|Home | About | Journals | Submit | Contact Us | Français|
The relationship between class and intergenerational solidarities in the public and private spheres calls for further conceptual and theoretical development. This article discusses the findings from the first wave of a qualitative longitudinal study entitled Changing Generations, conducted in Ireland in 2011–2012, comprising 100 in-depth interviews with men and women across the age and socioeconomic spectrums. Constructivist grounded theory analysis of the data gives rise to the following postulates: (1) intergenerational solidarity at the family level is strongly contoured by socioeconomic status (SES); (2) intergenerational solidarity evolves as family generations observe each others’ practices and adjust their expectations accordingly; (3) intergenerational solidarity within families is also shaped by the public sphere (the welfare state) that generates varying expectations and levels of solidarity regarding State supports for different age groups, again largely dependent on SES; (4) the liberal welfare state context, especially at a time of economic crisis, enhances the significance of intergenerational solidarity within families. We conclude by calling for research that is attuned to age/generation, gender and class, and how these operate across the family and societal levels.
Intergenerational solidarity has both a private and a public dimension. We have varying expectations regarding what is ‘due’ to and from younger and older family members, and harbour understandings of what different generations at societal level ‘owe’ to one another. Sociologists have generated concepts and data on intergenerational solidarity, conflict and ambivalence in family relations (Bengtson 2001; Luescher and Pillemer 1998), and in welfare states (Arber and Attias-Donfut 2000; Vanhuysse and Goerres 2012). In research on both family-level and societal-level intergenerational solidarity, the influence of Bengtson’s sixfold conceptualisation of parent–adult child solidarities is evident, comprising affectual (emotional closeness), structural (geographical closeness), associational (frequency of interaction), functional (help and care), normative (importance accorded to family), and consensual (agreement over values) solidarity.
Some researchers have sought to understand the relationship between intergenerational solidarity and class. At the societal level, class remains more influential than cohort differences in shaping expectations and beliefs about redistribution (Edlund and Svallfors 2012). Dykstra and Fokkema (2011) devise a fourfold typology of late-life families that taps into structural, associational, normative, and functional dimensions of solidarity. Employing measures of education and household income, they also analyse the relationship between family types and socioeconomic differentials, concluding that higher income older households engage in fewer intergenerational support exchanges, and that high socioeconomic status increases ‘the likelihood of individualism in late-life families’ (p. 563). In contrast, Ward (2001) argues that ‘socioeconomic status…[has] little relation to either family expectations or societal-level intergenerational attitudes’ (p. 204).
The relationship between intergenerational solidarities (public and private) and class calls for more refined conceptualisation and theoretical development. Szydlik (2012) advocates the life-course approach to understanding intergenerational solidarity and the place of social stratification in these processes, asserting that processes of cumulative advantage over time and generations are complex and require research methods capable of capturing such complexity. Kohli (2006) has drawn attention to the importance of the extent to which the putative ‘generational cleavage…masks the continued existence of the class cleavage between the wealthy and the poor’. He questions the extent to which ‘the new intergenerational conflicts [have] really crowded out traditional intragenerational ones’, and postulates that ‘[t]he discourse on intergenerational equity may function as an ideology: as a way to divert attention away from the still existing problems of poverty and exclusion within generations…based on class or gender’. Künemund (2008) suggests that ‘it seems quite plausible to assume that the interaction [with regard to intergenerational solidarity] between the state and the family is not identical across all social strata’ (2008, p. 117). We join these scholars in identifying the need for deeper insight into the processes that connect public and private intergenerational solidarities with class-based differences. Rapidly changing social and economic contexts are affecting intergenerational relations within families and welfare states across the globe, and intergenerational solidarities may help to sustain individuals, families and societies at a time of global economic challenges and growing inequality.
In response to a call for research on ‘broader, realistic contexts’ for intergenerational solidarity (Binstock 2010, p. 584), the Changing Generations study in Ireland is designed to interrogate the interface between family and societal solidarity and to enable re-conceptualisation of intergenerational solidarities. In order to understand solidarity between generations and inequality (especially cumulative (dis)advantage), one must include younger and older cohorts in the analyses, ideally within longitudinal research designs (Steinbach 2012). Consequently, the first wave of this qualitative longitudinal study, conducted in 2011–2012, examines everyday experiences and practices among people of different ages and socioeconomic backgrounds.
The Changing Generations study is contextualised within the Irish liberal welfare state (Considine and Dukelow 2009). We collected the data in the midst of one of the most severe recessions that has affected an OECD country since the 1930s Great Depression. The current economic crisis in Ireland was caused by a property market bubble that was fatefully combined with an overexposure of the banking system to the property sector. The financial collapse began in 2008, leading to a dramatic reduction in economic output in 2008–2010. Despite a recent recovery in the tradable sector and steep increase in exports, the domestic economy remains in severe recession in 2013. Tight fiscal policy, adopted to restore stability to the public finances and to fund the huge losses in the banking system, is manifesting itself in budget cuts which in turn are delaying recovery in domestic demand. Unemployment remains at an historically high level (Bergin et al. 2011). The traditional ‘safety valve’ of mass emigration has resulted in the exodus of over 250,000 people from Ireland since the start of the recession in 2008—a considerable number in view of the fact that the total population is around 4.4 million people (Central Statistics Office 2011). Ramifications of the property bubble that burst in spectacular fashion include negative equity, 10 % of all residential mortgages being in long arrears (Central Bank of Ireland 2012), and nearly 3,000 ‘ghost’ housing estates that were left uncompleted when the property market collapsed (Kitchin et al. 2012). The Changing Generations data therefore offer insights into intergenerational solidarity in the face of the challenges posed by a grave economic crisis.
Changing Generations adopted Charmaz’s (2006) constructivist grounded theory (CGT) method in the interest of enabling an inductive re-conceptualisation of intergenerational solidarity. In line with this method, our conceptualisation and theorising are constructed ‘from the ground up’, rather than with the view to testing or fleshing out any particular notions of or assumptions regarding solidarity. In everyday experiences and talk about intergenerational relations, manifestations and notions of solidarity feature. ‘Solidarity’ is an abstract concept that assumes content and meaning through human interactions and transfers, and our understandings of these. Changing Generations is a grounded theory study and hence did not use any a priori definitions or measurements of ‘intergenerational solidarity’. The team decided to avoid abstract or conceptual language when recruiting and introducing the study to participants. Rather, in order to allow understandings and experiences to emerge without any signposting of (the loaded and potentially misleading terms) ‘solidarity’ or ‘conflict’, the researchers adopted the more neutral language of ‘give and take’ between research participants, people close to them, and members of the broader society. Nonetheless, many research participants spontaneously used words such as ‘solidarity’, ‘justice’ and ‘fairness’—despite the fact that interviewers deliberately avoided introducing these words into the discussion.
Charmaz (2006) argues that ‘the combination of flexibility and control inherent in in-depth interviewing techniques fits grounded theory strategies for increasing the analytical incisiveness of the resultant analysis’ (p. 29). She advocates that researchers formulate an interview guide to direct the focus of data to topics specifically tailored to developing the theoretical frameworks as interviewing proceeds. The Changing Generations interview guide consisted of 10 questions. Two initial questions were designed to enable the participants to introduce key aspects of their lives to the interviewer (‘tell me about the stage you are at in your life now?’ and ‘who would you say are the people closest to you?’). Four ‘intermediate’ questions sought to tap into two experiences (‘giving’ and ‘receiving’ resources) and into two spheres (‘private’ and ‘public’):
As part of probing, interviewers introduced the temporal perspective into each of the above core questions, i.e. asked participants whether they ‘had received’ support in the past, ‘anticipated receiving’ support in the future, and so on. The interviewers also invited the participants to offer their reflections on how they felt about giving and receiving help and support as family members and as members of Irish society. The intergenerational dimension was not explicitly raised (unless introduced by the participant) until the middle of the guide, question five being phrased as ‘I’d like to hear about your thoughts on the State’s role in supporting the young and the old’. Ethical approval for the study was obtained in July 2011. Between September 2011 and July 2012, 100 in-depth semi-structured interviews were conducted with men and women across the age and socioeconomic spectrums to explore their experiences, practices and understandings of the give and take between family members and members of Irish society at large. The interviews ranged from 31 to 160 min in duration, averaging 72 min. All interviews were recorded and transcribed verbatim by two professional transcription services.
Table 1 summarises the sample composition by gender, age group and socioeconomic status (SES). We acknowledge that researchers use a wide range of definitions of SES, and that there is no consensus over the meaning of class. The study design entailed purposive sampling of participants from five areas with distinct levels of affluence and deprivation, selected on the basis of an index that measures education, principal economic status, housing, employment status and household composition of the population (Haase 2010). Data generated with participants across these five geographical sites indicated SES as an emergent category shaping intergenerational relations. This led the team to decide to sample purposively for balance in representation of high, middle and low SES participants to allow for theoretical exploration of the role of SES in shaping practices and perspectives. In the context of a qualitative study seeking considerable amounts of personal information and reflection on family relations and participants’ role as members of Irish society, we selected a pragmatic way of capturing individual’s SES. Occupational status categorisation used by the Central Statistics Office (CSO) of Ireland (Central Statistical Office 2009) to classify SES was selected and re-grouped into three categories of high, middle and low SES as presented here. If participants did not volunteer information about occupational status during the course of interviews, this information was elicited by interviewers. Formal assignment to a SES category was subsequently undertaken by the research team following CSO procedures. Participants who had never worked (including students and home-makers) were assigned to an SES category based on either their partners’ SES or that of their families of origin.
The interview data, spanning 3,737 pages and 1,172,503 words, were analysed following Charmaz’s approach to CGT building (Charmaz 2006; Charmaz and Belgrave 2012). As data collection progressed, the three researchers involved in interviewing wrote memos to each interview reflecting on emergent meanings and directions each interview added to the overall enquiry. Open-line and focused coding was also engaged. Emergent questions, directions of analysis and coding were discussed at regular team meetings where all four team members shared questions arising from their own disciplinary location. Insights from across the dataset illuminating specific questions were shared and debated for contrast and ‘fit’. A particular emphasis of this analytic process was whether the contentions being posed by the member proposing an interpretation fitted other members’ insights into ‘what was going on’ in the dataset.
Findings presented here were generated through a process of grounded theorising across the entire dataset. Themes emerging from the earlier interviews were pursued for further exploration and elaboration in subsequent interviews. Themes and social processes discovered within the data were built up into abstract categories and considered in light of the existing literature and extant theories to generate hypotheses that were then interrogated empirically in further data collection to arrive at the most plausible explanations. Excerpts from selected individuals are used to exemplify a finding and should be read as exemplars of a theme robustly confirmed across the extensive dataset. We first discuss findings pertaining to intergenerational solidarity at the societal level, then turn to discuss intergenerational solidarity at the family level, and analysing how solidarities in these two spheres are inter-linked. All first names and surnames used in the findings are pseudonyms.
Our study yielded little evidence of intergenerational conflict, either within the private or the public sphere. Across the dataset, older people were almost universally perceived as a deserving group that merited more and improved transfers and services from the State. ‘Upward’ solidarity towards older people was particularly marked among young people from the low and middle SES groups. For example, Noelle Lynch (19, low SES) reflects the general view among the younger research participants that older people’s benefits and free health care are deserved, should be universal, and must not be cut:
[Older people] get the pensions but they work[ed] for it, it is their money. The way I see it, they paid enough money over their life time into the State so they should get something back. I think the government cutting their pensions…is a disgrace. That is absolutely disgusting. (…) They have worked all their lives. (…) They should be able to sit back and have no financial worries. They should enjoy life.
Age therefore matters in Ireland as an ‘automatic’ demarcation of deservingness. The liberal Irish welfare state is a powerful driver of this finding—older people’s welfare entitlements have tended to enjoy strong legitimacy as ‘deserved’ benefits in liberal welfare states, the United States in particular (Pampel 1994; Campbell 2009). The only exception to strong approval and call for further supports for older people in our dataset was constituted by older people from high SES groups; the well-off older adults within this group stated that some benefits they enjoy (free medical care in particular) were excessive. Martha O’Flynn (69, high SES) encapsulates this viewpoint:
[Entitlement to free medical care] should be means tested…that blanket [cover for] over seventies was a magnanimous gesture…without too much thinking behind it. (…) I think you have got to look after yourself from the cradle to the grave.
These older adults who reject the age-based entitlement to free medical care espouse an ideology of self-sufficiency and independence, and endorse social policies that are reserved, on the basis of means-testing, for those who cannot make provision for themselves and their families. The moral dimension of State intervention—individuals should assume personal responsibility for looking after their health and social care needs, with reliance on the State representing a form of weakness—also resonated among this group. The entitlement to free medical care was contrasted by several well-off older participants to the difficulties that low- and middle-income families with children have in meeting the costs of primary health care (GP visits). This demonstrates a process observed in the data wherein solidarity at societal level can be strongly oriented away from one’s own age and social class group, and towards people of a markedly different age and life stage. It also aligns with Goerres and Tepe’s (2012) allusion to the ‘socially constructed nexus between the family and the welfare state’, which evinces ‘motives of reciprocal exchange between generations rather than pure age-based self-interest’ (pp. 199–200).
Solidarities do not have an ‘either/or’ quality. Older people who demanded improvements in old age-related benefits also tended to call for improvements in supports for (some) younger groups. Tommy Keeley (71, middle SES) is an example of an older rural dweller whose sense of solidarity traverses both social class and geographical boundaries as he expresses strong support for increased investment into the life chances of young people in deprived city areas:
[I]nner city and rundown areas…need extra help. (…) I think it is terrible to see that they were going to cut down on teachers in those areas. (…) [Helping children and young people in deprived areas] would be my priority.
Participants saw no need to juxtapose public expenditure on older people with expenditure on younger people. In fact, with only three exceptions, the idea of taking from one generation in the interest of another does not feature across the sample (although the idea of redistributing to some groups within younger or older generations does feature, e.g. well-off older people suggesting that money spent on their benefits should go to low- and middle-income families). Our analysis suggests that surveys and media portrayals that pit one generation or age group against another are using constructs and ideas that are far removed, indeed abstract, for ‘ordinary people’ in Ireland for whom the idea of generational conflict at the welfare state level is foreign.
However, for many participants the State was a remote entity. Especially in the younger middle/high SES individuals’ accounts of the current and expected future role of the State in their lives, almost complete dissociation from the State (other than as tax payers) featured prominently. The locus of risk management is therefore strongly oriented away from the State, with minimal expectations and even stark pessimism about future entitlements, and increasingly shifting towards the expectation of reliance on private resources, earned through employment and entrepreneurship, and derived from within family networks. Indeed, the richest and most detailed accounts of the ‘give and take’ between generations pertained to the family level.
We now turn to discussing practices of intergenerational solidarity at the family level, including the process we identify as ‘generational observing’, to interrogate more deeply the relationship between solidarity and class, and the interface between private and public solidarities. Generational observing is a concept that emerged from the Changing Generations dataset through the process of inductive analysis. It resembles the concept of generational intelligence developed by Biggs et al. (2011) who define it as ‘an ability to reflect and act, which draws on an understanding of one’s own and others’ life-course, family and social history, placed within a contemporary social climate’. The term generational observing is used here to denote a somewhat narrower process of witnessing the practices of younger or older family generations, and adjusting one’s expectations of intergenerational solidarity in the light of personal interpretations of these practices.
Many older participants, especially those in the middle and high SES groups, offered their observations of the lives of their adult children and grandchildren as grounds for low expectations regarding future care from younger family members. This illustrates how generational observing shapes understandings of intergenerational solidarity. Older interviewees observed the nuclear family of today closing in on itself in a manner that leaves little time for older generations. The primary drivers of this pattern are the labour market participation of both parents, and the extreme care and attention that is accorded by parents to the youngest generation, i.e. children and young adults. Expectations regarding future family care from adult children and their families were low among middle and high SES older adults whose adult children and children-in-law were in employment:
[My daughter] has a house, a partner and two kids. She will have to get back to work. She is working as it is, three/four days a week. (…) I see a pressure on [both parents] that they have to work. They have no time. I do not think there is much time.
Eileen Garvey, 74, high SES
The generational orientation is seen to be increasingly ‘downward’—investment into children becomes heavier and heavier, leaving less time for investment into (care of) older family members. The low expectations regarding family care therefore reflect well-off older adults’ observation of the heavy investment of time (and money) by their adult children and children-in-law into their grandchildren, and also their ability and inclination to pay for care out of private resources. Research on solidarity within families has largely ignored this single most important dimension of familial intergenerational solidarity, namely the intergenerational resourcing of children within families, practiced both by parents and grandparents (directly or via the parents) (Arber and Timonen 2012). Long-lasting differences in life chances arise from this kind of intergenerational solidarity within families, yet both research and policy are weakly attuned to these behaviours and their long-term consequences at individual and societal level.
Our data indicate that resourcing, or formation, of children is perhaps the single most influential and long-lasting generational family practice that influences life chances and key outcomes throughout the life-course, and shapes intergenerational solidarity. Striking accounts of extremely careful formation of some children, and erosion of the capabilities and confidence of others, feature in our data. Older women, especially those of high SES, gave accounts of the focused, time-consuming attention they had given to guiding their children through the education system in a manner that enabled them to get the most from their schooling (in most cases leading on to higher education), and educating them through hobbies and travel. As is evident from Lorna Gogan’s account, these practices are generationally transmitted and replicated by parents in the descending generations:
I brought [my children] always to museums and places of historical interest. I was always trying to broaden their education, I suppose. I made sure that they were always aware of current affairs and things like that. (…) my daughter…is doing all the things that I did with her, with her kids now…[our emphasis]
Lorna Gogan, 79, high SES
The interview data pertaining to people currently parenting reflect the ‘generational observations’ of the older participants, and also illustrate how life chances are stratified by class. Accounts evince heavy investment of time and money into hobbies and education by middle and high SES parents (mothers in particular—these practices are still highly gendered in Ireland), for whom entry into higher education is a ‘taken-for-granted’ assumption:
[I]n terms of schooling he [son] has to go to secondary [school] and he has to go to college. I think college is like doing your Leaving Cert [final secondary school exams] nowadays, you just go to college and that’s it, you know.
Caroline Fogarty, 40, high SES
Contrasting accounts featured among parents from the low SES group. Despite shared commitment and ambition to see their children go to third-level education, lower educational achievement and resources placed them at a significant disadvantage in helping their children, and by extension put their children at a disadvantage relative to children from middle or high SES families. For example, Melissa O’Regan (45, low SES) who left school at the age of 13, expresses a tentative wish that her children stay in education (‘please God’ = Irish expression denoting hope)—far from the taken-for-granted approach of higher SES groups:
My stage now in life is I cannot see much prospects happening for me. I am just looking forward to my own children’s future and trying to better them and trying to get them to stay in secondary school and please God they go to college (…)
The role of class in shaping life chances illustrated by these parents was corroborated by the data generated with younger participants. Based on the accounts given by the most marginalised young people in the sample, the Irish education system is performing poorly in levelling out the unequal resourcing of children and young people by their families (for instance literacy and numeracy problems are not addressed, or this is left until a very late stage in compulsory schooling). The heavy reliance on the family, and the belief in the autonomy of families in looking after their own members, constitute the deep roots of inequality in Ireland (and other familialist societies—see also Saraceno 2008, p. 4). The orientation of many welfare states towards self-reliance of individuals/families augurs movement towards or reinforcement of the intergenerational transmission of (dis)advantage that our data illustrate.
Within the high SES bracket, we find considerable (and perhaps increasing) stated and aspired-for independence from both younger family members and the State. High SES participants tended to express the view that they expect very little support from the State in terms of care services in old age. Most high SES older participants, like Fred Peterson (age 73), had made provision towards or expected to meet the costs of private help and support in their own home in old age, on the basis of the assumption that neither the State nor the family will or should be called upon to meet these costs:
Fred’s example illustrates how the liberal model of a welfare state is deeply rooted in many participants’ narratives. In contrast to the practice of increasingly ‘contracting out’ care by high SES groups, individuals and families with lower socioeconomic resources remain considerably more reliant on ‘free’ family care. In many cases, this is facilitated by the geographical propinquity of at least one adult child, made more likely in the Irish context by the large number of children in the older cohorts, and by these adult children’s tendency to establish households in close proximity to their parents (Barrett et al. 2011).
The reliance on and expectations of family care was markedly higher among low SES research participants and continues to shape expectations and beliefs of what is ‘good’ intergenerational practice. The pattern of young people’s involvement in transferring resources upwards through the generations was notable among low SES youth. For many, these transfers constituted the deepest meaning of family solidarity, and reflected very low expectations of State involvement. The most striking example of this pattern is Stacey Kennedy, a 19-year-old unemployed woman who cares fulltime for her octogenarian grandmother who has dementia. Stacey undertakes her caring role at the expense of pursuing her own education and employment opportunities, yet rejects the thought of any State financial compensation for this care work:
I do everything, the washing, the drying, the feeding, the bathing…I have to do everything for her (…) Your family is your family. (…) She is my Nanny [grandmother], I should not be paid to look after her I feel. (…) when it is your family I feel like it is kind of your duty.
Stacey is expressing her unquestioning obligation to care for her grandmother (see also Calkins 1972). Her intergenerational solidarity encompasses both her grandmother, whom she ‘love[s] to bits’, and her mother, a lone parent on very low income, whom she helps to the best of her ability with household bills. In similar vein, Kylie Quinn (25, low SES lone parent) expressed her belief that it is her unquestioning obligation to look after her mother despite the fact that her mother, an alcoholic, had been extremely neglectful towards her throughout her youth, because ‘[s]he’s your mother and you’ll always only have one mother’. Stacey’s and Kylie’s obligations—current and anticipated, real and perceived—embody strong moral capital (internalised social norms that obligate (grand)children to care for and support their older (grand)parents—Silverstein et al. 2012) and are in stark contrast to the expectations of middle and higher SES youth. For example, Brian Lyons is in his early 20s and, despite having an ageing parent with a disability, feels free to pursue his higher education and career in Ireland or abroad, thanks to the freedom that his family’s wealth affords through ‘contracting out’ care, a practice that he is currently witnessing in his grandparents’ care arrangements:
…money or whatever that the family kind of says is there should…cover that. I mean…the family is sort of wealthy enough…there is quite a bit there [to fund care].
Brian’s expectations align with those of Fred, the high SES older man quoted above: the older generation gives the licence to the younger to focus on self-development, career and offspring, and the younger generation takes the opportunity to develop work and personal interests instead of limiting them as Stacey does.
Based on these findings, we argue that perceptions and practices of intergenerational solidarity in Ireland are strongly class-based. Social class is the central variable that, together with gender, either ‘binds’ or ‘frees from’ provision of care (we are not considering emotional support here, or financial transfers, which we did not measure and which are inevitably influenced by income/wealth). The welfare state context matters too: the restricted availability and low public expenditure on long-term care in contexts such as Ireland is a central driver of private expenditure on care by those who can afford it and recourse to family care by those who cannot.
Based on in-depth interviews in a single country, at a specific point in time, the Changing Generations study cannot aspire to constitute a full picture of intergenerational solidarity across all contexts. However, our empirical findings highlight ways in which research and theorising on intergenerational solidarity might usefully be re-oriented. They are particularly apposite for contexts that are comparatively unequal, have weak welfare states, and place considerable reliance on the family as a source of security, as Ireland does. Ireland’s economic crisis also frames our findings, and provides insights into intergenerational processes and practices in the midst of great economic challenges.
Based on CGT analysis of Changing Generations data, we have formulated the following four postulates. First, that intergenerational solidarity at the family level is strongly contoured by SES. Second, that intergenerational solidarity evolves as family generations observe each others’ practices and adjust their expectations accordingly. Third, that intergenerational solidarity within families is also shaped by the public sphere (the welfare state) that generates varying expectations and levels of solidarity regarding State supports for different age groups, again largely dependent on SES. Fourth, that the liberal welfare state context, especially at a time of economic crisis, enhances the significance of intergenerational solidarity within families. The Irish welfare state (among many other welfare states) does not counteract the family-level inequalities effectively, which makes individuals’ life chances and sources of security highly dependent on the socioeconomic resources and solidarity of their family.
Public and private intergenerational solidarities and class are inter-twined in complex ways that we have only partially illuminated and theorised here. Intergenerational solidarities are simultaneously being constructed at the family level and at the societal level, and are at both levels powerfully shaped by socioeconomic resources that influence the attitudes and practices we adopt towards other generations. Hence, we echo and substantiate Kohli’s (2006) call for research on intergenerational solidarity that engages seriously with class (and gender—we recognise that gender norms and roles play an important role but cannot engage in a fuller discussion of gender in this article). We hope that our findings are relevant for future research and debates that are attuned to age/generation, gender and class, rather than according age and generation the exclusive focus that they do not deserve in discourses and research on intergenerational solidarity.
Kathy Charmaz is a member of the scientific advisory board for the Changing Generations study and has been closely involved in guiding the research team in its methodological approach, for which we are deeply grateful. We also thank the other members of our Scientific Advisory Board (Alan Barrett, Pat Dolan, Ricca Edmondson, Helen Johnson and Paula Mayock), and our funder, the Atlantic Philanthropies. The research team is especially grateful to the 100 people who participated in the study. Thanks are also due to the participants in the session on intergenerational solidarity at the ESPAnet 2012 conference, two anonymous reviewers and the Editor of this journal for their constructive comments.