BookPDF Available
Grandparenting in Europe:
family policy and grandparents’ role in providing childcare
July 2013
By Karen Glaser, Debora
Price, Giorgio Di Gessa,
Eloi Ribe, Rachel Stuchbury
and Anthea Tinker
Dr. Karen Glaser, PhD is the Director of the Institute of Gerontology, and Reader in Gerontology in the Department
of Social Science Health & Medicine (SSHM), King’s College London (KCL). Karen is also the Programme Director for
the Postgraduate Taught Programmes in Gerontology at KCL. Her research interests include: comparative research on
family structure, proximity and support across a variety of settings (e.g. the UK, Southern Europe and Latin America); the
relationship between the multiple roles of mid-life individuals (focusing on work and family commitments) and well-being
at older ages (i.e. physical, social and material); the living arrangements of older people in Britain and changes in residence
patterns over time; and the relationship between disruptions in key life-course events, in particular family disruption due
to divorce, death, or re-partnering and social support in later life. More recently she has focused on intergenerational
relationships in ageing societies.
Dr. Debora Price, PhD, a former barrister, is a Senior Lecturer in Gerontology SSHM, KCL, where she specialises in the
study of pensions and poverty of older people, the legal regulation of the nancial consequences of family formation and
dissolution, and survey methodology. She led an ESRC funded project investigating household money management within
older couples, quantitative analysis of nancial inequalities among older couples using the English Longitudinal Study
of Ageing, and analysis of social policy relating to nance for older people. She has also been involved in analysis and
commentary on recent UK
pension reform.
Dr. Giorgio Di Gessa, PhD is a Research Associate in the Institute of Gerontology in SSHM, KCL. Giorgio completed
his PhD in Demography at the London School of Hygiene and Tropical Medicine, and has a keen interest and expertise
in quantitative methods with a focus on statistical techniques for hierarchical data analysis. His academic interests in
the eld of social gerontology include the demographic and social determinants of health and well-being in later life in
Europe.
Eloi Ribe, MA is a research assistant in the Institute of Gerontology in SSHM. With a background in Sociology, Social
Gerontology and Public Policy, his research interests include intergenerational and social family care practices, public
pension reforms in Europe, poverty and exclusion of older people and transforming social care arrangements for older frail
individuals. He was recently involved in a cross-country research project on intergenerational relationships and alternatives
to institutional long-term care.
Rachel Stuchbury, MSc is a quantitative social researcher who specialises in the analysis of longitudinal survey
data. Current and recent research projects have been based at University College London, King’s College London and
the London School of Hygiene & Tropical Medicine. Among the datasets of which she has substantial experience are
the Ofce for National Statistics’ Longitudinal Study (which links census and vital registration returns for a sample of
individuals in England and Wales), the English Longitudinal Study of Ageing and the British Household Panel Survey for
1991 to 2006. Her research interests include partnerships, family relations, household structures and the life course more
generally.
Professor Anthea Tinker, CBE, PhD, FKC, AcSS, FRSA has been Professor of Social Gerontology at KCL since 1988.
She has been on the staff of three universities and three government departments and a consultant to the WHO, EU and
OECD. She chaired the King’s College Research Ethics Committee from 2001–2011. She is the author of 31 books and over
300 articles. She has carried out research on housing, assistive technology, family care, older workers, community care,
older women, very old people, elder abuse, falls and accidents.
About the authors
ISBN 978-0-9573281-6-7
3
Acknowledgements
This report has been produced by the Institute of Gerontology at King’s College London for Grandparents Plus in
association with the Beth Johnson Foundation. We are grateful for the support of the Calouste Gulbenkian Foundation
which enabled us to carry out the research from April 2012 to June 2013 and who funded the initial scoping study in
January- April 2010. We wish to express our thanks to all the contributors to the report for their hard work, advice and
expertise. We stress that this was independent research which does not necessarily represent the views of our partner
organisations.
We wish to give special thanks to members of the European Advisory Group, that is Dr. Paula Albuquerque (Portugal),
Professor Karsten Hank (Germany), Marta Korintus (Hungary), Professor Cornelia Muresan (Romania), Dr. Jim Ogg
(France), Dr. Dolores Puga (Spain), Professor Tine Rostgaard, (Denmark) Dr. Fleur Thomese (the Netherlands) and Dr.
Cecilia Tomassini (Italy) for their contributions and valuable feedback on earlier versions of the report. We also wish to
extend special thanks to Sarah Wellard, Director of Policy, Research and Communications at Grandparents Plus for her
excellent guidance and management of the project; Annabel Knight, Programme Manager at the Calouste Gulbenkian
Foundation for her invaluable support and advice; and Alan Hatton-Yeo, CEO of the Beth Johnson Foundation for his
sponsorship and encouragement.
4
Contents
Tables 6
Figures 8
Executive Summary 10
1 Introduction 18
1.1 Overview 18
1.2 Aims and objectives 18
1.3 Datasets and methodology 18
1.4 Report outline 20
2 Summary (and Brief Update) of Literature Review 21
3 Trends in Prevalence of Grandparent Households: Selected European Countries and the US 24
3.1 Evidence of trends 24
3.2 Characteristics of grandparent households 24
3.3 Evidence for Europe 24
3.4 Reasons for caregiving and policies 24
3.5 Summary 25
3.6 Grandparent households: Prevalence and characteristics 25
3.7 Grandparent characteristics and household type 31
3.8 Summary 31
4 Investigating Grandparent Characteristics in 12 European Countries 32
4.1 Prevalence of grandparenthood 32
4.2 Demographic and socioeconomic characteristics 32
4.2.1 Age 32
4.2.2 Gender 34
4.2.3 Family structure 34
4.2.4 Education 38
4.2.5 Main activity status 39
4.2.6 Wealth 39
4.3 Health and well-being 40
4.3.1 Self-reported health 40
4.3.2 Depressive symptoms 40
4.3.3 Cognitive function 41
4.3.4 Functional limitations 42
4.4 Grandparental childcare 42
4.4.1 Grandparental childcare and associated characteristics 44
4.5 Summary 44
5 Investigating Parent Characteristics: SHARE 46
5.1 Describing family data in SHARE 46
5.2 Parent demographic and socio-economic characteristics 48
5
5.2.1 Age 48
5.2.2 Gender 49
5.2.3 Family structure 49
5.2.4 Distance 52
5.2.5 Main activity status 53
5.3 Summary 56
6 Family Policy in Europe and Grandparenting 57
6.1 Family policy and patterns of Grandparental Childcare 57
6.1.1 Family Policies and Grandparents 57
6.1.2 Framework for Analysis 58
6.2 Grandparenting in practice 60
6.3 Family and Care Policies 60
6.4 Work and gender cultures and structures 63
6.4.1 Employment of mid-life women 65
6.4.2 Women’s Employment and Intensive Grandmaternal Childcare 65
6.5 Childcare cultures and structures, and intensive grandmaternal care 67
6.5.1 Summary 69
6.6 Attitudes and Preferences 70
6.7 Summary and Conclusion 71
7 Grandparent Characteristics Associated with Grandparental childcare 72
7.1 Grandparent characteristics associated with intensive, non-intensive and no care 72
7.2 Grandparent characteristics associated with intensive grandchild care 73
7.3 Multilevel model 74
7.3.1 Country-level characteristics associated with intensive grandparental childcare 75
7.4 Summary 76
8 Conclusion 77
8.1 Summary of ndings 77
Appendix A European Expert Advisory group A-1
Appendix B Detailed Methodology B-1
Appendix C Additional Tables and Figures Chapter 3 C-1
Appendix D Additional Tables and Figures Chapter 4 D-1
Appendix E Additional Tables Chapter 7 E-1
Appendix F Family Policies by Country F-1
Appendix G Policy Tables G-1
Appendix H References H-1
6
Tables
Table 3-1 Percentage of people aged 35 and over residing in three-generation households by selected characteristics,
weighted data. 27
Table 3-2 Percentage of people aged 35 and over residing in skipped-generation households by selected
characteristics, weighted data. 28
Table 3-3 Characteristics of grandparents in the household, grandparent households (three and skipped-generation),
row percentages, weighted 29
Table 3-4 Characteristics of grandchildren in the household, grandparent households (three and skipped-generation),
row percentages, weighted 30
Table 4-1 Mean and median age of grandmothers and grandfathers by country 33
Table 4-2 Mean number of children and grandchildren (and 95% condence intervals), by gender and country
(analyses restricted to grandparents only) 35
Table 4-3 Mean age of women at the birth of their rst child and the total fertility rate (TFR), by country -selected
years (1960, 1970, 1995) 36
Table 4-4 Percentage distribution of marital status among grandmothers and grandfathers by country 37
Table 4-5 Percentage of grandparents whose mother and father were still alive, by country 37
Table 4-6 Percentage distribution of level of education (ISCED-97) among grandparents, by gender and country 37
Table 4-7 Percentages of grandparents with self-rated health (SRH) reported as fair or poor, who had four or more
depressive symptoms and who are in the lowest quintile of the cognitive function, by gender and country 41
Table 4-8 Percentages of grandparents with one or more ADL limitations, by gender and country 42
Table 4-9 Percentages of grandfathers and grandmothers looking after grandchildren, by frequency, gender and
country 43
Table 4-10 Percentage (and absolute number) of grandparents providing intensive childcare, and mean (and median)
number of hours of childcare provided, by country 44
Table 4-11 Distribution of grandparental childcare by socio-demographic, economic, health indicators: only SHARE
countries (row percentages) 45
Table 5-1 Absolute number of parents by whether they have a child that is looked after by a grandparent, by country
(row percentages) 47
Table 5-2 Percentages of parents (and absolute numbers) who have a child that is regularly looked after by a
grandparent (among those who have any children looked after by a grandparent) within each age group and by country 48
Table 5-3 Percentage of parents (and absolute numbers) who have a child that is regularly looked after by a
grandparent (among those who have any children looked after by a grandparent), by gender and country 49
Table 5-4 Percentages (and absolute numbers) of parents who have a child that is regularly looked after by a
grandparent (among those who have any children looked after by a grandparent), by the age of the youngest child
and country 50
Table 5-5 Percentage of parents who have a child that is regularly looked after by a grandparent (among those who
have any children looked after by a grandparent), by marital status, gender and country 51
Table 5-6 Percentage (and absolute number) of parents who have a child that is regularly looked after by their
grandparents (among those who have any children looked after by a grandparent), by distance to the grandparents’
home and country 52
Table 5-7 Percentage of mothers and fathers who have a child that is regularly looked after by a grandparent (among
those who have any children looked after by a grandparent), by whether or not in paid work, and by country 55
Table 6-1 Indicators for policies, family & gender cultures and structures, labour market cultures and structures 59
Table 6-2 Percentages of grandmothers providing care for grandchildren in the absence of parents 61
Table 6-3 Percentage or working mothers by work status and age of youngest child, and gender pay gap, by country
(2008), in parentheses the percentage of mothers with any dependent child 63
Table 6-4 Institutional childcare usage, childcare expenditure, quality and satisfaction in various countries 67
Table 7-1 Overview of cultural-contextual factors by country 74
7
Table C-1 Characteristics of data samples used in the analysis C-1
Table C-2 Characteristics of people aged 35 and over associated with co-residence in three-generation or skipped-
generation grandparent households. Odds ratios from a multinomial logit model C-2
Table D-1 Logistic regression analysis of grandparenthood by gender D-1
Table D-2 Logistic regression analysis of grandparents being in the 50-64 age group D-2
Table D-3 Mean number of grandchildren and great-grandchildren (and 95% condence intervals) among
grandparents D-2
Table D-4 OLS regression analysis of number of children among grandparents D-3
Table D-5 OLS regression analysis of number of grandchildren and great-grandchildren among grandparents D-3
Table D-6 Logistic regression analysis of having at least one grandchild aged 0-2 among grandparents D-4
Table D-7 Logistic regression analysis of being married among grandparents (unadjusted and age-adjusted odds
ratios) D-5
Table D-8 Logistic regression analysis of grandparents being in the lowest educational group (unadjusted and
age-adjusted odds ratios D-6
Table D-9 Logistic regression of grandparents being in paid work by gender D-7
Table D-10 Logistic regression analysis of grandparents being in the lowest 20% of the wealth distribution D-8
Table D-11 Percentages and logistic regression models of various health outcomes (self-rated health SRH as poor
or fair, depressive symptoms, cognitive function and 1 or more ADLs) among grandparents (unadjusted and
‘age-adjusted’ odds ratios
D-9
Table D-12 Percentage of grandparents looking after grandchildren by frequency and country D-9
Table D-13 Average number of hours grandparents looked after grandchildren, by frequency and country D-10
Table E-1 Grandparent characteristics associated with grandparental childcare (no care; non-intensive care;
intensive care) by gender. Odds Ratios and 95% CIs obtained from a generalised ordinal logistic model (with
partial proportional odds) E-2
Table E-2 Grandparent characteristics associated with ‘intensive’ grandparental care by gender. Odds Ratios and
95% CIs obtained from fully adjusted Logistic Regressions. E-3
Table E-3 Multilevel logistic regression results predicting intensive grandparental childcare E-4
8
Figures
Figure 3-1 Percentage of people aged 35 and over residing in three-generation grandparent households: Selected
European countries and the US, 1980s–2000s 25
Figure 3-2 Percentage of people aged 35 and over residing in skipped-generation grandparent households: Selected
European countries and the US, 1980s–2000s 25
Figure 4-1 Percentage of older adults who are grandparents by gender and country 33
Figure 4-2 Age prole of grandparents by gender and country 35
Figure 4-3 Gender prole of grandparents by country, SHARE 36
Figure 4-4 Percentage of SHARE grandfathers and grandmothers with youngest grandchild aged 0-2 or 3-5 years, by
country (SHARE grandparents) 38
Figure 4-5 Employment distribution of grandparents, by gender and country 39
Figure 4-6 Percentage of grandparents in the lowest 20% of the wealth distribution for people aged 50 and over, by
gender and country 40
Figure 5-1 Tree diagram describing adult children in SHARE 46
Figure 5-2 Age prole of parents by whether they have a child that is looked after by a grandparent, by country 48
Figure 5-3 Percentages of parents who have a child that is looked after by a grandparent, by gender and country 49
Figure 5-4 Percentage of parents who have a child that is looked after by a grandparent, by age of the youngest child
and country 50
Figure 5-5 Percentage of parents who have a child that is looked after by a grandparent, by marital status and country 51
Figure 5-6 Percentage of parents who have a child(ren) that is looked after by a grandparent, by distance to the
grandparent’s home and by country 52
Figure 5-7 Percentage of parents who have a child that is looked after by a grandparent, by main activity status and
country 53
Figure 5-8 Percentage of fathers who have a child that is looked after by a grandparent, by main activity status and
country 54
Figure 5-9 Percentage of mothers who have a child that is looked after by a grandparent, by main activity status and
country 54
Figure 5-10 Percentage of mothers who have a child(ren) that is looked after by a grandparent, by abbreviated main
activity status and country 54
Figure 6-1 Framework for understanding grandparental care of grandchildren 58
Figure 6-2 Percentage of mothers aged 25-49 not in paid employment and proportion of grandmothers looking after
their grandchildren daily 65
Figure 6-3 Percentage of women aged 50-64 in paid work and grandmothers looking after grandchildren daily 66
Figure 6-4 Association between children aged 0-2 in formal childcare and grandmothers providing intensive childcare 69
Figure 6-5 Proportion considering that pre-school children suffer with a working mother, and intensive grandmaternal
childcare 71
Figure D-1 Percentage of adults aged 50 or over who are grandparents D-1
Figure D-2 Age prole of grandparents D-2
Figure D-3 Percentage of grandparents with at least one grandchild aged 0-2 D-4
Figure D-4 Marital status of grandparents D-5
Figure D-5 Distribution of educational level (ISCED-97) of grandparents D-6
Figure D-6 Percentage of grandparents in paid work D-7
Figure D-7 Percentage of grandparents in the lowest 20% of the wealth distribution D-8
9
10
Executive Summary
Overview
Younger grandmothers who are t, healthy
and with younger grandchildren are the
most likely to be providing care for their
grandchildren, yet they are also the very
women that governments across Europe
are aiming to encourage to stay in paid
work for longer, in order to grow our
economies and fund pensions, social care
and other welfare provision in later life.
Their vital but invisible role in providing
childcare, whether intensive, regular,
or occasional, is likely to conict with
their ability to self-nance their old age,
especially as widows’ benets in both
state and employer pension schemes are
eroded. The risk is an emerging care gap
as older women remain in work longer,
become less available to provide childcare
and so adversely affecting mothers’ labour
market participation.
Across Europe increased life expectancy means it is
now quite common for a child to grow up while their
grandparents and even great-grandparents are alive.
Grandparents have always provided nancial, emotional
and practical care and support to their children and
grandchildren. However, this role has generally been taken
for granted by families and governments, and grandparents
have little recognition and few rights. Ageing populations,
more mothers in the labour market and higher rates of
divorce and relationship breakdown all indicate that the
role grandparents play in family life is likely to become
increasingly signicant. In many countries austerity
measures and cuts to public services are likely to lead to
an expectation that grandparents will step in to ll care
gaps for children and adults. Yet our understanding of
grandparenting and how policy environments inuence
the role which grandparents play is limited. This research
seeks to address this knowledge gap and inform debate on
policy inuencing the grandparental role.
Lower fertility and increased life expectancy mean that
over the next two decades a fth to a quarter of the
population in many European countries will be aged over
65.1 Population ageing is leading to increased emphasis
on the health and well-being of older people, with an
expectation that older men and women participate for
longer in paid work. At the same time there is often an
implicit assumption that older people will continue to play
a vital caring role within their families. Grandparents are
important providers of childcare, enabling mothers to
1 Commission of the European Communities 2005. Green paper,
‘Confronting demographic change: a new solidarity between the
generations.’ Brussels.
enter or remain in paid work. They may also need to step
in to take on the full-time role of raising grandchildren
in difcult and distressing circumstances if parents are
unable to do so, for example due to death, physical
or mental health problems, drug or alcohol misuse, or
imprisonment.
How far grandparents’ informal caring roles can be
combined with paid work is highly relevant for public
policy, not only in relation to family and the labour market
but also pensions and retirement, and for understanding
inequalities across the life course. As we understand more
about the role that grandparents play across Europe, we
realise that it is important to implement social policies
that help sustain these important, complex and potentially
fragile social relationships.
This study examines international data from European
countries on grandparenting from SHARE (Survey of
Health, Ageing and Retirement in Europe), ELSA (the
English Longitudinal Study of Ageing) censuses and other
data sources in addition to mapping data on parental and
grandparental policies for leave and exible work, family
support from the state in the form of childcare and family
benets, retirement and adult care policies, and labour
market, childcare and family cultures and structures, to
address the following questions:
1. How do the living arrangements of grandparents vary
within and across European countries and how have
they changed over time?
2. How do the characteristics of grandparents vary across
Europe in terms of age, living arrangements, socio-
economic status, education, marital status, participation
in paid work, retirement status and health?
3. How does the level of involvement of grandparents
with their grandchildren vary across Europe in terms
of contact, help and care? What characteristics
of grandparents help to explain the diversity of
arrangements?
4. How do family policies vary, and how are these
variations in policy related to observed diversity in
the levels of involvement of grandparents with their
grandchildren?
11
Key ndings
Our study shows that across Europe grandparents, and
grandmothers in particular, are playing a major role in
providing both intensive and occasional care for their
grandchildren. 44% of grandparents in the 11 European
countries2 studied provide grandparental childcare
without the child’s parents present, while in Britain the
British Social Attitudes (BSA) survey showed that 63% of
grandparents with a grandchild under 16 do so.3
Younger grandmothers who are t, healthy and with
younger grandchildren – the most likely to be providing
care for their grandchildren – are the very women that
governments across Europe are aiming to encourage
to stay in paid work for longer, in order to increase
productivity and pay for their own pensions, health and
social care in later life. Their vital but invisible role in
providing childcare, whether intensive, regular and/or
occasional, is likely to conict with their own ability to self-
nance their old age, especially as widow’s benets in both
state and employer pension schemes are eroded.
England and Wales, like the US, has experienced an
increase in the prevalence of skipped-generation
households – households consisting of grandparents and
grandchildren but without the parents. This rose from
0.25% of adults aged 35 and over living in such households
in 1981 to 0.42% in 2001. These households are likely to
experience poverty and disadvantage. No other European
country studied so far follows this pattern.
Our study shows considerable variations in the
characteristics of grandparents across the European
countries studied. English grandparents are relatively
young, more likely to be in paid work and have more
grandchildren on average than grandparents in the
remaining 11 European countries. In England one in four
(23%) grandparents aged 50 and over are in paid work,
compared with an average of just one in seven across the
other 11 countries studied. Only Denmark and Sweden
have a higher percentage of working grandparents.
While overall grandparents in the European countries
studied provide high levels of childcare, there are
striking variations in the intensity and frequency of the
care provided. In France, Denmark, Sweden and the
Netherlands between 50% and 60% of grandparents
provide some childcare compared with just 40% in the
Southern European countries. However, regular and
intensive grandparental childcare is more common in
Southern Europe, with 20% of grandparents in Italy
providing almost daily childcare compared with just 2% of
grandparents in the Netherlands.
Across the European countries studied grandparents
who are younger, with higher educational levels, in better
health, and whose youngest grandchild is under age six are
more likely to provide childcare.
Differences in the characteristics of grandparents in the
2 The 11 SHARE countries are Austria, Belgium, Denmark, France,
Germany, Greece, Italy, the Netherlands, Spain, Sweden and
Switzerland.
3 Wellard, S. 2011. Doing it all? Grandparents, childcare and
employment: An analysis of British Social Attitudes Sur vey Data from
1998 and 2009. London: Grandparents Plus.
different countries (such as age and marital status) explain
some of the differences in grandparental childcare across
the 12 European countries, however there are signicant
differences between countries too. The research nds that
different family policy contexts are associated with varying
patterns of grandparental childcare.
In countries such as Sweden and Denmark (and to a
lesser extent, France) where parents are expected to work
full-time, formal childcare is widely available, and there
is generous maternity pay and support for mothers who
stay home - grandmothers play a far more limited role
in providing intensive childcare, but are still signicantly
involved in providing occasional and less intensive care for
grandchildren.
In Portugal, Spain, Italy and Romania, where welfare
payments to parents and mothers at home are limited,
there is little formal childcare and few opportunities
for mothers to work part-time, grandparents provide a
great deal of intensive childcare for their grandchildren.
Moreover, in these countries, mothers who do work often
do so for 40 plus hours a week, and since there is little
affordable formal childcare, there is greater reliance on
intensive care by grandmothers. With the exception of
Romania, in these countries there is less of a role for
grandparents providing occasional or less intensive care
without the parents present.
In the UK, Germany and the Netherlands where public
support for families is varied but less universal, childcare
coverage is patchy and often provided by the market
rather than the state, and the norm is that women work
part-time, grandparents generally play a middling role in
both intensive childcare and occasional/less intensive
childcare. In these countries a smaller proportion of those
mothers in full-time work do so for long hours, leading to
less reliance on intensive childcare by grandmothers. In
the Netherlands, which has by far the highest proportion of
mothers working part-time and very few mothers working
full-time, and where formal childcare is widespread,
there is very little intensive grandparental childcare by
grandparents.
In general, countries with the lowest usage of formal
childcare, Hungary, Portugal and Romania, have the
highest percentages of grandmothers caring intensively for
their grandchildren, and countries with the highest usage,
Sweden and Denmark, have the lowest percentages of
grandmothers providing intensive childcare.
In countries with higher percentages of older women in
paid work there is less involvement of grandmothers in
intensive childcare.
Given that grandmothers aged 50 to 69 who are not in paid
work are the most likely to provide childcare, the plans
of European governments to extend retirement ages and
increase female labour force participation at older ages
are likely to conict with their role in providing childcare,
and therefore has signicant implications for labour
market participation by younger mothers and for pension
acquisition and the nancial security of mid-life women.
12
Grandparent-headed households:
three-generation and skipped-
generation households
The study looks at trends over time in the prevalence
of adults living in grandparent households (both three-
generation and households with the parents’ generation
absent) in England and Wales, France, West Germany,
Romania and Portugal.
In England and Wales, France and West Germany
there has been a decline in the percentage of adults
aged 35 plus living in three-generation households.
In England this declined from 3.3% in 1981 to 1.5% in
2001, the latest period for which data is available. In
Romania, and also the US, there has been an increase over
the same time frame.
In England and Wales, like the US, there has been
an increase in skipped-generation households, from
0.25% of adults over 35 living in such households
in 1981 to 0.42% in 2001. This most likely reects the
increase in kinship care (wider family members raising
children) identied by Nandy and Selwyn’s analysis of
Census microdata.4
Both three-generation and skipped-generation grandparent
households are associated with poverty and socio-
economic disadvantage in all the countries studied.
Adults living in grandparent households are more likely to
be women, divorced, widowed or separated, with lower
educational levels, and economically inactive, and this
is particularly marked for those in skipped-generation
grandparent households.
Grandparent characteristics in 12
European countries
The study looks at the characteristics of grandparents
aged 50 and over from the following countries:
England and France
Denmark and Sweden (Scandinavia)
Germany, the Netherlands, Belgium, Austria and
Switzerland (Western Europe)
Spain, Italy and Greece (Southern Europe)
Percentage of older adults who are
grandparents
In all 125 European countries studied the majority of
women over 50 are grandmothers, ranging from 72% in
Denmark to 53% in Switzerland. In most countries the
majority of men over 50 are grandfathers, ranging from
62% in Belgium to 42% in Greece. In England 67% of
women over 50 are grandmothers and 58% of men over
50 are grandfathers.
Overall the highest percentage of older adults who
are grandparents are in Scandinavia and Belgium,
4
Nandy, S., Selwyn, J., Farmer, E. and Vaisey, P. (2011) Spotlight on
kinship care: Using Census microdata to examine the extent and
nature of kinship care in the U.K., London: University of Bristol.
511 SHARE countries plus England
followed by England and France, and the lowest
are in Southern Europe. Higher rates of grandparents
in Scandinavia and Belgium are likely to reect both
higher fertility among adult children and younger ages at
childbearing.
Age, gender and marital status
of grandparents
The youngest grandparents are in Denmark (mean age
67) and the oldest are in Greece (mean age 70). The
highest percentage of working-age grandparents (50 to
64) is in Scandinavia, with Denmark at 50%. The lowest
percentages are in Southern Europe, with around a third
aged 50 to 64 in Spain and Italy. The percentage of
working-age grandparents is relatively high in England
(41%).
In all countries the majority of grandparents are
women, ranging from 56% in Sweden and 57% in England
to 61% in Greece.
Marital status as well as age and gender is likely to be a
factor in whether grandparents provide childcare. The
highest percentage of still-married grandparents is in the
Netherlands, at 70%, with 69% in England. Grandmothers
are more likely than grandfathers to be widowed in all
countries.
Children and grandchildren
Even though in the Netherlands and Spain, grandparents
have more children (a mean of almost 3.0) compared with
2.7 in England, English grandparents have the most
grandchildren – an average of 4.9 compared with
4.2 across the other countries studied. The lowest
numbers of grandchildren are in Germany and Austria
(3.7) and Greece (3.8). English grandmothers have on
average 5.2 grandchildren, and English grandfathers have
4.6 grandchildren, more than in any other country.
Grandparental involvement in children’s lives is likely to
depend on the number of grandchildren, and also their
age. Among the European countries studied over half of
grandparents have at least one grandchild under the age
of 6. In the Netherlands, 40% of grandparents have a
grandchild under age 3, compared with just 18% in Austria.
Sandwich generation grandparents
Grandparents in the Scandinavian countries and France
show the highest percentages who are in the sandwich
generation with at least one of their own parents still alive,
at around 22%. The lowest is in Italy (12%). Relatively
fewer grandparents in England are in the sandwich
generation (15%). Our analysis includes all grandparents,
not just those with grandchildren under the age of 16.
Among these grandparents, as the BSA survey analysis
shows, the percentage of grandparents with their own
parents is much higher at 28%.
Education, economic activity and
wealth
There is a wide range of educational levels across Europe,
with over 80% of grandparents reporting a low educational
level in Southern European countries compared with just
13
25% in Germany. On average across the 11 countries
in SHARE 59% report a low educational level (56% in
England), 28% a middle level (28% in England) and 13% a
high educational level (16% in England).
Only in Sweden do grandmothers report higher
levels of education than grandfathers. There is
also wide variation in the percentage of grandparents
in paid work, from 29% in the Scandinavian countries
to 9% in Italy. Almost one in four (23%) of English
grandparents are in paid work, compared with the
average across the 11 countries in SHARE of one in
seven.
Across Europe grandmothers are poorer
than grandfathers, in part reecting the fact that
grandmothers tend to be older and are more likely
to be widowed than grandfathers. The percentage of
grandmothers who are in the poorest 20% of the wealth
distribution for people over 50 ranges from 23% in
Denmark to 32% in Germany, while for grandfathers those
in the poorest 20% range from 16% in France to 24% in
Italy.
Health and well-being
There is a wide range in the percentage of grandparents
across the countries studied reporting their health as
fair or poor, from just 12% of grandfathers and 16% of
grandmothers in Sweden, to 45% of grandfathers and 48%
of grandmothers in Germany. English grandparents rate
their health better in comparison to the average across the
11 countries in SHARE, with 31% of grandfathers rating
their health as fair or poor compared with the average of
38%. 30% of English grandmothers rate their health as fair
or poor compared with the average of 44%.
English grandparents, along with those from
Denmark, are least likely to report four or more
depressive symptoms (18%) while Spanish, French and
Italian grandmothers report particularly high levels (over
40%).
However, English grandparents have the highest
levels of health or disability-related limitation in
activities in daily living across the study, with almost
one in four reporting one or more limitation, compared
with 12% of grandmothers and 14% of grandfathers
elsewhere.
Grandparents across Europe tend to have poorer rates of
cognitive function than over 50s who are not grandparents,
reecting the fact that they tend to be older. After taking
age into account, differences between countries in
grandparents’ cognitive functions are small.
Grandparental Childcare
The research shows a high level of grandparental
involvement in childcare across Europe. 44% of
grandparents in the SHARE countries have looked after a
grandchild without the presence of the parents in the last
12 months. The highest level of grandparents providing any
grandparental childcare is in the Netherlands and Denmark,
with around 57% of grandparents looking after a grandchild
in the past 12 months, and the lowest rates are in Germany,
Austria, Switzerland and the Southern European countries,
at around 40%.
In Britain, the British Social Attitudes (BSA) survey
shows that 63% of grandparents with grandchildren
under 16 reported that they had ever looked a
grandchild in the last 12 months, compared with 50%
elsewhere in Europe who had provided some type of care
for a grandchild under 16 without the parents present.6
11% of grandparents across the 11 countries in
SHARE provided daily or almost daily care, ranging
from 20% in Italy and Greece to 2% or lower in the
Scandinavian countries and the Netherlands.
The BSA survey shows that 19% of grandmothers and
14% of grandfathers in Britain with grandchildren under
16 reported providing 10 hours a week or more of
childcare for one or more of their grandchildren. 6% of
all grandparents in Britain with a grandchild of any age
looked after a grandchild in the past week, averaging 30
hours a week.
Who are the grandparents
providing childcare?
The analysis found that grandparents providing
childcare are likely to be female, younger, with
a partner, with a higher educational level and in
higher wealth quintiles, and with better health and
younger grandchildren. Overall grandparental childcare
is associated with socio-economic advantage and being
younger.
Which parents are more likely to
receive childcare from a grandparent?
Parents7 in northern European countries are more likely
to have a child looked after by grandparents than those in
Austria, Switzerland and southern European countries.
However, for regular childcare the situation is the reverse:
parents in Scandinavian countries are least likely to have
their children regularly looked after by grandparents, while
parents in Italy, Greece and Belgium are most likely to
have children looked after regularly.
Mothers, especially those who have never been
married are more likely to have a child looked after by
grandparents. The younger the parent the more likely it is
that their child is looked after regularly by grandparents.
Parents whose youngest child is under six are more likely
to have a child looked after by a grandparent. Overall 55%
of parents whose youngest child is aged between nought
and two receive grandparental childcare for their children,
59% of those whose youngest child is aged three to ve,
and 48% of those whose youngest child is aged six to
11. Only 11% of parents whose child is aged 12 or older
receive grandparental childcare. Parents who live closer
to grandparents are more likely to have their child looked
6 Hank, K. & Buber, I. 2009. Grandparents caring for their
grandchildren ndings from the 2004 Survey of Health, Ageing, and
Retirement in Europe. Journal of Family Issues, 30, 53-73.
7 Please note that we do not have a representative sample of parents
in SHARE. What we do have are the selected characteristics of up to
four adult children given to us by the older person. We know which of
the adult children the older person identies as being a parent and we
also know whether these parents have been identied (that is by their
older mother or father) as being given grandparental childcare.
14
after by a grandparent. 38% of parents who live within ve
kilometres of a child’s grandparent received grandparental
childcare, compared with 20% of those who live more than
100 km away.
Overall a higher percentage of mothers who work part-
time have a child looked after by a grandparent than those
who work full-time. However the pattern varies across
countries.
For mothers, overall a higher percentage of those in paid
work receive grandparental childcare compared with those
who are not in paid work, however the reverse is true for
mothers in Scandinavia where those who are not in paid
work receive more help from grandparents than those who
are.
Family policy and patterns of
grandparenting
The report considers the extent to which differences in
the ways that grandparents care for grandchildren across
Europe might be accounted for by differences in family
and childcare policy, as well as related work and childcare
settings and cultural attitudes. Countries differ markedly in
the extent to which women and mothers participate in paid
labour and the extent to which people have access to and
use formal childcare. Cultural factors also shape different
preferences and norms for childcare, with variation across
Europe in beliefs about what is best for families and
children.
This element of the research focuses on care by
grandmothers, since grandfathers rarely provide childcare
in the absence of parents without grandmothers present.
Outcomes are examined in eleven countries, selected
for this analysis to provide clear examples of countries
with different policy environments, labour force and
childcare structures and varying family, care and work
cultures: Denmark, France, Germany, Hungary, Italy, the
Netherlands, Portugal, Romania, Spain, Sweden and the
United Kingdom.
Across Europe there is increased participation in the
labour market by women and mothers, with all countries
providing some support for leave from paid work and
childcare. Nevertheless, signicant gender differences
remain. Unstable and inexible labour markets and
underfunded or fragmented childcare are factors which
push mothers to nd alternative forms of care, or to leave
paid work or work fewer hours in order to care for their
children.
We have developed a framework for cross-country data
analysis to examine the relationships between family
and care policy and outcomes, labour market structures
and participation (especially of mothers), and family and
gender cultures. The objective is to analyse political,
cultural and employment settings in different countries
that help explain the level and intensity of grandmaternal
childcare. Across the three spheres of “policies”, “labour
markets” and “family and gender cultures” we explored
a raft of approximately 250 indicators for each of the
eleven countries on all kinds of parental and non-
parental leave, cash benets, childcare and elder care
and retirement policies, as well as data on female labour
market participation and attitudinal data. We classify
policies according to the extent to which the state
encourages or assumes a role for grandparents. We used
a qualitative constant comparative method suggested by
existing theoretical understanding of family policy and
labour markets to examine which variables (indicators)
were associated with each other and how, and how these
associations and interactions varied between countries. We
then used this analysis to cluster our countries according
to similarities and differences between them on these
indicators, and in the ways that these indicators were
associated with each other. We then considered these
clusters in detail, narrowing down our variables to those
that seemed most important in explaining how and why
grandparental care varied from country to country.
Findings from the policy analysis
Our analysis indicates that there is a close relationship
between the family and care policy context and the
likelihood that grandmothers are providing intensive
childcare. In terms of constellations of policies, we found
that our countries clustered into three groups. In the
rst group, exemplied by Sweden and Denmark, the
Scandinavian countries, and to a lesser extent, France,
the state organises and provides childcare, there is no
assumption that grandparents will provide care and all
transfers and benets are available only to parents. In
these countries where both parents expect to work full-
time, formal child care is well provided and there are good
maternal benets, fewer grandmothers provide intensive
childcare.
In the second group, there is an assumption that
grandparents will provide care – the southern and eastern
European countries studied fall into this group. In Hungary,
Portugal and Spain, this assumption is explicit, but Italy
and Romania are also considered part of this group since
policy vacuums leave a childcare gap that in practice
can only be lled by grandparents – the assumption that
they will provide care is implicit. In these countries there
are few part-time jobs, limited formal childcare and only
limited in-kind family benets, and more grandmothers
provide intensive childcare.
In a third group of countries public support is varied but
less universal, childcare coverage is patchy and provided
more by the market than the state, and women are
more likely to work part-time. Here grandparents play a
moderate role in both intensive childcare and occasional/
less intensive childcare. The UK, Germany and the
Netherlands are examples of these countries, although
the Netherlands, for the reasons given below, has very
low percentages of grandmothers providing intensive
childcare.
The family and care policy environments are however only
one part of the picture. The pattern of female labour force
participation in a country is associated with childcare by
grandmothers, independently of the policy context. Long
working hours for mothers and little institutional childcare
mean more grandmothers providing intensive childcare.
In countries where a high percentage of mothers with
young children do not work, those mothers who do work
are particularly reliant on intensive grandmother childcare.
Also, lower labour force participation among women aged
50 to 64 is associated with more intensive grandmother
childcare.
15
Use of formal childcare for young children is inversely
related to intensive childcare by grandmothers.
Furthermore, in those countries where maternal care
for pre-school children is the preferred norm, childcare
patterns suggest that grandmothers are regarded as the
best care substitute for those mothers who work in the
paid labour market.
Finding from the multivariate analysis:
Grandparental characteristics
associated with childcare
We used a wide variety of multivariate techniques as
appropriate to investigate which individual and country-
level characteristics are related to grandparental childcare.
Such analyses have several advantages. They permit us to
explore the relationship of each characteristic in relation
to grandparental childcare while taking into account the
potentially confounding inuence of other characteristics.
For example, in our descriptive analyses we found
signicant differences in the percentage of grandparents
in paid work across countries; such differences may help
to explain variations in grandparental childcare. However,
we also know that this is confounded with age, that is, an
older grandparent is less likely to be in paid work. Thus
we need to know whether it is being in paid work or age
(or both) that is driving the relationship to grandparental
childcare. Our presentation of analyses in the following
sections considers these questions with respect to all of
the characteristics discussed so far.
Intensive, non-intensive and no
childcare provision
First, we present our ndings for the three types of
grandparental childcare simultaneously, that is intensive
grandparental childcare, non-intensive grandparental
childcare, and no grandparental childcare. This is because
we want to understand the relative importance of
grandparent characteristics for each level of care and how
they relate to each other. We used a generalised ordinal
logit model (in our case with partial proportional odds).
Multivariate analysis shows that the grandparents
most likely to provide any (intensive and non-
intensive) childcare are female, young, married,
retired, and in the higher wealth quintiles. Married
grandparents are more than one and half times as likely
to provide any grandparental childcare as unmarried
(i.e, never married, widowed or divorced) grandparents.
Grandparents with lower levels of education are
signicantly less likely to provide any childcare; however,
they are more likely than those with high educational
levels to provide intensive grandparental childcare.
Grandparents with several grandchildren are signicantly
more likely to provide any grandparental childcare than
those with just one grandchild, but having more than one
grandchild is not signicantly associated with providing
intensive grandparental childcare. Grandparents with a
youngest grandchild between the ages of three and ve (in
comparison to ages one to two) are the most likely to be
providing any grandparental childcare. Grandparents whose
youngest grandchild is aged over six are signicantly less
likely to be providing care in comparison to grandparents
with a youngest grandchild between ages one and two.
Grandparents with better cognitive function are more likely
to provide any type of grandparental childcare but the
effect is greater for more intensive care. A similar pattern
is found when severity of health or disability related
functional limitations are considered.
We used our model to examine whether different policy
environments still retain some explanatory power once we
have taken into account the extent to which the personal
characteristics of grandparents differ across countries.
Multivariate analysis shows that even when we account
for the widely varying characteristics of grandparents
across Europe, different national policy contexts are still
associated with different levels of grandparental childcare.
For example, Danish and Swedish grandparents (which
fall into our category of countries where no grandparental
care is assumed by the policy context) are signicantly
more likely to provide some grandparental childcare, but
signicantly less likely to provide intensive grandparental
childcare than those countries with more neutral policy
regimes towards grandparental childcare, such as Germany
Grandparents in countries that fall into our category
of having policy contexts that assume grandparental
childcare, (e.g. Spain, Italy and Greece) are less likely to
provide some grandparental childcare but more likely to
provide intensive grandparental childcare than countries
with more neutral policy regimes, like Germany.
Grandparents in the countries where the policy context is
relatively neutral toward grandparents (i.e. Germany, the
Netherlands, Austria and Belgium), fall into a middle group
when considering the provision of intensive childcare –
providing more than in the Scandinavian countries but less
likely to provide intensive care than those countries where
policy assumes a grandparental role. In the provision of
any care, there is a much more even picture across all the
countries studied, with grandparents quite similar across
the SHARE countries in providing at least some care for
their grandchildren. However this analysis does show that
Germany and Austria are similar to Italy and Spain with a
lower likelihood that grandparents will provide some care,
while grandparents in the Netherlands and Belgium have
the highest likelihood of grandparents helping out with
care at least some of the time.
Intensive childcare provision
Multivariate logistic regression analysis was conducted to
explore which grandparental characteristics are associated
with intensive grandparental childcare – i.e. daily or at
least 30 hours a week of care. Characteristics considered
were gender, age, marital, employment and health status
and number of grandchildren.
Grandmothers are one and a half times more likely
to provide intensive grandparental childcare than
are grandfathers. Younger grandparents, and those
who are married or cohabiting are also more likely
to be providing intensive grandparental childcare.
Grandparents with lower educational levels and retired
grandparents are more likely to provide intensive
grandparental childcare. Retired grandparents are one
and a half times more likely to provide intensive childcare
than those grandparents in paid work (even taking age into
account). Wealth and the number of grandchildren are
16
not signicantly associated with providing intensive grand
parental childcare.
Among the various health indicators considered, functional
limitations and cognitive function are signicantly (and
negatively) associated with the provision of intensive
grandparental childcare. Grandparents without health
or disability related limitations are almost twice as
likely as those with such conditions to be providing
intensive childcare. However there is no signicant
relationship between self-rated health and providing
intensive childcare.
Our policy context classications help us to a large extent
to understand the hierarchy of countries when considering
the extent to which grandparents provide intensive
childcare for their grandchildren even after taking account
of other differences in grandparents’ characteristics across
countries. Grandparents in Sweden and Denmark (in
our classication of countries where policies assume no
grandparental care) for example were only around half
as likely as grandparents in England to provide intensive
childcare. England, the Netherlands, and Switzerland are
quite similar to each other in the provision of intensive
grandparental childcare, whereas the likelihood of
grandparents providing intensive childcare in France,
Germany, Austria and Belgium is between one and a half
and three times as high as in England. With the exception
of France, these latter countries are all classied into our
middle group of neutral countries; Spain, Italy and Greece
stand out however as having much higher likelihood of
grandparents providing intensive care – three to ve times
higher than in England, countries where policies assume
grandparental care.
Multilevel analyses taking country
indicators into account
So far, analyses above considered the different policy
contexts and their relationship to different levels of
grandparental childcare. In this model we examined
whether it is the policy context or the cultural and
institutional factors which these contexts produce and
reect that has more explanatory power in explaining
variation in grandparental childcare, again taking into
account the variation in individual characteristics of
grandparents across Europe. We nd that considering
the policy context groupings does get us a long way in
understanding grandparental childcare, but we can explain
even more of the variation when we look at the extent
to which differences in the cultural-contextual factors
across European countries are related to grandparental
childcare (while still taking grandparental characteristics
into account). We use multilevel logistic regression models
to look at intensive grandparental childcare, taking four
key country-level variables into account: the percentage
of mothers aged 25-49 who are not in paid employment
and the percentage of women aged 50-64 in paid work,
capturing the two-generation structure of the labour
market; the percentage of individuals who strongly agree
with the statement that “pre-school children suffer with a
working mother” capturing societal attitudes towards care
and gender; nally, the percentage of children under the
age of three who are enrolled in formal childcare, used as
an indicator of the use of formal childcare.
These models show that policies and cultural-structural
factors all shape the extent to which grandparents
provide intensive childcare in European countries. In
particular, certain country characteristics seem to provide
arrangements in which grandparents are more likely to
engage intensively in providing intensive childcare, even
when all the variation in grandparents’ characteristics
is taken into account. The extent to which mothers in a
country are not in the paid labour force is associated with
the degree of policy focus on providing formal, affordable
childcare, particularly for very young children. Similarly,
in countries where mothers are expected to stay at home
to care for their families there is also a belief that pre-
school children would suffer with working mothers. In
such ‘pro family care’ countries, opportunities for young
mothers (aged 25 to 49) to work exible hours also tend
to be limited; mothers who do work in countries where the
normative expectation is to stay at home to care for their
families tend to work full-time. Hence, mothers who work
in such countries need the co-operation of grandparents,
and grandmothers in particular. However, the availability
of grandmothers to offer such help is reduced in countries
where employment rates for women 50 to 64 are
comparatively high.
17
Conclusions
Our analysis indicates that across Europe grandparents
are playing a major role in providing childcare for
grandchildren.
We have found that in countries where formal childcare is
limited and benets for families and stay at home mothers
are not generous, grandparents are providing intensive
levels of childcare. In Italy and Greece for example almost
a quarter of grandparents look after their grandchildren,
without the parents there, for around 30 hours a week, and
more than one in ve grandmothers are providing almost
daily care. In these countries there are fewer opportunities
for mothers to work part-time, and those mothers who are
in work tend to work full-time.
On the other hand in countries where there is extensive
provision of formal childcare, generous maternity and
family benets and support for stay-at-home mothers,
grandparents are much less likely to be providing intensive
childcare, but much more likely to be providing occasional
care without the parents present.
In France, Denmark, Sweden and the Netherlands up
to 60% of grandparents provide some childcare, and in
Britain the gure is 63% for those with a grandchild under
16. In these countries mothers are much more likely to be
working, and grandparents are acting as a ‘reserve army’
of care. In many instances grandparents are likely to be
providing care to support working mothers, for example,
during school holidays and when children are ill and in
other family emergencies, or providing less intensive
regular childcare to complement formal childcare.
Across all countries our analysis shows that grandparents
who provide childcare tend to be younger, healthier,
married and to have higher educational levels, and also
to classify themselves as retired. These are the very
women whom governments across Europe are seeking to
keep longer in the labour market to grow our economies
in response to ageing populations, with fewer younger
workers entering the labour market and an increased life
expectancy. This conict between grandmothers’ role in
providing childcare and increased participation in paid
work both to protect their own retirement incomes and to
grow our economies has major implications for the future
paid employment of mothers of young children, as well as
for their own nancial security in later life.
As our populations age the role of grandparents in
family life is likely to become even more signicant.
Already, 17% of grandparents across Europe are in the
sandwich generation with their own parents still alive.
As life expectancy increases further this percentage is
likely to increase. Younger grandparents, most likely to
have younger children and grandchildren are of course
more likely to still have a parent alive. In Britain, 28% of
grandparents with a grandchild under 16 have a parent
still alive, six in 10 are still working and nearly eight in 10
are providing some care for grandchildren. This group
of grandparents is already under pressure to provide
work and care up and down the generations. Austerity
programmes leading to cuts in provision for both elder
and childcare risk putting yet more pressure on these
younger grandparents. Policymakers need to consider the
implications for the future nancial security of this mid-
life generation, as well as the implications of work, care
and retirement policies for those in mid-life on younger
working parents.
When we consider the experiences of other countries
in Europe it is clear that the UK faces a stark choice.
We can either prioritise grandmothers remaining in the
labour market for longer and thus supporting their own
retirement, but acknowledge that over time this is likely to
create a care gap for working parents, largely impacting
on mothers’ employment; or we can invest in universal,
affordable formal childcare which will meet, at least in
part, that emerging childcare gap and retain both older
women and working mothers in the labour market. A third,
and arguably the least attractive option would be to decide
to reverse the trend for working longer and rely heavily
on our ageing population to provide the childcare. Doing
so would create an even bigger pensions and care funding
gap for older generations and would quickly prove to be
unsustainable.
18
Introduction
1.1 Overview
Across Europe increased life expectancy means that it
is now quite common for a child to grow up while their
grandparents and even great grandparents are living. Our
ageing populations, and other demographic changes such
as more mothers in the labour market and higher levels
of relationship breakdown, indicate that grandparents
are likely to play an increasingly signicant role in family
life. The austerity measures and cuts to public services
being implemented in many countries in response to the
current international nancial crisis are likely to lead to
a greater expectation that grandparents will step in to ll
the care gap. Yet our knowledge and understanding of
grandparenting, and how different policy environments
inuence the role which grandparents play is limited.
This research seeks to address this signicant gap in our
knowledge, and to inform debate about the policy issues
surrounding the grandparental role.
Our main focus is on grandparenting in terms of
engagement in childcare. We recognise that there are other
important aspects of grandparenting which we were not
able to explore in detail here (e.g. support for parents by
looking after children in their presence, gifts to help young
adult grandchildren get a start in life, go to university or
buy a property, etc.). In addition to childcare, we also
examine intergenerational co-residence in this report, as
this is likely to involve childcare, for example in the case of
grandparents co-residing with adolescent or young adult
grandchildren. However, we recognise that this also may
capture households where co-residence may be due to
the older person’s need for support. In this report we also
briey touch on some of the more complex aspects of
grandparenting, for example, with respect to the ‘sandwich’
generation, that is those with potential commitments
across generations such as grandparents with their own
parents alive (see Chapter 5).
All countries in Europe face population ageing, the result
of declining fertility and increasing life expectancy. Within
the next fteen to twenty years, a fth to a quarter of the
population in many European countries will be aged 65
and over (Commission of the European Communities,
2005). An ageing population is placing greater emphasis
on improving health and well-being at more advanced
ages. As retirement ages are put forward, older people
are expected to participate in paid work for longer, but at
the same time also to undertake critical roles in caring for
children and adults. The question of how far these two
activities can or should be pursued simultaneously and
how far they must be regarded as alternatives is highly
relevant to policy formulation, and yet this informal
contribution is usually unpaid and unrecognised in policy.
Understanding the role of grandparents in supporting
and maintaining families is an important element of the
evidence base, not only for family and labour market
policies, but also for pension and retirement policies, and
for understanding inequalities across the lifecourse.
1.2 Aims and objectives
Our project investigates variations across Europe in the
diversity of grandparents, how grandparents contribute
to childcare, and how policies are related to patterns of
grandparenting (with particular reference to childcare).
We are grateful for the support of the Calouste Gulbenkian
Foundation which enabled us to carry out a scoping study
in January–April 2010. This study revealed that despite
its growing importance as a matter of policy, there is little
research examining what grandparenting looks like across
Europe, or how policies and contextual-structural factors
in different European countries inuence grandparenting.
Grandparents have always provided nancial, emotional
and practical care and support to their children and
grandchildren, and this support has generally been taken
for granted by families, communities and governments
alike. Grandparents are particularly important where they
become the primary carers for their grandchildren in
difcult and distressing circumstances because the child’s
parents are unable to do so (for example, due to death,
physical or mental health problems, drug or alcohol misuse
or imprisonment), or when the parents are still very young.
They are also important as informal providers of childcare
enabling mothers to enter the paid labour force – a specic
policy aim across the European Union. As we learn more
about the grandparental role around Europe, we realise
that to achieve caring and productive societies, it is
important to implement social policies that help to sustain
these important, complex and potentially fragile social
relationships. The role and contribution of grandparents is
currently little acknowledged in policy and the law accords
grandparents few rights.
By comparing different European countries, we can
develop a clearer understanding of what types of family
policies help to support the family including the extended
family, and in what circumstances. To do this, we address
the following questions:
1. How do the living arrangements of grandparents vary
within and across European countries and how have
they changed over time?
2. How do the characteristics of grandparents vary across
Europe in terms of age, living arrangements, socio-
economic status, education, marital status, participation
in paid work, retirement status and health?
3. How does the level of involvement of grandparents
with their grandchildren vary across Europe in terms
of contact, help and care? What characteristics
of grandparents help to explain the diversity of
arrangements?
4. How do family policies vary, and how are these
variations in policy related to observed diversity in
the levels of involvement of grandparents with their
grandchildren?
1.3 Datasets and methodology
We used a wide variety of data sources to address the
research questions including samples of census responses,
the English Longitudinal Study of Ageing (ELSA), and
SHARE (Survey of Health, Ageing and Retirement
in Europe). In addition, we used data from Eurostat,
the OECD (Organisation for Economic Co-operation
and Development), the European Social Survey (ESS),
Eurobarometer and other national and international
19
sources, as well as published work on maternal, parental
and other child-related benets and leave, to inform the
policy analysis. Consequently, the geographical scope of
the study varies as it was not possible to address all the
research questions using the same set of countries (for
more information on data sources see Appendix B).
For the rst research question, we examine patterns of
co-residence between grandparents and grandchildren
over time (with or without the parents being present) in
5 European countries and the US using: the Integrated
Public Use Microdata Series International (IPUMS)
for France, Portugal, Romania and the US, the ONS
Longitudinal Study (ONS LS) for England and Wales, and
the German Socio-Economic Panel Study (SOEP) for West
Germany. The IPUMS offers samples of census data which
have been cleaned (that is, checked for anomalies) and
harmonised. These countries were selected because they
had compatible data over three decades and because they
allowed grandparents and grandchildren to be identied
within households.
For the second and third research questions, we investigate
the characteristics of grandparents and grandparents’
involvement with their children (addressing the second
and third research objectives) across 12 European
countries using the English Longitudinal Study of Ageing
(ELSA) and the Survey of Ageing, Health and Retirement
(SHARE) which includes Austria, Germany, Sweden,
the Netherlands, Spain, Italy, France, Denmark, Greece,
Switzerland, and Belgium. Both surveys are based on
people aged 50 and over and their partners and are
comparable. We use the rst wave of data collected in
2002/03 for ELSA and 2003/4 for SHARE. ELSA has
information on close to 12,000 people and SHARE’s
sample size in this wave was 29,917 people aged 50 and
over (ranging from 1,707 in Denmark to 3,193 in France).
These data sources permit the detailed study of
grandparenthood and grandparenting as they ask
respondents whether they have grandchildren and how
many and whether they regularly or occasionally looked
after their grandchild(ren) (without the children’s parents
being present). So far, only limited analysis on this topic
has been carried out and published (Albertini et al., 2007,
Hank and Buber, 2009). While studies show considerable
variation in grandparental childcare across countries, few
have considered both contextual characteristics (such as
the policy environment) as well as individual ones (Igel
and Szydlik, 2011, Jappens and van Bavel, 2011). Even the
few studies that have taken contextual factors into account
use mostly broad indicators of country-level factors (for
example, expenditure on families), consider this issue from
the parents’ rather than the grandparents’ perspective;
and have largely not taken cultural factors into account
(Igel and Szydlik, 2011, Jappens and van Bavel, 2011).
Our analysis in this report examines both individual-level
characteristics and country-level indicators capturing
family and labour market cultures and structures (from the
perspective of both parents and grandparents) to explain
variations across Europe in grandparental childcare. See
Appendix B for our detailed methodology.
The nal research question addresses the policy strand
of the study and consists of two steps; the policy analysis
itself and a combination of the policy and demographic
analysis. The rst step, the policy analysis, involves 11
European countries using three criteria: rst, geographical
spread; second, representation of different types of
welfare regimes and of economic characteristics; and third,
inclusion in most cross-national data sources. Thus, the
countries chosen for this analysis are: Sweden, Denmark,
The Netherlands, Germany, France, the UK, Spain,
Portugal, Italy, Hungary and Romania.
Drawing on data from national ministries of work and
family together with international policy sources such as
MISSOC (Mutual Information System on Social Protection)
and the International Network on Leave Policies and
Research (INLPR), this analysis draws on established
methodologies for studying family policy across Europe,
inuenced in particular by the approach advocated by
the EU Government Expert Group on Demographic
Issues (2009).8 In previous studies family policies have
not been analysed according to their implications for
grandparental childcare and intergenerational relations;
the analytical focus has almost always been on mothers’
paid employment. We used desk- and internet- based
research to map family policy across three spheres for
each European country studied: (i) parental policies, (ii)
child benets, family allowances and childcare services
and (iii) policies impacting directly on grandparents as
entitled persons. Our policy mapping has been checked
with the project’s advisory group of European experts
(see Appendix A). As expected this mapping reveals that
countries cluster into regimes of grandparental childcare,
reecting different family cultures, national policies and
other country-specic contexts. This novel analytical
approach to European family policy increases our
understanding of the relationship between family policy
and family structure.
In the second step, we feed our policy analysis into
the demographic models which involve ELSA and
SHARE in two ways. First, we add country controls to
the multivariate analysis to capture the extent to which
variation in demographies of grandparenting across
Europe remains related to the nation state itself, even
after individual and family characteristics are taken into
account. Second, we use our policy mapping to group
countries into categories reecting different policy regimes
regarding their probable impact on grandparental care.
These categories are represented by key indicators which
are then used in the statistical models to investigate the
relative importance of family policy within a country to
its grandparenting demography of care and therefore to
inter-generational relations across Europe. This analysis
also reveals which policy regimes are associated with
which demographic patterns, which is of critical interest to
policymakers and lobby groups.
Thus we do not evaluate specic family policies
within each European country (which would require
very thorough evaluation research to examine the
circumstances before and after the introduction of a
specic policy) but rather we consider how the general
framework of family policy within each country relates
to grandparenting. In this type of research precise timing
of small policy changes becomes less important, as it
is the overall pattern of convergence over a number of
policy spheres that becomes the most important unit of
analysis. Welfare regime research over the last twenty
8 Towards a Framework for Assessing Family Policies in the EU.
20
years has shown that broad policy patterns change very
slowly within nation states, even if there are changes to
individual policies. However, we use our advisory group
of European experts to capture whether there have been
very substantial shifts in family policy regimes since the
demographic data were collected and to ensure that these
are taken account of in the analysis.
1.4 Report outline
Chapter 2 presents a brief update of the literature in this
area (as well as any key references) since the publication of
our scoping study in 2010 (Glaser et al., 2010). Chapter 3
focuses on our analyses of the prevalence of grandparent
households in selected European countries and the
US between the 1980s and 2000s. Chapter 4 describes
grandparent characteristics across the 12 European
countries in ELSA and SHARE. Chapter 5 presents parent
characteristics in the 11 European countries in SHARE.
Chapter 6 describes our analysis of grandparent policy
regimes and their relationship to intensive grandparental
care. Chapter 7 discusses the relationship between the
grandparent policy regimes, country-level cultural-
structural indicators and grandparental childcare. Finally
in Chapter 8 we summarise and discuss the policy
implications of our results.
21
2 Summary (and Brief Update) of Literature Review
A thorough review of the literature contributing to
knowledge of this research topic was carried out in the
earlier scoping study and has been published separately
(Glaser et al., 2010). To summarise:
Grandparents are likely to become more significant in family
life as populations age
During most of the 20th century western
societies have experienced a series of rapid socio-
demographic changes. Improvements in survivorship
mean that three-generation families are no longer
an exception (Post et al., 1997, Watkins et al., 1987).
The number of grandparents in England has doubled
over the past 50 years (Department of Health, 2009).
A child under age 5 born in the rst half of the 19th
century was likely to have just two grandparents
alive but this rose to 3.5 grandparents alive by 2010
(Murphy, 2011). Children today have at least three
living grandparents for most of their childhood
(Murphy, 2011).
Other changes, such as the growth in mothers’ paid
employment and rises in divorce and step-families,
are causing a considerably increased need for extra-
parental child care, in which grandparents can play
a major role (Wheelock and Jones, 2002, Herlofson
and Hagestad, 2012).
Dutch research investigating changes in the provision
of childcare for two cohorts of grandparents between
1992 and 2006 showed a signicant increase in
grandparents providing care for the children of an
adult daughter (Geurts et al., submitted). Among
the possible reasons for this change was the higher
labour force participation of mothers and increases
in lone parenthood (Geurts et al., submitted).
Co-residence between grandparents and grandchildren
Recent work by Nandy and colleagues (2011) using
microdata from the 2001 UK censuses estimated
that the number of children living with relatives
but without their biological parents (that is in
‘kinship care’ as inferred from co-residence9) was
approximately 173,000; the proportion of such
children had doubled between 1991 and 2001.
There is a lack of evidence about families headed
by grandparents in Europe, although evidence from
the UK suggests that grandparents form the largest
group among family and friends awarded kinship
care of children (Farmer and Moyers, 2008, Nandy et
al., 2011).
Increasing co-residence between grandparents
and grandchildren in the US (from 3.2% in 1970
to 5.5% of children by 2003) suggests a rise in the
share of grandparents raising or helping to raise
grandchildren; especially signicant is the rise in
skipped-generation households; those comprising
grandparents living with their grandchildren without
the child’s parents (Casper and Bryson, 1998, Pebley
and Rudkin, 1999, U.S. Census Bureau, 2004).
9 What in the US literature are considered skipped-generation
households.
Our work in this report suggests a smaller but
notable similar rise in skipped-generation households
in England and Wales. In the UK and the US a range
of reasons for this rise have been suggested including
parental neglect or abuse, drug or alcohol misuse,
and mother’s imprisonment or death (Goodman and
Silverstein, 2001, Jendrek, 1993, Nandy et al., 2011)
– illustrating the vital social role that grandparents
are playing. It has also been suggested that in Europe
the rise in intergenerational households containing
a grandparent may be linked to rising poverty
(Lyberaki and Tinios, 2005 ).
Nandy and colleagues also found greater poverty
to be associated with children living in households
with relatives other than their birth parents (most
often with a grandparent) (Nandy et al., 2011). This
is in line with studies in the US which have also
found poverty to be greater in skipped-generation
households (Mutchler and Baker, 2004). This was also
found to be the case in European countries such as
Portugal, where skipped-generation households are
more likely to be found in the bottom of the income
distribution in comparison to other households with
co-resident grandparents (Albuquerque, 2011).
Grandparents providing help to families
Research shows that in northwest Europe and the
US there is frequent contact between older parents
and their adult children; however there is less
involvement in regular transfers of nancial and
social support than in southern Europe (Albertini et
al., 2007).
This is partly due to the greater availability of state
support in the former countries including welfare
benets, public housing, eldercare and childcare, as
well as to different cultural norms.
Most transfers are down the generations, with
nancial and practical support provided by older
parents to their adult children and grandchildren. It is
only when grandparents reach the age of 75 or older
that they are more likely to receive than to give help
(Albertini et al., 2007, Attias-Donfut et al., 2005).
Analysis of Europe-wide data shows that older
people with more resources, for example those
with a partner, or with higher levels of wealth or
educational attainment are more likely to provide
help, while those who are in poor health or single are
less likely to provide support (Albertini et al., 2007).
There is also a gender difference, in line with the
role of women as perceived ‘kinkeepers’: women
are more likely to provide assistance and help in
comparison to men (Albertini et al., 2007).
Grandparenting and family breakdown
Rises in divorce and step-families means that the role
of grandparents in families is likely to increase as
studies have shown the importance of grandparental
involvement at times of family breakdown (Dench
and Ogg, 2002).
On the one hand, studies have found that
22
grandparents are more likely to provide help with
care of grandchildren if the parents are separated
than if they are together (Dench and Ogg, 2002).
On the other hand, there is increasing evidence to
suggest that paternal grandparents are less likely
than maternal grandparents to be involved in care of
grandchildren after their child’s divorce or separation
(Hank and Buber, 2009, Herlofson and Hagestad,
2012, Igel and Szydlik, 2011).
In addition, higher divorce rates across all
generations (including middle and older generations)
mean that grandparents themselves are more likely
to experience divorce or to have experienced it in the
past (Brown and Lin, 2012). Divorced grandparents,
or those who have previously been divorced and
since remarried, tend to have fewer contacts with
their grandchildren, take part in fewer shared
activities with them, and say they feel less close to
their grandchildren than grandparents who have
never been divorced (King, 2003) . This probably
reects less close relationships between older people
who have ever experienced divorce and their adult
children.
These negative effects are stronger for grandfathers
and paternal grandparents, probably reecting
fathers’ loss of contact with their children following
divorce (King, 2003).
The likely increase in future numbers of divorced
older people may have negative implications for the
closeness of future generations of grandchildren and
grandparents. However, as divorce and separation
become more common it is likely that the effects on
family relations may also change.
Grandparents providing childcare
Around 40% of parents with children under 16
in the 2009 Childcare and Early Years Survey
reported using informal care (that is, care outside
any regulated or formalised system) (Rutter and
Evans, 2011). Among those using informal childcare,
grandparents are the most common providers of
such care, enabling parents more easily to reconcile
work and family responsibilities (Rutter and Evans,
2011).
It is estimated that in Britain there are currently
14 million grandparents (Wellard, 2011). From the
grandparents’ perspective we also know that the
majority play an important role in looking after
grandchildren. For example, nearly two thirds (63%)
of grandparents in Britain with grandchildren under
16 provide some grandparental childcare and 17%
provide at least 10 hours a week (Wellard, 2011) .
Evidence also shows that over half of grandparents
in selected European countries provide childcare
to a grandchild under the age of 16 (Albertini et
al., 2007, Ware et al., 2002, Hank and Buber, 2009,
Igel and Szydlik, 2011). However, there are striking
differences across Europe in the frequency of
grandparental childcare.
In Italy, Spain and Greece roughly 40% of
grandparents who provide any grandparental
childcare are regularly looking after a grandchild
younger than 16 (that is almost weekly or more
often), compared with 20% of their counterparts
in France, Denmark, Sweden, France, and the
Netherlands (Albertini et al., 2007, Ware et al., 2002,
Hank and Buber, 2009, Igel and Szydlik, 2011).
On the other hand, more grandparents provide
childcare for a grandchild younger than 16 in
Sweden, France, the Netherlands and Denmark
(around 60%) than in the southern European
countries where this is just over 50% of
grandmothers and 40% of grandfathers (Albertini et
al., 2007, Ware et al., 2002, Hank and Buber, 2009,
Igel and Szydlik, 2011).
The literature investigating factors associated with
grandparental childcare is particularly extensive in
the US. Gender, age, marital status, health, education,
and employment have all been shown to be
signicantly associated with grandparental childcare
(Fuller-Thomson and Minkler, 2001, Minkler and
Fuller-Thomson, 2005).
Recent studies in Europe have begun to examine the
socio-economic and demographic characteristics
associated with grandparental childcare from
a comparative perspective. Such studies show
that younger, healthier, and nancially better-off
grandparents are more likely to provide any as well
as regular grandparental childcare (Albertini et al.,
2007, Hank and Buber, 2009, Igel and Szydlik, 2011).
This is in contrast to much of the US literature
which shows that grandparents with ‘primary
care’ responsibilities for grandchildren (many of
whom are co-resident) or who undertake intensive
grandparental roles are often among the most
disadvantaged (Fuller-Thomson and Minkler, 2001,
Minkler and Fuller-Thomson, 2005). For example,
they are more likely to be black, female and living
on low incomes or below the poverty line (Fuller-
Thomson and Minkler, 2001, Minkler and Fuller-
Thomson, 2005). Half of all US grandmothers
providing intensive childcare live in the same
household as their grandchild.
This difference between Europe and the US is most
likely due to the different denitions of ‘intensive’
childcare used. In the US data are routinely
collected on whether grandparents have a ‘primary
responsibility’ for raising a grandchild, whereas to
our knowledge no survey in Europe collects these
data. In Europe intensive grandparental childcare
usually refers to a less intensive form of childcare
than that measured in the US.
US studies that have investigated grandmothers
providing childcare (rather than primary care) also
nd that, for example, they are younger, healthier,
report higher educational levels, and are more likely
to be married and to live with their spouse (Baydar
and Brooks-Gunn, 1998). These ndings are more
in line with the European studies suggesting that
when studies are not restricted to very intense
grandparenting the availability of material and
personal resources partly conditions the giver’s
ability to provide assistance (Baydar and Brooks-
Gunn, 1998).
In Europe studies that have explored grandparental
childcare have especially looked at the relationship
between grandparents’ participation in paid work
and care. For instance, grandmothers aged 50 to
23
65 in paid work were found to be less likely to be
providing regular grandparental childcare (Zamarro,
2011). This nding is consistent with other European
evidence (Albertini et al., 2007, Hank and Buber,
2009, Igel and Szydlik, 2011).
Grandchild characteristics have also been found
to be important: Igel and Szydlik (2011) found
grandparents are more likely to provide any
grandparental childcare for children aged 4 to 6,
whereas intensive grandparental childcare is more
likely for children under 3 years of age.
Grandparenting and mothers’ participation in paid work
The focus of recent European studies has been on
the importance of the intergenerational link and
grandparental childcare in particular with regard to
mothers’ labour force participation. These studies
have shown that for some countries mothers
are more likely to engage in paid work when
grandparents are providing grandparental childcare
(Arpino et al., 2010, Ware et al., 2002, Wistow and
Hardy, 1999).
Contextual-structural factors as explanations for patterns of
grandparenting across Europe
While recognising that European countries differ in
terms of policies and cultural-contextual structures
(that is with respect to welfare state provision,
demographic and socio-economic behaviours
and family norms) few studies have attempted to
directly measure how these factors inuence the role
grandparents play in family life.
Some authors have suggested that the greater
reliance on substantial grandparental support in
southern Europe is related to the lower availability of
formal childcare (Albertini et al., 2007). Welfare state
systems have thus been pointed to as an important
factor for understanding the extent and intensity of
intergenerational relations.
To date research has found two country-level factors
in particular to be signicantly associated with
grandparental childcare: public childcare provision
and family norms. Igel and Szydlik (2011) (also using
SHARE data) found that cross-national differences
in public expenditure on childcare and other family
services (families and maternity and parental leave)
showed a signicant association with grandparental
childcare: where public expenditure on childcare was
higher, grandparental childcare was less likely (Igel
and Szydlik, 2011).
Jappens and Van Bavel 2012 examined the
association between family norms in a country and
the provision of grandparental childcare. While
they also found the supply of formal childcare
to be important, their work showed a signicant
association between country-level attitudes and
grandparental childcare. For example, grandparental
childcare was more common in those European
regions with more conservative attitudes toward
gendered family roles (Jappens and van Bavel, 2011).
24
3 Trends in Prevalence of Grandparent Households:
Selected European Countries and the US
In this chapter our aim is to examine how residence in
grandparent households (that is, households that include
a grandparent-grandchild dyad) by middle-aged and
older adults varies across particular European countries
and how its prevalence has changed over time. Given the
important role that grandparents play in family life, a better
understanding of grandparent households is likely to shed
new light on a key aspect of grandparental childcare: co-
residence with grandchildren as a proxy for kinship care
(Lewis et al., 2008, Nandy et al., 2011). However, as we
include households with young adult grandchildren we
recognise that not all grandparents in these households are
necessarily providing care for grandchildren.
In the US as grandparents are more frequently involved in
childcare arrangements involving co-resident care, data is
routinely collected on whether grandparents have ‘primary
responsibility’ for raising a grandchild (Fuller-Thomson et
al., 1997). ‘Custodial households’ can be identied where
living with a grandchild is combined with a grandparent
acting as primary carer (Mutchler and Baker, 2004). These
studies have shown that the vast majority of co-resident
grandparents whether in three-generation or skipped-
generation households have primary care responsibilities
(Fuller-Thomson and Minkler, 2001). To our knowledge, no
national surveys in Europe or the U.K. collect these data
but the practice of ‘kinship care’ is generally inferred from
co-residence (Nandy et al., 2011)
Therefore, we study adults aged 35 and over and
investigate trends in the likelihood of living in a
grandparent household between the 1980s and 2000s
in England and Wales, France, West Germany, Portugal,
Romania and the United States.10 We also identify
the socio-economic and demographic characteristics
associated with variations in such household residence. A
distinction is made between ‘three-generation households’
(comprising grandparents and grandchildren, with at least
one of their parents) and ‘skipped-generation households’
(consisting of grandparents and grandchildren but without
the parents) (Casper and Bryson, 1998, Mutchler and
Baker, 2004). The data sources used are the Integrated
Public Use Microdata Series International (IPUMS),
the Ofce for National Statistics’ Longitudinal Study for
England and Wales (LS), and the German Socio-Economic
Panel Study (SOEP).11
3.1 Evidence of trends
As summarised in Chapter 2, in England, Wales and
Scotland using 1991 and 2001 census data, Nandy and
colleagues (2011) showed an increase in kinship care from
the perspective of children under the age of 18 (that is
children living in households where no parent is present)
(Nandy et al., 2011). Similar trends are also apparent in
the US. For instance, the US shows evidence of increasing
co-residence between grandchildren and grandparents
suggesting a rise in the share of grandparents raising or
helping to raise grandchildren (Casper and Bryson, 1998,
U.S. Census Bureau, 2004).
10 Analysis of trends in grandparent households for the other countries
in our study was not possible given the lack of appropriate data.
11 More information on these datasets can be found in Appendix B.
3.2 Characteristics of grandparent
households
As stated in Chapter 2, grandparents living in households
with their grandchildren are more likely to be in poverty
when compared to other grandparents (Albuquerque, 2011,
Casper and Bryson, 1998, Fuller-Thomson and Minkler,
2001, Minkler, 1999, Minkler and Fuller-Thomson, 2005,
Mutchler and Baker, 2004). In addition our earlier report
showed that grandparents in these household types in the
US are more likely to female, African American, and less
educated (Glaser et al., 2010). It should be noted, however,
that children in mother-only households are usually
worse off economically than those in households where a
grandparent co-resides (Mutchler and Baker, 2009).
Our earlier report also showed that three-generation
households are less likely to fall below the poverty line
than skipped-generation households (Mutchler and Baker,
2004). Furthermore grandparents in multi-generation
households are also more likely to be younger than those
in skipped-generation households, and both grandparents
are more likely to be present in the latter. In skipped-
generation households grandchildren are more likely to
be older in comparison to multi-generation households
(Mutchler and Baker, 2004).
3.3 Evidence for Europe
As mentioned in Chapter 2 there is a lack of evidence
about grandparent households in Europe, although recent
work showed an increase in the rise of three-generation
grandparent households in Portugal (Albuquerque, 2011).
Nevertheless few studies have examined the characteristics
of grandparent households in Europe (Albuquerque, 2011,
Hank and Buber, 2009, Smith Koslowski, 2009).
3.4 Reasons for caregiving and policies
Grandparents may take on the role of a parent, either
legally or informally, for a range of reasons including
(as mentioned in Chapter 2) parental neglect or abuse,
drug or alcohol misuse and a mother’s imprisonment or
serious illness or death (Nandy et al., 2011). In the UK,
the increase in kinship care in the 1990s is thought to be
due to growing problems with parental substance misuse,
rising imprisonment, and increasing use of formal kinship
care since the introduction of the Children Act 1989
and reinforced by subsequent legislation (Nandy et al.,
2011). In the US recent policy changes have also greatly
contributed to enhancing the role of grandparents in
childcare. For example, following the decline of licensed
foster homes in the US in the 1980s and 1990s placing
children with relatives (often grandparents) reected a
major policy shift (Berrick, 1998, Smith and Beltran, 2001).
As a result not only is kinship care increasing in the US,
but in some states nearly as many children are being
placed in kinship as in foster care (Berrick, 1998, Smith and
Beltran, 2001). It is also important to note that in Europe
it has been suggested that the rise in intergenerational
households containing a grandparent may also be linked to
rising poverty (Lyberaki and Tinios, 2005 ).
25
3.5 Summary
In the UK there was a rise in the number of children
growing up with grandparents in kinship care households
in the 1990s (Nandy et al., 2011). In the US there was a
similar increase; however, this involved both household
types, that is those where three generations are living
together, and those where the parent is absent or unable to
full their parental role and the grandparent is the primary
caregiver. Grandparents (and therefore grandchildren)
in these latter household types are more likely to be in
poverty than other grandparents.
Grandparents may take on the role of a parent, either
legally or informally, for a range of reasons including
parental neglect or abuse, drug or alcohol misuse and
mothers’ imprisonment or death (Nandy et al., 2011).
Evidence from the UK suggests that grandparents form the
largest group among family and friends awarded formal
kinship care of children (Nandy et al., 2011). However,
there is generally a lack of evidence about families in
grandparent households in Europe (Nandy et al., 2011)
although there is some evidence to suggest that rises in
intergenerational households including a grandparent may
be a response to poverty rather than issues relating to
parental problems (Lyberaki and Tinios, 2005 ).
3.6 Grandparent households:
Prevalence and characteristics
First, we report on trends over time in the prevalence
of adults living in grandparent households (that is three-
generation and skipped-generation households). As noted
above, the European countries studied in answering this
question are England and Wales, France, West Germany,
Romania and Portugal.12
We distinguish between ‘three-generation households’
(comprising grandparents and grandchildren, with at least
one of their parents) and ‘skipped-generation households’
(consisting of grandparents and grandchildren but without
the parents). There is no limit on the numbers either of
grandparents or of grandchildren in a single household,
although the number of grandparents is unlikely to
exceed four. The grandchildren can be of any age; where
a grandchild is aged (for example) over 20 it may well
be that care and support between grandparent(s) and
grandchild is mutual rather than solely from the older to
the younger generation. The presence of other people
in the household besides grandparent-parent-grandchild
triads is ignored; their numbers are in any case relatively
small.
Figure 3-1 and Figure 3-2 show the percentage of
people aged 3513 and older residing in three-generation
households. These people may be grandparents, parents,
grandchildren (in a few cases) or indeed other resident
relatives or friends. With the exception of Romania, and
to a lesser extent Portugal, the other European countries
studied (that is England and Wales, France, and West
Germany), show a decline in the percentage of adults aged
35 and older residing in three-generation grandparent
12 France, Romania and Portugal are the countries with suitable data
for 3 time points in IPUMS.
13 We also undertook this analysis for individuals aged 40 and over