The decrease in fertility and the increase in life expectancy has significantly altered the age structure of the population. Many developed countries have experienced an increase in the average age of the population (where the over 65 years-olds prevail), known as aging populations. Longer life expectancy means that more children have the opportunity to live with their grandparent(s), although this trend might be counterbalanced by a delay in fertility, especially in those countries where this phenomenon is particularly pronounced (Tomassini and Wolf, 2000). However, it should be noted that demographic changes and socioeconomic shifts, also occurring in the last 3 decades, have modified the meaning of grandparenting.

In the literature, it is well established that grandparents play an important role in providing family support, in terms of emotional, financial, and practical care, specifically for both their child(ren) and their grandchild(ren) (Fuller-Thomson and Minkler, 2001; Hank and Buber, 2009; Aassve et al., 2012; Dunifon et al., 2014; Di Gessa et al., 2016a; Bordone et al., 2017). Their supporting role as childcare providers is even stronger in countries with lower levels of paid employment among older women, limited public childcare services, and more conservative attitudes toward gendered family roles, such as in Southern European countries (Igel and Szydlik, 2011; Jappens and Van Bavel, 2012; Di Gessa et al., 2016a).

Although a growing body of literature has investigated the involvement of grandparents in grandchild care (Hank and Buber, 2009; Aassve et al., 2012; Di Gessa et al., 2016a; Bordone et al., 2017; Glaser et al., 2018; Zamberletti et al., 2018), some topics have received little attention. According to Glaser et al. (2018), one of the main gaps in the literature is the analysis of trends in co-residence between grandparent(s) and grandchild(ren). Indeed, most of those studies are widely concentrated in the United States (Casper and Bryson, 1998; Goodman and Silverstein, 2002; Dunifon et al., 2014), which has experienced a relevant increase in the prevalence of multigenerational households since the 1970’s (Casper and Bryson, 1998; Taylor et al., 2010; Dunifon et al., 2014), in some cases associated with socio-economic disadvantage (Dunifon et al., 2014).

For Europe, studies on multigenerational households and on co-residence between grandparents and grandchildren are not as abundant as they are for the United States, also because of data quality (Glaser et al., 2018). Although in Western European countries intergenerational co-residence declined dramatically over the course of the 20th century, they are less common in Northern than in Southern and Eastern European countries (Palloni, 2000; Tomassini et al., 2004; Albuquerque, 2011; Selwyn and Nandy, 2014).

So far, the assumption has been that all grandparents are equally likely to be involved in grandparental childcare; however, previous studies have stressed the preponderance of women as care providers (Wall et al., 2001; Thomese and Liefbroers, 2013; Zamberletti et al., 2018). In Western cultures, women, and in particular grandmothers, are considered as kin keepers (Rosenthal, 1985; Hagestad, 1986; Friedman, et al., 2008; Sear and Mace, 2008).

Several studies have also examined the relationship between grandparental childcare and grandparents’ health in different countries, often yielding ambiguous results (Hank et al., 2018). Some of them suggest that grandparents who look after their grandchild(ren) occasionally are more likely to report better physical and psychological health than grandparents with intensive commitments or grandparents who do not provide childcare at all (Fuller-Thomson and Minkler, 2001; Minkler and Fuller-Thomson, 2001; Hank and Buber, 2009; Arpino and Bordone, 2014). These grandparents are also more likely to report positive effects or no major effects once other characteristics are taken into account (Hughes et al., 2007; Di Gessa et al., 2016b; Ates, 2017). Conversely, other studies have highlighted a negative relationship between grandparents’ childcare and health, for example depressive symptoms and worse self-rated health as well as physical health problems (Minkler et al., 1997; Lee et al., 2003; Blustein et al., 2004). With a particular focus on co-residing grandparents, some studies found that looking after grandchild(ren) may have negative consequences on grandparents’ health, who are more likely to experience health declines and to report depressive symptoms compared to those who provide lower levels of grandchild care or no grandchild care at all (Minkler et al., 1997; Blustein et al., 2004; Hughes et al., 2007; Chen and Liu, 2012).

More generally, some studies underline that the familial context of grandparental childcare is central to understanding the health implications and that the context of grandparenting (e.g., whether they are custodial or non-coresiding) may cause possible selection effects (Danielsbacka et al., 2019).

Di Gessa et al. (2016b), analyzing the longitudinal association between grandparental childcare and health in eleven European countries, and also taking into account individuals’ advantages or disadvantages throughout the life-course, found that looking after grandchildren affects the physical health of grandmothers and grandfathers differently: it appears to be beneficial for grandmothers who look after their grandchild(ren) intensively or non-intensively as opposed to those who do not provide any childcare, while this pattern result was not significant for grandfathers. Other studies also found gender differences in the effect of the provision of grandchild care, in particular on mental health, and attributed such a difference to the different roles, expectations, and desires which men and women have concerning care and family involvement (Blustein et al., 2004; Grundy et al., 2012). Grandmothers may perceive grandchild care and may experience and perform care differently from grandfathers, thus, childcare in turn may affect their health differently (Waldrop et al., 1999; Stelle et al., 2010).

In addition, other studies stress that providing care may have both positive and negative consequences on well-being (Walker et al., 1995; Pinquart and Sörensen, 2003; Stanca, 2012). Ruppanner and Bostean (2014) found that women report worse well-being than men in countries with greater attitudinal support for co-residential familial caregiving. Nevertheless, reciprocally, elderly dependents may provide help in the home for the caregiver’s family, especially when children are present (Ingersoll-Dayton et al., 2001).

Although the aforementioned studies investigated grandparents’ characteristics, gender differences in childcare, the effect of childcare on grandparents’ health, and (marginally) grandparental childcare in multifamily households in Europe, most childcare works focus on the overall population without considering migrant families as potential consumers of childcare (Williams and Gavanas, 2008).

During the last decades, migration flows have shifted and still continue to shift European societies into multicultural societies. As migrants settle in the destination countries, childcare emerges as a new exigency. The few studies available on migrants’ childcare preferences suggest that they use more informal than formal care (Ryan, 2007; Barglowski et al., 2015).

Studying the Italian case is interesting for multiple reasons. First, the Italian care regime has long been marked by the male breadwinner model, with a traditional gender division of work and family responsibilities (Mencarini and Solera, 2004), which generally conceives childcare as a female issue (Naldini and Saraceno, 2011). Second, Italy is characterized by a “familistic model” of care based on solidarity between generations, resulting in a lack of public services (Esping-Andersen, 1990). Accordingly, in Italy, where grandparents play a key role in informal childcare, being a migrant may constitute a further constraint.

“As a matter of fact, the implicit familialism with its weak direct support for the family’s caring function and its lack in service provision does not directly intervene in gender relations. It resembles a laissez-faire model of family policy where the state seeks to exercise no influence on the family at all. Nevertheless, this type affects gender relations since it simply reproduces and thus confirms the status quo of gendered care provision within the family” (cf. Harding 1995: 183–6).

The present study, using two Italian surveys carried out by the Italian Institute of Statistics (ISTAT) and comparing Italian and migrant families, sheds light on grandparental childcare in those households called “sandwich generation” families (Bryson and Casper 1999), which are a specific form of multigenerational family where there is at least one grandparent, his/her child(ren), and his/her grandchild(ren). To our knowledge no research is devoted to this particular kind of family as far as the Italian context is concerned.

Italy is characterized by a “familistic” welfare model (Esping-Andersen, 1990), where the family are responsible for the wellbeing of members, and family policies (based on the three pillars: family allowances, parental leave, and care services) are still scarce, not universal, and not generous (Costa and Sabatinelli, 2011; Presidency of the Council of Ministers–Department of Family policies, 2020). In addition, some studies show that among the European countries, Italy is characterized by the strongest intergenerational exchange (Di Gessa et al., 2016a; Zamberletti et al., 2018; Santero and Naldini, 2020).

In Europe, although Italy appears to be the “oldest country”, it has the lowest percentages of grandparents (53%) compared to England, France, the Scandinavian countries, and Belgium (between 62 and 67%) (Glaser et al., 2013). Nonetheless, Italy is one of the countries where grandparents are more likely to provide a higher intensity and regular daily form of care. The share of grandparents looking after their grandchild(ren) at least once a week is 20% in Northern countries, around 30% in France, and 45% in Spain and Italy. In Southern European countries, more interesting is the share of grandparents who provide care on a daily basis: 30% in Italy and Spain compared to only 2% in the Northern countries (Arpino et al., 2010).

A recent study by Di Gessa et al. (2016a) found that grandparents are more likely to provide intensive childcare for working mothers in countries where women enter less frequently in the labor market, such as in Italy. This finding shows that the probability of grandparents being involved in intensive childcare not only depends on individual characteristics, but also on the needs of childcare and on country characteristics.

As already mentioned in the previous paragraph, the relationship between providers and users of care among the elderly is also influenced by gender. Zamberletti et al. (2018) found that gender is crucial for the probability of providing childcare: grandmothers are more than three times more likely to provide intensive childcare than grandfathers. However, even in this context, the inverse relationship between childcare provider and care recipient is evident; those in poor health are less likely to take care of grandchildren. Another study by Del Boca et al. (2005) found that the presence of a grandmother who lives nearby and in good health is determinant for the choice to shift from public to informal childcare, in particular for very young child(ren).

The structure of the Italian population has been modifying for some decades, in part due to the constant increase of both migration flows and migrant settlements (Bonifazi, 2017; ISTAT, 2018). The migrant population in its settlement process has formed (or reunited) and enlarged families with children’s birth and subsequently has shown a need for childcare (Barbiano di Belgiojoso and Terzera, 2018). Mostly migrants live together with their descendants (Terzera and Barbiano di Belgiojoso, 2019), indeed, the cohabitation of several generations is more frequent among foreigners than among Italians.

According to two national surveys conducted by ISTAT, in Italy, among the Italian households the share of multigenerational households (i.e., families with at least one parent and one co-resident grandparent) with children aged 0–13 is 7.5% (ISTAT, 2018) ¹ , while among foreigners the percentage is higher (11.5%; ISTAT, 2018) ² .

The Italian law establishes strict criteria for the reunification of parents and relatives other than partner/spouse and minor children: parents aged under 65 can be reunified only if they do not have other children who can look after them in their own country, while parents aged 65 and more can be reunified only if any other relatives can’t support them in the country of origin due to serious health problems (Article 23 L 189/02). Moreover, in 2008 the Legislative decree n. 160 added the request for private health insurance (or privately funded registration with the National Health System) as a requisite for the reunification of elderly aged 65 or above (Bonizzoni, 2015). These requisites limit the number of family reunifications for relatives other than partner/spouse and minor children. The rare reunification of parents or relatives, as well as the low share of first generations of immigrants already grandparents, reduce the family network compared to the conditions in their home country (Barbiano di Belgiojoso and Terzera, 2018) and compared to non-migrants.

Italy began to experience considerable immigration during the 80’s, thus the aging of the immigrant population is still a marginal phenomenon (Cela and Barbiano di Belgiojoso, 2019); actually, the share of elderly migrants is rather small and therefore little studied. However, older populations are increasing in relative and absolute numbers and in the future, elderly people (both natives and migrants) will play a more important role in their family and in society in general, as long as their life expectancy increases. Although family and society will have to care for some of the elderly population, at the same time, they will benefit from the help of others, as grandparents looking after grandchildren and supporting working mothers and fathers.

Our contribution to the literature is fourfold: first, while most of the studies on grandparental childcare focus on the whole population, we analyzed and compared grandparental childcare between Italian and migrant households. Second, we analyzed grandparental childcare provided by co-resident grandparent(s), which is an understudied topic in the European context focusing on a segment of the population with specific characteristics. Third, while several studies investigated the effect of childcare on grandparents’ health, we analyzed co-resident grandparents in Italian and migrant families to examine whether such cohabitation is related to health status, thus analyzing whether grandparents need care, or conversely, they represent a family support system in providing childcare. Fourth, besides grandparents’ characteristics, the household is used as the analysis unit, taking into account the role of the household composition, the mother’s employment status, and the perception of the economic condition of the household, while the majority of previous studies focused only on the relationship between grandparents’ childcare and mother’s participation in the labor market, neglecting the role of the household composition (Del Boca et al., 2005; Aassve et al., 2012; Arpino et al., 2014). Therefore, this study has three aims. First, it analyses and measures the differences in co-resident grandparents’ childcare between Italians and migrants. Subsequently, by considering grandparents’ self-rated health, it explores whether there are differences in co-resident grandparents’ role between the two populations; finally, it examines gender differences in grandparental childcare, by verifying whether they also persist within migrant families.

We created a pooled dataset merging two different national surveys conducted by ISTAT. The first (and, so far, unique) survey is “Social Condition and Integration of Foreign Citizen” (hereafter SCIFC), collected during 2011–2012, which contains information on a sample of nearly 9,500 households with at least one foreign-born member. The second survey is “Multiscopo–Aspects of Daily Life” (hereafter AVQ), conducted during 2011, which contains information on almost 20,000 Italian households. In both surveys information about childcare were asked only to households with at least one child aged 13 or less, thus we restricted our sample to those families with at least one parent (4,085 households in the AVQ sample and 3,389 in the SCIF sample). In addition, from the SCIF survey we excluded mixed couples (i.e., with an Italian parent) and single-parent families with Italian children (around 32%), because the surveys do not include information on the latter as, being Italians, they did not meet the criterion for being in the target group (2,298 households in the SCIF sample). Since our study focuses on co-resident grandparents’ childcare in Italy, we further selected households with at least one co-resident grandparent (303 households in the AVQ and 219 in the SCIF survey). Finally, the study analyses the role played by grandparents’ self-perceived health, thus, among this sample we selected grandparents for whom health information is available. Hence, the final sample is composed of 517 families (302 Italians and 215 foreigners) for a total of 718 co-resident grandparents.

From the data, two possible caveats emerge. First, in the AVQ survey we do not consider naturalized families (i.e., those who have acquired Italian citizenship) because the sample does not provide this information. Second, as stated beforehand, for migrant households, the presence of grandparents is strictly limited by the law. However, as regards the naturalization issue, three considerations are needed here: on the one hand, in Italy the migration phenomenon is relatively recent, the real increase was around the end of the 1990’s and the beginning of the 2000’s. On the other hand, Italy’s citizenship policy (n. 91/1992) requires 10 years of residence for naturalization (Paparusso, 2019). Finally, the bureaucratic process may take up to 3 years, meaning that the actual naturalization rate is the result of what happened 10–13 years before. Until the year of the survey (2012) the naturalization rate registered in Italy was low and negligible (1.6%; ISMU, 2015).

Socio-demographic changes of the family in modern societies have modified the meaning of grandparenting, driving the attention towards three-generation interactions and highlighting the key role of grandparents as an important source of help for their (grand)children. Grandparental childcare in multigenerational households, in which children reside with at least one parent and grandparent, is an understudied topic in the European context. In addition, the existing literature that addresses grandparental childcare focuses on the whole population without distinguishing between migrant and non-migrant populations. This study addresses this gap, and it is the first one to investigate differences in grandparental childcare in multigenerational households between Italian and migrant families, by taking into account the role played by grandparents’ health and by gender.

As previously mentioned, studying the Italian context is very interesting. On the one hand it is because in Italy childcare is usually conceived as a female issue (Mencarini and Solera, 2004); on the other hand, it is because the country is characterized by a “familialism by default model” (i.e., characterized by scarce public child-care services and parental leave; Saraceno and Keck, 2010) where grandparents play a central role in providing childcare. Therefore, being a migrant may constitute a further constraint.

In particular, the study sheds light on co-resident grandparents to better understand if the “choice” of co-residence is due to a poor health status, thus the grandparents need care, or conversely, they represent a family support in helping parents by providing childcare.

From an overall analysis, we found that migrant co-resident grandparents are more likely to look after their grandchild(ren) than Italian co-resident grandparents do. This finding might suggest that, in migrant families, co-resident grandparents support and help the family in rearing a child, providing childcare in the host country.

By combining grandparents’ self-rated health and their attitude towards looking after their grandchild(ren), results display that, net of all controls (gender, age group, mother’s occupational status, number of grandparents in the household, grandparents’ occupational status, and the perception of the household’s economic condition), on the one side, migrant co-resident grandparents are significantly less likely to declare bad SRH and no-childcare and, on the other side, they are more likely to declare good health and to provide childcare than Italians. This result confirms the previous one, suggesting that in migrant families, the co-habitation of grandparents is principally linked to childcare support. However, caution should be applied in the interpretation of these results. The selection of individuals into multigenerational households may be linked to other factors. Financial hardship is seen as another important reason for drawing on the support of extended family in the form of intergenerational co-residence (Goodman and Silverstein, 2002; Baker et al., 2008; Dunifon et al., 2014). This might be the case among Italians, where a higher share of households declared a poor/insufficient perception of their economic condition with respect to migrants. Such a result might also reflect a higher percentage of multigenerational households with a single parent among Italians than migrants. Indeed, some studies found that, even in the case of single father or a single mother, those with co-resident grandparents are usually better off financially (Goodman and Silverstein, 2002; Mutchler and Baker, 2004). As explained in the background section, for migrant families, the “choice” of co-residence can be influenced by the Italian family reunification policies.

When gender disparities are considered, the real role is revealed. We found that being a woman, regardless of migrant status, is associated with higher probability of declaring poor health and to looking after grandchild(ren) than men. Such a significant role of gender could suggest that the “familistic” pattern is adopted also by foreigners: grandmothers may be asked to help their daughters and sons in childcare even if their health is poor.

When studying grandparental childcare, we must consider that it requires decision-making at two levels: that of grandparents in addition to parents. Some scholars argue that decisions about childcare and who is the “caregiver” are embedded in contextualized moralities: what individuals consider the “right thing to do” within the social framework in which they live (Duncan et al., 2003; Lewis, 2009). The gender differences observed in the analysisare well-known even in the literature. Actually, the gender division of families’ responsibilities between parents, who conceive childcare as a female issue, is likely to also pertain in the grandparental generation, as a result of traditional role ideology and divergent male and female care and employment trajectories over the life-course (Leopold and Skopek, 2014). The result is that grandmothers would be more likely than grandfathers to provide childcare support, and to add this “work” to their activities.

SRH is a good predictor of morbidity, use of healthcare services, and mortality (Idler and Benyamini 1997). However, health perception differs according to individuals’ characteristics, aspirations, and culture (Jylhä et al., 1998). As regards comparisons across cultures, evidence suggests that ethnic groups differ in their self-perception of health (Bombak and Bruce, 2012). Therefore, SRH may suffer from individual reporting heterogeneity (Bago d’ Uva et al., 2008), and its comparability among native and immigrant populations may be questioned (Jürges et al., 2008). Despite this issue, the literature has validated its use across ethnic groups, genders, and ages, showing that a poorer SRH is associated with higher disease prevalence rate (Chandola and Jenkinson, 2000; Gerritsen and Devillé, 2009; Crimmins et al., 2011).

This study is not without limitations. Our data are cross-sectional and cannot identify causality. We do not know how many grandchildren are cared for by grandparents and we have no information about the frequency of grandparental childcare or about grandparents’ gender role, which is an important factor, especially among migrants characterized by different cultures. Finally, we have no information about whether the dwelling was owned or rented, identified as an important determinant of co-residence (Stephens et al., 2015).

Despite these limitations, our findings provide new information on childcare provided by co-resident grandparents distinguishing between Italian and migrant families. The study suggests that grandparents’ cohabitation is the difference resulting from the need for care or offer of care. Furthermore, we found that, in addition to migrant status and self-report health, gender is a determinant of grandparents’ childcare: women look after grandchild(ren) more than men even with poor health conditions.

Given the increase of the elderly population and the migrant populations, including the aging of the migrant population, and the widespread economic austerity in all European countries, this would lead us to expect an increase in multigenerational households for the same reason as in the United States, especially in familistic countries. Further works are needed to explore the observed patterns in different contexts.