Preferences for informal care and at-home care shape the potential expansion of formal care services and increased expenditure on such services.2 The success of policy initiatives in maintaining or enhancing the supply of formal care requires an understanding of the determinants of the use of such care. Institutional factors, such as the availability, accessibility, affordability, and quality of alternative care services to the informal care provided at home factor into families’ decision making around the use of one type of care over the other. Moreover, the weights assigned to different objectives fulfilled by formal care systems — such as ensuring educational readiness of children, health and dignity of elders, and/or promoting women’s labor force participation—can affect the shape of resulting care institutions. In many countries across ECA, the supply of appropriate care services remains thin, often outstripping demand (as will be discussed below). Finally, household composition and income-earning opportunities for household members figure into families’ decisions about the use of formal care services. Household members with limited earning potential may add to their current labor market participation earnings disincentives, the care needs their households have, and prefer not to invest in formal care provision, especially if the cost of these providers becomes prohibitive compared to their expected labor income. These factors in combination result in a low use of formal care providers in the region, as households rely heavily on informal care providers instead. 5. The demand for childcare services for children aged 0–5 years will remain significant as countries aim to reverse fertility declines and encourage higher female labor force participation and productivity. Although pure demographic projections (such Figure 1a) suggest the shift in the structure of care demand from children to the elderly, in reality pressure will continue to build for the provision of accessible, affordable, and quality childcare services. This can be explained by several factors. First, there is the existing unfulfilled demand for these services in many ECA countries (for example, the Russian Federation) caused, at least in part, by relative neglect of these institutions during the fertility decline of the 1990s transition period. Second, there is mounting evidence that increases in fertility in aging European societies can be achieved only by providing women with a favorable environment for combining work and motherhood (Luci-Greulich and Thevenon 2013). Thus, the availability of accessible, affordable, and quality childcare options are an essential ingredient of successful pro-natalist policy design both in terms of their push for women to have more children and to start the process earlier than what the trend has been to date. In addition, to sustain economic growth in the context of aging, even without a push for increased fertility, aging societies need to invest in the skills of their young population, which starts with early childhood education replacing or at least complementing informal provision of childcare. 2 And of home care as a potential intermediate arrangement (see for example, Tarricone and Tsouros 2008). 8 6. Care recipients’ preferences and fiscal sustainability analyses tend to align in favor of informal care provision. There is a vast literature demonstrating the relationship between informal support and the economic, physical, psychological and social well-being, and social status of the recipients of this support (Albertini 2014). Children benefit from their parents’ care starting before their birth and well beyond their entry into adulthood: the quantity and quality of parental support is correlated with educational opportunities and labor market outcomes, as well as such correlates of well-being as access to housing and credit and geographical mobility. Moreover, through provision of informal grandparental childcare, parents can enable their children to remain in the labor market, which can also affect children’s fertility decisions. For the elderly, receiving informal support enables healthier and more active aging, lower risk of social isolation, and delayed entry into institutionalized care facilities (Albertini 2014). Informal support from children is also necessary to connect frail parents to appropriate formal care services (Choi 1994; Litwin 1997). It is thus not surprising that across different country contexts, the majority of elderly prefer receiving care in their homes or in community-based settings (Colombo et al. 2011; Keenan 2010). At the same time, the realization of the potential impact of population aging on the fiscal sustainability of health care and Long-Term Care (LTC) systems has resulted in a shift away from reliance on expensive institutionalized care and toward more decentralized, privately provided care. In most developing countries, where formal care systems for the elderly are still exclusively focused on institutionalized care, these preferences of care recipients and policy trends translate into a greater demand for informal, mostly family-provided care. 7. There is relatively scarce recognition of the tradeoffs and risks faced by providers of informal care. The importance of ensuring that elders can live their lives in dignity and that children are in an environment conducive to development should not be underestimated. Yet, the second imperative to care about care—the well-being of caregivers—has been largely overlooked in the economics literature and in policy discussions on the relative effectiveness of various eldercare and childcare models, including those regarding early childhood education and care and long-term care. Indeed, Albertini (2014) finds relatively little comparative research about the effects of caregiving on the well-being and life chances of caregivers. He emphasizes the importance of adopting the ‘donor’s perspective’ in analyzing the consequences of intergenerational support, especially as “help with personal care and household chores is time-consuming, physically demanding and, most importantly, usually is highly needed at a point of the donor’s life course in which her working and career maturity is at its peak” (Albertini 2014). The current paper aims to fill this important gap in the literature and to assess the state of policy discussion on care systems from the perspective of caregiver well-being. 9 Figure 1: Projections of demand for childcare and eldercare in ECA, 2010 –2060 a) Care demand by type, average ECA country b) Aggregate care demand, average ECA country 40.0% 40.0% 35.0% 35.0% 30.0% 30.0% 25.0% 25.0% 20.0% 20.0% 15.0% 15.0% 10.0% 10.0% 5.0% 5.0% 0.0% 0.0% Adults (20+) with severe ADL limitations / Healthy adults (20+) Aggregate care burden 1 (children + severe ADL limitations) / healthy adults Adults (20+) with moderate or severe ADL Aggregate care burden 2 (children + severe & limitations / Healthy adults (20+) moderate ADL limitations) / healthy adults Children (0-5) / Healthy adults (20+) c) Change in aggregate care demand, by country 15% % point change (2010-2060) 10% 5% 0% TUR TKM MNE HUN AZE BIH RUS MKD UZB BGR GEO UKR LTU ARM ALB LVA EST CZE SVN SVK KGZ BLR SRB MDA TJK KAZ ROU HRV POL -5% -10% -15% -20% Aggregate care burden 1 (children + severe ADL limitations)/healthy adults Aggregate care burden 2 (children + severe & moderate ADL limitations)/healthy adults Source: Authors’ calculations based on European Health Interview Survey (EHIS) and Russia Longitudinal Monitoring Survey (RLMS) data on the age profile of ADL limitations and UNPP data on population projections. Note: (i) The analysis assumes that the age-sex disability profile remains stable from 2008 (EHIS) or 2005 (RLMS) to 2060, and thus the changes in the care burden over time are driven purely by demographics. (ii) Unweighted average of projections for Albania *, Armenia+, Azerbaijan+, Bulgaria, Bosnia and Herzegovina*, Belarus+, the Czech Republic, Estonia#, Georgia+, Croatia*, Hungary, Kazakhstan+, the Kyrgyz Republic+, Lithuania#, Latvia, Moldova+, FYR Macedonia*, Montenegro*, Poland, Romania, the Russian Federation, Serbia*, Slovakia, Slovenia, Tajikistan+, Turkmenistan+, Ukraine+, Uzbekistan+. More details on the methodology behind these projections is available in the Annex on Major Data Sources for this paper. * Age profile of ADL limitations imputed from Slovenia. + Age profile of ADL limitations imputed from the Russian Federation. # Age profile of ADL limitations imputed from Latvia. 10 8. Within families, the burden of increasing informal and at-home care falls disproportionately on women of all ages. Similar normative and household factors that shape the balance between formal and informal care provision also affect the gender division of care duties within families. It is well documented that childcare duties fall disproportionately on women in all of the ECA countries as well as in the rest of the world. For the case of eldercare, while filial obligation on the part of the child might rest equally on daughters and sons, those more likely to act upon it are daughters and daughters-in-law. In an expanded generational view, as mothers are expected to be the main childcare provider, grandmothers are often expected to provide care for grandchildren when mothers need support. Ciani (2012) suggests that the selection and decision on who is to provide informal care to the elderly is often made before the need arises. For example, children who opt for geographical proximity to parents before they retire might be opting in for eldercare and deciding to forgo better labor opportunities that will require moving further from them.3 Patterns of elderly relocation (Litwak and Longino 1987; Speare et al. 1991; Smits et al. 2010) also suggest that emerging care needs are one of the factors for relocating closer to those who can provide informal care or to cohabitation with an adult child, particularly in the absence of public provision of home care or long- term care options (Pezzin et al. 1996). Finally, women’s higher life expectancy, as well as their oftentimes lower labor market attachment and earning potential (partly as a result of their childcare role), frequently results in their higher propensity to become caregivers at one point or another in the life cycle. 9. The burden of eldercare and childcare responsibilities can have serious negative repercussions on women’s economic outcomes, leading to increased vulnerability and exacerbating gender-based inequalities. As women spend more time engaging in unpaid, informal care work, they have less time to work in the market. Studies looking at the relationship between caregiving and labor market outcomes show negative impacts both on the extensive and intensive margins and reduced human capital accumulation (Becker 1985; Behrman and Wolfe 1984; Ribar 1995; Jaumotte 2003). There is also evidence that caregivers receive lower wages, further discouraging labor force participation (Correll et al. 2007; Carmichael and Charles 1998, 2003; Heitmueller and Inglis 2007). Together, these may contribute to reduced lifetime earnings for caregivers, leading to a disadvantageous position in terms of financial status, lower pension accumulation, and long-term economic vulnerability. 10. Yet, if care demands are addressed by appropriate policy interventions, there is an opportunity to increase labor force participation and productivity, especially for women, goals which are vital for sustainable development in aging countries. The rising demand for care services provides an opportunity to develop a formal care industry, which can contribute to the active aging objective by recruiting younger old to care for older old, as well as increase female labor force participation, in particular for women with low skills. This is critical, as the reduction in the size of the working age population in aging societies will require more individuals to enter the labor force, particularly in developing countries. As productive and reproductive years overlap for women, and single women and those without children already exhibit a labor force participation rate comparable with their male counterparts, support for working mothers (and fathers) needs to be provided to prevent women to drop out of the labor force due to childcare demands. This target cannot be attained without improved care services that not only free women to take part in paid work but also ensure adequate human capital investment in the young generations. 3 Rainer and Siedler (2010) look at siblings in the United States and find these results. Stern (1995) suggests that family members make the decision about long-term care well in advance of the need. 11 11. We examine the provision of childcare and eldercare in ECA countries and suggest policy priorities that address these risks, maximizing the potential of aging societies. An aging society obliges households to juggle multiple imperatives: to have more children, participate longer in the labor force, increase savings, and provide care of the rising ranks of the elderly. This paper examines how households in Western Europe and ECA currently fare on fulfilling these roles and how policies can assist households in attaining healthy, active, and prosperous lives for themselves and the people they care for. Specifically, the paper assesses the normative, institutional, and household-level factors shaping the balance between formal and informal care provision, the characteristics of informal caregivers, and the relationship between caregiving and labor market outcomes. The paper draws on several sources of information as the evidence base—from quantitative surveys, including the Generations and Gender Survey (GGS), Time Use Surveys, the Survey of Health, Ageing, and Retirement in Europe (SHARE), to a new mixed-methods dataset (see Table 1 for a summary of data sources by ECA countries, Box 1 for detailed information on the independent mixed-methods dataset, and Annex on Major Data Sources for additional details). The paper is structured as follows: Section II addresses the split between formal and informal childcare and eldercare provision in ECA. Section III focuses on informal care, discussing how women at different stages of life provide childcare and eldercare. Section IV focuses on the challenges that caregiving demands pose in terms of labor market outcomes and consequent economic well-being. Section V concludes the paper by examining what we know in terms of policies that can support families in informal care provision in a sustainable and incentive-compatible manner. 12 Table 1: Summary of data sources by ECA countries Indepen- National ECA Countries SHARE GGP ESS HETUS RLMS dent Data TUS Albania X Armenia X Azerbaijan Belarus Bosnia and Herzegovina X Bulgaria X X X Croatia X Czech Republic X X X Estonia X X X X X Georgia X Hungary X X X Kazakhstan Kosovo X X Kyrgyz Republic X Latvia X X Lithuania X X X FYR Macedonia X X Moldova X Montenegro Poland X X X X Romania X X Russian Federation X X X Serbia X X Slovakia X Slovenia X X X Tajikistan Turkey X Turkmenistan Ukraine X X Uzbekistan Note: The full titles of the abbreviated data sources listed above are as follows: the Survey of Health, Ageing, and Retirement in Europe (SHARE), The Generations and Gender Programme (GGP), the European Social Survey (ESS), the Harmonized European Time Use Surveys (HETUS), National Time Use Surveys (TUS), and the Russian Longitudinal Monitoring Survey (RLMS). 13 Box 1: A new, independent mixed-methods dataset on care, aging, and gender To better understand the context of childcare and eldercare provision in the region, a survey on the distribution of formal and informal care was designed and implemented in seven ECA countries. The field work was divided broadly into two components: (a) a demand assessment, which investigated the care needs and preferences of families with children and/or elders and (b) a supply assessment of available care services, with a focus on their quality, accessibility, and affordability. The demand assessment targeted households with children and/or elders and included an investigation of time use, care needs, perceptions, and preferences about care responsibilities, as well as barriers in access to formal child or elder care services. Whenever possible, it followed the dynamics of care demand and supply at the household level, with women and their labor force engagement at the center. This assessment included quantitative individual-level questionnaires as well as qualitative focus group discussions. The supply assessment investigated the types of child and eldercare services available to households, both public and private, and explored their quality, cost, and accessibility, as well as the social perception and normative views around care and the use of the different available alternatives. This included site visits, mixed-methods interviews, and, when appropriate, quantitative observational checklists. Both demand and supply assessments were conducted in each of seven countries: Ukraine, the Kyrgyz Republic, Armenia, Bosnia and Herzegovina, Kosovo, FYR Macedonia, and Serbia. In each country except Bosnia and Herzegovina, we chose three sites: a rural community, a small city, and a middle-class neighborhood in the largest urban center of the country. Due to its unique political structure, in Bosnia and Herzegovina, we selected four sites— two urban and two small cities— to maintain balance across the Federation of Bosnia and Herzegovina and Republika Srpska. For the demand assessment, individual interviews were conducted with 734 individuals, all of whom were invited to one of the 66 focus group discussions. These participants were one-third male and two-thirds female, between 25 and 65 years of age, who had different levels of engagement in the labor market (employed, unemployed, and inactive). The supply assessment was a census-type study of all care services available in the sites we targeted for the demand assessment. Overall, 156 childcare facilities, 33 eldercare facilities, and 21 care intermediaries were interviewed and assessed. Country Individuals FGDs held Childcare Eldercare Intermediaries interviewed facilities facilities assessed assessed assessed Ukraine 99 9 51 2 10 Kyrgyz Republic 94 9 73 7 0 Armenia 121 9 30 3 1 Bosnia and Herzegovina 107 12 8 5 0 Kosovo 102 9 9 3 3 FYR Macedonia 103 9 20 5 3 Serbia 108 9 18 8 4 Total 734 66 209 33 21 14 II. How is care organized in ECA? Formal and informal supply 12. The organization of care—for children or elderly—is closely linked to three interrelated factors: household-level capacity to provide care, social norms about the distribution of responsibilities, and the extent of available services and support structures. Informal care in this study refers to unpaid and generally unregulated care, usually provided by family members, whereas formal care is defined as care that is paid and is thus regulated by some type of a contractual arrangement (see Table 2). In most countries, formal care tends to emerge as a response to support families in their caregiving role when that role cannot be fulfilled within the family. An interaction between prevailing social norms and institutional environment determines each society’s reliance on particular modalities of formal support for caregiving, such as leave arrangements, financial support, and in-kind services. Table 2: Typologies of care Type of Care Definition Relevance for Childcare Relevance for Eldercare This refers to care for which See specific examples of See specific examples of recipients or family members pay. It both types of formal both types of formal Formal Care can include institutional (center- childcare under eldercare under institutional based) care, as well as residential (at- institutional care. care. home) care. Examples include early Also referred to as center-based care, childhood education and Examples include nursing Institutional this is a type of formal care. It care (ECEC) programs homes for long-term stays Care includes paid care which occurs out provided in kindergartens, and day-care centers for of the home. crèches, and other day-care elderly. facilities. Also referred to as home-based or at- Residential home care, this is a second type of Examples include a nanny Examples include an at- Care formal care. It includes paid care or babysitter. home nurse. which occurs in the home. Within informal care, Unlike informal childcare, mothers are seen as there is no ‘natural’ primary This refers to unpaid care. Informal ‘natural’ primary caregiver for eldercare. This Informal caregivers are usually family caregivers. Others, such as role is often, though not Care members, friends, or relatives of the grandparents, fathers, and always, taken by the elder's care recipient. siblings, can also be children, spouse, and/or informal caregivers. household members. Source: Author’s based on Kraus et al. (2010). 13. The particular mixes of formal care policies and households’ caregiving decisions in the context of these policies can, in turn, generate different ‘care regimes’ that exhibit varying prevalence of formal and informal care provision. Work by Kraus et al. (2010) classifies 14 European Union (EU) 15 countries according to use and financing of care (Table 3). Indicators for this classification include income and needs-corrected spending, share of private expenditures, formal care use, role of informal care, support for informal caregivers, accessibility, targeting, and importance of cash benefits. They find a variety of care systems that provide a range of levels of support to informal care providers. These care regimes also differ in the extent to which formal care substitutes for or complements family-provided care. For instance, Bettio and Plantenga (2004) classified Mediterranean countries, such as Italy, Greece, and Spain, in a cluster with high reliance on informal care for both children and elderly and underdevelopment of any formal care support policies. At the other extreme, in Nordic countries, families play a modest role in care provision, given the broad range of publicly provided care services for the youngest and oldest members of society. In between the two ends of the spectrum, formal care support in the U.K.-Netherlands regime is limited to the elderly, while that for children is considered to be the natural domain of the family; in the Austria- Germany cluster, informal care by the family is financially subsidized by the state; and in the Belgium- France cluster, formal care policies focus on services and financial support rather than leave arrangements, resulting in greater reliance on formal rather than informal care. The aim of this section is to identify the prevailing care regimes in ECA and to discuss the forces that determine the shape of these care regimes. Table 3: Typology based on use and financing of care Nature of the system Countries Characteristics Oriented toward informal care, Belgium*, Czech Republic, Low spending, low private funding, high low private financing Germany, Slovakia informal care use, high informal care support, modest cash benefits Generous, accessible, and Denmark, the Netherlands, High spending, low private funding, low formalized Sweden informal care use, high informal care support, modest cash benefits Oriented toward informal care, Austria, England, Finland, Medium spending, high private funding, high private financing France, Spain high informal care use, high informal care support, high cash benefits High private financing, informal Hungary, Italy Low spending, high private funding, high care seems a necessity informal care use, low informal care support, medium cash benefits Source: Author’s based on Kraus et al. (2010). Note: *Medium spender. 14. The interaction of formal and informal care is a widely acknowledged feature of the childcare part of the care regime. Societies provide different types of formal support to households for them to better decide how to allocate household’s members’ time between care and work. The type of support available varies, with some countries supporting family-provided care and others providing options that substitute family care with center-based care. There is a widespread recognition that access to formal affordable and quality childcare—mostly center-based—allows informal care providers to reallocate their time to other activities, largely paid employment (Gornick and Meyers 2004; Esping-Andersen 2009), enabling them to accumulate valuable labor market experience, attain economic independence through income generation, as well as achieve future income security through pension contributions. 16 15. The links between formal and informal eldercare are complex and may depend on the intensity of recipients’ care needs. Formal caregiving can compensate, substitute, complement, or reinforce informal caregiving by family members (Chappell and Blandford 1991; Denton 1997; Van Houtven and Norton 2004). Different ‘care regimes’ play a role in the distribution of care tasks between formal and informal provision. In the same vein, families do not give up providing care when formal care is available. However, while childcare is likely to complement informal care and alleviate some of the informal caregivers’ burden, this link is not as direct when it comes to eldercare, as the degree of substitutability of the two types of care depends on the extent of the care recipient’s needs (Bonsang 2009). Whereas recipients with low levels of disability may be cared for either formally or informally, as the level of disability rises, the resources (in terms of time, knowledge, and money) of informal care providers may become insufficient; in this case, informal care turns from a substitute to a complement to the available formal care. This interaction suggests a ‘task-specific model’ where formal and informal care are complements focusing on different tasks, with informal care satisfying recipients’ basic care needs (such as bathing and feeding), while formal care focuses on tasks that require specialized knowledge (such as health monitoring and physical therapy maintenance). However, even as care needs of the elderly increase, the decision to use formal care remains dependent on the actual and perceived accessibility, quality, and affordability of formal care, as well as on the social norms surrounding care obligations. 16. The confluence of multiple factors results in a higher prevalence of informal childcare in ECA, compared to their Western European benchmarks. A combination of service availability and intra-household decision-making processes underlies the relatively low utilization of formal care services in the ECA region. As one proxy for formal childcare, pre-primary school enrolment shows large variation across countries (Figure 2). Young countries in Central Asia and Turkey, independent of income levels, still exhibit low levels of coverage, while aging countries have a much higher participation of children in early education, although all but three (Ukraine, Belarus, and the Czech Republic) have enrollment rates below the EU average. As Figure 3 shows, utilization of formal care also varies by child’s age, with children over the age of 3 years being more likely to be in formal care than younger children. Indeed, on average, 63 percent of children in the EU (28 countries) under the age of 3 years receive only parental care (going as high as 95 percent in Bulgaria). Other surveys specifically looking at intra-household care arrangements such as the GGS demonstrate that households with young children are more likely to use some form of informal care—unpaid care by a family member, friend, or another person—to complement parental care compared to using formal/institutional care (Figure 4). Interestingly, the split between unpaid care and formal institutional childcare is high even in EU-13 countries as well as Eastern Partnership countries, suggesting that the two forms of care might be used as complements in these sub regions. 17 Figure 2: School Enrolment, pre-primary (gross % of relevant age groups) Source: Authors’ calculations based on World Development Indicators 2014. Figure 3: Children in formal childcare by age group (children with at least 1 hour of formal care per week as % of the population in the age group) Children from 3 to minimum Children age 3 or less compulsory age to start school Source: Authors’ calculations based on EU-SILC 2012. 18 Figure 4: Percent of households with at least one child under 7 years who use institutional, paid at home, and unpaid childcare 100% 90% 80% 70% 63% 65% 57% 60% 50% 41% 37% 40% 30% 20% 10% 0% EFTA, EU15 EU13 Western Balkans Eastern Parternship Young Countries in ECA Paid childcare inside the home (nanny/babysitter) Institutional/paid childcare outside the home Unpaid childcare Source: Authors’ calculations based on GGS data (most recent wave for Bulgaria, the Russian Federation, Georgia, Romania, Lithuania, Poland, the Czech Republic, Germany, France, the Netherlands, Norway, Austria, and Belgium) and fieldwork data (2014 data for Armenia, Bosnia and Herzegovina, Kosovo, the Kyrgyz Republic, FYR Macedonia, Serbia, and Ukraine). 17. In ECA, utilization of any care arrangements is much lower for elders than for children; and most eldercare remains informal. Compared to data on childcare, data on eldercare arrangements is much more sparse and difficult to interpret. SHARE provides some suggestive evidence on the topic of utilization of different forms of eldercare. In all the countries included in the study, less than a quarter of older (50+ years) adults with self-reported limitations in activities of daily living received any help with personal care (Figure 5). Of the ones that did receive help, the vast majority relied on informal assistance from someone in the household, a friend, or a neighbor. Less than two percent of individuals are receiving formal care in the form of long-term stay in nursing homes, which would represent the highest level of substitution of informal care for formal care. The much greater reliance on informal/unpaid care is confirmed by the mixed- methods data collection in a subset of countries (Figure 6). Notably, some Western Balkan countries (FYR Macedonia, Kosovo, and Bosnia and Herzegovina) have already developed a menu of eldercare options, while in other countries (Serbia, Armenia, Ukraine, and the Kyrgyz Republic), unpaid care appears to be the exclusive mode of eldercare arrangement. 18. Given the constraint on available time, families consider normative, institutional, and labor market factors when allocating resources to childcare and eldercare. When deciding between formal and informal care, household members are faced with the choice between three kinds of activities—work for the market, work for the household, and leisure (Pagani and Marenzi 2008; Becker 1965). Decisions about allocating time to each activity are shaped largely by a set of forces coming from prevailing social norms, specific household-level factors, as well as institutional arrangements that can affect both the demand for and supply of formal and informal care. 19 Figure 5: Share of adults 50+ years with self-reported limitation in activities of daily living that receive care 30% 25% 20% 15% 10% 5% 0% Informal care Formal care Source: Authors’ calculations based on SHARE Wave 4. Notes: Informal care is defined here as help during the last 12 months with personal care such as washing, getting out of bed or dressing, or practical household help from someone in the household, any family member from outside the household, any friend or neighbor. Formal care is defined here as an overnight stay in a nursing home in the last 12 months. Figure 6: Percent of households with at least one elder 65+ years who use institutional, paid at home, and unpaid eldercare 100% 83.3% 75.0% 80% 60.0% 54.5% 60% 40% 25.0% 29.6% 17.4% 20% 0% Serbia FYR Macedonia Bosnia and Kosovo Armenia Ukraine Kyrgyz Herzegovina Western Balkans Eastern Partnership Young Paid eldercare inside the home Institutional/ paid eldercare outside the home Unpaid eldercare Source: Authors’ calculations based on fieldwork data (2014). 20 Social norms and views 19. The decisions regarding time allocation for market work and care duties take into account norms with respect to care obligations. These norms and societal views on how care should be provided and who should be providing it also shape the types of formal services and caregiver support that become available, through the mediating influence of the social policy environment. 20. Social norms in ECA countries place a high expectation on the provision of intergenerational support, in particular around women acting as caregivers for both children and the elderly. In virtually all countries in the world, women face much higher expectations than men when it comes to providing time for both childcare and eldercare. In fulfilling the roles of both sons and fathers, men are more likely to be expected to help with financial matters and specific practical tasks such as home repairs or moving heavy objects (Campbell and Martin Matthews 2003; Gallagher 1994; Bender 1994; Finch and Mason 1990; Hugentobler 2003). Prevailing social norms, such as the degree to which a mother’s employment is damaging to her children’s well-being, can alter a woman’s calculus in terms of combining work and motherhood and can also shape the availability of services for those mothers (Crompton and Lyonette 2006; Morgan 2008). Indeed, as can be observed in Figure 7a, respondents often connect the welfare of the care recipient (in this case, preschool age child) with women foregoing care duties. 21. While younger generations, which have been exposed to more flexible norms and higher levels of female labor force participation, are somewhat less likely to have a gendered view on childcare issues, this is not the case for all countries. Notably, the data identify a conservative cluster of nine countries (where more than half of the respondents both above and below 50 years associate mothers’ employment with negative outcomes for children), consisting of Armenia, Georgia, the Russian Federation, the Kyrgyz Republic, Bulgaria, Ukraine, Poland, Lithuania, and Kosovo. At the other end of the spectrum is a more liberal cluster of Norway, Serbia, and Estonia, where 20 percent or less have this association. More generally, Eastern Partnership and the Russian Federation countries appear to have the most conservative norms related to women’s involvement in childcare duties, and European Free Trade Association (EFTA)/EU-15 countries the most liberal ones, with EU-13 and Western Balkan countries falling in between. There is also evidence that social norms related to eldercare appears to be less gender- specific. In virtually all countries sampled in the GGS and fieldwork, the agreement with the statement “When parents are in need, daughters should take more caring responsibility than sons,” falls below 50 percent (except older respondents in Kosovo and the Kyrgyz Republic) (Figure 7b). However, relative to most EFTA/EU-15 countries (except Italy), countries in ECA were more likely to agree with this statement, implying a higher burden of care placed on daughters. 21 Figure 7: Agreement with the following statements by respondents’ age and country a. “A preschool child is likely to suffer if his/her b. “When parents are in need, daughters should take mother works” more caring responsibility than sons” EFTA, EU15 EU13 EFTA, EU15 EU13 Western Balkans Eastern Partnership & Russia Western Balkans Eastern Partnership & Russia Young Countries in ECA Young Countries in ECA 100% 60% y=x y=x 90% 50% 80% Armenia Georgia Kosovo Kyrgyzstan 70% Russia 40% Kosovo Bulgaria 60% Russia Kyrgyzstan Up to 49 years Up to 49 years UkrainePoland Lithuania 50% 30% Georgia Bulgaria Armenia Romania Romania Czech Bosnia & 40% Republic Czech Herzegovina Bosnia & France Serbia Herzegovina 20% Republic FYR Belgium 30% Italy Macedonia Germany Ukraine Poland Lithuania 20% Serbia FYR Estonia 10% Macedonia Belgium 10% Norway Germany NorwayFrance 0% 0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20% 30% 40% 50% 60% 50 + years 50 + years Source: Authors’ calculations based on GGS data (most recent wave for Bulgaria, the Russian Federation, Georgia, Romania, Lithuania, Poland, the Czech Republic, Germany, France, the Netherlands, Norway, Austria, and Belgium) and fieldwork data (2014 data for Armenia, Bosnia and Herzegovina, Kosovo, the Kyrgyz Republic, FYR Macedonia, Serbia, and Ukraine). 22. The acceptability of state involvement in childcare and eldercare is much lower in ECA relative to Western European benchmarks. As shown in Figure 9, two countries—Norway and the Netherlands—have relatively high shares of respondents who believe that care for children and elderly is mainly a task for society rather than the family. At the other end of the spectrum are most ECA countries, where the opinion that the primary responsibility of society for the care of the elderly and preschool children is shared by less than 15 percent and 10 percent of respondents, respectively. A similar picture emerges from SHARE data based on opinions of respondents aged 50+ years. The majority of respondents in Nordic countries assign the main responsibility for supporting older people to the state, including with personal care and household chores; on the opposite end, in Poland and the Czech Republic (the only two ECA countries included in SHARE at the time), the vast majority of respondents see more of a role for the family, especially when it comes to help with everyday tasks and chores (Figure 8). Regardless of levels, however, across countries formal care gains relevance when limitations increase to bathing, dressing, or needing support at night. 22 23. These normative perceptions about state responsibility in care are important both as a driver of care-related policy and as a reflection of the availability of formal care. The stronger the view that eldercare or childcare should be done by the family, the more likely it is that the family will fulfill this obligation. Thus, the earlier observation of greater reliance on informal childcare in ECA compared to EU- 15 countries could stem, at least in part, from the normative assignment of responsibility for care to the families being reflected in social policies promoting informal rather than formal care. Moreover, these opinions are not independent of the availability, diversity, and quality of support provided by the state to individuals in need. Interestingly, Figure 9 demonstrates that in most ECA countries, there is more general acceptance of state involvement in provision of eldercare, relative to childcare (except in three Western Balkan countries—Bosnia and Herzegovina, Kosovo, and FYR Macedonia—as well as Germany). This could potentially be explained by the expectation of state support for care based on older people’s lifetime contributions. 24. The evolution of norms around formal childcare in different settings suggests the possibility of policy driving gradual adjustment in norms around eldercare as well. The norms around provision of childcare have evolved, partly due to the growing recognition of the importance of the educational element in this service, which can be better monitored when childcare is provided formally. This combines with the expansion of pre-primary education in many countries (for example, Turkey, Romania, and Bulgaria, among others). A similar logic can apply to the health aspect of eldercare provision. If the state can take on quality control in the provision of eldercare in the same manner as it provides it for preschool education, the expectation that families (and in particular, women) should bear the sole responsibility to ensure comfort to their elderly relatives can slowly change. The sharing of the care burden between the state and the family will then allow caregivers to devote more time to other productive activities, such as paid employment. 23 Figure 8: Percentage of respondents who think that the state (as opposed to the family) should bear responsibility for older persons who are in need, in each of the following categories 100% 90% 80% 70% 60% 50% 40% 30% Personal Care 20% 10% Help with Chores 0% France Austria Italy Spain Netherlands Denmark Poland Germany Czech Republic Ireland Switzerland Belgium Sweden ECA EFTA, EU15 Source: Authors’ calculations based on SHARE Wave 2. Figure 9: Opinions about state involvement in provision of childcare and eldercare EFTA, EU15 EU13 Western Balkans Eastern Partnershup & Russia Young Countries in ECA 0% 5% 10% 15% 20% 25% 30% 35% 40% 90% Think that caring for older persons in need of care at NLD 80% their home is more of a task for society NOR 70% 60% 50% 40% 30% BGR 20% BEL RUS LTU CZE FRA ROM DEU POL 10% GEO SRB ARM KGZ UKR BIH 0% MKD KOS Think that caring for pre-school children is more of a task for society Source: Authors’ calculations based on GGS data (most recent wave for Bulgaria, the Russian Federation, Georgia, Romania, Lithuania, Poland, the Czech Republic, Germany, France, the Netherlands, Norway, Austria, and Belgium and fieldwork data (2014 data for Armenia, Bosnia and Herzegovina, Kosovo, the Kyrgyz Republic, FYR Macedonia, Serbia, and Ukraine). 24 Social policies and supply of formal care 25. The lower availability of acceptable options for formal childcare and eldercare can explain the higher prevalence of informal care in ECA. While the question of what should come first, demand or supply, has often plagued policymakers involved with childcare provision, norms about the acceptability of formal care options are more likely to change in the presence of adequate supply of such options. In many ECA countries, demand for formal childcare and eldercare services indeed outstrips supply, with some countries experiencing long queues for both kindergartens as well as social care services targeted to the elderly. In contexts where formal care services are not available, families are forced to provide informal care even when such provision is inefficient or exposes households to economic vulnerability (due to taking an income earner out of the labor market or via finding private arrangements at home or via paid care support). 26. Accessibility, affordability, and quality of formal care options figure prominently in the decision-making process for households with care needs. While care policies will be discussed in detail in the policies section of this document, a few elements are important when assessing the allocation of care between formal and informal. The first has to do with accessibility of services. This refers not only to location of care providers but also to the capacity at each location. The second is the affordability of these services—the explicit or implicit cost associated with utilization of formal care services, and conversely, the available financial support provided for informal caregivers. Third, the quality of care services can not only determine current utilization of formal care but also shape the norms surrounding such utilization. The quality of care services depends, in a large part, on regulations and standards, which ideally would reflect priorities and expectations of households consuming these services. Javornik (2012) collected information on these three aspects of childcare services for eight former socialist countries, finding important differences in the priority given to different elements (Figure 10). Some, like Poland and the Czech Republic, scored poorly across the board. Slovenia was the only country receiving full marks for setting uniform quality standards at the central level while quality standards in other countries were more decentralized. Accessibility (in terms of service hours, age at entry and exit, and allocation of places) was highest in Lithuania, Hungary, and Slovenia. Affordability was highest in Hungary, where childcare services were provided free of charge in 2007, while Slovenia used a sliding scale taking into account family incomes and household size.4 4 Complete details on the metric used can be found in the paper. Data is compiled from different sources including the Mutual Information System on Social Protection in the Member States of the European Union, the OECD Family Database, Eurostat and the European Commission, and national administrative sources, and others. 25 Figure 10: Assessment of childcare policies in selected ECA countries, 2007 8 7 Index (8=highest) 6 5 4 Quality 3 Cost 2 Accessibility 1 0 Source: Authors’ calculations based on Javornik (2012), with accessibility index calculated as a simple average of 4 indexes (compulsory schooling age, compatibility of service hours with working hours of parents, admission age to day care, and allocation of places). 27. Capacity to meet demand—actual and potential—emerges as one of the key constraints in the provision of formal childcare and eldercare services in ECA. Insufficient accessibility of formal care services, both for children and the elderly, appears to be a common phenomenon in ECA. In 2010, of the women in the EU-27 countries who were out of the labor force or working part-time for reasons related to child or eldercare, 25 percent did so due to a lack of availability of childcare services, but this went up to almost 50 percent in the Czech Republic and Croatia.5 Observing the availability of care services in selected ECA countries via data collected specifically for this study, we find large variations in the number of available care facilities across countries and between different areas within a country. Childcare and (the very few existing) eldercare services are largely concentrated in urban areas, with few, if any, care services available in other localities. Services for the elderly are particularly sparse, and even when present in urban areas, they cover a larger geographical location (that is, the entire city) compared with childcare services, which generally focus on a narrower catchment area (Table 4). Participants in focus group discussions in each of the seven countries confirmed this limited availability. Another piece of evidence on capacity constraints is provided by data from questionnaires filled out at care facility sites in the selected countries. Almost two-thirds of childcare providers visited do not accept new clients right away, with nearly half referring clients to a waitlist; indeed, only Bosnia and Herzegovina and Serbia appear to have available spots in most childcare facilities (Figure 11a). Eldercare services in some countries seem to have more spare capacity, as both facilities in Armenia and the one live-in eldercare facility identified in Ukraine and the majority of facilities in Serbia and the Kyrgyz Republic can accept new clients without a waitlist (Figure 11b). On the other hand, the two identified live-in facilities in Kosovo had a waitlist for new clients. 5 Source: Mills et al. (2014) based on EU Labor Force Survey (LFS) 26 Table 4: Number of care facilities in the field data sample Childcare Eldercare Country Urban Small City Rural Total Urban Small City Rural Total Armenia 19 11 0 30 3 0 0 3 Bosnia and Herzegovina 2 3 3 8 2 1 2 5 Kosovo 8 1 0 9 3 0 0 3 Kyrgyz Republic 52 19 2 73 3 3 1 7 FYR Macedonia 17 2 1 20 2 2 1 5 Serbia 8 7 3 18 7 1 0 8 Ukraine 22 27 2 51 1 1 0 2 Total 128 70 11 209 21 8 4 33 Source: Independent field data (2014). Note: The sample of facilities include all public, private, and community providers of childcare and/or eldercare services in the catchment areas. Childcare services include day care, kindergarten, preschool, and other types of care, and eldercare services include day care, long-term care, social clubs run by an administrator, permanent care, and living facilities. Orphanages, hospitals, and schools are not included (unless there is a specific after-school care program). Figure 11: Availability of places at care facilities in selected ECA countries, 2014 a) Childcare b) Eldercare 100% 100% 12 13 13 8 90% 90% 20 26 26 Percent of Childcare Providers 33 Percent of eldercare providers 80% 80% 25 43 70% 50 70% 60 60% 50 62 60% 40 50% 47 42 44 92 50% 100 100 100 40% 40% 30% 63 20% 30% 57 32 38 50 22 27 27 20% 40 40 10% 0% 10% 0% Yes, and there is no waitlist Yes, but there is a waitlist No Yes, and there is no waitlist Yes, but there is a waitlist No Source: Independent field data (2014). Note: The graph shows responses to the question “Is this care provider currently accepting new clients?” For childcare, facilities were included that serve younger children (5 years and under). For eldercare, live-in facilities were included. 28. Besides sheer coverage and capacity constraints, accessibility of formal care services is negatively affected by limited service hours and age limits for clients. The compatibility of formal care services and the needs of households requiring care can affect utilization above and beyond the availability of vacant spots. For example, if childcare facilities require drop-offs and pick-ups during normal working hours, their usefulness to parents working full-time diminishes dramatically. The data collected for this 27 study reveal that while most childcare providers in the selected ECA countries tend to open sufficiently early (between 6 a.m. and 7:30 a.m., except in Armenia), very few remain open beyond 6 p.m. Moreover, in most countries in the study, many providers, especially in rural areas, close their doors during the summer months (July–August), which can send working parents in a search of an alternative mode of childcare provision. Another aspect that can explain the lower prevalence of formal care in ECA compared to Western European benchmarks is the lowest age limits for admission in many childcare facilities. According to the data collected for this study, children under 2 years of age are served by less than half the providers in all countries in the study except Serbia and Kosovo; indeed, in Ukraine and Armenia, fewer than 5 percent of the providers cater to these young children. When paid maternity leave ends before children reach the age of admission to most childcare facilities, mothers face the choice between leaving the labor force altogether or opting for another form of informal care (such as that provided by grandmothers). 29. The second channel via which policies can affect utilization of formal and informal care is the financial demands imposed on households to access and use care services. If we follow Ettner (1995), the decision to care for an elderly person and the decision to have a child and bear the burden of childcare are very similar in terms of the time and financial commitment they represent. Many countries in the ECA region attempt to address some of these constraints by providing a package of leave and financial support policies related to care duties. Some of these policies, such as care-related leave and family allowances, can encourage families to provide dependent care by themselves. In the latter case, what matters for families is how long a household member can sustain full-time caregiving without becoming a burden on the household’s economic well-being. As can be seen in Figure 12, in many European countries there is a significant gap between the duration of regulated parental leave and the period of effective leave when caregivers are compensated for the foregone earnings (at least at 60 percent of their income). Other policies can stimulate utilization of existing formal care options by providing vouchers or subsidies. The affordability of formal care, including the net effect of the aforementioned policies, is a key factor in the balance between formal and informal care. According to the analysis of EU-LFS data in Mills et al. (2014), more than half of all working-age women in EU-27 who cited care-related reasons for being inactive or working part-time reportedly blamed the cost of childcare for their inability to work full-time, although cost figured more prominently (compared to accessibility and quality) in EU-15 countries, such as Ireland (85 percent), United Kingdom (73 percent), and Netherlands (71 percent), and also in Romania (80 percent). 30. Affordability of care services does not appear to be the major constraint in most countries selected for special data collection in this study. Indeed, in two countries—Armenia and Ukraine—more than 60 percent of childcare providers offered full-day childcare for free or for a voluntary donation, as can be seen in Figure 13. In some countries where free childcare was not available (such as Kosovo and FYR Macedonia), more than half of childcare providers offered some flexibility with respect to payment for their services. Eldercare in the few live-in facilities identified in Kosovo and Ukraine was provided free of charge, while in Serbia all live-in eldercare providers offered payment flexibility. The Kyrgyz Republic appears to be one country where affordability of care services might be an issue, as more than two-thirds of childcare providers and 80 percent of live-in eldercare providers were inflexible about payment for care services. 28 Figure 12: Length of parental leave and effective parental leave in selected European countries 30 40 35 25 Months of total parental leave 30 20 25 Weeks 15 20 15 10 10 5 5 0 0 Portugal Germany Denmark Latvia Greece Sweden Russia Austria Slovakia Slovenia Spain Romania Poland Luxembourg France Norway Lithuania Belgium Netherlands Hungary Georgia Ireland United Kingdom Italy Finland Bulgaria Estonia Czech Republic Effective parental leave Length of well paid leave parental leave total duration Source: Multilinks database. Note: Effective parental leave: duration of paid parental leave (including maternal leave) by the income replacement rate of the parental benefit. Well-paid leave: Length of leave (parental and maternal) with an income replacement benefit at 60 percent or more. 31. Concerns about the quality of formal care options figure prominently in households’ calculation of costs and benefits of formal and informal care. Existing data on the quality of formal care services are very limited. There is wide cross-country variation on the assignment of responsibility, existence of standards, and other measures of quality. For eldercare, many of these quality control mechanisms are not present or are in the process of development. An additional challenge is the distance between the letter of regulation and actual practice. Data from site visits to care facilities was used to construct quality indexes, focusing on aspects of infrastructure, human resources (that is, training and tenure of caregivers), activities and materials, and specifically for eldercare, health-related support. In line with the greater consensus on the quality standards for childcare, there appears to be less variation among countries in the quality index constructed from field visit questionnaires, compared to that of eldercare. As can be seen in Figure 14, Armenia appears to have relatively high quality of childcare and eldercare services, whereas the Kyrgyz Republic is close to the bottom on both. On the other hand, in some countries, quality appears to be relatively high in one but not the other type of care services: FYR Macedonia has the highest quality index in childcare and the lowest in eldercare, while Bosnia and Herzegovina is at the bottom in terms of childcare quality and near the top in terms of quality of eldercare services. Figure 13: Affordability of care facilities in selected ECA countries, 2014 a) Childcare b) Eldercare 29 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Not free but offer some payment flexibility Not free but offer some payment flexibility Offer free full-day care Offer free full-day care Source: Independent field data (2014). Note: Facilities are considered ‘free’ if they offer full-day care to the majority of their clients without charge, or with a voluntary donation. When calculating whether a facility offers ‘some payment flexibility’, the free facilities were excluded. For childcare, facilities were included that serve younger children (aged 5 years and under). For eldercare, live-in facilities were included. 30 Figure 14: Quality of care services in selected ECA countries, 2014 a) Childcare b) Eldercare 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Special needs, Healthcare, and Support sub-index Materials, Curriculum, and Learning sub-index Schedule, Activities, and Materials sub-index Infrastructure sub-index Infrastructure & Safety sub-index Human Resources sub-index Human Resources sub-index Source: Independent field data (2014). Note: The overall quality of childcare facilities was evaluated using three sub-indices: Materials, Curriculum, and Learning (composed of 8 binary variables); Infrastructure (composed of 19 binary variables); and Human Resources (composed of 4 binary variables). Each sub-index is weighted equally in the overall quality assessment. The total childcare index is on a scale from 0 to100, so each sub-index is from 0 to 33.33. The overall quality of eldercare facilities was measured in a similar way. It is composed of 4 sub-indices: Special needs, Health care, and Support (composed of 14 binary variables); Schedule, Activities, and Materials (composed of 16 binary variables); Infrastructure and Safety (composed of 24 binary variables); and Human Resources (composed of 8 binary variables). As with childcare, each sub-index is weighted equally in the overall quality assessment. The total eldercare index is on a scale from 0 to 100, so each sub-index is from 0 to 25. For childcare, facilities were included that serve younger children (5 years and under). For eldercare, live-in facilities were included. 32. Intergenerational co-residence is correlated with the provision of informal care within the household. Household size and composition, including the number of actual and potential ‘care recipients’ and ‘caregivers’ in the household is an important consideration. Time is a limited resource, and the smaller the household size, the fewer the individuals to rely on for support in terms of care. The move from independent living or living with a partner to cohabitation appears to be related with an increased need for support. Widowhood, for example, is one of the factors that might push women into cohabiting with their children or other family members, in particular if declining health or need for help with daily activities is present. Lack of alternative options for elders—such as long term-care facilities or professional support at home—will increase the prevalence of informal care via intergenerational co-residence (Dykstra et al. 2013). Countries in Eastern Europe, such as Romania, have higher numbers of elderly living with adult children compared to Western European benchmarks—a trend that can be traced back to the 1990s (De Vos and Sandefur 2002), but more widely, increases in co-residence can be observed in other countries in the region. Care from household members is the main form of care for those who cohabit with other family members. Looking specifically at those elders who report experiencing limitations due to health, care intensity increases as these limitations become more severe. SHARE data (Figure 15) show that regardless 31 of the severity of the limitation, for those who live with family members who can act as care providers, care by a household member is the more frequent form of care. Figure 15: Distribution of source of care for those with severe and moderate limitations by living arrangement (self-reported health) Living with children/grandchildren and partner limitations Severe Living with children/grandchildren Living with partner only Living alone/other living arrangements Living with children/grandchildren and partner limitations Moderate Living with children/grandchildren Living with partner only Living alone/other living arrangements 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Receiving care from HH members Receiving care from non-HH members Receiving formal care Source: Authors’ calculations based on SHARE, Wave 4. 33. Households are more likely to choose formal care when other demands on the time of ‘potential caregivers’ take a more prominent role. Regression analysis examining the relationship between formal care utilization and household-level factors was conducted using GGS and the data collected from a special study of selected ECA countries, using a sample of female respondents, aged 20– 64 years, who live in a household with at least one child. The analysis identified correlates of using institutional/paid childcare (that is, either day-care center, nursery or preschool, after-school care center, self-organized childcare group, or from another institutional or paid arrangement) for households with children under 14 years and under 7 years.6 As can be observed in Figure 16, the results confirm that the presence of multiple children in the household and the gainful employment of the respondent, who is likely the primary caretaker, are positively correlated with utilization of institutional/paid childcare. Finally, the positive correlation with regular unpaid childcare help suggests that utilization of institutional/paid childcare is a complement rather than a substitute to informal provision of childcare by people other than the child’s mother. 6 Regression tables are in Tables 8–10 in Annex 2. 32 Figure 16: Correlates of using institutional/paid childcare arrangements in selected ECA and EU-15/EFTA countries Receives regular unpaid help with childcare *** *** Receives regular paid care services at home (nanny/babysitter) Respondent is employed *** *** Respondent currently has a partner/spouse *** *** Respondent has less than tertiary education *** *** Respondent is 50-64 years old *** Respondent is 35-49 years old *** *** At least one young child (under 5 years) in the *** household *** Two or more children (under 7 years) in the *** household Two or more children (under 14 years) in the household *** Elder (65+ years) in the household * *** -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 HH with children under 7 HH with children under 14 Source: Authors’ calculations using GGS (most recent wave for Bulgaria, the Russian Federation, Georgia, Romania, Lithuania, Poland, the Czech Republic, Germany, France, the Netherlands, Norway, Austria, and Belgium) and fieldwork data (2014 data for Armenia, Bosnia and Herzegovina, Kosovo, the Kyrgyz Republic, FYR Macedonia, Serbia, and Ukraine). The models are estimated using Ordinary Least Squares (OLS) regressions, with the sample constrained to female respondents, aged 20–64 years, in households with at least one child (either under 14 years or under 7 years). A full set of country dummies were also included in each regression. 34. Factors that are associated with a lower likelihood of using institutional care largely speak to the care capabilities within the home. For example, presence of an elder in the household is negatively correlated with the use of paid or center-based childcare, confirming that intergenerational co-residence is correlated with informal care (Figure 16). A similar relationship emerges if the respondent has a spouse, possibly because of a division of paid and unpaid labor between the man and women in the couple. If the respondent has not completed tertiary education, the household is less likely to use institutional childcare, which may speak to the woman’s lower opportunity cost of labor. Not surprisingly, in higher-income countries with a more developed childcare system, such as Germany, France, and Norway (but also Hungary, and in some specifications, Armenia and Ukraine), women are more likely to use institutional childcare. In other countries, the opposite is true. These include some EU-15 countries, such as the Netherlands and Belgium, some EU-13 countries, including Romania and Poland, as well as Georgia and Kosovo. 33 35. In sum, the choice between formal and informal care for children and elderly depends on the social norms and the institutional environment as much as household composition and labor market forces. During the focus group discussions held in countries selected for special data collection, men and women alike, independent of the country, location, or labor market status, weighed normative, institutional and household factors when discussing the reasons for choosing informal over formal care or vice versa. Box 2 provides a sample of their reasoning to illustrate how, depending on context, some factors gain a more prominent role in relation to others. Box 2: Different countries, different factors Men in a small city in FYR Macedonia noted that quality, cost, and accessibility of childcare and eldercare services were important in their decision making around use of these services. “There was nobody who could take care of my child. I am employed, my wife is also going to work, and we had to find a solution. After we saw the conditions in the kindergarten we decided to hire a nanny.” “Finances are the biggest reason [to opt for the use of formal childcare]. In a situation where there is high employment and better economic standard it would be easier for us to pay for the services, otherwise there is no income source for such an expense.” “I would send my father-in-law to a retirement home, but it’s far away, I cannot make that decision, I’m not his daughter. If he were here and went to a home, it would be different, we would visit him, take him home for the weekend, etc. ” Employed women from an urban center in Bosnia and Herzegovina also discussed cost as well as quality when it came to childcare and eldercare: “I wanted to take my children to the kindergarten, but my husband did not want to. He said that it was wasting money.” “We don’t have finances for the elderly. My mother says that she would like to go to a nursing home, but the finances don’t allow it.” “I could never imagine myself in a nursing home. Maybe it would be good if there were good conditions in it, but I would never place my mother in a nursing home.” For women from rural villages in Ukraine and the Kyrgyz Republic, social norms—and strong sanctioning of other people’s choices for formal care—played the larger role in their decision making. When asked why they had reported a low use of the eldercare home, Ukrainian women argued that, “You won’t do that to a relative [place him in a nursing home]” because of reasons of ‘conscience’ and love. They also noted a specific regulatory challenge: “It’s not so easy to put someone into the home for elderly people. They don’t take people if they have children. Children have to sign a refusal (so parent’s pension payments will go to the eldercare home) before their parents can be considered.” Their views remained equally strong when discussing childcare for infants. “No. A sane mother won’t take her baby there.” “You don’t need daycare services if you have a grandmother around, grandfather, or other family. I’d rather starve than give away my baby to a nursery.” However, some signs of change appeared when observing the lack of adequate quality services: “But there are young couples with children who don’t have par ents to help.” “You earn social benefits till the child is 3 years old.” “…Maybe if I had a more gainful employment than my current state work.” “… I wish we had twenty-four-hour kindergarten, like in big cities.” In rural Kyrgyz Republic, women agreed that no one in their village would leave their parents in the eldercare home. “On the contrary, such things should be prevented,” with the only acceptable situation under which it would be acceptable for an elderly person to turn to these services are, “if they live alone and they are unattended, then it is possible; but if they have children, it is unacceptable.” They saw that opening retirement homes might give the wrong signal and “Some children may think that now it is only the government’s duty to care fo r elderly. If you have children, then they should not leave parents in a home for elderly.” 34 III. Women are the main informal care providers in ECA 36. On average, women allocate more time to care-related and home production tasks than men, and they devote more time to these activities regardless of whether they are working for pay or not. This pattern of gendered time use is illustrated in Figure 17 and Figure 18, which depict patterns of time use over the course of a day (from 6 a.m. until midnight), using the most recent available nationally representative time-use data from Serbia, Moldova, and Estonia. Figure 17 highlights differences in the percentage of men and women who engage in three different types of activities: paid work, household- related tasks, and care throughout the day. In both Serbia and Moldova, men are more likely to be engaged in paid work than women at all times of the day, with the difference peaking in the early afternoon in both countries (at 14 percentage points and 10 percentage points, respectively). In Estonia, men and women participate in paid work at nearly identical rates in the morning, but fewer women than men return to paid work after midday. Clearly reflecting a reallocation of time, a larger share of women in Estonia engage in household production as the day continues until it peaks just after the end of a typical work day. Women’s care duties, which are relatively constant during daylight hours, peak in the evening for all countries. The gender difference on care and household production is consistent. On the former, men and women’s participation differs by as much as 22 percentage points in Serbia (at 1:30 p.m.); 15 percentage points in Moldova (at 12:30 p.m.); and 12 percentage points in Estonia (at 6:30 p.m.). Furthermore, for nearly the entire day in Serbia, a higher percentage of women are engaging in household work than in paid work. 37. Thus, though women are spending less time working for pay—and presumably, accruing less income—they spend more time each day engaging in work than men. This is clearly illustrated by Figure 18, which depicts gender differences in paid work and total work (including unpaid household production and care). Women are consistently more likely than men to engage in care, and in all three countries, women are more likely than men to engage in some form of work at every point of the day due to a much higher percentage of women who engage in unpaid work. This division of tasks repeats in other countries in Europe, as documented by Aliaga (2006) using a set of National Time Use Surveys from a larger group of European countries. 38. The role of women as the main childcare providers of has been well documented, as mothers are seen as the natural primary childcare providers. Either as sole providers or with support from formal services, paid help, or other family members, having children increases the opportunity cost for women on their time. However, motherhood also triggers a set of social norms and expectations about women’s role as mothers, which are normally mirrored in the support social policies provide to women (Gornick et al. 1997). While the distribution of childcare tasks between partners has become less skewed as female labor force participation increases and norms change, it has not seen significant changes for the past two decades in most ECA countries. Data from national Time Use Surveys (Figure 19) show how, on average, married women perform the majority of care work, with a higher time investment when children are below the age of 6 years. This inequality remains regardless of women’s employment status (Fisher and Robinson 2009). The demand for maternal care is more intense before children enter the school system, either primary or pre-primary education, and more so for women whose children are younger than the age of three years. 35 Figure 17: Proportion of working age population Figure 18: Proportion of working age population (aged 20–64 years) undertaking activities by gender (aged 20–64 years) engaging in work by gender and time of day and time of day 50% 80% 70% 40% 60% Serbia 2010/2011 30% 50% 40% 20% 30% 10% 20% 0% 10% 6:00 9:00 12:00 15:00 18:00 21:00 24:00 0% 6:00 9:00 12:00 15:00 18:00 21:00 24:00 Paid work, Men Paid work, Women Total work, Men Total work, Women Care, Men Care, Women HH Produc, Men HH Produc, Women Unpaid work, Men Unpaid work, Women 50% 80% 70% 40% 60% 30% 50% Moldova 2011/2012 40% 20% 30% 10% 20% 0% 10% 6:00 9:00 12:00 15:00 18:00 21:00 24:00 0% 6:00 9:00 12:00 15:00 18:00 21:00 24:00 Paid work, Men Paid work, Women Care, Men Care, Women Total work, Men Total work, Women HH Produc, Men HH Produc, Women Unpaid work, Men Unpaid work, Women 50% 80% 70% 40% 60% 30% 50% 40% 20% Estonia 2010 30% 10% 20% 10% 0% 0% 6:00 9:00 12:00 15:00 18:00 21:00 24:00 6:00 9:00 12:00 15:00 18:00 21:00 24:00 Paid work, Men Paid work, Women Care, Men Care, Women Total work, Men Total work, Women HH Produc, Men HH Produc, Women Unpaid work, Men Unpaid work, Women Source: National Time Use Surveys (2010–12). 36 Figure 19: Average daily minutes spent on childcare, by sex, country, and age of the youngest child Source: Eurostat, Time Use Survey - Collection round 2000. Note: Sample includes only persons in a couple with children. Time spent on childcare includes teaching, reading, and talking with child. 39. Women’s care provider role extends well beyond caring for their own children, to care for other family members. Within a household, women will be expected to provide care for those in need, such as elders, small children, and others (Becker 1981, 1985; Akerlof and Kranton 2000).7 This process is replicated throughout women’s life cycles. Among elderly couples, women are more likely to care for their partners. Daughters will provide care for their elderly parents, and grandmothers will provide care for their grandchildren. A good number of women are likely to be engaged in caregiving at different stages of their life, and this care will go in one or multiple directions at the same time (Figure 20), with different intensities in terms of time and frequency (Agree et al. 2003; Dykstra et al. 2013). Figure 20: Care flows throughout women’s life cycles 7 Indeed, Becker suggests women choose to specialize in family care and devote more time to household tasks. 37 40. Aging societies increase the demand for upward care flows besides the traditional downward care. Women will give more time to intergenerational support, from social support and practical assistance to direct caring for a parent or grandchild or helping in the household (Glaser et al. 2010). As longevity increases and the number of children decreases, women who have had breaks in their careers due to childcare and are already out of the labor force, have lower education (and expected lower returns in the market to their skills), and/or have reached a point when some of their care responsibilities have ended (that is, have adult children) are more likely to become eldercare providers if the need arises (Carmichael et al. [2010] find this is the case in England). Even when households have the capacity to pay for care, income cannot completely substitute for social norms around filial obligation and women are more likely to be the household member expected to become a care provider (Mentzakis et al. 2009), as besides opportunity costs, deciding to become a caregiver is related both with earlier experiences of care of any sort and proximity to the person that requires care. In all five ECA countries participating in SHARE, more social support is provided upward (from children of SHARE respondents to their parents) than from parents to children, but downward flows in the form of grandparental childcare are not meaningless. 41. Associated with the intensity patterns of care observed above, co-residence with an elderly person is more frequent in Southern Europe and some of the ECA countries included in SHARE (Figure 22). More than half of Polish and Slovenian respondents aged above 50 years lived with at least one child, and more than 12 percent lived with at least one parent, compared with less than two percent of Danish or Swedish respondents who report living with a parent at the time of the survey. This finding is in line with the documented difference between Southern and Northern European countries, with the former being more of a ‘familial nature’ (Leitner 2003; Bettio and Plantenga 2004), under which higher levels of intergenerational reciprocity are expected. Figure 21: Participation and intensity of social support 25 Denmark 400 Sweden Intensity of social support among participants (median Spain Italy 350 Participation in social support to children (%) 20 300 Belgium Netherlands Germany 250 15 Czech Rep.** hours) 200 Poland** Czechia** Greece* 10 Belgium FranceAustria 150 Switzerland Greece* Italy Poland** Austria Hungary*** 100 Germany Estonia*** Switzerland 5 Slovenia*** Spain FranceNetherlands Portugal*** 50 Sweden Portugal*** Denmark 0 0 0 5 10 15 20 25 0 5 10 15 20 25 Participation in social support to parents (%) Participation in social support to children (%) Source: Albertini (2014) based on SHARE, all waves. 38 Figure 22: Co-residence patterns for adults over 50 years of age 20 Percent co-residing with at least one parent 18 Italy Spain 16 Greece* 14 Slovenia*** Portugal*** Poland** 12 10 Hungary*** Estonia*** Germany Austria 8 Czech Rep.** 6 Switzerland 4 France Belgium 2 Sweden Netherlands Denmark 0 0 10 20 30 40 50 60 70 Percent co-residing with at least one child Source: Albertini (2014) based on SHARE data (all waves). The role of grandparents as childcare providers 42. Formal care will not completely substitute informal care, and depending on the design of the service, it might not serve the needs of working parents, who in the absence of regular childcare for some periods of the day, rely on grandparents as care providers (Settles et al. 2009; Igel and Szydlik 2011; Gray 2005). Intergenerational support with childcare is frequently provided by the mothers of the child’s parents. Grandmothers are expected to have higher disposable time—because of retirement and changes in their household composition and care needs—particularly if they live near the grandchild or in the same household. Buber-Ennser (2014) finds that in 23 countries from Europe and Central Asia included in SHARE and GGS, 63 percent of grandmothers and 53 percent of grandfathers aged 50–79 years provided some care for at least one grandchild aged 15 years or younger, with the highest prevalence and intensity to be found among Russian women, with 81 percent of grandmothers reporting caring for grandchildren (Figure 23).8 These findings are confirmed in the supply and demand fieldwork in an additional set of ECA countries, where also grandmothers and mothers-in-law appear as the most frequent care providers supporting mothers. Participation rate in grandparenting across European countries follows the distribution between North and South family support divide across Europe. 43. Sex of the grandparent, age, presence of a competing care demand such as a co-resident partner, distance from the grandchild, and age of the child are among the factors that will affect the provision of grandparental care. Drawing on the findings by Buber-Ennser (2014), we can observe that each of these factors plays a role in the frequency and intensity of grandparental care. The author uses the 8 Countries included Austria, Belgium, Bulgaria, the Czech Republic, Denmark, Estonia, France, Georgia, Germany, Greece, Hungary, Ireland, Italy, Lithuania, the Netherlands, Poland, Portugal, Romania, the Russian Federation, Slovenia, Spain, Sweden, and Switzerland. 39 most recent available data from SHARE and GGS, focusing on the 50–79 age group. Following Hank and Buber-Ennser (2009), a logit model is estimated for grandparents who reported to have looked after a grandchild, including variables on the grandparent characteristics, age of the grandchild, and if available, information about the parent of the grandchild who is related to the grandparent. Grandparent characteristics are age (three categories: 50–59, 60–69, and 70–79 years); partnership status (indicating whether the respondent lives in a union); employment status (working vs. non-working);9 and health (a binary indicator indicating any ADL limitations or not). The model is estimated for grandmothers and grandfathers and for two classifications of care: the provision of any help and the intensity of such help, with a focus on regular help or help provided ‘almost weekly or more often’ compared with ‘any help’ or help provided less frequently than that. Full tables are in Annex 2, Tables 8–10. Figure 23: Grandmothers and grandfathers who provided any childcare by country (in percent) 100 90 81 77 80 74 75 75 76 71 70 64 64 65 65 66 66 63 60 56 57 57 58 59 59 Percent 49 49 50 46 40 40 30 20 10 0 Austria Romania Spain Georgia France Italy Greece Denmark Poland Hungary Czech R. Netherlands Ireland Germany Bulgaria Slovenia Russia Total Lithuania Estonia Belgium Portugal Switzerland Sweden Grandmothers Grandfathers Source: SHARE Wave 4 (AT, BE, CH, CZ, DE, DK, EE, ES, FR, HU, IT, NL, PO, PT, SE, SI) and wave 2 (GR, IE). GGS Wave 1 (GE, RO, RU) and Wave 2 (BU, LT). N = 27,708 grandparents aged 50–79 years with at least one grandchild under the age of 16 years. Note: Countries are sorted in ascending order for grandmothers. 44. Across the 23 selected countries, one of three grandparents provided regular care for a grandchild in the year before the interview. Four out of ten grandparents never looked after a grandchild, and one out of four helped occasionally. Among grandmothers, only 37 percent never looked after a grandchild. Grandfathers’ providing childcare is less frequent and for smaller periods. About half of them report not providing any childcare and 25 percent do so only occasionally. Age is a critical factor for 9 Not working includes unemployed, housekeepers and others. 40 grandparental care for both men and women. Younger grandparents might be active in the labor market, and older grandparents’ frailty and health limitations might shift them into the category that needs care. Before 60 years of age, between 70–60 percent of grandfathers and 60–40 percent of grandmothers were employed. While younger grandparents have a higher involvement in childcare than those over 70 years of age, before retirement age (63–65 years depending on the country) involvement is substantially higher among grandmothers than grandfathers. However, grandmothers in their sixties provide significantly less frequent childcare as those in their fifties, but higher than grandfathers at all ages. 45. Grandparents with grandchildren under the age of 6 more often provide care compared to those with grandchildren at school age. The finding of descriptive analyses revealing that grandparents with grandchildren below school age more often provide regular care compared to those with grandchildren at school age is confirmed in the multivariate framework. Intensity of childcare sees almost no variation by the age of the youngest grandchild, but provision of grandparental childcare is the highest for grandchildren aged 3–5 years, probably reflecting shortages of formal childcare supply. 46. Under all circumstances, grandmothers are more likely to provide grandchild care than grandfathers, and they will do so regardless of the presence of and competing demands on their time. Employment has no statistically significant effect on the provision of grandparental help among women, regardless of the intensity of such care. Employment status is significantly associated with provision of regular care among grandmothers. Employed grandmothers provide care to about the same extent than non- employed grandmothers, but labor market affects the intensity of their care. Employed grandmothers show a lower frequency of regular care than non-employed grandmothers; in particular for younger grandmothers (50–59 years of age), labor force participation is important for providing regular care. The presence of a co-residing partner (as another factor affecting women’s time availability) will affect the levels of childcare provision by grandmothers, but they continue to provide more care than grandfathers in the same situation (67 vs. 58 percent). Distance from grandchildren’s house, as another factor affecting time, reduces the levels of grandparental care, but the difference between grandmothers and grandfathers remain. Among grandparents living up to 5 kilometers away, 72 percent of grandmothers and 62 percent of grandfathers report caring for a grandchild, incidence that goes down to 67 and 58 percent, respectively, as distance increases between 5–25 kilometers. 47. Georgia, Austria, Greece, the Russian Federation, Bulgaria, Italy, and Romania are the countries were the highest intensity of grandmothers’ care can be observed. In these countries, grandmothers help out with grandchildren almost weekly or more often (Figure 24 shows the estimated country coefficients for the two types of provision of care identified). Nordic countries, France, and Estonia are at the opposite end in terms of care frequency and intensity.10 In Sweden and Denmark, grandmothers are more likely to provide sporadic than frequent care. Notably, the Russian Federation has by far the highest share of grandmothers looking after grandchildren (81 percent), regardless of intensity of the care. Romania, on the other hand, sees lower frequencies of grandmother’s care but at high intensity when it happens. Georgia has the highest share of grandparents living with a grandchild (45 percent), and although 10 Nordic countries as well as France have a good structure of family-friendly policies, from extended leave to childcare, and grandparents might be complementing institutional care but not substituting it. 41 three out of four provide childcare if co-residing, Georgian grandmothers have a significantly lower propensity to provide grandparental childcare than other countries. Figure 24: Estimated country coefficients for grandmothers’ care provision (any and regular grandchild care) 1.5 1 *** *** *** ** 0.5 *** ** * ** 0 * Any care *** -0.5 *** *** *** *** *** Regular care -1 *** -1.5 Germany (DE) ***Denmark (DK) Czech R. (CZ) Ireland (IE) ***Bulgaria (BU) Netherlands (NL) ***Romania (RO) Belgium (BE) Lithuania (LT) Greece (GR) ***France (FR) Switzerland (CH) Spain (ES) Italy (IT) ***Estonia (EE) Slovenia (SI) Poland (PO) Austria (AT) Hungary (HU) Portugal (PT) Georgia (GE) ***Russia (RU) ***Sweden (SE) Source: Buber-Ennser (2014) based on SHARE Wave 4 (AT, BE, CH, CZ, DE, DK, EE, ES, FR, HU, IT, NL, PO, PT, SE, SI) and Wave 2 (GR, IE). N = 27,708 grandparents with at least one grandchild under the age of 16 years and N = 16,360 that provide any care. Note: Estimated coefficients refer to the multivariate model controlling for age, partner status, employment status, health, co- residence with grandchildren and youngest grandchild’s age. ***, **, * indicate statistical significance of the estimated coe fficient for grandmothers (* p < 0.05; ** p < 0.01; *** p < 0.001). Significance marks next to the country name indicate that coefficients are significant for both specifications. Eldercare and upward generational flows of care 48. Children’s support to elderly parents is the most frequent informal care flow from younger to older generations. Most informal care of older people is provided by their partners and children, a decision determined by filial obligation, the family’s capacity to provide care, and the elder’s own preferences (Kalwij et al. 2014; Murphy, Martikainen and Pennec 2006; Litwin 1997; Finch and Mason 1990). As with childcare, and mirroring the gender distribution of tasks, women are the most frequent providers of eldercare—from wives to daughters and daughters-in-law (Bettio et al. 2006; Saraceno and Keck 2010). Even in the presence of siblings, when elderly parents require support with activities of daily living or nonprofessional health care, daughters are more likely to be the care provider than sons (Dwyer and Coward 1991; Lee et al. 1993; Haberkern and Szydlik 2010). 42 49. Of the households surveyed in the supply and demand assessment, respondents (of which two- thirds were women) were the majority of eldercare providers, and elder mothers were the majority of eldercare receivers. Even as, on average, we also find that there is more than one person in the household providing care for the elder relative (the ratio of caregivers per eldercare recipient is 1.61 for the group of countries), the additional care provider is likely to be a woman. The percentage of female caregivers is 63 percent among respondents and 43 percent among caregiving partners of respondents looking after elder household members. Respondent and partner caregivers are more likely to fall in the 40–49 year age group (32 percent). Regarding those who receive eldercare, for 32 percent of households, the care recipient is female. 50. Women who are eldercare providers are different from childcare providers in a series of characteristics, with age being the most noticeable. Following the original life-cycle diagram, eldercare needs reach women at a later stage in life. Overall, analyses from the supply and demand assessment (Figure 26) are in line with evidence from GGS for a set of countries (Figure 25), indicating that respondents who provide eldercare fall into older age groups, as parental support starts increasing with age (the reverse of what happens to care provided to other family members, which tends to decline). Figure 25: Care provided to parents and/or parents-in-law among women who report providing help to someone in the previous year. Female respondents, by age 80 providing any care to someone else 70 % of women in the age group 60 50 Russia 40 Bulgaria Georgia 30 Romania 20 10 0 24 and less 25-35 35-44 45-54 55-64 65+ Source: GGS (Wave 1 for the Russian Federation and Romania, 2 for Bulgaria and Georgia). 51. Demographic and household structure also tends to differ between respondents who report providing elder care to relatives and those who report not providing care (see Figure 26). A larger percentage of single respondents are elder caregivers (compared to non-caregivers) in our supply and demand assessment. A smaller percentage of respondents in a union or those in cohabitation with a partner provide care for an elder relative. Household size and the percentage of elder members are greater among households with a respondent who provides elder care, confirming that cohabitation increases the likelihood of care. Households with eldercare provision also have a larger number of adult women in the household as well as a larger share of employed adults. 43 52. In terms of time allocation, as with childcare, individuals who have fewer competing demands for their time are more frequently found as eldercare providers. Respondents who are not in the labor force are more likely to be eldercare providers, and those who also provide childcare are less likely to provide eldercare. Figure 26: Difference of demographic and household characteristics between respondents who are non-elder caregivers and elder caregivers Percentage of adult employed ty Characteristics Employment nsibili Household Percentage of adult women Percentage of children (14 and younger) Percentage of elder (55 and over) Household size respo Respondent provides adult care Care Respondent provides child care Full time work Paid work Status Not in Labor Force Unemployed Employed Cohabitant partner Marital status Not in union In Union Single Tertiary Se Educatio Secondary/High school n No education/Primary Male x 60 and over 50-59 yrs old Age 40-49 yrs old 30-39 yrs old 21-29 yrs old -0.2 -0.2 -0.1 -0.1 0.0 0.1 0.1 0.2 0.2 0.3 0.3 Source: Child and Elder Care Survey. Individual Survey. Note: Green represents that t-test is statistically significant. Orange represents that there is no statistically significant difference. 53. SHARE allows us to see how, for a different group of countries, eldercare looks from the perspective of the care recipient and not just the caregiver. Besides sex of the recipient, age, health, and marital status correlate with higher likelihood of receiving social support and higher intensity of that support. Women are more likely to receive some care and support (nonfinancial) from their adult children, so do parents that are older, do not have a partner, and have some limitations with their daily activities. Recipients’ socioeconomic status is also an important correlate for receiving social support from children. Parents with lower educational levels and fewer economic resources are more likely to receive informal care and they also tend to receive more hours of support. Figure 27 and Figure 28 present the 44 results in terms of characteristics of the respondent for being eldercare receivers from a child and the intensity of such care.11 54. Parents in poor health are more likely to receive support across all countries. Country differences can also be observed, with the Czech Republic having the highest likelihood of receiving care from children and high intensity of such care. The intensity of care provision is particularly high in Spain where the median value in the sample is of more than 600 hours of support per year. Poland sees 308 hours per year, while in Austria, Germany, and France, the median respondent receiving support from children receives around half of that, with 150 hours in the twelve months before the survey. 55. While most respondents of SHARE are above 50 years of age, they do provide eldercare support. Albertini’s (2014) analysis of SHARE finds that among the donor characteristics positively associated to provide support are sex and income. Daughters are significantly more likely to provide help to parents than sons, with daughters from lower-income elders being more likely to provide informal care to their elderly parents. More than one-third of those whose parents are still alive provided some level of care in Germany, Belgium, the Netherlands, Denmark, and Sweden. In France, Switzerland, and Italy this rate goes down to about one-fourth of the respondents. Among new countries in the last wave of SHARE, including Estonia, Hungary, and Slovenia, less than 10 percent of respondents provided help to their parents. 11 Results obtained from random effects logistic regression models on the unbalance sample of individuals taking part in at least one of the three regular waves of SHARE. For more details see Albertini (2014). 45 Figure 27: Participation of children in social support to Figure 28: Intensity of social support provided to SHARE respondents SHARE respondents by participating children (hours) Estonia ** *** Poland *** * ** Slovenia ***** *** Czechia ** Portugal *** * *** *** Hungary Belgium * ** Poland *** Switzerland *** *** Czechia Country (ref. Austria) Country (ref. Austria) *** *** *** Greece Belgium *** Switzerland *** Denmark *** *** *** *** *** Greece France ** Denmark ***** *** Italy ** France ****** * *** *** Italy *** *** Spain ** * *** *** Spain *** *** Netherlands *** *** *** Netherlands *** ***** Sweden *** *** Sweden *** * * *** Germany *** *** Germany *** LN real wealth ** Income & wealth *** *** *** LN real wealth *** *** Income & wealth *** *** LN financial wealth ** *** LN financial wealth *** *** LN other welfare transfers *** *** *** *** * *** LN other welfare transfers LN pension incomes *** *** *** *** LN pension incomes *** * ** LN other incomes *** *** LN other incomes Employ Other condition status Health Educational level status ment *** 50-65) p status th Educational level ent status * Employm Employed, self-employed,… *** *** ** Other condition ** *** ** Employed, self-employed,… *** *** Missing info ****** *** *** *** ISCO 5-6 ****** * Missing info * *** *** ISCO 3-4 ****** * *** ISCO 5-6 ** *** *** * *** ISCO 1-2 *** *** * * *** ISCO 3-4 *** *** *** **** Respondent has limitations *** ****** ISCO 1-2 ** *** *** * ****** Partner *** ****** Age (ref Partnershi Heal Has partner who needs… *** *** Respondent has limitations *** ship *** *** *** Has not partner ****** *** *** *** *** ** *** Has partner who needs… ** *** *** Age (ref 50-65) *** *** ** *** 81+ *** *** *** Has not partner *** *** *** 66-80 ****** *** ** *** *** *** 81+ *** *** *** *** *** 66-80 ** * Female (ref. male) *** ** *** Female (ref. male) -2.0 -1.0 0.0 1.0 2.0 3.0 -2 -1.5 -1 -0.5 0 0.5 1 1.5 ECA_EU13 EC15_Southern EC15_NCE Women Men Full sample EC15_Southern EC15_NCE Women Men Full sample Source: Albertini (2014). Note: Results are conditioned on having at least one child alive and residing closer than 500 kilometers. Intensity could not be estimated for ECA_EU13 countries due to sample size. For employment, the reference is ‘pensioner’. For education, the reference is ‘no education’ (ISCO 0). For partnership status, the reference is ‘has partner with no limitations’. 46 Caught in the middle: Simultaneous intergenerational flows of care 56. Population aging, and specifically, the combination of increased longevity and delayed onset of fertility, has given rise to what has been called a ‘sandwich generation’—women who are expected to provide care simultaneously to multiple generations, normally their parents and their children and/or grandchildren. The existence of a group of women, normally in their middle age has been documented for the United Kingdom and the United States (Grundy and Henretta 2006) using data from 1988 for the United Kingdom and 1998 for the United States, and they find that one-third of women between 55 and 69 years reported providing help to members of both generations. Using SHARE data, Buber-Ennser (2014) shows that multiple tasks are a common feature among many individuals over 50 years of age. In the age group 50–59 years, eight out of ten grandmothers provide either grandparental childcare, help other relatives or friends, or even provide both these types of social support. Thirty-three percent of women between 50 and 59 years provide childcare as well as help to others; a little lesser than two-thirds of those women are, at the same time, employed and provide additional help to others; and 23 percent of women in their sixties also provide care to more than one person; while at lower levels, 24 percent of men in their fifties provide multiple care, and 18 percent of those in their sixties do so (Figure 29). 57. Across countries, women over 50 years of age providing care in both directions range from 21 percent in Slovenia and Spain for the 50–59-year age group, to over 40 percent in the Netherlands and Belgium. Albertini (2014), using all waves of SHARE, notes that sandwich care obligations for the 50+ age group are higher in Nordic countries. Even when excluding co-residence, between one-third and one-fourth of respondents provide intergenerational care up and down in Sweden, Denmark, the Netherlands, and Belgium. In other countries—mostly continental Europe as well as the Czech Republic and Poland—the size of the sandwich generation is between 10 and 20 percent of those who have at least one surviving child and parent. However, notably, Figure 30 shows, when focusing on specific age groups, women in their fifties in the Czech Republic are the most ‘sandwiched’, with 36 percent of them providing care both ways. 58. Married women out of the labor force and higher-educated, higher-income women are at the highest risk of being caught in a care sandwich. Different from the case for other forms of care, those women who are more educated, have higher market incomes (that is, incomes not from pensions or welfare transfers), and belong to richer quintiles in terms of income distribution among respondents are at higher risks of being caught in between intergenerational demands of care from younger and older generations (Albertini 2014). This finding is particularly relevant since as care demands increase, the cost of devoting time to other activities increases as well. Reduction in hours of work or labor force dropout might happen, even more so if caregivers are close to retirement age. 47 Figure 29: Multiple tasks of grandparents by age and gender (in percent) Women 50-59 19% 19% 5% 8% 14% 18% 3% 13% Men 50-59 19% 19% 10% 16% 5% 11% 3% 17% Women 60-69 3% 4% 2% 20% 38% 8% 23% Men 60-69 2% 6% 5% 16% 36% 8% 26% Women 70+ 13% 26% 10% 51% Men 70+ 11% 29% 12% 47% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Source: SHARE Wave 4 (AT, BE, CH, CZ, DE, DK, EE, ES, FR, HU, IT, NL, PO, PT, SE, SI) and Wave 2 (GR, IE). N = 19,090 grandparents aged 50+ years, with at least one grandchild under the age of 16 years. Figure 30: Multiple tasks of women aged 50–59 years with young grandchildren by country (in percent) Austria (AT) 15% 21% Belgium (BE) 31% 15% Czech R. (CZ) 21% 20% Estonia (EE) 19% 10% France (FR) 21% 12% Hungary (HU) 17% 24% Italy (IT) 6% 39% Netherlands (NL) 26% 14% Slovenia (SI) 10% 39% Spain (ES) 6% 17% Switzerland (CH) 20% 7% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Employed, providing childcare, providing further help to others Employed, providing childcare, not providing further help to others Employed, not providing childcare, providing help to others Employed, not providing childcare, not providing further help to others Not employed, providing childcare, providing further help to others Not employed, providing childcare, not providing further help to others Not employed, not providing childcare, providing help to others Not employed, not providing childcare, not providing further help to others Source: SHARE Wave 4 (AT, BE, CH, CZ, EE, ES, FR, HU, IT, NL, SI). N = 2,787 grandmothers with at least one grandchild under the age of 16 years. 48 IV. How does caregiving affect women at different stages of the life cycle? 59. Given the substantial time demands involved in regular informal provision of childcare and eldercare, negotiating the balance of care responsibilities and labor force participation is challenging for most women. Household production theory divides the time of household members into three categories: market work, household work, and leisure (Becker 1965). Expansion of household-related responsibilities due to rising childcare and eldercare needs implies less time for the other two activities. Women bear a disproportionate share of the burden in terms of household-related activities, and since their leisure time is relatively scarce (in comparison to men), the time for paid market work can decrease to compensate for the rise in care duties. Fulfilling the expected role of primary caregiver for children and/or elderly can thus affect women’s ability to be engaged in paid work, both on the extensive and intensive margins. It can also reduce their productivity due to potential effects on health and human capital accumulation. Career interruptions and reduction in labor market attachment due to periods of intensive caregiving can imply significant drops in caregivers’ economic well-being due to lower lifetime earnings and higher vulnerability to poverty in later years due to lower pension accumulation.12 60. While childcare and women’s labor supply has been discussed more widely, the specific effect of eldercare on care provider’s labor market outcomes has been less so. However, the effects on caregivers’ employment has been documented recently by Geyer and Korfhage (2014) for Germany, where they report that half of caregivers in 2010 had stopped altogether or reduced their paid work in favor of care work; by Carmichael and Charles (2003) for the United Kingdom; by Lilly et al. (2010) for Canada; and Van Houtven et al. (2013) for the United States. All of them, as well as Ettner (1995), observe that women’s labor supply is the most affected.13 Caregiving and labor force attachment: childcare 61. The impact of rising care duties on the time women devote to paid work can take the form of lower labor force participation or lower work intensity. The effect of rising care duties on female labor supply can take on numerous forms. Women can withdraw from the labor force altogether, thereby being affected on the extensive margin, or they can reduce working hours (for example, by starting to work part time or by requesting flexible work arrangements) or switch to jobs that are less time intensive, implying an intensive margin effect. 62. There is a well-documented decrease in labor supply for women with young children, and in some countries, this reduced labor supply of mothers persists for an extended period. Presence of young children in a household is often associated with a reduction in women’s time spent working for the market (Browning 1992; Hill et al. 2004; Nakamura and Nakamura 1992). There is evidence of lower labor force participation rates for young mothers in many contexts around the world, including in ECA. As can be observed in Figure 31, ECA countries have lower employment probabilities for mothers of young children compared to Western European countries. Moreover, while overall European mothers’ 12 For example, multiple-generation households may distribute care duties in a way that substitutes paid work of some female members (for example, grandmothers) for that of others (for example, mothers of young children). 13 Recent studies also suggest that availability of intergenerational social support (such as grandparental childcare) can also factor in individuals’ fertility decisions (Aassve et al. 2012a; Harknett et al. 2014; Raymo et al. 2010; Hank and Kreyenfeld 2002). 49 employment rates increase as the age of their youngest child increases, this pattern is far more pronounced in ECA countries, such as the Czech Republic, Hungary, Georgia, and Romania. Although employment rates of mothers with young children could be affected by the length of maternity leave available in each country, mothers’ labor supply does not always bounce back when children graduate from infants to toddlers. In fact, according to Keck and Saraceno’s (2013) analysis based on EU Statistics on Income and Living Conditions (SILC) data for mothers whose youngest child is aged 3–12 years, in some European countries (Austria, Cyprus, the Czech Republic, Germany, Ireland, Italy, Luxembourg, the Netherlands, Spain, and the United Kingdom), being a mother is correlated both with lower likelihood of being in paid work as well as with fewer working hours. In other countries, including Nordic countries as well as Belgium, France, Greece, and Hungary, the reduced labor supply effect related to motherhood is only on the intensive margin. Figure 31: Employment rate of mothers by age of their youngest child 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0-3 years 4-6 years 7-9 years 10-12 years Bulgaria Georgia Germany France Hungary Italy Netherlands Romania Norway Austria Belgium Lithuania Poland Czech Source: Authors' calculations based on GGS Wave 1 data. Note: The sample includes female respondents with at least one child in the household. The employment rate is the share of those who report that they are employed or self-employed. 63. The relationship between female labor supply, education, and caregiving is complex and context-specific. On the one hand, one would expect women with higher earning potential (for example, those with relatively higher education) to be less likely to trade market work for care duties due to higher opportunity cost of time (Fong and Lokshin 2000). On the other hand, women with lower education might not be able to afford to withdraw from the labor market, especially after they have children. For example, in Turkey, Aran et al. (2008) find, based on Demographic and Health Survey (DHS) data, that more urban 50 high-skilled women drop out of the labor force after giving birth than rural women or low-skilled women from the same urban areas. Moreover, for the high-skilled women, another significant drop in labor force participation rate can be observed after the birth of the second child, a substantial increase in the employment rate of rural women occurs after the birth of each child. Similarly, 2010 Eurostat data shows most European countries having higher shares of lower-educated women aged 25–49 years who do not stop work (beyond the mandated maternity leave) to care for their youngest child under 8 years, compared to higher-educated women. Indeed, women with less than tertiary education are also more likely to experience extended career interruptions (over 12 months) due to childcare duties (see Figure 32), with the highest rate of extended work stoppages among three EU-13 countries—Romania, Slovakia, and Estonia.14 Figure 32: Career interruptions of over 12 months due to childcare, by education Source: Eurostat (2010). 64. Access to formal childcare (or lack thereof) correlates with female labor force participation in ECA. As can be observed in Figure 33, childcare in ECA is mostly used by mothers who are working full-time or part-time, although there is considerable variation across countries. On one side of the spectrum, there is Slovenia, where 75 percent of mothers aged 25–49 years using childcare are working full-time and another 9 percent are working part-time and FYR Macedonia, where slightly more than 40 percent of childcare users are working. From the supply and demand assessment data, we can also see strong differences between Kosovo (where 90 percent of surveyed facilities had at least 80 percent working mother clientele) and Armenia (where very few facilities had a majority of working mothers as clientele). Whereas working women tend to use childcare services, some women who are not working or are working part-time cite childcare-related reasons for their reduced labor supply. Eurostat data demonstrate that mothers in EU- 14 In their analysis of European countries, Keck and Saraceno (2013) find that in the majority of countries in the study, lower-educated women do not experience an additional penalty in terms of labor force participation upon becoming mothers; however, in a few countries (Cyprus, Italy, Portugal, Spain), there was this additional penalty for the extensive margin and in the Netherlands, for the intensive margin. 51 13 countries are more likely to stay out of the labor force due to childcare availability and quality and more likely to be in part-time employment due to availability and cost compared to EU-15 benchmarks (Figure 34). Availability concerns are especially prominent in Latvia (cited by 6 percent of mothers as the reason for not working) and the Czech Republic (mentioned by 6 percent of part-time employed mothers); quality problems are highest in Hungary (reported by 2.5 percent of inactive and unemployed mothers); while cost issues prevail in Romania (where 11 percent of mothers cannot afford the childcare needed to increase their labor supply). 65. There is rich evidence that increased availability of formal childcare options results in improved labor force participation of women in many different contexts—in Brazil (Deutsch 1998; de Barros et al. 2011); in rural Colombia (Attanasio and Vera-Hernandez 2004); in urban Argentina (Berlinski and Galiani 2005); in Japan (Asai et al. 2014); and in Canada (Lefebvre and Merrigan 2008). Closer to the ECA region, Del Boca and Locatelli (2006) used data from the European Community Household Panel to show that female labor force participation is affected by the availability, and even more importantly, affordability of childcare. The issue of childcare cost is especially relevant in former Communist economies in which this cost has changed dramatically over the preceding decades; indeed, it may be the determining factor of whether young mothers return to work in these countries (Todd 2013). Fong and Lokshin (2000) examined the relationship between female labor supply and the cost of paid childcare in Romania between 1989 and 1995 and found that both female labor force participation and the decision to use paid childcare were sensitive to the price of childcare. In the Russian Federation, Lokshin (1999) used policy simulations based on panel household survey data to show that providing subsidies for paid childcare increased maternal employment by almost twice as much as comparable wage subsidies. Besides this extensive margin effect, childcare subsidies increased the amount of time working mothers spent at work and were more effective in raising the overall family income than any other policy intervention examined in the study. It is important to note that access to childcare can affect male labor market outcomes as well as female labor supply. Calderon (2012) examined the impacts of a Mexican government-provided childcare program and found that it not only increased female labor employment rates and earnings but also enabled men to spend time searching for better paid jobs. 52 Figure 33: Working mothers as clients of childcare facilities a) Percent of working mothers in all childcare facilities Share employed full-time Shared employed part-time 90% childcare services weekly for their 80% % of women aged 25-49 who use 70% 60% 50% youngest child 40% 30% 20% 10% 0% Slovakia Greece Spain France Iceland Latvia Austria Denmark Hungary Romania Italy Norway Finland Poland Croatia Bulgaria Slovenia Netherlands Ireland Cyprus Germany Czech Republic Malta Luxembourg Estonia Belgium Lithuania Sweden Portugal FYR Macedonia United Kingdom ECA non-ECA Source: Eurostat (2010). b) Percent of facilities with different intensities of working mothers 100 90 80 70 60 50 81-100% of clients are working mothers 40 61-80% 30 41-60% 20 10 21-40% 0 1-20% 0% of clients are working mothers Source: Independent field data (2014). Note: The percentages of clients who are working mothers are based on responses from representatives of childcare facilities to the following question, “What percentage of mothers (whose children receive care here) are employed (‘working mothers’)?” 53 Figure 34: Childcare-related reasons for reduced labor supply Not working Working part-time 6.00% 7.00% % of inactive and unemployed women aged 25-49 that cite childcare-related reason for not working % of part-time employed women aged 25-49 that cite childcare-related reason for working part- 6.00% 5.00% 5.00% 4.00% 4.00% time 3.00% EU-13 EU-13 3.00% EU-15 EU-15 2.00% 2.00% 1.00% 1.00% 0.00% 0.00% Childcare Childcare Childcare Childcare Childcare Childcare Cost Availability Quality Cost Availability Quality Source: ‘Employment and Unemployment (Labour Force Survey)’, Source: ‘Employment and Unemployment (Labour Force Survey)’, Eurostat (2010). Eurostat (2010). Note: The sample is restricted to women, aged 25–49 years, who are Note: The sample is restricted to women, aged 25–49 years, who are inactive or unemployed. For the bars representing ‘childcare cost’, employed part-time. For the bars representing ‘childcare cost’, the the numerator is the number of these women who reported that they numerator is the number of these women who reported that they work do not work because of childcare costs, and the denominator is the part-time because of childcare costs, and the denominator is the total total sample. For the bars representing ‘childcare availability’, the sample. For the bars representing ‘childcare availability’, the numerator numerator is the number of these women who reported that they do is the number of these women who reported that they work part-time not work because of childcare costs, and the denominator is the total because of childcare costs, and the denominator is the total sample. For sample. For the bars representing ‘childcare quality’, the numerator the bars representing ‘childcare quality’, the numerator is the number of is the number of these women who reported that they do not work these women who reported that they work part-time because of because of childcare quality and the denominator is the total sample. childcare quality and the denominator is the total sample. 66. Informal childcare support, usually provided by grandmothers, may also enable young mothers to remain productively employed in the workplace. As shown in the section on care flows, young mothers may sometimes have an alternative to formal childcare by relying on time transfers from their family members, most often their mothers. GGS and fieldwork data demonstrate that informal care provision by relatives and other caregivers for whom childcare is not a job is used mostly by women who are either working or searching for a job (Figure 35). Moreover, in every country group, mothers with unpaid informal childcare help have a higher activity and employment rate than those without any (formal or informal) help. Of course, the challenge in identifying the impact of informal grandparent-provided childcare on mothers’ labor supply is the endogeneity of childcare provider choice. Many studies attempt to overcome this issue by using instrumental variables and different estimation strategies, and there is evidence that grandparent-provided childcare allows young mothers to return to the labor force and to spend more time in paid work. Posadas and Vidal-Fernández (2012) use instrumental variable analysis to estimate that grandparent-provided childcare improves mothers’ labor force participation in the United States by 15 percentage points. Dimova and Wolff (2011) use SHARE data to find a strong positive impact of grandparent-provided childcare on mothers’ labor force participation and propensity for full-time work across Europe. Unobserved preferences with regard to grandparent-provided childcare make it challenging to determine a causal relationship between this type of care and mother’s labor force participation, as these preferences can also be correlated with women’s preferences for engaging in paid work. Aassve et al. (2012) use instrumental variables to get around the endogeneity problem; based on the GGS, they find that 54 grandchild care increases young mothers’ labor supply in some countries (Bulgaria, France, Germany, and Hungary) but not in others (Georgia, the Russian Federation, or the Netherlands). Figure 35: Labor force status for mothers with and without informal care Employed Unemployed Inactive in LF 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Unpaid help No help Unpaid help No help Unpaid help No help Unpaid help No help EFTA, EU15 EU13 Eastern Parternship W. Balkans Source: Authors’ calculations based on GGS data (most recent wave for Bulgaria, the Russian Federation, Georgia, Romania, Lithuania, Poland, the Czech Republic, Germany, France, the Netherlands, Norway, Austria, and Belgium) and fieldwork data (2014 data for Armenia, Bosnia and Herzegovina, Kosovo, FYR Macedonia, Serbia, and Ukraine). Note: Sample restricted to female respondents, who live in a household with at least one child under 14 years of age. For the bar on unpaid help, the sample was restricted to those who receive regular help with childcare from relatives or others for whom caring is not a job. For the bar on no help, the sample was restricted to those who do not receive regular help with childcare either from relatives or others for whom caring is not a job or from a day-care center, nursery or preschool, after-school care center, self- organized childcare group, or other institutional or paid arrangement. 67. However, provision of grandchild care can also affect the labor supply of grandparents, in particular grandmothers. While grandparent-provided childcare can increase the time allocated to paid work by young mothers, this can come at the expense of labor supply of grandparents themselves. This is particularly the case in some countries in the ECA region with an ongoing or only recently completed fertility transition (implying a still relatively young age at first birth) combined with relatively low statutory retirement ages for women. These two factors imply that grandmothers in these countries are quite young, often in their mid-50s. Whether the substitution of their labor supply for that of young mothers is a positive or a negative phenomenon depends on the relative productivity of the two groups. In the United States context, Zamarro (2011) finds that grandparents’ labor force participation and grandchild care are negatively correlated, and Ho (2008) shows that retirement decisions of grandmothers take into account the needs for grandchild care. Using the European Social Survey and Multilinks database for 22 countries in Europe, Van Bavel and De Winter (2013) found that becoming a grandparent speeds up retirement, especially for women aged 55–60 years. Buber-Ennser (2014) compared employment rates of potential providers of grandparental childcare (that is, individuals with grandchildren under 16 years of age) to people without young grandchildren, finding employment was higher for the latter group, especially for men aged 50–54 years and for women aged 55–59 years (Figure 36). Focusing on grandparents, she finds a negative 55 association between the employment of grandfathers aged 60–69 years and their provision of any grandparental childcare; while employed grandmothers in the 50–59 years age group are less likely to provide regular (almost weekly or more frequent) care for grandchildren (See Tables 8–10 in the Annex 2). Figure 36: Employment rate of older individuals with/without grandchildren, by sex and age 90% 80% 70% 60% % employed 50% 40% 30% 20% 10% 0% 50-54 55-59 60-64 65-69 70-74 75-79 50-54 55-59 60-64 65-69 70-74 75-79 Men - Women With grandchildren under 16 Without grandchildren under 16 Source: Author’s calculation based on Buber-Ennser (2014). Caregiving and labor force attachment: eldercare 68. As with childcare, intensive eldercare duties can reduce female labor supply during the most productive years. The negative effect on (mostly) middle-aged women’s labor supply starts to predominate when their parents become frailer and less able to take care of not only grandchildren but also their own activities of daily living. Indeed, Albertini et al. (2007) document that the largest amount of intergenerational time transfers are provided by middle-aged (45–65 years old) women to both the younger and older generations within the family. This peak of inelastic care responsibilities usually coincides with women’s most productive middle-age years, implying that reduction in labor supply during this time can leave the biggest dent in lifetime income. In contrast with childcare duties, the lack of predictability in timing, duration, and intensity of eldercare (Keck and Saraceno 2009) makes it more difficult to negotiate a combination with paid employment, particularly full-time. There is a substantial body of evidence, from a variety of contexts, that intensive, time-demanding care, such as that requiring more than 20 hours per week, has significant negative effect on the likelihood of staying in the labor force (Jacobs et al. 2014a; Gabriele et al. 2011; OECD 2011; Lilly et al. 2010; Bolin et al. 2008; Heitmueller and Inglis 2007; Henz 56 2006; Johnson and Lo Sasso 2000; Sarasa 2006; Carmichael and Charles 1998).15 Focusing on the cohort effect, Jacobs et al. (2014a) find that the impact of caregiving on labor force participation was statistically the same for pre-Baby Boomer and Baby Boomer women. Some researchers find that the adverse effect of informal caregiving on employment is stronger for countries like Sweden and Norway due to a lower societal acceptance of family care, which in turn reduces the acceptance by employers of a care-work combination. In Central European countries, caregiving has an impact on the number of hours women work but not on their labor force attachment (Bolin et al. 2008). Also focusing on the intensive margin and using European Community Household Panel data, Spiess and Schneider (2003) demonstrate that a negative effect on work hours for women who start or increase caregiving does not reverse when caregiving is reduced. 69. Residential proximity and intensity of care needs mediate the relationship between caregiving and labor supply. The effect of care demands on women’s labor supply varies depending on co-residence and proximity as well as the health status of the elder in need. OECD (2011), Crespo and Mira (2010), and Casado et al. (2011) show that opportunity costs in terms of paid work are higher for women who co-reside with the elderly than for those who care for someone who does not live with them and changes depending on the intensity of the care demanded. Casado et al. find more significant impacts in specific combinations of daughter characteristics and parental disability conditions, such as low education levels of daughters and mental frailty of the elderly. The extent of the care provided also matters—longer periods of care have negative results on labor force participation as well as on intervals of starting and stopping care. Jacobs et al. (2014b) examine the association between informal caregiving and women’s decision to retire in the United States context; they find no significant relationship for all caregivers but a 3 percentage point higher retirement propensity for women who provide 20 or more hours of informal care per week. 70. Greater availability of formal eldercare options can be expected to affect female labor force participation, although evidence on this topic is so far limited. To date, given the incipient state of formal eldercare in many countries around the world, there is far less empirical evidence on the effectiveness of formal eldercare options in raising female labor force participation rate or work intensity. Heger (2014) uses SHARE data to look at caregivers’ employment and finds caregiving decreases employment rates in countries with low supply of formal care (or ‘family care countries’) by 34 to 60 percentage points depending on the frequency of care but has no impact on caregivers’ employment probability in countries with more established care systems. Earlier, Viitanen (2007), using the European Community Household Panel to simulate the effect of greater public expenditure on formal residential care and home-help services for the elderly, found a positive effect on the employment rate of 45–59-year-old women by 9–13 percentage points across Europe. Loken et al. (2014) examine a 1998 expansion of local, home-based care for the elderly in Norway, which resulted in a significant reduction of extended absences from work for adult daughters of single elderly. Geyer and Korfhage (2014) examine long-term care support in Germany and conclude that cash benefits discourage care providers from engaging in paid work, while 15 As often is the case, the relationship between caregiving and labor force status is complicated by the simultaneity problem (with the decision to provide care and to participate in paid work decided jointly) as well as unobserved individual heterogeneity problem (with individuals selecting into caregiving potentially having different labor- market profiles compared to individuals selecting into paid work). Indeed, Carmichael et al. (2010) find that participation in paid work and the level of earnings have a negative impact on the willingness to supply informal care. Thus, while caregivers are likely to exhibit lower labor force participation compared to non-caregivers, this is not necessarily due to caregiving per se (Lilly et al. 2007; Heitmueller 2007). 57 benefits given in kind (and as such better substituting for the specific time commitment of the informal caregiver) provide incentives to already caring household members to increase labor supply. These findings confirm Todd’s (2013) analysis that there are still few acceptable market-based options for eldercare in developing countries (Todd 2013) compared with childcare. 71. Across Europe, less than 15 percent of people aged 65 years and above receive home-based care and less than 7 percent reside in care institutions 72. Figure 37). Utilization rates of these formal care services are correlated with labor force participation rates of the population that is most likely to be providing informal caregiving—women aged 40–64 years, with the correlation stronger for home-based care relative to institutionalized care prevalence. This suggests that there may be a link between access to formal eldercare and female labor force participation. The fieldwork conducted for this paper was able to collect data on the prevalence of working daughters in the clientele of eldercare facilities. In contrast with findings on childcare facilities (Figure 35), working daughters are not the largest client of formal eldercare services, except in two Western Balkan countries (Serbia and FYR Macedonia), where half of eldercare facilities surveyed reported that at least four out of five of their clients have working daughters (Figure 38). Indeed, in Kosovo and Ukraine, all clients in most facilities did not have working daughters; this could mean either that all the elderly were childless or only had (unmarried) sons or that their daughters were not working. Figure 37: Formal eldercare utilization and middle-aged female labor force participation Residence in care institutions Receipt of home-based care 90.0 90.0 Labor force participation of females aged 40 to 64 Labor force participation of females aged 40 to 64 Estonia Sweden Estonia Sweden FinlandNorway Finland Norway 80.0 Denmark 80.0 Denmark Lithuania Germany Germany Czech United Czech Latvia Slovak Latvia Portugal Republic Kingdom Netherlands Portugal Netherlands Republic 70.0 Republic France 70.0 France United Kingdom Bulgaria Slovenia Slovenia Spain Spain Hungary Luxembourg Hungary Luxembourg Romania 60.0 Poland Austria 60.0 Poland Austria Ireland Belgium Ireland Belgium Greece Greece 50.0 Italy 50.0 Italy y = 1.1541x + 60.32 y = 0.9514x + 57.894 40.0 R² = 0.0614 40.0 R² = 0.1574 30.0 30.0 0 2 4 6 8 0 5 10 15 20 % living in care institutions (of persons aged 65+) % receiving home-based care (of persons aged 65+) Source: Multilinks for care prevalence (most recent wave used); Eurostat for labor force participation rates. 58 Figure 38: Working daughters as clients of eldercare facilities (% of all clients) 100 90 80 70 81-100% of clients are working daughters Percent of facilities 60 61-80% 50 41-60% 40 21-40% 30 1-20% 20 0% of clients are working daughters 10 0 Serbia FYR Bosnia and Armenia Kosovo Ukraine Macedonia Herzegovina Note: The percentages of clients who are working daughters are based on responses from representatives of childcare facilities to the following question, “What percentage of women whose parents or parents-in-law receive care here are employed (‘working women’)?” 73. The double-duty care burden on the ‘sandwich generation’ of women can have significant implications on their labor supply. The high care burden can leave little time for engaging in paid work. Indeed, sandwiched women have been found to experience higher levels of work-related stress, absenteeism as well as a greater mismatch between actual and preferred working hours (Buffardi et al. 1999; Starrels et al. 1997; Couch et al. 1999; Keene and Prokos 2007; Chapman et al. 1994 as cited in Albertini 2014). These outcomes can push women to reduce their labor supply either on the intensive or extensive margin. One empirical study finds a more nuanced situation for women in a care sandwich. Pagani and Marenzi (2008) find that in the context of Italy, being in the sandwich generation has both a positive and a negative impact on female labor force participation: while eldercare duties hinder female labor supply, the informal help received by the older relatives alleviate some of the other household responsibilities, enabling women to engage in paid work. As longevity increases in Europe beyond the delays in fertility onset, the care duties of the sandwich generation go from caring simultaneously for children and parents to caring for grandchildren and oldest-old (85+ years) parents. Examining the characteristics of the sandwiched generation for SHARE respondents aged 50 years and above, Albertini (2014) finds that higher- educated pre-retirement-age (50–65 years) women are at the highest risk of performing the double care duties. At the same time, sandwiched individuals are less likely to be in the labor force, which implies that sandwich responsibilities are associated with early retirement of highly skilled women (See Table 10 in Annex 2). 74. Caregivers’ productivity in market work can be affected by health impacts associated with the provision of informal care. Even when caregivers are participating in paid work, their productivity can be affected by the toll that intensive caregiving can take on their health. There is extensive literature on the potential health and well-being consequences of caregiving, or what has been termed the ‘caregiving effect’ (Bobinac et al. 2010; Brouwer et al. 2006), suggesting that overall, the time demands and physical and emotional stress caregivers experience can negatively affect their health. Psychological well-being of women in the sandwiched generation is at a particular risk due to the need to cope with demands of care 59 from both older and younger generations, and the need to combine work and family (Hammer and Neal 2008; Voydanoff and Donnelly 1999; Rubin and White-Means 2009; Opree and Kalmjin 2012 as cited in Albertini 2014). Caregiving and income 75. As population aging increases the value of every working-age adult, ECA countries cannot afford to underutilize a large share of women whose lifetime productivity in the labor market is currently reduced by informal care provision. As discussed above, women tend to reduce their labor supply on either the extensive or intensive margin when market, normative, and institutional forces push them toward fulfilling their caregiving mandate in the household. Career breaks during the most productive middle-age years and potential health effects of caregiving can result in decreased productivity upon return to the workplace, which, in aggregate, can depress output and future growth. As shown above, women are expected to provide care to younger and older generations throughout their adult lives. Population aging implies that the peak of these care responsibilities coincide with women’s most productive middle -age years. Career interruptions or reductions in work hours can have a permanent negative impact on women’s lifetime income, affecting their households’ current living standards and human capital investments as well as future well-being due to reduced pension wealth and damaged health. 76. There is evidence that caregivers obtain lower wages, which can discourage their labor supply. There is substantial empirical evidence of the ‘motherhood penalty’ in terms of perceived competence and salary offers. Becker (1985) proposes that this penalty arises due to young mothers’ lower work effort or productivity as a response to childcare duties. However, Anderson et al. (2003) cast doubt on this hypothesis by finding a non-monotonic relationship between women’s skill level and the magnitude of the penalty; the authors posit that the penalty arises more due to working schedule inflexibility. Moreover, Correll et al. (2007) use a candidate-rating lab experiment that controls for information on past work performance to demonstrate that, relative to equivalent childless candidates, mothers are perceived to be less competent and less committed, less likely to be recommended for promotion or for management roles, and likely to be offered lower starting salaries. Given such prospects, young mothers may preemptively withdraw from the labor force or reduce their career ambitions until their children grow older. Informal eldercare providers may also experience a wage penalty. Carmichael and Charles (1998) find that in the United Kingdom, provision of informal care for more than 20 hours a week reduced the hourly wage rate by about 10 percent, which is comparable to the wage impact of caring for one dependent child. Carmichael and Charles (2003) update the previous results and find that the wage penalties are much higher (18 percent for women and 23 percent for men) who act as main co-resident caregivers. Using wage decomposition based on data from the British Household Panel Study, Heitmueller and Inglis (2007) demonstrate that the wage gap between caregivers and non-caregivers increased in the decade from the early 1990s to 2000s and that this gap cannot be explained by differences in observable characteristics. Instead, the authors suggest that caregivers may request more flexible working arrangements, which may be perceived as signals of their lower job commitment or reliability. Thus, the penalties for young mothers 60 and informal caregivers for the elderly could have similar roots in the necessity for flexible work arrangements.16 77. Career breaks or slowdowns due to care duties can be associated with reduced permanent labor incomes. Women who need to interrupt their employment or reduce their work hours to care for their children, their parents, or both, are likely to have lower permanent labor incomes than their non-caregiver peers. This could be, for example, due to lower human capital accumulation and even its depreciation during prolonged breaks in labor market attachment. In the context of Canada, Phipps et al. (2003) find that duration of job interruptions explains part of the ‘motherhood penalty’ in terms of labor income; indeed, the magnitude of this depreciation is fairly large—with one year of inactivity resulting in 37 percent loss of the marginal effect of an additional year of experience. There is evidence that such income losses due to breaks in employment are very difficult to regain later on. Jacobsen and Levin (1995) find that U.S. women who took breaks from employment of 6 months or longer were never able to catch up to their peers in terms of earnings due to losses in seniority, gaps in on-the-job training, and depreciation of job skills. 78. Female caregivers can find themselves at a particular risk of poverty, both during caregiving spells as well as later in life. Reduced incomes from employment and/or higher dependency ratios in caregiver households can imply a higher poverty incidence, especially for female caregivers.17 An OECD (2011) examination of data from Australia, the United Kingdom, the United States, Northern Europe, Central Europe, and Southern Europe finds that being a female caregiver is associated with a significantly higher poverty risk everywhere except Central and Southern Europe, whereas male caregivers are more likely to be poor only in Australia and Central Europe. Besides the immediate economic well-being, caregiving can be associated with a higher poverty risk later in life: Wakabayashi and Donato (2006) use the U.S. Health and Retirement Study to demonstrate that females who provided informal care to their parents in 1991 were, by 1999, more likely to live in poverty and receive public assistance; the authors posit that this enduring decline in economic well-being was intensified by employment interruption and worsening health. Part of the increased risk of post-retirement poverty faced by female caregivers can stem from their lower accumulated pension wealth both due to their lower lifetime earnings and also potentially due to shorter contribution history in the absence of pension credits for caregivers (PCC). For example, Kingson and O’Grady-LeShane (1993) show that caregiving for children and elderly significantly reduced U.S. women’s social security benefits, with a stronger effect observed for women in the bottom and middle of the earnings distribution. While the majority of OECD and EU countries have some type of PCC systems in place, their coverage, targeting, and generosity has a tremendous degree of variation, with implications on the programs’ success in keeping female caregivers out of poverty (Fultz 2011; Monticone et al. 2008). Women’s longer life expectancy exacerbates the disadvantaged economic position during retirement as they may have to depend on their own pension entitlements and survivor benefits, leading to a significant deterioration in living standards upon widowhood. 16 Notably, in their study using SHARE data in Europe, Bolin et al. (2008) do not find any significant wage differentials between caregivers and non-caregivers. The authors postulate that differences with earlier studies finding caregiver wage penalty could be explained by different age ranges in the samples (with SHARE only including respondents aged 50 years and above). 17 Indeed, there may be a vicious cycle whereby women from lower-income households are more likely to become informal caregivers, which in turn results in even lower household incomes (Lee et al. 2014). 61 V. Policy discussion 79. From the public policy perspective, the reasons for directing public resources to support child and eldercare are not the same; however, the focus on the care recipient remains constant. In terms of policies, there are two sides when it comes to care, those intended at improving the outcomes of the recipient. For children, early childhood development via education and care to reduce inequalities later in life; for the elderly, the main focus is to protect them from increased vulnerability after retirement and to limit the effects of age-related functional limitations on the elderly quality of life, respecting their preferences. From the care provider side, the main focus is to support them in their care responsibilities and duties, so these responsibilities do not affect their access to opportunities and do not generate unintended effects such as reducing fertility levels in already old societies.18 80. While there is now greater consensus around the set of policies that will increase child well- being, these policies have not always taken into account their mothers or other care providers’ situation if not in relation to the child.19 However, adaptations such as extended hours for childcare services to mirror working-day hours show that the role of the parent or other care provider and parental constraints related to labor market participation are being considered with more frequency. For the case of eldercare, a more recently developed policy discussion suggests that postponing entry to institutional care— the idea of aging in place—while providing services that allow for individuals to adapt their current living situation to support their needs is preferred to institutionalization and long-term care in medical institutions in order to secure quality of care. As with children, the focus has been the recipient of care (in particular elderly without family members) and less so the family member providing care or supporting existing formal care provision. 81. Very few countries have developed care systems that look at care needs as a whole, including the needs of the receivers, the formal and informal care providers (services and institutions and the family members or other people providing care), and the financing aspects of such a system. The variation across countries in patterns of implementation, age and type of population covered, and financing arrangements does not only depend on public resources availability but also on a set of normative factors that will reflect societal preferences and needs. The policies presented below show the array of what is today available in ECA countries, with a focus on those that would affect informal care providers ’ well- being, especially women’s economic participation and labor market outcomes. Providing accessible, affordable, quality formal care services 82. Households’ decisions about how to meet their eldercare and childcare needs balance multiple imperatives, constraints, and opportunities, and the outcomes of these choices can have significant repercussions for societies’ economic trajectories. The projections of future care demand, presented in Section I, foreshadowed the rising ranks of care dependents in ECA, mostly consisting of people with moderate ADL limitations. Where will these people receive the non-medical help they need 18 Greulich and Thevenon (2013) show how fertility trends across OECD countries respond to family policies. Similar analysis has been carried out for Italy by Del Boca (2002), and Norway by Rindfuss et al. (2008among others. 19 See, for a summary of this approach, World Bank (2013). “What Matters Most for Early Childhood Development: A Framework Paper.” Saber Working Paper Series. Number 5, January. 62 with activities such as dressing, bathing, and feeding? As Section II demonstrates, choices on formal and informal care for children and elderly dependents are embedded in social norms on care obligations, policies and institutions supporting formal and informal care, as well as household-level factors and market forces. Norms in ECA emphasize the role of female family members in caring for children and elders, bolstering the use of informal care and placing a disproportionate burden on women. While these views may negatively affect the use of formal care, they can also reflect the inadequate formal care options available to households. The accessibility, affordability, and quality of formal care options are critical elements in households’ decisions about how to meet their care needs, and in many ECA countries, there is much scope for improvement in terms of formal childcare, while formal eldercare options, if they exist, are scarce and poorly regulated. Holding these factors constant, household-level factors, such as the household composition and size, can influence households’ abilities to meet their care needs informally and thus, also play an important role. Scant income-earning opportunities outside of the home decrease household members’ opportunity cost of their time, making them more willing (and thus , more likely) to engage in informal care. Relative to the Western European comparators, the combination of these factors leads to the higher prevalence of informal care in ECA. As established in Section III, informal care in ECA, as in other regions around the world, is primarily undertaken by women: mothers (in the case of childcare) as well as wives and daughters (in the case of eldercare). Grandparents, especially grandmothers, also play a role in childcare, and women in the ‘sandwich generation’ may bear the responsibility of caring for both older and younger generations. Section IV argues that the substantial burden of informal care provision can result in caregivers’ lowered labor force participation at the extensive and intensive margins as well as d ecreased productivity, depressed wages, and increased vulnerability to poverty. 83. Accessible formal care services are needed to boost women’s labor force participation. According to the Women, Business and the Law database, public provision of childcare for children under the age of primary education exists in 95 countries, or 66 percent of the 128 countries included in the study. In almost all EU countries, demand for ECEC outpaces its supply, particularly for younger children(European Commission/EACEA/Eurydice/Eurostat 2014). Indeed, the gap between adequately compensated leave20 and the formal childcare coverage for children aged 0–3 years can be very substantial (Figure 39). This gap places the burden by default on families to provide informal care or to finance formal care (Figure 39). Increasing the coverage of formal childcare can attenuate the negative impact of young children on female labor supply and allow mothers to return to work (as noted by Gornick and Meyers 2004; Esping-Andersen 2009). There is rich evidence that increased availability of formal childcare options results in improved labor force participation of women in many different contexts—in Brazil (Deutsch 1998; de Barros et al. 2011); in rural Colombia (Attanasio and Vera-Hernandez 2004); in urban Argentina (Berlinski and Galiani 2005); in Japan (Asai et al. 2014); and in Canada (Lefebvre and Merrigan 2008). Closer to the ECA region, Del Boca and Locatelli (2006) used data from the European Community Household Panel to show that female labor force participation is affected by the availability, and even more importantly, affordability of childcare. Though data is thinner on eldercare, a similar argument applies, as formal eldercare options allow for care providers (often women in their peak productive ages) to reallocate their time to other activities, largely paid employment, allowing them to remain in labor markets, gain economic independence and income generation, including pension contributions that will reflect later into their income as they reach old age themselves. 20 The EU report defines adequately compensated leave as 65 percent of prior earnings. 63 Figure 39: Early childcare gap in the EU, 2003–2007 100% 90% 80% 70% Percent of time 60% 50% 40% 30% 20% 10% 0% DK* GR EE ES NO NL HU LV IE LU PL PT LT IT FR SI FI RU BE SE AT CY SK DE BG CZ UK Effective leave Early care gap Coverage 0-3 Source: Saraceno and Keck (2008). Multilinks. Calculation on the basis of various sources. Note: Effective leave is calculated as the number of weeks paid at the average level; coverage is calculated as the number of weeks per child available on the total number of children aged 0–3 years. * Denmark is the only country in which there is even an overlap between childcare service provision and effective leave time. 84. Formal eldercare systems are generally much less accessible than childcare systems and will require more investment. In the OECD, public spending on long-term care varies from 0.1 percent of GDP in Portugal to 3.6 percent of GDP in Sweden (OECD 2011). Long-term formal eldercare is common across countries, but the level of provision and use vary widely from less than one percent in Poland and Portugal to around six percent in France, Belgium, and the Netherlands (Source: Multilinks 2009). A developed formal care market improves the labor market status of the formal caregiver and that of family members who have more time for other employment. Both the formal caregiver and family members are paid more for their work, and taxes and social contributions are higher. Additionally, there is more room for professionalization and increasing quality in care provision (Angermann and Eichhorst 2012). In the past two decades, elder care in Europe has moved away from institutionalized care and public provision of care and toward home care and private or mixed services supported by cash transfers. The trend has been increasing services that complement rather than replace informal care (Bettio and Verashchagina 2010). 85. Ensuring that families can afford to take advantage of available care services is critical. In a United Nations Children’s Fund (UNICEF) review of 25 OECD countries, 11 had subsidized and regulated childcare services for 25 percent of children under the age of 3 years.21 Of the 128 countries examined in Women, Business and the Law, only 15 offer tax deductions for childcare payments. In the EU, only eight countries—the four Nordic states, Germany, Estonia, Slovenia, and Malta—guarantee a legal right to ECEC 21 UNICEF. 2008. “The Childcare Transition: A League Table of Early Childhood Education and Care in Economically Advanced Countries.” 64 for every child soon after birth, usually immediately following the end of childcare leave. This is usually full-time care, and while parents are typically expected to contribute financially, the fees are low and means- tested reductions are possible to obtain for low-income families. In the rest of the EU and in ECA, the costs of formal childcare vary considerably by country. Within the EU, public funding from the central and local government usually supports ECEC for older children while families pay a larger share of ECEC for younger children (Figure 40), and families will only pay if the services provided match their childcare needs in terms of hours of operation and quality of care. The issue of affordability is especially relevant in former Communist economies, where this cost has changed dramatically over the preceding decades. Indeed, childcare costs may be the determining factor of whether young mothers return to work in these countries (Todd 2013; Fong and Lokshin 2000; Lokshin 1999). The first column of Table 5 exhibits the full range of government financial support related to care, from extensive in Armenia to virtually no support in Ukraine except in very particular circumstances. Figure 40: Free ECEC provision in the EU, by age and weekly hours, 2012/13 Source: Eurydice, European Commission/EACEA/Eurydice/Eurostat (2014). 65 86. The quality of formal care services is also a key factor into households’ decisions about how to meet their care needs, and improving and maintaining quality will require strengthening regulations and standards. In Table 5, columns (2) and (3) present the wide cross-country variation in the main features of quality assurance around childcare—registration, guidelines, role of the private sector— for countries included in the special data collection in this study. In some countries, such as Bosnia and Herzegovina, the Kyrgyz Republic, and Ukraine, there are no requirements for registration and accreditation for childcare institutions, while other countries, like Serbia, impose specific quality standards, including on the qualifications of staff providing childcare. On eldercare, even within the EU, there is substantial variation on the assignment of responsibility, existence of standards, and other measures of quality of care (Dandi et al. 2012). Some countries, such as Hungary and Austria, have policies on quality of formal long-term care that focus on monitoring inputs rather than outcomes, whereas Estonia and Slovakia have policies and indicators on quality of both residential and home-based formal long-term care that focus on outputs, and Poland and Slovenia lack policies and indicators on quality of care altogether. Table 5: Summary of regulatory environment around childcare in selected countries (1) (2) (3) How extensive is government Are childcare Are the regulations broad or financial support? institutions required to specific? be accredited or registered? Armenia Extensive Yes - clear standards Broad, but include specific staffing requirements Bosnia and All personal funds must be spent No Very broad Herzegovina before any public assistance is available. Kosovo Some level of government Yes Fairly specific, including assistance staffing and safety in building Kyrgyz Tax code benefits No Broad Republic health care at reduced rates. Additional costs are on the beneficiary. Macedonia Some assistance available, but Yes Fairly specific, such as staffing very insufficient to cover needs qualifications Serbia Assistance is available, Yes Specific, including staffing depending on socioeconomic qualifications status of beneficiary (from full to no support). Ukraine No government financial No Broad assistance unless there is no family Source: Authors’ summary of Care Policy and Regulatory Environment Reviews, completed by firms as part of the independent data collection (2014). Each review is at the country level and is based on the text of specific legislations, regulations, and governmental action plans. 66 87. An additional challenge is closing the distance between the letter of regulation and actual practice. Table 6 shows the share of facilities in each selected ECA country that reported being officially accredited by the appropriate government institution. While accreditation levels in Ukraine and Serbia were high across the board, none of the three eldercare facilities in Armenia were accredited, although the majority of childcare facilities were. Overall, it can be observed that for the selected ECA countries accreditation is more frequent for childcare facilities than for eldercare facilities. Improving regulations and standards and ensuring that actual practice meets these standards can promote quality formal care options and make them more appealing to potential consumers. Table 6: Percent of facilities that are formally accredited Childcare Childcare Eldercare Eldercare Country % Total N % Total N Armenia 85.71 29 0.00 3 Bosnia and Herzegovina 75.00 7 75.00 5 Kosovo 77.78 8 50.00 3 Kyrgyz Republic 58.82 20 66.67 7 FYR Macedonia 100.00 20 60.00 5 Serbia 100.00 18 100.00 8 Ukraine 80.85 50 100.00 2 Total 83.56 152 68.97 33 Source: Authors’ calculations based on Independent Field Data (2014). Note: Accreditation was self-reported by representatives of the care facilities. Acknowledging and improving informal care 8. Policies supporting informal care providers can ensure their well-being during caregiving spells as well as smooth the transition to (full-time) employment after such spells. As argued in Section II, formal care and informal care are not perfect substitutes, and families will continue to provide and consume informal care to some extent even in the presence of optimal supply of formal care options. For both childcare and eldercare, different policy objectives underlie different priorities in supporting caregivers. Schemes that aim to alleviate poverty usually target caregivers with low earnings, allow for long periods of care provision, and permit caregivers to perform other work. Schemes that aim to improve female labor force attachment provide high replacement rates for short caregiving periods, and those that try to achieve better gender equity include incentives for a better gender balance of caregiving responsibilities (Fultz 2011). Care systems that support informal care may include policies concerning paid and unpaid care leave, flexible work arrangements, allowances for care givers and/or care recipients, and tax incentives. Leave policies 88. Parental leave supports the family as childcare providers, with policy design potentially affecting the distribution of childcare duties between parents. Leave can be unpaid or paid (fully or partially) by the government, employers, or both. According to Women, Business and the Law, almost all countries (139 of 143) mandate paid or unpaid maternity leave. Much fewer mandate paternity leave (64 67 countries) or parental leave (47 countries). Well-designed leave policies can help shape the way informal childcare is distributed within a household both within the first year after birth and beyond. For example, Iceland offers nine months of parental leave (with 80 percent of wage replacement), whereby the mother and father must each take three months and decide how to allocate the remaining three months. Women take most of this additional leave, but overall this policy has resulted in high paternity leave uptake among men with some promising changes in gender relations at home and in the workplace (World Bank-IFC 2014). 89. The design of paid care leave can influence labor market outcomes. Slightly less than half of all OECD countries have mandated paid (elder) care leave. In most cases (except Belgium and Japan), this leave is of very short duration—less than one month or the final stages of a terminal illness. The generosity of paid care leave is highest in Nordic countries, where up to 100 percent of the full wage could be provided. In Germany, the part-time work/part-time paid leave scheme proposed in 2010 provided 75 percent of the wage (Bettio and Verashchagina 2010). According to the OECD (2011), some studies using longitudinal data found that flexible hours and family leave are associated with higher retention of female employees but with potential negative effects on career prospects for women taking advantage of such schemes (Pavalko and Henderson 2006 cited in Bettio and Verashchagina 2010). There is also a risk that paid care leave policy can incentivize early retirement, especially if it offers high replacement rates and social insurance contributions. This has occurred in Belgium with the ‘Crédit temps’ or ‘Time Credit’, which can last up to five years of partial or full leave. 90. Unpaid care leave is provided in many countries, including in most OECD countries, although its uptake may be hindered by low awareness and the financial vulnerability associated with it. In some countries (for example, Belgium, France, Spain, and Ireland), it can be relatively long (one or more years), although in the case of Spain and Ireland, it can be refused by the employer on business grounds, and in the case of France, there is strict eligibility criteria in terms of dependents’ severity of disability (80 percent autonomy loss). In Anglo-Saxon countries and the Netherlands, only short unpaid leave is available (up to three months), whereas Austria and Germany lie in the middle (with leave up to 6 months). Austria’s ‘family hospice leave’ was introduced in 2002 for individuals taking care of dying family members or severely disabled children. The leave lasts up to three months, with the possibility of extension up to six months. Job protection is provided from the application time to 4 weeks after the termination of leave. According to Bettio and Verashchagina (2010), the scheme can be expected to negatively affect female labor force participation due to the gendered division of labor and care, and high firing risk after the end of the protected period. Very few people take up this leave due to limited awareness as well as high financial insecurity related to the scheme. 91. Across Europe, the length of leave and its wage replacement rate vary widely, which affects how parents decide to use it to provide informal care, with corresponding labor market implications. Del Boca et al. (2009) show that the length of parental leave has a nonlinear relationship to maternal employment: availability of maternity leave allows women to remain attached to their jobs, while extended leave (above 3 years) is associated with a lower probability of returning to work, with the effect being stronger for less-educated women. Policies to promote female employment therefore need to balance supporting families with not contributing to labor market detachment for mothers. 68 92. In some countries, retirement contributions during care-related career interruptions continue to be paid. This accrual of social insurance benefits during either eldercare or childcare leave provides a number of benefits. It is used in the majority of OECD countries to prevent female poverty, improve gender equality, recognize the value of care-related work, incentivize formal female employment, and to raise fertility (Fultz 2011). In some countries, pension credits for caregiving can be combined with other paid work; this has the advantages of supporting current family incomes while addressing a future source of vulnerability and enhancing labor market attachment (Fultz 2011). Countries use a range of ways to set the credited amount. Germany and Sweden set the value of pension credits based on the prevailing average wages, which can imply providing more credit than what would be earned by low-skilled caregivers had they stayed in the labor market. Alternatively, the United Kingdom and Canada implement it as a reduction in years required for pension qualification, thus implicitly basing it on each caregivers’ labor market earnings. Finally, France sets it based on the minimum wage, which can make these schemes less attractive to highly skilled caregivers. Flexible work arrangements 93. Flexible work arrangements are an alternative to unpaid leave, which maintain caregivers’ labor market attachment and sometimes the number of hours worked while allowing them to attend to their care duties. According to the OECD (2011), flexible working hours were associated with less of a reduction in hours for caregivers in Australia and the United Kingdom. In the United States, female employees who began care duties while working in companies with flexible hours had significantly higher likelihood of remaining employed within the two-year period studied than those who did not have that flexibility (Pavalko and Henderson 2006). Two examples of regulations that encourage employers to provide flexible work arrangements are Slovenia’s 2007 law on labor relations and the United Kingdom’s 2007 Flexible Working Regulations. The former required employers in Slovenia to provide justification for refusing to accommodate a worker’s request for a change in the distribution of working hours to improve work-family balance (Kanjuo-Mrcela 2010). The latter extended the 2002 Employment Act’s application of employees’ right to request flexible or reduced working hours from just parents to individuals with care responsibilities for dependent adults (Bettio and Verashchagina 2010). The U.K. Employment Task force reported that “in a 2007 CBI survey (Confederation of British Industries) over 95 percent of employers felt the Right to Request (RTR) legislation had a positive or neutral impact on productivity, recruitment, retention, employee relations and absence rates” (2008: 4). Yet, while many workers with long-term care responsibilities would want to take advantage of flexible working hours (Smeaton et al. 2009), the RTR may have more limited use for these workers due to their adult dependents’ unpredictable needs being incompatible with request applications, which are only allowed to be made once a year, with any alteration to hours constituting a permanent change of contract. 94. Part-time work arrangements can also maintain labor market attachment for women with care responsibilities, but such jobs are often substandard in terms of hourly wages and job stability as they are often offered on a temporary contract basis. Using cross-country bivariate probit regressions, Del Boca et al. (2009) find that for women aged 21–45 years who are married or live with a partner, the regional availability of part-time employment is negatively associated with the propensity for employment; however, ‘high quality’ part-time jobs—those that pay as much per hour as full-time jobs—have a positive association with employment. Additionally, when care duties end, reentering full-time work may present a 69 challenge. The right to revert to full-time work after a part-time spell exists in less than half of the 14 OECD countries with part-time work for care rights. According to the European Working Time Survey, in Eastern European countries and Portugal, it is virtually impossible for a part-timer to move to a comparable full- time job in the same establishment (OECD 2011). Allowances and tax incentives 95. Allowances paid to caregivers try to recognize the hard work that is performed by people (usually women) who take on informal care duties. Of the 21 EU countries reviewed by Riedel and Kraus (2011), 18 had some type of cash benefit for informal care providers (Table 7). Within these, variation exists in who can access these benefits. This reflects the tension between coverage and generosity of these allowances, with some countries opting for small benefits covering all caregivers (even those who are formally employed and providing low-intensity care) and other countries providing more generous but means-tested benefits (OECD 2011). Often benefits are not large enough to be considered remuneration but rather are considered a sign of support and recognition of the care provider’s efforts (Figure 41). In Nordic countries, municipalities employ family caregivers directly and provide them with wages that are similar to those obtained by paid caregivers on the market. This policy recognizes the effort of family caregivers, but it is relatively expensive and it does not do much to develop a market for professional well-regulated caregiving industry. Anglo-Saxon countries tend to provide means-tested benefits to high-intensity full- time family caregivers, who are assumed to be out of the labor force due to care duties. In England, caregiver cash benefit eligibility requires 35 hours a week of care activities (Riedel and Kraus, 2011). Although more targeted to protect those who really need assistance, these benefits have the disadvantage of potentially discouraging caregivers, especially those with low skills, from working outside the house. Indeed, means- tested allowances in Australia and the United Kingdom generate incentives for caregivers to reduce hours of non-care-related work (OECD 2011). The impact depends on the skill level, especially for women, and the availability of formal care options. Low-skilled women are more likely to be recipients of cash transfers and tend to have less intensive caregiving responsibilities when in-kind benefits are provided instead of cash transfers (Sarasa 2006). Thus, means-tested allowances for informal caregivers may cement provision of care as a low-wage, low-status activity. 96. An allowance provided to the person in need of care, in theory, puts the beneficiary of care services in charge of his or her care arrangements (OECD 2011). In most OECD countries, care recipients can use these benefits, often calculated based on care needs, to support their caregivers or to hire them formally. Such benefits can stimulate home care for the elderly and increase the supply of trained caregivers for the growing long-term care market. However, they can also inhibit the emergence of a professional care industry as families continue to rely on intra-household care; worse, as has happened in Italy, they can depress the quality of care and stimulate a grey market in care services, as family members substitute low-paid untrained care providers whose quality of care cannot be monitored. Finally, the exchange of cash for care between family members can have a negative effect on intra-family relationships and inadvertently trap caregivers in that role; for this reason, Japan has not targeted family caregivers with assistance. Within benefits for care recipients, the main variations are in the benefit amount (Figure 41), form of the benefit, restrictions on the use of the benefit, and the eligibility criteria (for example, whether the benefit is means-tested). In a review of 21 EU member states, Riedel and Kraus (2011) find that 14 countries have benefits for informal care recipients (Table 7). Of these, all but one had a requirement for 70 an assessment of medical or nursing needs. Benefits in half of the countries were not means-tested. The research also noted that countries with a stronger tradition of insurance were more likely to have cash benefits without means-testing. Most countries do not restrict whether the benefits could be used to pay for formal or informal care. Figure 41: Range of monthly allowances to care recipients or providers Poland Luxembourg Italy Hungary Estonia Bulgaria Austria 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Euro Source: Author’s calculations based on Multilinks 2009. 97. Countries use child allowances to support family incomes, which can be a way of promoting fertility; however, these allowances can have a negative effect on maternal employment. Allowances are often delivered in the form of cash benefits such as periodic child/family allowances, birth grants, tax allowances, and other benefits targeted to families with children. Such benefits are often featured as part of pro-natalist policy packages, designed to reduce families’ cost of having children. Regular and/or generous child-related cash benefits can theoretically decrease maternal employment through the income effect. Gonzalez (2011) uses a Regression Discontinuity Design (RDD) to evaluate the impact of a new generous one-time child benefit in Spain and finds a significant effect on women’s postponement of return to work. According to Del Boca et al. (2009), many studies find a consistent negative effect of cash benefits on women’s employment. 98. Higher potential economic vulnerability of caregivers suggests the importance of time-bound targeted assistance to their households to prevent inefficient coping strategies. As discussed in Section IV, career interruptions and lower permanent incomes due to caregiving duties can result in caregivers’ disadvantaged position in terms of financial status and economic vulnerability. This implies that public financial transfers to their families are likely to have an important redistributive effect, assuming both that their household consumption is low due to the need to support an elderly as well as because of the reduced earnings brought in by working adults. Fevang et al. (2012), for example, find that adult children of Norwegian elderly reduce their labor supply and increase dependence on sickness insurance and other social 71 benefits in the years before parents’ death, but employment and earnings rebound shortly after the parent’s demise. Thus, public transfers could help households smooth consumption over such caregiving periods without resorting to inefficient and costly coping strategies such as sales of productive assets or decreases in health care utilization, which can plunge caregivers into a poverty or vulnerability trap. Table 7: Availability of cash benefits Country Monetary benefit for: Care recipient Informal care provider Belgium Yes Yes Denmark No Yes France Yes/no No Germany Yes Yes Netherlands Yes No Sweden No Yes Austria Yes Yes England Yes Yes Finland Yes Yes Italy Yes Yes Latvia Yes Yes Slovenia Yes Yes Spain Yes/no Yes Bulgaria No Sometimes Czech Republic Yes No Estonia No Some regions Slovakia No Yes Hungary No Yes Lithuania No No Poland Yes Yes Romania yes Yes Source: IHS HealthEcon compilation 2010, Typologies: Kraus et al. (2010). 99. Some countries also use tax deductions as well as in-kind transfers and services to support care providers. For example, Italy’s care allowance is paid either in cash or in the form of tax credit (Bettio and Verashchagina 2010). Only a few countries have explicit tax incentives aimed to reward caregivers for their work, although some countries do exempt caregiver allowance from income taxes (for example, the Czech Republic and Ireland) (OECD 2011). In Canada and the United States, where care-related tax credits are available, they have quite stringent eligibility criteria (either in terms of income thresholds or in terms of the severity of dependents’ disability) and high transaction costs in claiming these credits, which limit their coverage. Moreover, the credit amounts often represent small shares of household income (OECD 2011; Keefe and Fancey 1999). In-kind transfers often include technology to support home self-care such as for people with limited mobility. Support for caregivers is also provided through respite care, which relieves caregivers of their role for a period. Countries also provide counseling and information services as resources for informal care (Riedel and Kraus 2011). 72 Conclusion 100. Without appropriate policies to address the expected rise in the care burden, population aging can reduce women’s access to economic opportunities and decelerate future growth, thereby threatening the agenda of poverty reduction and shared prosperity in ECA. Moreover, failing to meet the challenge of rising care burden can undermine the achievement of other policy objectives, such as increasing women’s investment in human capital, stimulating fertility by promoting mothers’ employment , and enabling the combination of family and working life, as well as extending the working life. Based on the analysis of existing and newly collected quantitative and qualitative data, several key policy recommendations can be formulated for policymakers’ consideration.  Improvement of the accessibility, affordability, and quality of formal childcare and eldercare options offers a way to address challenges related to excessive reliance on informal care and to capitalize on current opportunities. Enhanced formal care options can allow would-be informal caregivers (who are mostly women) to reallocate their time to formal labor market activities, thereby contributing directly to economic output and providing tax revenues that can ease the fiscal burden commonly associated with population aging. Moreover, a well-regulated childcare sector can imply improvements in the school readiness for children via better coverage of early childhood education; this, in turn, can translate into higher human capital accumulation, which is vital for sustaining economic growth in the context of population aging. Finally, quality provision of formal eldercare can potentially improve health outcomes of the elderly through prevention, early detection, and consistent maintenance of chronic diseases, which may imply long-term cost savings in the health care sector.  The design of future demographic, health, and education policy reforms should take into account any potential effects on informal care providers. Formal care and informal care are imperfect substitutes, and households will continue to consume and provide informal care to some extent even in the presence of optimal formal care supply. Future decisions on social policies may affect, intentionally or unintentionally, the balance between formal and informal care, and the potential repercussions should be recognized and discussed from the outset. For instance, pro-natalist policies in some countries in ECA attempt to stimulate higher birth rates through provision of cash or near-cash transfers, without providing the necessary environment for mothers to continue participating in the labor market. Such policies, aimed to delay aging, can have serious, unintended effects on the ability of the country to sustain economic growth as women withdraw from the labor force to become informal caregivers to their children. In the area of health, the projected significant increases in fiscal spending on formal long-term care are often interpreted as evidence to recommend switching to a less formal or community-based long-term care model. While this may be the optimal care model in some contexts, it is important to consider the impacts of such a care regime not only on care recipients and the state but also on informal (often, family) care providers whose caregiving intensifies when the state steps out, and to ensure that, at the very least, informal caregivers are equipped with adequate resources to provide quality care and to transition back to their previous roles after the caregiving spell is over. 73  Care leave (both paid and unpaid) can shape families’ choices about care and market work . The design, duration, and replacement rate of parental leave and care leave can affect uptake rates and effectiveness of protection against income shocks associated with caregiving, and prevent sub- optimal coping strategies. In a sense, care leave should be sufficiently long and generous to allow the caregiver the opportunity to fulfill the care obligations that are expected by the prevailing social norm and that are made necessary by the availability of formal care options. However, the duration and generosity of the care leave should not provide the caregiver with a disincentive for returning to the labor market at the earliest opportunity, as lengthy breaks in the work history can lead to human capital depreciation and thus a significant reduction in permanent income. Of course, the precise design of care leave policies has to depend on the state priorities and the local context.  Flexible work arrangements can function as effective alternatives to unpaid leave. Compared to Western European benchmarks, the incidence of part-time work and flexible work arrangements is relatively low in most ECA countries. In the long run, the ability of caregivers to decrease labor supply on the intensive rather than extensive margin can increase the likelihood of transition back to full-time work after caregiving responsibilities abate. Policies supporting uptake of flexible work can take the form of RTR regulations, or if necessary, temporary subsidization of such arrangements in certain circumstances, where they can prevent costly labor market detachment of caregivers.  Care-related allowances (both in-kind and cash) aim to promote quality care for children and elders and recognize the work of caregivers but may have negative repercussions on caregivers’ labor force outcomes. Allowances for informal caregivers recognize the hard work they do but can also discourage caregivers from working outside of the home. When allowances are given to the person in need of care, it can empower them but may also negatively impact intra-family relationships. Child allowances intend to promote fertility but may also depress women’s employment. The design of such policies should carefully consider both the explicit and the implicit incentives related to the receipt of these allowances. 101. Increased recognition of the critical role of care in aging societies and careful review of the policy environment related to formal and informal care provision can help governments to harness the full potential of demographics, thereby promoting poverty reduction and shared prosperity. Due to different policy objectives and prevailing social norms, the design of the optimal package of care-related policies would be context-specific. That said, explicit consideration of objectives surrounding care issues and prioritization of the needs of caregivers and care recipients can ensure better human development outcomes for those who need care without sacrificing the current and future well-being of caregivers as well as remaining on the right trajectory in terms of sustainable economic growth and progress on the twin goals of poverty reduction and shared prosperity. 74 References Aassve, A., B. Arpino, and A. Goisis. 2012b. “Grandparenting and Mothers’ Labour Force Participation: A Comparative Analysis Using the Generations and Gender survey.” Demographic Research 53–84. Aassve, A., E. Meroni, and C. Pronzato. 2012a. “Grandparenting and Childbearing in the Extended Family.” European Journal of Population 28: 499–518. Agree, E., B. Bisset, and M.S. Rendall. 2003. “Simultaneous Care for Parents and Care for Children among Mid-life British women.” Population Trends 112: 29–35 Akerlof, G., and R. Kranton. 2000. “Economics and Identity.” Quarterly Journal of Economics 115 (3): 715–75 Albertini, M. 2014. “Intergenerational Support in Europe.” Background paper pr epared for the World Bank Regional Aging Report. Albertini, M., and L. Mencarini. 2014. “Childlessness and Support Networks in Later Life. New Pressures on Familistic Welfare State? Journal of Family Issues, 35(3): 331-357. Albertini, M., M. Kohli, and C. Vogel. 2007. Intergenerational Transfers of Time and Money in European Families: Common Patterns, Different Regimes?” Journal of European Social Policy 319–334. Aliaga, C. 2006. “How is the Time of Women and Men Distributed in Europe?” Statistics in Focus, Population and Social Conditions 4. Luxembourg: EUROSTAT. Anderson, D.J., M. Binder, and K. Krause. 2003. Motherhood Wage Penalty Revisited: Experience, Heterogeneity, Work Effort, and Work-Schedule Flexibility. Industrial and Labor Relations Review 56: 273–294. Angermann, Annette, and Werner Eichhorst. 2012. “Eldercare Services - Lessons from a European Comparison.” Working Paper No. 6 of the Observatory for Sociopolitical Developments in Europe/ IZA Research Report No. 45. Aran, M., S. Demir, O. Sarica, and H. Yazici. 2008. “Poverty and Welfare Changes in Turkey 2003–2006.” The World Bank Turkey Office. Mimeo. Asai, Y., R. Kambayashi, and S. Yamaguchi, S. 2014. “Childcare Availability, Household Structure, and Maternal Employment.” Discussion Paper Series A, No. 611. Tokyo: Instituteof Economic Research. Attanasio, O., and M. Vera-Hernandez. 2004. Medium and Long Run Effects of Nutrition and Child Care: Evaluation of a Community Nursery Programme in Rural Colombia. The Institute for Fiscal Studies. Becker, G. 1965. “A Theory of the Allocation of Time.” Economic Journal 493–517. ———. 1981. A Treatise on the Family. Cambridge, MA: Harvard University Press. Becker, G.S. 1985. “Human Capital, Effort, and the Sexual Division of Labor.” Journal of Labor Economics S33– S58. Behrman, J.R., and B.L. Wolfe. 1984. “Labor Force Participation and Earnings Determinants for Women in the Special Conditions of Developing Countries.” Journal of Development Economics 15 (1): 259–288. Bender, W.N. 1994. “Joint Custody: The Option of Choice.” Journal of Divorce and Remarriage 115–131. Bengston, V.L. 2001. “Beyond the Nuclear Family: The Increasing Importance of Multigenerational Bonds.” Journal of Marriage and Family 1–16. Berlinski, S., and S. Galiani. 2005. The Effect of a Large Expansion of Pre-Primary School Facilities on Preschool Attendance and Maternal Employment. The Institute for Fiscal Studies. Bettio, F., A. Simonazzi, and P. Villa. 2006. “Change in Care Regimes and Female Migration: The ‘Care Drain’ in the Mediterranean”. Journal of European Social Policy 16(3): 271-285. Bettio, F., and J. Plantenga. 2004. “Comparing Care Regimes in Europe.” Feminist Economics 10 (1): 85–113. doi:10.1080/1354570042000198245. 75 Bettio, Francesca, and Alina Verashchagina. 2010. “Long -term Care for the Elderly. Provisions and Providers in 33 European countries.” EU Expert Group on Gender and Employment. http://ec.europa.eu/justice/gender - equality/files/elderly_care_en.pdf. Bobinac, A., N. van Exel, F.F Rutten, and W.B. Brouwer. 2010. “Caring For and Caring About: Disentangling the Caregiver Effect and the Family Effect.” Journal of Health Economics 29 (4): 549–556. Bolin, K., B. Lindgren, and P. Lundborg. 2008. “Your Next of Kin or Your Own Career? Caring and Working Among the 50+ of Europe.” Journal of Health Economics 27 (3): 718–738. Bonsang, E. 2009. “Does Informal Care from Children to their Elderly Parents Substitute for Formal Care in Europe?” Journal of Health Economics 143–154. Brandt, M., K. Haberkern, and M. Szydlik. 2009. Intergener ational Help and Care in Europe.” European Sociological Review jcn076. Brouwer, W., J. Van Exel, B. Van Gorp, and W.K. Redekop. 2006. “The Carer Instrument: a New I nstrument to Measure Care-Related Quality of Life of Informal Caregivers for Use in Economic Evaluations.” Quality of Life Research 15: 1005-1021. Browning, M. 1992. “Children and Household Economic Behavior.” Journal of Economic Literature 1434–1475. Buber-Ennser, Isabella. 2014. “Grandparental Childcare across Europe.” Background paper prepared for the World Bank Regional Aging Report. Buffardi, L.C., J.L. Smith, A.S. O’Brien, and C.J. Erdwins. 1999. “The Impact of Dependent Care Responsibility and Gender Work Attitudes.” Journal of Occupational Health Psychology 4: 356–367. Calderon, G. 2012. The Effects of Child Care Provision in Mexico. Stanford: Stanford University. Campbell, L.D. and A. Martin-Matthews. 2003. “The Gendered Nature of Men’s Filial Care.” Journal of Gerontology: Social Sciences. 58B(6): S350-S358. Carmichael, F., and S. Charles. 1998. “The Labour Market Costs of Community Care.” Journal of Health Economics 17: 747–765. ———. 2003. “The Opportunity Cost of Informal Care: Does Gender Matter?” Journal of Health Economics. 22: 781-803. Carmichael, S. Charles, C. Hulme. 2010. “Who will care? Employment participation and willingness to supply informal care.” Journal of Health Economics 29(1): 182 -190. Casado-Marín, D., P. García-Gómez, and Á. López-Nicolás. 2011. “Informal Care and Labour Force Participation Among Middle-aged Women in Spain.” SERIEs 2 (1): 1–29. Chapman, N., B. Ingersoll-Dayton, and M. Neal. 1994. “Balancing the Multiple Roles of Work and Caregiving for Children, Adults and Elders.” In Job Stress in a Changing Workforce: Investigating Gender, Diversity and Family Issues, edited by G.P. Keita and J.J. Hurrell. Washington, DC: American Psychological Association. Chappell, N.L., and A. Blandford. 1991. “Informal and Formal Care: Exploring the Complementarity.” Ageing & Society 11: 299–315. Choi, N. G. 1994. “Patterns and Determinants of Social Service Utilization: Comparison of the Childless Elderly and Elderly Parents Living With Or Apart From Their Children. ” The Gerontologist 34: 353-362. Ciani, Emanuele. 2012. “Informal Adult Care and Caregivers' Employment in Europe.” Labour Economics (0927- 5371) 19 (2): 155. Colombo, F., J. Llena-Nozal, J. Mercier, and Tjadens, F. 2011. “Help Wanted? Providing and Paying for Long-Term Care.” OECD Health Policy Studies. Correll, S.J., S. Benard, and I. Paik. 2007. “Is There a Motherhood Penalty?” American Journal of Sociology 1297– 1339. Couch, K.A., M.C. Daly, and D.A. Wolf. 1999. “Time? Money? Both? Allocation of Resources to Older Parents.” Demography 36: 219–232. 76 Crespo, L., and P. Mira. 2010. “Caregiving to Elderly Parents and Employment Status of European Mature Women.” Review of Economics and Statistics. Just accepted MS. Posted online November 12, 2013. Crompton, R., and C. Lyonette. 2006. “Work-life ‘Balance’ in Europe.” Acta Sociologica 379–393. Dandi, Roberto, Georgia Casanova, Roberto Lillini, Massimo Volpe, Antonio Giulio De Belvis, Maria Avolio, and Ferruccio Pelone. 2012. “Long-term Care Quality Assurance Policies in European countries.” ENEPRI Research Paper No. 111. http://aei.pitt.edu/34070/1/Quality_Assurance_Policies_for_LTC_in_in_the_EU.pdf. de Barros, R.P., P. Olinto, T. Lunde, and M. Carvalho. 2011. The Impact of Access to Free Childcare on Women's Labor Market Outcomes: Evidence from a Randomized Trial in Low-income Neighborhoods of Rio de Janeiro. De Vos, S., and G. Sandefur. 2002. Elderly Living Arrangements in Bulgaria, the Czech Republic, Estonia, Finland, and Romania. European Journal of Population 18: 21–38. Del Boca, D. 2002. “The Effect of Child Care and Part Time Opportunities on Participation and Fertility Decisions in Italy.” Journal of Population Economics 15: 549-573. Del Boca, D., and M. Locatelli. 2006. The Determinants of Motherhood and Work Status: A Survey. Bonn: The Institute for the Study of Labor (IZA). Del Boca, D., and R. M. Sauer. 2009. “Life Cycle Employment and Fertility across Institutional Environments.” European Economic Review 53: 274-292. Denton, M. 1997. “The Linkages Between Informal and Formal Care of the Elderly.” Canadian Journal on Aging, 30-50. Deutsch, R. 1998. Does Child Care Pay? Labor Force Participation and Earning Effects of Access to Child Care in the Favelas of Rio de Janeiro. Washington, DC: Inter-American Development Bank. Dimova, R. and F.C. Wolff. 2011. “Do Downward Private Transfers Enhance Maternal Labor Supply? Evi dence from around Europe.” Journal of Population Economics 911–933. Dwyer, J.W. and R.T. Coward. 1991. “A Multivariate Comparison of the Involvement of Adult Sons versus Adult Daughters in the Care of Impaired Adults.” Journal of Gerontology: Social Sciences 46(5): S259-69. Dykstra, P.A., T. van den Broek, C. Muresan, M. Haragus, P.T. Haragus, A. Abramowska-Kmon, and I.E. Kotowska. 2013. “State-of-the-art Report: Intergenerational Linkages in Families.” Families and Societies Working paper series 1:2013. Esping-Anderson, G. 2009. “The Incomplete Revolution: Adapting to Women's New Roles.” Journal of Social Policy 483–484. Ettner, L. 1995. “The Impact of ‘Parent Care’ on Female Labor Supply Decisions.” Demography 63–80. European Commission/EACEA/Eurydice/Eurostat. 2014. “Key Data on Early Childhood Education and Care in Europe.” 2014 Edition. Eurydice and Eurostat Report. Luxembourg: Publications Office of the European Union. Fevang, E., S. Kverndokk, and K. Røed. 2012. “Labor Supply in the Terminal Stages of Lone Parents’ Lives.” Journal of Population Economics 25 (4): 1399–1422. Finch, J., and J. Mason. 1990. “Filial Obligations and Kin Support for Elderly People.” Ageing & Society 151–175. Finley, N.J. 1988. “Motivators and Inhibitors of Attitudes of Filial Obligation Toward Aging Parents.” The Gerontologist 28: 73–78. doi:10.1093/geront/28.1.73. Fisher, K., and P. Robinson. 2009. “Average Weekly Time Spent in 30 Basic Activities Across 17 Countries.” Social Indicators Research 93 (1): 249–254. Fong, M., and M. Lokshin. 2000. Child Care and Women's Labor Force Participation in Romania. The World Bank. Fultz, E. 2011. “Pension Crediting for Caregivers: Policies in Finland, France, Germany, Sweden, the United Kingdom, Canada, and Japan.” Washington, DC: Institute for Women’s Policy Research. 77 Gabriele, S., P. Tanda, and F. Tediosi. 2011. “The Impact of Long-Term Care on Caregivers' Participation in the Labour Market.” ENEPRI Research Report No. 98. Gallagher, S. 1994. “Social Support in the Context of Caregiving: Husbands' Provision of Support to Wives Involved in Parent Care.” The Journals of Gerontology 43–52. Gans D., and M. Silverstein. 2006. “Norms of Filial Responsibility for Aging Parents across Time and Generations.” Journal of Marriage and Family 68: 961–976. doi:10.1111/j.1741-3737.2006.00307.x. Gerstel, N., and S.K. Gallagher. 2001. “Men’s Caregiving: Gender and the Contingent Nature of Care.” Gender & Society 15 (2): 197–217. Geyer, J. and T. Korfhage. 2014. "Long-Term Care Insurance and Carers' Labor Supply: A Structural Model," Discussion Papers of DIW Berlin 1421, DIW Berlin, German Institute for Economic Research. Glaser, K., E. Ribe, U. Waginger, D. Price, R. Stuchbury and A. Tinker. 2010. “Grandparenting in Europe.” London: Grandparents Plus. Gonzalez, L. 2011. “The Effects of a Universal Child Benefit.” Discussion Paper Series IZA DP 5994. Gornick, J.C., and M.K. Meyers. 2004. “Welfare Regimes in Relation to Paid Work and Care.” In Changing Life Patterns in Western Industrial Societies. Netherlands: Elsevier Science Press. 45–67. Gornick, J.C., M.K. Meyers, and K.E. Ross. 1997. “Supporting the Employment of Mothers: Policy Var iation across Fourteen Welfare States.” Journal of European Social Policy 7(1): 45-70. Gray, A. 2005. “The Changing Availability of Grandparents as Carers and Its Implications for Childcare Policy in the UK.” Journal of Social Policy 34: 557-577. Grundy, E., and J.C. Henretta. 2006. “Between Elderly Parents and Adult Children: A New Look at the Intergenerational Care Provided by the ‘Sandwich Generation’.” Ageing & Society 26 (5): 707–722. Haberkern, K., and M. Szydlik. 2010. “State Care Provision, Societal Opinion and Children’s Care of Older Parents in 11 European Countries.” Ageing & Society 30 (2): 299–323. Hammer, L.B., and M.B. Neal. 2008. “Working Sandwich -Generation Caregivers: Prevalence, Characteristics, and Outcomes.” The Psychologist-Manager Journal 11: 93–112. Hank, K. 2007. “Proximity and Contacts between Older Parents and their Children: A European Comparison.” Journal of Marriage and Family 69 (1): 157–173. Hank, K. and M. Kreyenfeld, “A Multilevel Analysis of Child Care and the Transition to Motherhood in Western Germany.” DIW Berlin Discussion Paper 290, German Institute for Economic Research. Hank, K., & Buber, I. (2009). Grandparents Caring for their Grandchildren: Findings from the 2004 Survey of Health, Ageing, and Retirement in Europe. Journal of Family Issues, 30: 53-73. Hank, K., and I. Buber. 2009. “Grandparents Caring for their Grandchildren Findings from the 2004 Survey of Health, Ageing, and Retirement in Europe.” Journal of Family Issues 30 (1): 53–73. Harknett, K. and B. Francesco and C. Medalia. 2014. “Do Family Support Environments Influence Fertility? Evidence from 20 European Countries,” European Journal of Population, Springer 30(1): 1-33. Harknett, K., F. Billari, and C. Medalia. 2014. “Do Family Support Environments Influence Fertility? Evidence from 20 European Countries.” European Journal of Population 30: 1–33. Heger, D. 2014. “Work and Well-Being of Informal Caregivers in Europe.” Ruhr Economic Paper No. 512. Heitmueller, A. 2007. “The Chicken or the Egg? Endogeneity in Labour Market Participation of Informal Carers in England.” Journal of Health Economics 26 (3): 536–559. Heitmueller, A., and K. Inglis. 2007. “The Earnings of Informal Carers: Wage Differentials of Opportunity Costs. ” Journal of Health Economics, 26(4): 821-41. Henz, U. 2006. “Informal Caregiving at Working Age: Effects of Job Characteristics and Family Configurations.” Journal of Marriage and Family 68: 411–429. 78 Hill, E.J., V.K. Märtinson, M. Ferris, and R. Zenger Baker. 2004. “Beyond the Mommy Track: The Influence of New-Concept Part-time Work for Professional Women on Work and Family.” Journal of Family and Economic Issues 121–136. Ho, C. 2008. “Grandchild Care, Intergenerational Resource Allocation and Retirement: A collective approach” mimeo. Hugentobler, V. 2003. “Intergenerationale Familienbeziehungen und Pflegebedurftigkeit im Alter [Intergenerational family relations and dependency on long-term care in old age].” In Gesundheit – Eine soziale Frage [Intergenerational Family Ties and Dependency on Long-Term Care in Later Life: A Social Issue] . Lucerne, Switzerland. 151–65. Igel, C. and M. Szydik. 2011. “Grandchild Care and Welfare State Arrangements in Europe.” Journal of European Social Policy 21(3): 210-224. Jacobs, J.C., C.H. Van Houtven, A. Laporte, and P.C. Coyte. 2014a. “Baby Boomer Caregivers in the Workforce: Do They Fare Better or Worse Than Their Predecessors?” Working Paper 2014-02, Canadian Centre for Health Economics, January. Jacobs, J.C., C.H. Van Houtven, A. Laporte, and P.C. Coyte. 2014b. “The Impact of Informal Caregiving Intensity on Women’s Retirement in the United States.” Working Paper 2014-09, Canadian Centre for Health Economics, April. Jacobsen, J., and L. Levin. 1995. “Effects of Intermittent Labor Force Attachment on Women's Earnings.” Monthly Labor Review 14–19. Jaumotte, F. 2003. “Labour Force participation of Women: Empirical Evidence on the Role of Policy and Other Determinants in OECD Countries.” OECD Economic Studies 51–108. Javornik, J. 2012. “Rethinking Comparative Childcare Policy Analysis. Example of Central and Eastern Europe.” Welfare Studies Working Paper Series No. 10/2012, Department of Sociology, Umeå University. Johnson, R.W., and A.T. Lo Sasso. 2000. “The Trade-Off between Hours of Paid Employment and Time Assistance to Elderly Parents at Midlife.” Mimeo, The Urban Institute. Kalwij A., G. Pasini and M. Wu. 2014. “Homecare for the Elderly: Family, Friends and the State.” Review of Economics of the Household 12: 379-404. Kanjuo Mrčela, A. 2010. “Elderly Care in Slovenia. Provisions and Providers.” External report commissioned by and presented to the European Commission Directorate-General for Employment, Social Affairs and Equal Opportunities, Unit G1 ‘Equality between women and men’. Keck, W., and C. Saraceno. 2009. “Balancing Elderly Care and Employment in Germany (No. SP I 2009-401).” Discussion Paper, Wissenschaftszentrum Berlin für Sozialforschung (WZB), Research Professorship Demographic Development, Social Change, and Social Capital. ———. 2013. The Impact of Different Social-Policy Frameworks on Social Inequalities among Women in the European Union: The Labour-Market Participation of Mothers.” Social Politics 297–328. Keefe, J. and P. Fancey. 1999. “Compensating Family Caregivers: An Analysis of Tax Initiatives and Pension Schemes.” Health Law Journal 7: 193-204. Keenan, T. A. 2010. “Home and Community Preferences of the 45+ Population.” AARP Research and Strategic Analysis, 1-19. Keene, J., and A. Prokos. 2007. “The Sandwiched Generation: Multiple Caregiving Responsibilities and the Mismatch between Actual and Preferred Work Hours. Sociological Spectrum 27: 365–387. Kingson, E.R., and R. O'Grady-LeShane. 1993. “The Effects of Caregiving on Women's Social Security Benefits.” The Gerontologist 33 (2): 230–239. Kraus, Markus, Monika Riedel, Esther Mot, Peter Willeme, Gerald Rohrling, and Thomas Czypionka. 2010. “A Typology of Long-Term Care Systems in Europe.” ENEPRI Research Report No. 91. 79 Lee, G.R., J.W. Dwyer, and R.T. Coward. 1993. “Gender Differences in Parent Care: Demographic Factors and Same-Gender Preferences.” Journal of Gerontology 48 (1): S9–S16. Lee, Y., F. Tang, K.H. Kim, and S.M. Albert. 2014. “The Vicious Cycle of Parental Caregiving and Financial Well - being: A Longitudinal Study of Women.” J Gerontol B Psychol Soci Soc Sci (January 31). Lefebvre, P., and P. Merrigan. 2008. “Child -Care Policy and the Labor Supply of Mothers with Young Children: A Natural Experiment from Canada.” Journal of Labor Economics 26 (3): 519–548. Leitner, S. 2003. “Varieties of Familialism: The Caring Function of the Family in Comparative Perspective.” European Societies 5 (4): 353–375. Lilly, M.B., A. Laporte, and P.C. Coyte. 2007. “Labor Market Work and Home Care's Unpaid Caregivers: A Systematic Review of Labor Force Participation Rates, Predictors of Labor Market Withdrawal, and Hours of Work.” The Milbank Quarterly 85 (4): 641–690. Lilly, M.B., A. Laporte, and P.C. Coyte. 2010. “Do They Care Too Much To Work? The Influence of Caregiving Intensity on the Labour Force Participation of Unpaid Caregivers in Canada.” Journal of Health Economics 29 (6): 895–903. Litwak, E., and C. Longino. 1987. “Migration Patterns Among the Elderly: A Development Perspective.” The Gerontologist 266–272. Litwin, H. 1997. “The Network Shifts of Elderly Immigrants: The Case of Soviet Jews in Israel.” Journal of Cross- Cutting Gerontology 12: 45-60. Loken, K., S. Lundberg, and J. Riise. 2014. “Lifting the Burden: State Care of the Elderly and Labor Supply of Adult Children.” Institute for the Study of Labor (IZA) Discussion Paper (8267), June. Lokshin, M. 1999. Household Childcare Choices and Women's Work Behavior in Russia. Washington, DC: The World Bank. Luci-Greulich, A., and O. Thevenon. 2013. “The Impact of Family Planning Policies on Fertility Trends in Developing Countries.” European Journal of Population, 29(4): 387-416. Mentzakis, E., P. McNamee, and M. Ryan. 2009. “Who Cares and How Much: Exploring the Determinants of Co - residential Informal Care. Review of Economics of the Household 7 (3): 283–303. Mills, M., Patrick Prag, Flavia Tsang, Katia Begall, James Derbyshire, Laura Kohle, Celine Miani, and Stijn Hoorens, 2014. Use of Childcare Services in the EU Member States and Progress towards the Barcelona Targets: Short Statistical Report No. 1, RAND Europe. Monticone, C., A. Ruzik, and J. Skiba. 2008. “Women's Pension Rights and Survivors' Benefits: A Comparati ve Analysis of EU Member States and Candidate Countries.” ENEPRI Research Reports (53) April. Morgan, K. 2008. “The Political Path to a Dual Earner/Dual Carer Society: Pitfalls and Possibilities.” Politics and Society 403–420. Murphy, M., P. Martikainen, and S. Pennec. 2006. “Demographic Change and the Supply of Potential Family Supporters In Britain, Finland And France In The Period 1911 –2050.” European Journal of Population 22(3): 219– 40. Nakamura, A., and M. Nakamura. 1992. “The Econometrics of Female Labor Supply and Children.” Econometric Review 1–71. OECD (Organisation for Economic Co-operation and Development). 2004. Labor Force Participation of Women: Empirical Evidence on the Role of Policy and Other Derteminants in OECD Countries. Paris: OECD. ———. 2011. “Public Long-term Care Financing Arrangements in OECD Countries.” In Help Wanted? Providing and Paying for Long-Term Care. http://www.oecd.org/els/health-systems/47884942.pdf. Opree, S.J., and M. Kalmjin. 2012. “Exploring Causal Effects of Combining Work and Intergenerational Support on Depressive Symptoms among Middle-aged Women.” Ageing & Society 32: 130–146. O'Reilly, D., S. Connolly, M. Rosato, and C. Patterson. 2008. “Is Caring Associated with an Increased Risk of 80 Mortality? A Longitudinal Study.” Social Science & Medicine 67 (8): 1282–1290. Pagani, Laura, and Anna Marenzi. 2008. “The Labor Market Participation of Sandwich Generation Italian Women.” Journal of Family and Economic Issues 29 (3): 427–444. Pavalko, E.K. and K.A. Henderson. 2006. “Combining Care Work and Paid Work: Do Workplace Policies Make a Difference?” Research on Aging 28(3): 359-374. Pezzin, L., P. Kemper, and J. Reschovsky. 1996. “Does Publicly Provided Home Care Substitute for Family Care? Experimental Evidence with Endogenous Living Arrangements.” Journal of Human Resources 31: 650–676. Phipps, S., P. Burton, and L. Lethbridge. 2003. “In and Out of the Labour Market: Long-Term Income Consequences of Child-Related Interruptions to Women’s Paid Work.” Canadian Journal of Economics 34(2). Posadas, J., and M. Vidal-Fernández. 2012. Grandparents' Childcare and Female Labor Force Participation. IZA Journal of Labor Policy 2: 14. Rainer, H. and S. Thomas. 2010. “Family Location and Caregiving Patterns from an International Perspective .” IZA Discussion Papers 4878, Institute for the Study of Labor (IZA). Raymo, J., L. Mencarini, M. Iwasaw, and R. Moriizumi. 2010. “Intergenerational Proximity and the Fertility Intentions of Married Women.” Asian Population Studies 6(2): 1774-1730. Raymo, J., L. Mencarini, M. Iwasawa, and R. Moriizumi. 2010. “Intergenerational Proximity and the Fertility Intentions of Married Women.” Asian Population Studies (June 22): 193–214. Ribar, D. 1995 “A Structural Model of Child Care and the Labor Supply of Married Women.” Journal of Labor Economics 13(3): 558-97. Riedel, Monika, and Markus Kraus. 2011. Informal Care Provision in Europe: Regulation and Profile of Providers. ENEPRI Research Report No. 96. Rindfuss, R.R. and S.R. Brauner-Otto. 2008. “Institutions and the Transition to Adulthood: Implications for Fertility Tempo in Low-Fertility Settings.” Vienna Yearbook of Population Research 57-87. Rubin, R., and S. White-Means. 2009. “Informal Caregiving: Dilemmas of Sandwiched Caregivers.” Journal of Family and Economic Issues 30: 252–267. Saraceno, C. and W. Keck. 2008. “The Institutional Framework of Intergenerational Family Obligations in E urope: A Conceptual and Methodological Overview.” Multilinks project, WP1. Berlin: WZB Social Science Research Center. Sarasa, S. 2006. “The Women’s Conciliation of Adult Care and Employment in Different Welfare Regimes.” Paper presented at Equalsoc/Employ_Famnet Workshop, 27th January, Torino. Settles, B.H., J. Zhao, K. Doneker Mancini, A. Rich, S. Pierre, and A. Oudor. 2009. “Grandparents Caring for Their Grandchildren: Emerging Roles and Exchanges in Global Perspectives.” Journal of Comparative Family Studies 40(5): 827-848. Silverstein, M., D. Gans, and F.M. Yang. 2006. “Intergenerational Support to Aging Parents The Role of Norms and Needs.” Journal of Family Issues 27 (8): 1068–1084. Smeaton D, S. Vegeris, and M. Sahin-Dikmen. 2009. “Older Workers: Employment Preferences, Barriers and Solutions.” London: Equality and Human Rights Commission. Smits, A., R.I. Van Gaalen, and C.H. Mulder. 2010. “Parent–Child Coresidence: Who Moves in With Whom and for Whose Needs?” Journal of Marriage and Family 72: 1022–1033. doi: 10.1111/j.1741-3737.2010.00746.x Speare, A., R. Avery, and L. Lawton. 1991. “Disability, Residential Mobility, and Changes in Living Arrangements.” Journal of Gerontology: Social Sciences 46: S133–S142. Spiess, C.K., and A.U. Schneider. 2003. “Interactions between Care-giving and Paid Work Hours among European Midlife Women, 1994 to 1996.” Ageing & Society 23 (01): 41–68. Starrels, M., B. Ingersoll-Dayton, D. Dowler, and M.B. Neal. 1997. “The Stress Caring for a Parent: Effects of Elders’ Impairment on an Employed, Adult Child.” Journal of Marriage and Family 59: 860–872. 81 Stern, S. 1995. “Estimating Family Long-Term Care Decisions in the Presence Of Endogeneous Child Characteristics.” The Journal of Human Resources 30(3): 551-580. Tarricone, R., and A. Tsouros, eds. 2008. Home Care in Europe. Copenhagen: World Health Organization. Todd, P.E. 2013. How to Improve Women's Employability and Quality of Work in Developing and Transition Economies. United Nations Foundation. Van Bavel, J., and T. De Winter. 2013. “Becoming a Grandparent and Early Retirement in Europe. ” European Sociological Review, 29(6) 386-396. Van Houtven, C. H., and E.C. Norton. 2004. “Informal Care and Health Care Use of Older Adults.” Journal of Health Economics 1159–1180. Van Houtven, C., N.B. Coe, and M. M. Skira. 2013. “The Effect of Informal Care on Work and Wages.” Journal of Health Economics 32: 240-252. Viitanen, T.K. 2007. “Informal and Formal Care in Europe.” The Institute for the Study of Labor (IZA) (2648). Voydanoff, P., and B.W. Donnelly. 1999. “Multiple Roles and Psychological Distress: The Intersection of the Paid Worker, Spouse, and Parent Roles with the Role of the Adult Child.” Journal of Marriage and Family 61(3): 725– 738. Wakabayashi, C., and K.M. Donato. 2006. “Do es Caregiving Increase Poverty among Women in Later Life? Evidence from the Health and Retirement Survey.” Journal of Health and Social Behavior 47 (3): 258–274. WHO. 2013. “Mental Health and Older Adults.” Fact Sheet N. 381. World Bank-IFC. 2014. Women Business and the Law 2014. Removing Restrictions to Enhance Gender Equality . Washington, D.C. Zamarro, G. 2011. “Family Labor Participation and Child Care Decisions: The Role of Grannoes.” The RAND Labor and Population Working Paper Series January. 82