Sharing and Caring: Older Europeans' Living Arrangements

Iacovou, Maria Article Sharing and Caring: Older Europeans' Living Arrangements Schmollers Jahrbuch – Zeitschrift für Wirtschaftsund Sozialwissenschaften. Journal of Applied Social Science Studies Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Iacovou, Maria (2002) : Sharing and Caring: Older Europeans' Living Arrangements, Schmollers Jahrbuch – Zeitschrift für Wirtschaftsund Sozialwissenschaften. Journal of Applied Social Science Studies, ISSN 1865-5742, Duncker & Humblot, Berlin, Vol. 122, Iss. 1, pp. 111-142, https://doi.org/10.3790/schm.122.1.111 This Version is available at: https://hdl.handle.net/10419/292008 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Schmollers Jahrbuch 122 (2002), 111-142 Duncker & Humblot, Berlin Sharing and Caring: Older Europeans' Living Arrangements By Maria Iacovou Abstract This paper uses data from the European Community Household Panel to chart the living and caring arrangements of older men and women in 14 European countries. There are wide variations between men and women; large variations with age; and notable differences between countries. Older people in 'Southern' European countries are much more likely to live with their children, either with or without a partner, than those in 'Northern' European countries, who tend to live with just a partner, or to live alone. Older people in the Scandinavian countries and the Netherlands are the most likely to live alone; in these countries living with adult children is a rarity for older people. The paper also describes patterns of intra-household care where older people co-reside with their adult children. Older people who live with their children are more likely to receive care in the Southern than in Northern European countries, and the proportion receiving care increases with age. Among women, the giving of care is to a large extent reciprocal, with child care being provided within the household by the 'younger old', to almost the same extent as care is provided by other family members to the 'older old'. However, this reciprocity does not hold for men: older men living with their children provide very little child care, although they receive as much care as older women. JEL Classification: J12, J14 1. Introduction1 This paper deals with the living arrangements of older people in the European Union, and asks the following questions: Who do older people live with? How does this vary between the sexes, between age groups and between countries? To what extent do older people living with other family members receive and give care within the household? And how are living arrangements linked to older people's well-being? 1 This paper is part of a project funded by the European Union under the Targeted Socio-Economic Research Programme (TSER). The ECHP data on which the paper is based were provided by Eurostat. Thanks to two anonymous referees, to members of the EPAG group, and to colleagues at ISER for many helpful comments; all remaining errors and deficiencies are my own. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 112 Maria Iacovou This paper provides answers to these questions in a descriptive framework, covering almost all member countries of the European Union; in doing so, it provides the first comprehensive cross-European description of older people's living arrangements. As well as being interesting in their own right, cross-national descriptive statistics are essential if meaningful comparisons are to be made between other studies dealing with aspects of older people's lives. For example, there is a growing literature analysing the determinants of older individuals' living arrangements in a multivariate framework (Van Solinge 1994; Wolf and Pinelli 1989; Wolf 1990 and 1995; Iacovou 2000); but in order to interpret and compare the results of these studies meaningfully, a knowledge of the how the residential patterns of older people differ between countries is essential. Additionally, living arrangements are intrinsically linked to social exclusion. There is no generally accepted definition of the concept of social exclusion (Burchardt et al 1995), but if one thinks in terms of a process whereby certain citizens are excluded from many of the advantages and normal activities of life, as a result of (among other things) persistently low incomes, lack of access to education and labour markets, and reduced participation in social networks, it becomes clear that older people may be at greater than average risk of social exclusion. In most countries older people live on lower incomes than people of working age (Whiteford and Kennedy 1995); they are for the most part out of the labour market; and they may lack access to social networks and be socially isolated, either because of a degree of infirmity, because they live alone, or because many of their contemporaries have died. Living in the same household as other family members may mitigate some of the economic and social factors associated with old age but it may also be associated with feelings of dependency: hence, this paper investigates how older people's well-being varies with living arrangements. Finally, policy-makers are interested in older people because they are a group rapidly increasing in number; with improved life expectancy among older cohorts and declining fertility among younger cohorts, this increase in the proportion of older people in society is set to continue. Eurostat (1999: 202) predicts that over the next 20 years, the population of over-70s in the 15 states of the EU will increase from 42.8 million to 56.5 million, or from 11.4 per cent to 14.8 per cent of the total population. This demographic change is particularly important because, as has been documented in several studies (for example, OECD 1996 and 1996b) older people tend to receive more inputs from the state than younger cohorts, either in cash (in terms of pensions and social assistance), or in kind (for example, in social services and medical care). Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 113 This paper makes use of the European Community Household Panel (ECHP), a large and relatively new survey of households in the member states of the European Union. It exploits the fact that the ECHP is the first large-scale survey which has been administered in a comparable way across the EU, to households rather than to individuals. The survey thus provides an opportunity to compare household composition across the EU in a way which has not been previously possible. As well as describing variations in behaviour between countries, some of the analysis in the paper also describes differences in behaviour between groups of countries. The groupings used in this paper are as follows: a 'Nordic' group, consisting of Finland, Sweden, Denmark and the Netherlands; a 'Northern European' group, consisting of the UK, Belgium, Luxembourg, France and Germany; and a 'Southern European' group, consisting of Italy, Portugal, Spain, Greece, Ireland and Austria. These groupings correspond loosely to the typology of welfare states proposed by EspingAndersen (1990). The 'Nordic' group described above is close to EspingAndersen's 'social-democratic' regime, characterised by high levels of state support and an emphasis on the individual rather than the family. The 'Northern' group is close to Esping-Andersen's 'conservative' regime, characterised by an emphasis on insurance-based benefits providing support for the family rather than the individual. However, there are a couple of exceptions: Esping-Andersen includes Italy in his 'conservative' group, while here Italy is assigned to the 'Southern' group where it fits much better; additionally, Esping-Andersen designates the UK as a sort of associate member of the 'liberal' group of welfare states typified by the US, while here (in the absence of other 'liberal' countries) the UK is assigned to the 'Northern' group. A final difference between Esping-Andersen's typology and the groupings used here, is that Esping-Andersen does not assign most of the Mediterranean countries to any cluster: the inclusion of the third grouping of countries in this paper follows the proposal of Ferrera (1996), who proposed the addition of a 'Southern' group of welfare states to Esping-Andersen's typology. Additionally, there is the question of why Ireland and Austria, which have little in common geographically with other members of the 'Southern' group, should find themselves in this category. Blackman et al (2001) show that in many respects, provision for older people in Ireland has features in common with the Mediterranean countries; subsequent analysis also shows that there are empirical arguments for including Ireland and Austria in this group, since in many ways older people's living arrangements in these two countries have more in common with the Mediterranean countries than they do with the 'Northern' group. Additionally, in common with the Mediterranean countries, the proportion of Catholics in Ireland and Austria is very high2. Thus, it may be more useful Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 114 Maria Iacovou to think of this third group of countries as a 'Catholic' group rather than a geographically 'Southern' group. Throughout the paper, older people's living arrangements are defined in terms of the other people living in the same household as the older person, and five types of arrangement are defined: (1) a partner is present, but none of the older person's children, (2) both a partner and children are present, (3) children are present, but no partner, (4) the person lives alone, and (5) the person lives with other adults (relatives or otherwise), but not with a partner or children. For the analysis of care arrangements, older people living with their children are also divided into those living with sons and daughters, and by whether their children are single or partnered. One group who are not represented in the sample under consideration are older people living in residential institutions: in common with most largescale surveys, the ECHP samples only those people living at private addresses. Thus, the sample strictly represents those elderly people not in residential care, rather than the total population of elderly people. The proportions of older people in residential care are not large in any country3, but they will be higher for certain groups, notably the 'older old', for whom the inaccuracy brought about by the omission of residential homes from the sampling frame will be greater. The paper is structured as follows. Section 2 introduces the data set used, and Section 3 discusses the literature on older people's living arrangements. In Section 4 older people's living arrangements are examined in detail. Section 5 describes how multi-generational families living together assist each other with care. Section 6 examines how older people's well-being varies with living arrangements. Section 7 draws together the findings from the paper and suggests directions for future research. 2. Data: the European Community Household Panel The analysis in this paper is based on the European Community Household Panel (ECHP), a large-scale longitudinal survey co-ordinated by the European Union. The ECHP contains data on personal characteristics, incomes and expenditure, education, employment and unemployment, and various measures of life satisfaction. 2 Details of the proportion of Catholics in each country may be found in CIA (1998). 3 OECD (1996) puts the average OECD figure for over-65s in residential care at less than 4.5 per cent for Italy, Portugal, Spain and Greece; 4.5-5.4 per cent for Austria, Belgium, France, Germany, Ireland and the UK; 5.5-6.4 per cent for Denmark; and 9 per cent for the Netherlands. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 115 This data set has several advantages. Because it is a household survey, it collects information on all members of respondents' households, which is particularly useful in the analysis of living arrangements. Because (with minor exceptions) the same questions are asked in each country, results are directly comparable across countries. In addition, the ECHP is relatively large compared to some other data sets; Wave 1 contains information on over 15000 women and 11000 men aged 65 and over. More complete details of data, sampling procedures and so on are available in Eurostat (1992,1994). In Wave 1, collected in 1994, the following countries took part in the survey: Denmark, the Netherlands, Belgium, Luxembourg, France, the UK, Germany, Ireland, Italy, Greece, Spain and Portugal. Austria joined in 1995, Finland in 1996 and Sweden in 1997. Table 1 gives Wave 1 sample sizes for each country, for men and women aged 65-74, and for men and women aged 75 (sample sizes for late-joining countries are given for the first year of their survey). Table 1 Population and ECHP Sample Size Population Population ECHP ECHP ECHP ECHP (millions, age 65+ sample: sample: sample: sample: 1994 (thousands, Women Men aged Women Men aged LFS) 1994 LFS) aged 65-74 aged 75+ 75+ 65-74 Finland 4.9 639 341 296 189 146 Sweden 8.8 n/a 550 531 344 347 Denmark 5.1 838 348 304 278 195 Netherlands 15.1 1867 534 449 304 224 UK 57.3 8748 731 602 503 307 Belgium 10.1 1577 427 361 265 167 Luxembourg 0.4 53 105 81 50 33 France 56.1 8305 882 753 559 319 Germany 80.4 12724 638 437 340 176 Austria 7.9 1158 488 346 292 161 Ireland 3.5 386 462 431 283 259 Portugal 9.8 1434 862 766 514 353 Spain 38.8 6378 1126 949 839 484 Italy 56.3 8806 928 785 571 400 Greece 10.2 1764 870 686 550 419 Notes: Population figures in the first two columns are from the Labour Force Survey (LFS), published in Eurostat (1995). Figures in the last four columns are from the ECHP, except for the German data which is from the German Socio-Economic Panel (SOEP). All data are from 1994 except for Austria (1995), Finland (1996) and Sweden (1997). Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 116 Maria Iacovou Table 1 also gives the total populations, and the populations over age 65, of all the sample countries. These figures are given for information, and also because they are used to re-weight observations when analysis is performed using groups of countries rather than individual countries. It should be noted that because of this weighting procedure, these multi-country analyses will reflect behaviour in large countries to a greater extent than small countries. Although four waves of data are available for most countries, the majority of the analysis in this paper is cross-sectional and uses mainly the Wave 1 data, since this gives a more reliably representative (although somewhat less recent) sample of the population in question. For late-joining countries, the first available wave of data is used. For some analysis, particularly of within-family caring among older people living with their children, observations from subsequent waves are also used, but each individual only appears once in the analysis, in the first year they report living with children. A few issues relating to individual countries should be highlighted. First, the sample size for Luxembourg is so small that further analysis has not been undertaken. Second, the German ECHP does not provide full information on the ages of older respondents; data from the German Socio-Economic Panel (SOEP) have been used instead. 3. The Literature This paper covers living arrangements, intra-household care of and by older people, and the relationship between living arrangements and older people's well-being. Each of these areas has a well-developed and distinct literature, and there is not space in this section to describe in detail the literature in each of these areas, particularly as to do so one would have to cover 15 countries. Thus, this section will deal mainly with the literature on older people's living arrangements, and only briefly with the literature in the other areas. It is worth noting that in all the areas mentioned - living arrangements, caring arrangements, and well-being - the large majority of research focuses on a single country, or more rarely on a comparison of two or more countries, and that the essential purpose of this paper is to provide a range of descriptive statistics which are comparable across countries and which will serve to enable comparisons to be better made of other research relating to single countries or groups of countries. For example, the study by Blackman, Brodhurst and Convery (2001) of provision for the care of elderly people covers six European countries: Denmark, Norway, the UK, Ireland, Greece and Italy. Models of social care in Europe are discussed within a typology similar to the one used in this paper, and the authors show that while the framework of care in each country has Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 117 its own unique characteristics, there are also similarities across groups of countries in the provision of care to the elderly, with comprehensive provision of care by the State in Denmark and Norway, and much lower levels in Italy and Greece. The benefits to the reader of this and other studies, of having available as a reference detailed data on living arrangements in a single source, should be clear. A certain amount of information about elderly populations and their residence patterns is available from the statistical offices of each of the countries of the EU. These statistics are based on births, deaths and marriage data, as well as the national Censuses, and have the obvious advantage of much larger sample size and therefore superior accuracy to figures based on survey data. Cross-national statistics are also available in publications by international organisations such as the EU and the OECD. However, many official publications (e.g., OECD 1997, Eurostat 1996) relate explicitly to the working-age population; even where this is not the case, the information on living arrangements in both single-country and cross-national reports is rather limited. For example, Eurostat (2001) and Council of Europe (1996) contain details of the size of Europe's elderly populations, and various other statistics broken down by age - but no detailed description of living arrangements. For Britain, ONS (2000) contains details of the percentage living alone, and marital status, and health, by age and sex, but again no further information on older people's living arrangements. Several papers exist which collate data from different national data sources across Europe. Wall (1989) gathers together data from several national censuses in Europe and presents details of the percentages of elderly people outside private households, on household headship rates, on living alone, and on family structure and patterns of co-residence. Grundy (1996) also presents details of elderly people's living arrangements: the proportion married, the proportion living with a child and the proportion living alone. This paper builds on both these papers by expanding the range of countries to almost the entire European Union, by separating out men and women, and by presenting a more detailed breakdown of living arrangements. This paper draws on survey data spanning only a few years, and may therefore be usefully complemented by studies which offer a longer-term perspective. Mitterauer and Sieder (1982) describe the development of the European family over several centuries, tracing the trend for increasing numbers of older people to live alone back to the middle of the nineteenth century, and finding that (in Central Europe at least) the multi-generational family caring for its elders within the household is not the time-honoured Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 118 Maria Iacovou tradition which it is sometimes taken to be, but actually a relatively new phenomenon. Kirk (1981) presents much more recent demographic statistics for a wide range of countries, including information on marriage and fertility rates of different cohorts, which are clearly linked to living arrangements. Grundy (1996) also discusses the relationship between marital status, fertility and living arrangements in older cohorts, showing that for a selection of six European countries, women aged 65 in the mid-1990s were more likely to be childless than those ten or twenty years younger, with clear implications for residence patterns. Pampel (1992) analyses the proportions of older people living alone between 1975 and 1989, noting that the proportions living alone increased in all the countries studied, but that relative rankings did not change much over time. The literature analysing the determinants of living arrangements in a multivariate framework is in fact much better developed than the descriptive literature. This literature spans the disciplines of economics, sociology and social policy, and falls into three main groups. First there are articles which use aggregate data to explain changes in household structure over time (Michael, Fuchs and Scott 1980; Macunovich et al, 1995; Glick, Bean and Van Hook, 1997). The second group uses cross-sectional micro data to explain cross-sectional variations in living arrangements (Wolf and Soldo 1988; Van Solinge 1994; Burr and Mutchler 1992; Carliner 1975). A third and much smaller group uses panel data to study the dynamics of changes in household structure at a micro level (Mutchler and Burr 1991; Mutchler 1992). The majority of research in all groups is based on data from the United States ; however, a growing European literature also exists. Van Solinge (1994) finds that in the Netherlands the likelihood of living alone increases with age, with being divorced, with being an owner-occupier, and with income, and is higher in regions with low capacity in nursing homes. Wolf and Pinelli (1989) find that older women in Italy are more likely to live with others with increasing age after 75, and if they have a disability; they are more likely to receive help from the family with increasing levels of disability, though not simply as a result of increasing age; there are also strong regional effects. Wolf (1990) uses data from five countries, to examine the effects on living arrangements of kin availability, financial resources, and disability and health status, finding that those with larger families or severe disability are less likely to live alone, while those with higher income are more likely to live alone. Iacovou (2000) in a comparative study using the ECHP, finds that older women with higher incomes are more likely to live alone, but that the effect of income is concentrated in the lower part of the distribution in high-income countries, and in the higher part of the distribution in low-income countries. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 125 1998). There is an enormous difference between groups of countries in the proportions living with a partner plus children: in the Nordic countries only a small fraction of women (less than five per cent) live with a partner plus children after age 60; in the Northern countries living with a partner plus children is not uncommon at age 60, accounting for just under one in five women, and the proportion does not drop below five per cent until around age 70; while in the Southern countries, living with a partner plus children is the most common situation at age 60, accounting for two in five women, and it is not until age 80 that less than five per cent of women live in this situation. The proportion of women living with children but no partner is minute after age 50 in the Nordic countries; in the Northern countries it declines to 3 or 4 per cent in the mid-sixties and increases after this age, accounting for around ten per cent of women after age 80; and in Southern Europe it increases throughout the age range, accounting for around 30 per cent of women after age 80. The proportion of women living alone is higher at all ages in Nordic countries than in other groups, and lower in Southern countries, but the difference between groups becomes much more apparent towards the older part of the age range. In Southern countries, only around 40 per cent of women over age 80 live alone, compared with nearly 70 per cent in the Northern countries and approaching 80 per cent in Nordic countries. All the foregoing discussion has focused on women, but the same sort of evolution of living arrangements with age is visible for men, with a number of interesting differences. First, because men tend to have children later and keep their partners longer than women, all the curves for men are shifted to the right. Also because of lower rates of widowhood, men in both Northern and Southern countries are less likely than women at any age to live alone or to live with just their children. But, as for women, differences are observable between groups of countries: men in Southern Europe are more likely to live with their children at all ages, whether with or without a partner, and less likely to live alone, than men in Northern Europe or in Nordic countries. Note that for neither men nor for women is there any great difference between groups of countries in the proportion of older people who live with a partner. Rather, the differences arise because Southern Europeans are far more likely to live with their children into older age, whether or not they also live with a partner. This difference is particularly pronounced in those aged under 75 for women with partners, and more pronounced in those aged over 75 for those without partners. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 126 Maria Iacovou Dynamic Processes It has already been noted that the higher probability of older people living alone in Nordic countries is associated with earlier home-leaving by young people in these countries. However, Figure 2 shows that home-leaving is only part of the reason. For five-year age groups of men and women, the percentage in each age group who either stopped or (re-) started living with their children between subsequent waves of the ECHP is shown. Figure 2: Children transitions U M •ss ... .. Nordic: Last child leaves home —•— North: Last child leaves home » South: Last child leaves home ... - - Nordic: Children return or parents move in w ith children —•— North: Children return or parents move in w ith children —o— South: Children return or parents move in w ith children 45-49 50-54 55-59 60-64 65-69 Age Group 70-74 75-79 80+ Parents move towards an empty nest earliest in the Nordic group, where the peak age for becoming an 'empty nester' is the early fifties. The peak in the Northern countries is also in the early fifties, though the rate at which the last child moves out of the parental home is lower at all ages. In Southern Europe by contrast, the peak time for moving to an empty nest is the late fifties and early sixties. So children move out of their parents' home later in the South, but there are also differences in the rate at which older people and their children move in together later in life. Until the late sixties, those in Nordic and other Northern countries are more likely to 'regain' children than their Southern counterparts; at this stage in the life cycle, this is likely to be because children who have left home, maybe temporarily, are returning home for a period. Before the sixties, it is uncommon for Southerners to regain their children, possibly because so few of them have lost all their children to begin with. However, after the age of 60, older people in the Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 127 South 'regain' their children with increasing frequency, and at higher rates than in Northern European countries, until the age of 80+. Older people in the Nordic countries are extremely unlikely to 'regain' children at any age after the early sixties. The higher rates of living with children in Southern Europe may therefore be attributed to two factors: first that by the time people reach older age, it is more common to have children still at home in Southern Europe, and second that in Southern Europe there is a higher rate of older people moving in with children, or children moving back in with their parents, than there is in Northern Europe. Those who live with children: sons or daughters? This section investigates, in respect of older people who live with their children, whether they live with sons or daughters. An elderly widow may have an unmarried son still living with her, or she might have moved in with a married daughter and her family Which type of arrangement predominates? Are there variations between countries? These questions are interesting both in themselves, and in the context of the giving of care, which is discussed in the next section. For example, research in the UK suggests that the burden of care for elderly people rests mainly on their daughters (Tinker 1997, OPCS 1992, Walker and Warren 1996). Is this reflected in living arrangements, with older people living predominantly with daughters rather than sons? In this analysis, the following countries have been excluded: Luxembourg, because of small sample size, and Sweden, Denmark and the Netherlands, because the numbers of older people living with their children in these countries is too small to yield meaningful results. This section also attempts to focus on older people who live with their adult children out of some element of choice, rather than because their children are simply too young to have left home yet. Thus, it focuses on women: since men are able to father children into their fifties and beyond, older men are much more likely than older women to live with children who have not yet flown the nest. It also considers only women over age 70, since their children are likely to be aged 35 or over and in most cases will have already left home if they are intending to leave home at all. For the group of women over age 70 who are not currently in a partnership, and who live with one or more of their children, Table 3 shows the sex and marital status of these co-resident children. For each country, where one type of arrangement is particularly prominent (more than eight percenSchmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 128 Maria Iacovou tage points ahead of the next most common arrangement), it is highlighted in bold type. Table 3 Sex and Marital Status of co-resident Children (Women age 70+ who live with their Children but no Partner) Row Percentages Single Married Daughter/s Daughter/s Single Son/s Married Son/s Son/s and Daughter/s Sample Size Finland 19 10 53 16 1 64 UK 17 26 34 18 5 77 Belgium 22 24 44 8 2 69 France 29 25 29 13 4 135 Germany 17 25 32 17 8 95 Austria 13 24 29 32 2 194 Ireland 20 14 39 22 6 194 Portugal 23 43 17 15 2 289 Spain 21 40 18 15 5 513 Italy 16 25 18 36 5 349 Greece 15 23 14 46 1 370 All 19 29 25 22 5 2352 The sample consists of all those living with children, in the first year in which they are observed living with children. Sweden, Denmark and the Netherlands are omitted as hardly anyone in these countries lives with their children; Luxembourg is omitted owing to small sample size. First it should be noted that these proportions are not independent of age. The proportion living with single sons decreases with age, and probably indicates late home-leaving or marriage by sons; the proportions living with married sons and daughters increase with age. Although overall, an older person is more likely to live with a married daughter than any of the other options, this is certainly not the case in all countries. In Spain and Portugal older women are much more likely to live with married daughters (and are also relatively likely to live with single daughters, so that over 60 per cent in each country live with a daughter). By contrast, older women in Italy, Greece and Austria are more likely to live with married sons than with married daughters, and are also more likely to live with single sons than single daughters. Ireland displays a different pattern again, with 39 per cent of women who live with a child, living with a single son. This is very likely a result of the same phenomenon of (possibly infinitely) delayed marriage, which gives rise to the large numbers of never-married elderly men and women in this counSchmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 129 try. Another 20 per cent of older women in Ireland live with a married son, meaning that 60 per cent of women in Ireland live with sons rather than daughters. In Finland, the UK and Belgium, more elderly people live with single sons than with daughters or married sons. However, rather than reflecting any preference for living with sons, this probably indicates the presence of some sons who are extremely late in leaving home: taken as a proportion of all older women, not just those with co-resident children, the proportion of Northern Europeans living with single sons is lower than the proportion of Southern Europeans living with single sons. Although the results in Table 3 show no clear regional patterns in terms of whether older people live with sons or daughters, it is clear that in many countries older people are as likely, or more likely, to live with sons as with daughters, and therefore the fact that care for elderly people is predominantly given by women does not seem to be reflected in older women's living arrangements. The section which follows suggests some reasons for this apparent contradiction. 5. Giving and Receiving Care in the Extended Family Table 4 shows the age and disability status of elderly men and women, by whether they live with sons or daughters. It is clear that those who live with single sons tend to be younger and thus probably less in need of care than those who live with single daughters, married daughters, or married sons: the mean age of an older person living with a single son is 77.3 years, compared with over 80 years for those living with a single daughter, or a married son or daughter. Those living with single sons are also in better health than those living with children in other arrangements: only 22 per cent of those living with a single son have a severely limiting health problem, compared with 28 per cent for those living with a married daughter, and over 30 per cent for those living with married sons or single daughters. The figures shown are a weighted average for all countries, but the results are similar if data are broken down by countries or groups of countries. Significance tests show that for women (and for the sample as a whole), the mean age in column (3) is significantly lower than the mean age in all the other columns at the 1 % level. Additionally, the mean age in column (5) is significantly lower, again at the 1 % level, than the mean ages in columns (1), (2) and (4). There is no significant difference between the mean ages in columns (1), (2) and (4). Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 130 Maria Iacovou Table 4 Age and Health status, by Living Arrangement (Sample: Men and Women age 70+ not Living with a Partner) Live with: (1) (2) (3) (4) (5) Single Married Single Married Don't live Daughter/s Daughter/s Son/s Son/s with children Mean age (years) Women 80.3 80.2 77.2 80.2 78.2 Men 78.5 80.8 77.8 82.1 77.9 All 80.0 80.3 77.3 80.6 78.2 Percentage where health hampers activity severely Women 33.9 29.2 21.8 29.6 16.4 Men 16.6 21.7 20.6 33.9 15.7 All 31.1 27.8 21.6 30.5 16.2 Unweighted sample size: 6978 women (of whom 2221 live with children) and 1949 men (of whom 489 live with children). The sample consists of all those living with children, in the first year in which they are observed living with children. Statistics based on chi-squared tests: Test statistic for whether percent with severe disability varies between groups: chi-squared(4) = 122.4, P=0.000. Test statistic for whether percentage with severe disability varies between groups, excluding those not living with children: chi-squared(3) = 13.86, P = 0.003. For men, there is no significant difference between the mean ages in columns (1), (3) and (5), or between the means in columns (2) and (4). The means in columns (2) and (4) are significantly higher than those in columns (1), (3) and (5) [P < 0.001]. For women, the percentage with severe disability is significantly lower in column (5) than in all other columns [P < 0.01], and significantly lower in column (3) than in columns (1), (2) and (4) [P < 0.05 or better]. There is no significant difference in the percentages with a severe disability in columns (1), (2) and (4). For men, the percentage with a severe disability is significantly higher in column (4) than in all other columns [P < 0.05]. There is no significant difference between the percentages in other columns. Table 4 shows that one possible explanation for the fact that older people are often just as likely to live with sons, even though research finds that care is given by women, is that those living with single sons are less likely than others to need care. Table 5 suggests an additional explanation: that even when elderly people live with married sons, it is still in many cases women in these households (presumably their daughters-in-law) who take on the caring. The ECHP asks respondents: "Do your daily activities include looking after, without pay, another person who needs special help because of old age, illness or disability? The person could be living in the household or Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 131 elsewhere". Those who respond in the affirmative are asked "[Is] any looked-after person living in this household?" Thus, these questions allow the researcher to identify individuals in the household who are providing care for an elderly person also in the household, but where more than one elderly person lives in the household, one cannot identify which of them is receiving the care. This problem is overcome by focusing on people over age 70 who live with their children, but who do not live with a partner, and where there is no-one else over age 70 in the household. Table 5 shows the proportions of household members under age 65 in households where a single elderly person over 70 is present, who report looking after an older person in the same household. In each panel, the first row gives the percentages of older people who are cared for by a man under age 65 in the same household; the second row gives the percentages being cared for by a woman under 65 in the same household; and the third row shows the percentages being cared for by a man or a woman in the same household. As one would expect, where an older person lives with a single daughter, care is nearly always provided by a woman in the household, and where an older person lives with a single son, care is nearly always provided by a man. There is a gender difference in the proportions receiving care: those living with single daughters are much more likely to be receiving care from within the household than those living with single sons. This difference is particularly pronounced for the 'Northern' countries, where 29 per cent of older people living with a single daughter receive care, compared with only 13 per cent of older people living with single sons. Gender differences in the giving of care are also pronounced where older people live with married sons or daughters. In both cases, they are more likely to be cared for by a woman than by a man, and in most of the households where a man is providing care, care is also being provided by a woman. Weighted averages over all countries show that of older people who live with their married sons, 49 per cent are receiving some sort of care from within the household, 44 per cent are receiving care from a woman, while only 20 per cent are receiving care from a man (their son): thus, only 5 per cent are receiving care exclusively from the man. This difference is driven largely by the Southern countries, where the proportion of married sons giving care is particularly low (15 per cent, compared to 30 per cent in Northern countries). In Northern countries the proportion of married sons giving care is approximately equal to the proportion of daughters-in-law giving care. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 132 Maria Iacovou Table 5 Care for older People within the Household, by Living Arrangement (Women and Men age 70+, not Living with a Partner, who Live with their Children) Single Married Single Married Son/s All DaughDaughSon/s Son/s and ter/s ter/s Daughters 'Nordic' Countries Cared for by a man under 65 * * 21 * * 23 in the same household Cared for by a woman under * * 0 * * 19 65 in the same household Cared for by anyone under 65 * * 21 * * 34 in same household 'Northern' Countries Cared for by a man under 65 0 22 13 30 * 16 in the same household Cared for by a woman under 28 37 0 29 * 20 65 in the same household Cared for by anyone under 65 28 50 13 41 * 31 in same household 'Southern' Countries Cared for by a man under 65 0 12 19 15 18 12 in the same household Cared for by a woman under 26 53 1 50 32 37 65 in the same household Cared for by anyone under 65 26 54 20 52 36 41 in same household Weighted average, All countries Cared for by a man under 65 0 16 16 20 20 14 in the same household Cared for by a woman under 27 47 0 44 19 30 65 in the same household Cared for by anyone under 65 27 52 16 49 29 37 in same household Unweighted sample sizes: 'Nordic' countries: 110; 'Northern' countries: 653; 'Southern' countries: 2110. The sample consists of all those living with children, in the first year in which they are observed living with children. Cells marked with an asterisk contain too few observations to provide meaningful percentages; other cells all contain 50 observations or more. Test statistics: Single daughters are significantly more likely than single sons to provide care in the 'Northern' group of countries (P = 0.001), and over all countries [P = 0.000] though this difference is significant only at the 10% level in the 'Southern' group [P = 0.069]. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 133 Of course, these data fall well short of showing a full picture of care by older people's children: in particular, they give no information about the amount of care children are giving. However, there is strong evidence that more women than men care for older people: older people living with (single) sons are less likely to need inputs of care; single sons are less likely than single daughters to provide care; when an older person lives with a married daughter, the daughter is much more likely than the son-in-law to provide care; and when the older person lives with a married son, a good deal of care falls to the daughter-in-law. Two final questions are addressed in this section: how receiving care within the extended family is related to age, and how giving care may be seen as a reciprocal rather than a one-way arrangement. It has already been shown that many older people living with their children receive care from within the household, but this section examines how many of the older people provide caring services in the other direction, by looking after grandchildren. Figure 3(a) shows how the proportions of older people receiving care in the extended family changes with age. This chart is based on a sample of men and women living with one or more of their children, and not living with a partner. Again, this means that one can be reasonably sure that the individual in question, rather than another adult in the household, is receiving care, but it means that some of the cell sizes are rather small, particularly for the 'Nordic' countries, which are not shown on the graphs. For all groups, the proportion receiving care within the extended family increases with age, from under 10 per cent for the 60-64 age group, to between 40 and 60 per cent or more for those aged 80 and over. At the beginning and the end of the age range, the proportions of older people receiving care do not differ much between groups. However, towards the middle of the age range, especially among people aged 70-79, there is more of a difference between groups, with men and women in Northern countries less likely to be receiving care than those in Southern countries. The differences between groups of countries are statistically significant at the 1 per cent level (test statistics not shown), with Northern men less likely to receive care than Southern men, and Northern women less likely to receive care than Southern women. However, within groups of countries the differences between the proportions of men and women receiving care are not statistically significant. Figure 3(b) takes as its sample all older people living with their children (whether or not they also live with a partner, since the inclusion of older people with partners does not confuse the issue of whether an individual older person provides childcare), and shows the proportion who provide care for young children in the same household (it should be remembered Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 134 Maria Iacovou Figure 3(a): Proportion of older people living with their children but without a spouse, who receive care in the extended family 2 50 5 20 60-64 65-69 70-74 Age group 75-79 80+ Figure 3(b): Proportion of older people living with their adult children who provide childcare within an extended family 2 8 o> c > "5) 0) o> 3 c 8 k. 0) Q_ 60 50 30 20 -O- - Women - 'Northern' % Women 'Southern' -A- • • Men - 'Northern' -A—Men - 'Southern' 60-64 65-69 70-74 Age group Note: First panel of the chart based on 603 women and 137 men in 'Northern' countries, and 2728 women and 632 men in 'Southern' countries. Second panel based on 1044 women and 889 men in 'Northern' countries, and 4903 women and 4117 men in 'Southern' countries. In both panels, the sample includes those living with children, in the first year in which they are observed living with children. that not all these people will be in a position to be providing care, as not all will have grandchildren resident in the same household). For all groups, the proportion looking after grandchildren declines with age, reflecting the fact Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 Older Europeans' Living Arrangements 141 - (2000) Health, Wealth and Progeny: Explaining Older Europeans' Living Arrangements. ISER Working Papers. Paper 2000-8, Colchester: University of Essex. - (2000b) The Living Arrangements of Elderly Europeans. ISER Working Papers. Paper 2000-9, Colchester: University of Essex. Kirk, M. (ed) (1981) "Demographic and Social Change in Europe 1975-2000". Liverpool University Press, Liverpool. Macunovich, D. J. / Easterlin, R. A. I Schaeffer, C. M. / Crimmins, E. M. (1995) "Echoes of the Baby Boom and Bust: Recent and Prospective Changes in Living Alone among Elderly Widows in the United States". Demography 32, pp. 17-28. Michael, R. T./Fuchs, V. R. /Scott, S. R. (1980) "Changes in the Propensity to Live Alone: 1950-1976". Demography 17, pp. 39-56. Mitterauer, M./ Sieder, R. (1982) "The European Family: Patriarchy to Partnership from the Middle Ages to the Present". Basil Blackwell, Oxford. Mutchler, Jan E. (1992) "Living Arrangements and Household Transitions among the Unmarried in Later Life". Social Science Quarterly 73, pp. 565-580. Mutchler, Jan E./Burr, Jeffrey A. (1991) "A Longitudinal Analysis of Household and Nonhousehold Living Arrangements in Later Life". Demography 28, pp. 374-390. OECD (1996) Caring for Frail Elderly People: Policies in Evolution. OECD Social Policy Studies no. 19, Paris. - (1996(b)) Ageing in OECD Countries: a Critical Policy Challenge. OECD Social Policy Studies no. 20, Paris. - (1997) "Labour Force Statistics". OECD Statistics Directorate, Paris. ONS (2000) "Living in Britain: Results from the General Household Survey". Office for National Statistics, London: The Stationery Office. OPCS (1992) General household survey: Careers in 1990. Office of Population Censuses and Surveys, 1992. Pampel, F. (1992) "Trends in Living Alone among the Elderly in Europe", in: Rogers, A. (ed) Elderly Migration and Population Redistribution: a Comparative Study, Belhaven, London. Tinker, A. (1997) Older People in Modern Society (fourth edition). Addison Wesley Longman Limited, New York. Van Solinge, H. (1994) "Living Arrangements of Non-Married Elderly People in the Netherlands in 1990". Ageing and Society 14, pp. 219-236. Walker, A./Warren, L. (1996) Changing Services for Older People. Open University Press. Whiteford, P. /Kennedy, S. (1995) Incomes and Living Standards of Older People. Department of Social Security Research Report No. 34, HMSO 1995. Wall, R. (1989) "The Residence Patterns of the Elderly in Europe in the 1980's", in: Grebank et al (eds) Later phases of the family life cycle, Clarendon Press, Oxford. Wolf, Douglas A. (1990) "Household patterns of older women". Research on Aging 12:4, pp. 463-486. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42 142 Maria Iacovou - (1995) "Changes in the Living Arrangements of Older Women: an International Study". The Gerontologist 35:6, pp. 724-731. Wolf, Douglas A./Pinelli, Antonella (1989) "Living arrangements and family networks of older women in Italy." Research on Aging 11, pp. 354-37. Wolf, D. A./Soldo, B. J. (1988) "Houshold Composition Choices of Older Unmarried Women". Demography 25, pp. 387-403. Schmollers Jahrbuch 122 (2002) 1 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.122.1.111 | Generated on 2023-04-04 12:29:42