Living Arrangements of Older People and Determinants

Trinh Thai Quang*

Abstract: This paper^[1] describes the living arrangements of older people in Vietnam and factors that determine their living parttern using data from Vietnam National Ageing Survey 2011 with the sample of 2,789 older people. Findings indicate a majority of older people living in multigenerational households and a significant proportion of living alone. Older people who are in advanced age tend to live only with a spouse, only with a child or live alone; those who ever married tend to live in multigenerational households; household structure (large household, having grandchildren) facilitates living in multigenerational households while owning the home and having children living nearby encourage living alone.

Key words: Older People, Living Arrangement, Vietnam, Ageing, Intergenerational Relationship.

1. Introduction

Living arrangements are a critical aspect of older peoples’ lives because how they live and who they live with undoubtedly affect their daily life, health and well-being. Living arrangements can be understood as where you live and who you live with. There are numerous factors involved in an individual’s choice of a living arrangement, particularly for older people, including: the family structure, marital status, health, financial condition, housing conditions and, in particular, traditional familial values and norms (Velkoff, 2001). These all add complexity to the examination of living arrangements.

Under the long-term impact of Confucian cultural traditions, particularly regarding the roles and relationships among family members, Vietnamese elderly are expected to live in multi-generational households (MGHs), be householders and manage all family activities, be respected not only within the family but in society, and set good examples for later generations. At the same time, the family is regarded as the central institution providing care for the elderly, and adult children bear a strong obligation to provide care for their parents. Filial piety is considered one of the most important values that children have to follow.

Living arrangements can be perceived as a strategy the elderly adopt to secure their social security and well-being in later life. Gender role differentiation is firmly embedded in traditional Vietnamese society, and is reflected in older people’s living arrangements, with a preference for living with sons over daughters as they get older. This is a cultural effect of patrilineage, patrilocality and patriarchy in the kinship systems in Vietnam. It also reflects cultural aspects in providing care for family members, specifically the elderly. In recent years, older people’s positions and roles have changed dramatically, becoming more equal to the next generations’. With the impact of industrialisation, modernisation, migration and other social processes, patterns in older people’s living arrangements are changing accordingly to fit the new social and cultural context. A considerable literature in Western countries has identified factors that influence older people’s living arrangements; how applicable they are to the Vietnamese context is yet to be addressed.

This paper contributes to answering the central research question on how social changes affect older people’s living arrangements by first, focusing on describing living arrangement patterns of Vietnamese elderly, and second, identifying variations in living arrangements as they reflect an individual’s variabilities in a particular social and cultural context. It also specifically examines older people’s living arrangements in relation to structural context with regard to rural–urban differences. The following sections discuss Vietnamese older people’s living arrangements using data from the Vietnam National Ageing Survey 2011 with a sample of 2,789 individuals aged 60 and older. The analyses focus on older people’s family structure and their cultural preferences.

2. Family Structure and Living Arrangements

Dramatic economic, social and cultural changes in Vietnam since the 1990s have brought about profound changes in Vietnamese society and strongly affected family structure, ideologies and relationships among generations. Why and how does current family structure determine Vietnamese older people’s living arrangements? The following section discusses the relationship between family structure and various patterns of living arrangements.

Since the 1990s, the nuclear family has been seen as the primary structure in Vietnam (Hirschman & Vu, 1996); usually, two generations live together and the family is modest in size. A modified extended family structure has also emerged recently, in which family members of different generations live separately from each other but still keep ‘regular contact and maintain mutual support through visiting, the telephone, letters, emails and social networking sites’ (Browne, 2011, p. 92), Thus, they maintain close relationships. This type of family structure supports geographic mobility, especially as migration has become more common in Vietnam (particularly rural–urban migration). Demographic transition has promoted the rise of the nuclear family and the decline in multi-generation families. Additionally, there has been a decline in the average number of children per woman and a reduction in household size, both of which have led to variations in family living arrangements. Another arrangement is for parents to live adjacent to their children. In this type of family, older parents live in the same location as their children, but are relatively independent of each other. This can also be referred to as a modified extended family, as described above, because of the regular contact and direct mutual support arising from the close proximity.

The Vietnamese population has been involved in a transition process and experiencing many socio-economic changes. Two-generation households account for 63.4% of the total population, while MGHs have been declining though they are more common in urban areas; household size has declined over time. Single older person households increase as they tend to live alone or with their spouse only. However, the majority of older people are still living in MGHs or with at least one child (Truong et al., 1997).

In this paper, categorisation of older people’s living arrangements was based on familial relationships. The primary division of living arrangements includes living alone and living with others, with living with others further divided into seven sub-groups comprising (1) in multi-generational households, (2) with spouse and children, (3) with spouse and others, (4) with spouse only, (5) with children only, (6) with children and others and (7) with others. Children in categories (2), (5) and (6) will be identified by their age to assess whether there are any under 18 living with older parents.

Type of household and living arrangements

A nuclear household in this analysis contains an older couple without children or with children who may be still young and depend on their older parents, or adults who are married but have not had children yet at the time of the survey. The children living in nuclear households may also be adopted or children in-law, regardless of gender. The potential risks when grouping all these subjects into one group are the loss of information on differences between living with: a daughter versus a son, married children versus unmarried children, and dependent versus independent children. For example, if the older couple lives with a married, independent adult child, they may be the receiver of support or vice versa. Nevertheless, the number of older people living with at least one child under 18 years old is not significant. Thus, they were grouped together and considered a two-generation household, to be compared with a multi-generational household with three or more generations living in the same house.

Table 1. Living Arrangement Patterns (n = 2,789)

Two-generation households (an older couple and their children) in this analysis only accounted for about 20.2% of the total population, but considering the GSO’s definition of a nuclear household, the percentage would be 38.3% (including households with only an elderly couple). This contrast can be explained by the fact that the data used in this survey were from a sampling survey that targeted households with at least one older person. Thus, the possibility of having MGHs in the sample was much higher than for studies that investigated all households.

It should be noted that adult children in two-generation households include both unmarried and married but childless couples; it does not matter if they are biological children, adopted children or children in-law. This group consists of 352 older people, accounting for 12.7% of the sample, forming the third largest set of living arrangements. Older people in this group were concentrated among those aged 60–69, at 18.9%. This percentage decreases as age increases, implying that older people at more advanced ages tend to live in other living arrangements rather than with a spouse and children. This may occur because of life course effects, including change in marital status (death of a spouse) or children moving out of the house following marriage. Although the proportion of older people at more advanced ages living with a spouse and adult children is relatively small, it suggests a trend of delayed marriage and parenthood among the younger generations, which explains why they are still living with their older parents even after marriage.

Variations in this group were also found by gender and marital status. The percentage of older women in this group is significantly lower than their counterparts (7.8% compared with 19.8%), which may be the result of their higher life expectancy, which leads to a higher possibility of being widowed. Thus, they are more likely to live in MGHs, alone or only live with children rather than in this pattern of living arrangement. Also, those who are separated or divorced are not present in this sub-group; thus, the proportion of those who are no longer married living with a spouse and children is minimal.

Another type of nuclear household as defined above consists of an older person living only with a spouse (or an older couple household). They may have children or not; if they do, they may live near their children’s house or adjacent to a child. However, generally, they live relatively independent from their children. The circumstances of older people in this type of living arrangement are very different from those who live in two-generation households. For instance, when ill, it is easier for those who co-reside with children to get direct help from them than for those who live only with a spouse. The intergenerational relationship in this pattern of living arrangement also differs from those who live with children in regards to contact frequency, emotional support and intergenerational exchange. Thus, this paper analyses living only with a spouse and living with a spouse and children separately, providing more insights into different patterns of living arrangements.

The total number of older people living only with a spouse in this analysis is 505, accounting for 18.1% of the total sample. In contrast to the case of living alone, the pattern of living only with a spouse was reported as declining with age. This may be a result of a partner’s death at the older age, after which a number possibly chose to live alone, and hence, the number living alone increased. Alternatively, they may choose to move in with children or other relatives. Figure 1 indicates gender differences are marginal among older people aged 60–69 and 70–79, but become more significant among those aged 80 years and older. Particularly, among older people aged 60-69 who live only with a spouse, 44.6% are women and among those who 80 and older, this percentage reduces to 33.9%. As mentioned above, the life expectancy of Vietnamese women is higher than that of men, and this explains why the proportion of women aged 80 and older living only with their spouse is significantly lower than for male counterparts. Of course, a number would be women living with their husband and children in multi-generational families, as described above. Marital status is considered a crucial factor to the well-being of older people, with positive impacts for those living with a spouse. Living with a spouse also enables support for each other such as caregiving and financial assistance when required (Knodel & Chayovan, 2008; UNFPA, 2011a).

Figure 1. Living with a Spouse by Age and Gender (n = 505)

Source: VNAS 2011.

The dominance of the nuclear family helps explain the relatively large number of older people living only with a spouse. Adult children growing up, getting married and living apart from parents, together with a decrease in the average number of children per couple, leads to a reduction in the number of older adults living with children. These factors may also contribute to a higher possibility of older people living only with a spouse. The results from this analysis also found a correlation between the number of children and living only with a spouse among Vietnamese older people, but the influence of number of children on older people’s living arrangement patterns is analysed and discussed later in the thesis by using regression analysis.

Another result that demonstrates the trend towards family nuclearisation is the relatively high proportions of older people having sons and grandchildren but living only with a spouse. Similar to other patterns of living arrangements, this pattern has been influenced by multiple interrelated factors. For instance, living in a smaller family possibly decreases the likelihood of older people living with children and grandchildren, and the resources for support may decline as well. Thus, older people may need to keep working for their living. Conversely, having children living nearby could be the motivation for older people to live on their own because proximity to children may ensure proper immediate support from children, especially care provision when needed. These factors are discussed further in the section on older people’s resources and vulnerabilities.

An older parent living only with children is also considered a nuclear household. In this analysis, all children are included, regardless of age, gender, marital status and whether they are adopted or biological children (but adult children must be without children of their own). Thus, in general, this type of living arrangement consists of only one older person and their children living under the same roof. Older people’s marital status is closely related to this type of living arrangement, with the majority ‘no longer married’ (divorced, separated or widowed).

Generally, 7.4% of older people were reported to live only with children. Variations by age are present in this group, with the oldest elderly holding the largest share at 13.3% compared with 7.6% of those in the 70–79 age group and only 3.6% in the 60–69 age group. A significant difference between male and female older people was also reported, with 10.8% of older women compared with only 2.5% of older men. This may imply an increasing number of older women living only with children, or even living alone, because of their widowhood, separation or divorce. It is also likely connected to their higher life expectancy compared with male counterparts.

There is no significant difference between rural and urban areas, income adequacy, access to a pension or not, having a son, and children living nearby, in this group of elders. The most obvious variation is marital status. Those who are no longer married are more likely to be in this group, with married counterparts accounting for only 0.7%. None of the never-married older people are in this group, as they are unlikely to have children with whom to live. Regarding household size, contrary to the case of those who live in MGHs, the highest share of older people living only with children is among those who live in smaller households (contain 1–4 members), while older people tend to live in MGHs when they have larger households (more than five members).

Skipped-generation household is a pattern that includes older people living with spouse and others (accounting for 3.8% of the total sample), with the others in this category including grandchildren and great-grandchildren. In this type of household, older people take care of their grandchildren and the grandchildren’s parents are not living in the household. One of the potential reasons for their absence is migration to other areas (more often to urban areas or cities) or to other countries to work, which is a widespread social issue in Vietnam, especially in rural areas. Older people in this living arrangement were reported as more vulnerable in terms of financial resources, depending on remittances provided by their out-migrated adult children (UNFPA, 2011a). Grandparenting may become burdensome to these older people, particularly when they have financial and health difficulties. They also face lack of care and affection, and do not have adequate time for leisure or social networking activities (Tran, 2016).

Living in a multigenerational household is the traditional pattern of living, which remains the primary pattern for Vietnamese older people. This group contains sub-groups, the most common of which consists of those who live with a spouse, children and grandchildren in a typical three-generation household; a few cases include great-grandchildren in the household (four-generation households). The other sub-groups include older people’s parents, the older couple and children; older people’s parents, an older person, their children and grandchildren; and a horizontally extended traditional family that includes the older people’s siblings. Although there are differences among these compositions regarding the position of older people in the family and their related responsibilities (as both caregivers and care receivers), the main reason for grouping them as one category is that they live in a family with several generations where intergenerational exchange of support of all kinds may be more direct and frequent than in other types of living arrangements. Further, the shares of the sub-groups are minimal (1.8% with parents, 0.6% with parents-in-law, 1.4% with siblings), and would be insignificant if considered alone in the analysis. Thus, they were merged into the one group living in multigenerational households, as described in Figure 2 below:

Figure 2. Grouping Categories for Living in a Multi-generational Household

Source: Developed by author

Living in MGHs varies slightly by age. The share of older people aged 80 years or older living in a multi-generational family is slightly higher than for other age groups (47.8% compared with 46.6% among 70-79 years old, and 41.4% among 60-69 years old). Gender difference is not significant, with 47.2% of females and 41.0% of males found to live in MGHs. The proportion of no longer married older people in MGHs is the highest, at 54.9%. Conversely, a modest number of never-married older people were reported in this pattern of living (7.7%). This result presents the apparent impact of marital status on older people’s living arrangements, as no longer married older people have fewer options to choose from as they no longer have their spouse. Thus, they may choose to live in another setting after the death of their partner, and in this case, it is more likely to be in MGHs. Older people who have grandchildren were also reported to mainly live in MGHs (46.7%). Having grandchildren may increase the possibility of older people living in MGHs because when living in the same household, they can help their adult children care for their grandchildren. Alternatively, older people in larger households were found more in MGHs than in other patterns of living arrangements.

Living with children and others indicates an older person co-residing with children and other people of the same generation, such as siblings; this is another type of nuclear household. This group in VNAS 2011 is insignificant, accounting for only about 0.1%. This pattern of living arrangement is similar to those living with a spouse and children; however, the major difference here is older people’s marital status, which helps to determine this group. Older people in this group are likely to be no longer married, as is the case of older people living with others, which refers to an older people co-residing with other relatives including their parents, siblings, grandchildren or great-grandchildren. In this case, they possibly play the role of support providers rather than receivers, which may influence their well-being. A large share of this group includes those living in skipped-generation households, who are even more vulnerable as they are not accompanied by a spouse.

Living alone once referred to older people who live only with a spouse and unmarried older people living with no kin (Palloni, 2001). However, this definition is problematic as the case of those who live with a spouse is entirely different from those living on their own regardless of marital status. It is closely related to family structure and marital status of older persons. Living alone does not necessarily imply these people have never married or are childless. Living alone poses many challenges for older people, related but not limited to physical and mental health care, financial security and nutrition. Although living alone can promote ‘privacy, independence, and dignity for the elderly’ (Arokiasamy, 1997, p. 233), they are more in need of external support in case of illness or disability, and also have a higher risk of social isolation (United Nations, 2005). Much previous research has indicated a rapid increase in living alone for Vietnamese older people (GSO, 2011; UNFPA, 2011a; Vietnam Women's Union, 2011). The most recent research shows that 6.1% of older people were living alone in 2008 (UNFPA, 2011a); in this analysis, it was 9.4%. In this research, the term only refers to an older person who lives with no kin, regardless of marital status. This helps to identify more precisely the circumstances of these people because those living with a spouse can receive immediate support from their partner when needed, in addition to emotional benefits.

Living alone, as noted previously, is an increasing trend among older people in Vietnam, especially among the oldest age group (age 80 and older), and seen more in rural areas than urban areas (10.8% compared with 5.3%). The variation becomes more obvious when looking at marital status in relation to this pattern. Never-married older people tend to live alone (33.8%), compared with 20.7% of those who are no longer married and only 0.6% of those who are married. This trend was also seen more among female older people with no children or grandchildren. Data from VHLSS from 1992/93 to 2008 revealed a slight decline in the percentage of married older people from 64% in 1992/93 to 59% in 2008 and an increase in widowed elderly from 34% in 1992/93 to 39% in 2008. This implies a higher proportion of older people who live alone or with people other than their spouses. For instance, the proportion of older people who live alone almost doubled between 1992/93–2008. Conversely, the percentage of those who lived with children was sharply reduced from 80% in 1992/92 to 63% in 2008 (UNFPA, 2011a). UNFPA’s report did not count MGHs as a category of living arrangement, which is very different from other types of living arrangements in regards to intergenerational exchange and older people’s well-being.

Figure 3. Living Alone by Age and Gender (n = 261)

Figure 4. Living Alone by Marital Status and Gender (n = 261)

Source: VNAS 2011.

Figure 3 depicts gender differences among older people living alone in each age group. It is easy to see that the percentages of male elderly living alone tend to increase as they age (from 13.2% to 22% across three age groups), but decrease among women. In terms of marital status, among the never-married elderly who live alone, 91% are female compared with only 9.1% of male elderly. This proportion decreases sharply among married female elderly (66.7%) but significantly increases among their male counterparts (33.3%). While never-married people might have a higher possibility of living alone, no longer married participants seem to have more choice of living arrangements. They were reported more in MGHs than other patterns of living arrangement, which may suggest changes in their marital status influence their choice of living arrangement. In this analysis, it seems that they tend to choose to live with children and grandchildren, or live only with children or other relatives, from whom they receive support.

The number of children is a family structure indicator and a resource for older people in terms of support. It is assumed that older people with more children will have a greater possibility of co-residence with children than those who have fewer children. It can be seen that many older people who have no children live alone or with others (35.2% and 42.9% respectively). About 10% live only with a spouse and 4.8% live with a spouse and an “other”. Having no children, living alone and living with others may escalate vulnerabilities, as older people obtain inadequate support from their offspring when they are in need, affecting their well-being at more advanced ages.

Older people who live in smaller households have fewer opportunities for co-residence with children or other relatives. This is, as argued by Knodel et al. (2000), because when household size is smaller, the proportion of older people having only daughters may be higher, which reduces the possibility of co-residence. Household size is also closely related to intergenerational exchange between older people and their adult children, which, in the current research, is divided into three sub-groups, including households with 1–4 members, 5–7 members and 8 and more members. The results show significant variation between living in MGHs and other patterns of living arrangements. It can be said that the larger the family, the higher the likelihood of older people living in MGHs and vice versa.

Apart from family structure, cultural preference is no less important to consider when examining patterns of living arrangements among Vietnamese older people. It is influenced strongly by the traditional perception of how family should make their living arrangements related to different social contexts and family conditions. The most important factors are the kinship system and son preference, as discussed in the following section.

3. Cultural Preferences for Living Arrangements

From the literature review, living arrangements are strongly influenced by cultural preferences, including social customs (Hashimoto, 1991), traditional living patterns and son preference. Cultural traditions also influence living arrangements, even in Western countries, and they vary with specific cultural context. In some countries, older people prefer to live independently if they have sufficient resources. In others, they may live with children until the children get married. This finding from the literature is also reported in the Vietnamese context, as children usually live with their parents until they marry. Even after marriage, they may still live with parents until they can afford their own accommodation.

The existence of MGHs depends on land availability, housing condition and home ownership (Hirschman & Nguyen, 2002). This section mainly focuses on the kinship system and son preference in the current social context in Vietnam in regard to older people’s living arrangements.

The patrilocal kinship system and son preference in Vietnam encourage older people to live with sons rather than daughters, and this was confirmed in this analysis. The percentage of those who live with sons is more than twice the percentage of those who live with daughters. This is important empirical evidence proving the ongoing existence of son preference in Vietnam and supports the argument that Vietnamese families are inclined towards patrilocal rather than matrilocal residences. The son, in traditional culture, plays an important role in the family because he carries the family name, carries out religious and spiritual duties such as ancestor worship, and critically, provides care for older parents with both physical and emotional assistance. Elderly people without a son may live only with their spouse or alone, and may only move into their married daughter’s house when they become widowed.

Son preference in Vietnamese society remains strong, especially in rural areas. The proportion of older people who live with at least one son is significantly higher than those who live with at least one daughter. The higher percentage of older people having a son who live in MGHs (relative to those who have no son) (46.5% and 30.4% respectively) demonstrates the significant relationship between having a son and living arrangements. The traditional preference for living with a son is also partly reflected in the group of older people living with a spouse and children, in that among those who have a son 13.5% are living with spouse and children (the third largest group) compared with 5.4% of those who do not have a son. How do older people arrange their living if they have no son? The majority live in MGHs (30.4%), followed by alone (22.1%), with other people (17.9%) and only with a spouse (13%). Thus, more than 50% of older people who have no son are living in other living arrangements including living alone, living with others and living only with a spouse rather than in MGHs, reflecting the still-strong son preference in Vietnam.

Figure 5. Living Arrangements According to Whether or Not the Elderly Have a Son (n = 2,789)

Source: VNAS 2011.

4. Structural Context and Living Arrangements

Variations in older people’s living arrangements by structural context focus on differences between rural and urban areas in Vietnam. In 2014, the majority of Vietnam’s older population lived in rural areas (68%), compared with 32% in urban areas,[2] which was confirmed in this analysis as the data indicated 73.5% of respondents live in rural areas. Given the enormous differences in socio-economic conditions and cultural context between rural and urban areas in Vietnam, this section explores how older people choose their living arrangements in each of these two areas.

Table 2 provides information on the living arrangements of older people by area of residence. Initially, it was assumed that the proportion of older people living in MGHs in rural areas would be higher than in urban areas because of the strongly embedded traditional culture; the results, however, indicate the opposite. The proportion of urban elderly living in MGHs is higher than for those living in rural areas (52% compared with 42%). Many factors have led to the reduction of MGHs in rural areas, including the migration of young adults and housing constraints in the urban sector, contributing to the high proportion of MGHs in urban compared with rural areas (Lam, 2008).

Table 2. Living Arrangements of Older People by Area of Residence (n = 2,789)

Migration of adult children from rural to urban areas contributes to creating a pattern of living arrangements in rural areas referred to as ‘skipped-generation households’ (as discussed earlier), which implies that only the first (older people) and third generation (grandchildren) in the family live together. This pattern has specific impacts on the lives of the elderly. Missing the middle generation in the family may reduce the resources of rural older people, particularly when they need immediate support, for example, when they are sick. Moreover, it might be a burden for them to take care of grandchildren, especially if they have insufficient financial resources or poor health. Later, their well-being may be negatively influenced. Once adult children settle in an urban area, they may bring their older parents from rural areas to urban areas to live with them in the same house, which, in turn, increases the number of older people living in MGHs. Filial piety remains strong in Vietnamese society and adult children bear the responsibility of providing care for older parents. Thus, MGHs help to ensure that older people receive care directly and immediately from their children.

The proportions of older people living alone and living only with a spouse in rural areas exceeded those in urban areas. In the case of those living in rural areas, the share of older people living alone was significantly higher than in urban areas (10.8% compared with 5.3% respectively). Living only with a spouse is also more frequent among older people in rural areas (20.4%) than in urban areas (11.6%). The possible explanation for this is the closer proximity of older parents and non-co-resident adult children in rural areas, an arrangement that is popular in rural areas, as discussed above. When older parents have children living nearby or in the same village, it is not necessary to live together to ensure immediate support or help from each generation. However, this is not the case for older people in urban areas, where housing constraints are much more serious than in rural areas. Housing constraints may limit the possibility of both generations living independently of each other among the urban population; on the contrary, availability of land and proximity may facilitate living alone or living only with a spouse in rural areas. The proportion of older people living only with children in urban areas is slightly higher than for those who live in rural areas. No significant difference was found in other patterns of living arrangements.

Regarding age, there is a common trend across the three age groups that the proportion of older people living alone in rural areas is significantly higher than in urban areas. For instance, among those aged 60–69 years, 3.8% of urbanites live alone compared with 7.3% in rural areas. For those in urban areas aged 70–79 5.5% live alone compared with 12.1% of those in rural areas, and 7.2% of those aged 80 or older who live alone in urban areas, compares with 14.9% in rural areas. Results related to living with a spouse are similar in terms of the difference between the two areas by age. The same variation can also be observed between the two areas and age groups among those who are living in MGHs, but in reverse, so that urban older people are reported more frequently in living in MGHs than rural older people, as mentioned above. However, regional differences are not obvious, especially among those aged 80 and older.

There is no significant variation between urban and rural areas by age among elderly living only with children. The percentages of those aged 60–69 living only with children in both areas are minimal at 3.5%. At more advanced ages, these percentages increase, and are higher in urban areas (9.2% compared with 6.7% in rural areas for those aged 70–79, and 14.8% in urban areas compared with 12.4% living in rural areas, for those aged 80 and older). The increase in percentage of older people living only with children is directly related to marital status. With age people are more likely to face widowhood, and once this happens, different options for living arrangements are pursued. If they are living separately from children, they may move in with them. In the case where they are already living with children, they live only with children when their partner passes away. The other option is living alone in their own home or in a nursing home, which is not popular in Vietnam yet as families still play the central role in caring for older people.

Table 3. Gender Differences by Area of Residence (n = 2,789)

Source: VNAS 2011.

Another effect of marital status can be seen in the case of those living with a spouse and children. Among those aged 60–69, no difference between rural and urban areas was found (18.9% in both areas). However, this changes sharply among those aged 70–79 and 80 and older; at these ages, the likelihood of widowhood is much higher. The percentage living with a spouse and children is slightly higher in urban areas than in rural areas for those in the 70–79 age group and doubles for the 80 and older group. This may be because of housing difficulties in urban areas and higher rates of extended delayed marriage among urban young people (both male and female) than among their rural counterparts (UNFPA, 2011c). This allows the urban young people to stay longer in their parents’ home. Never married or no longer married older people living alone are found more in rural than urban areas. Gender difference is also present. In rural areas older women live alone more frequently than men and this trend is even more pronounced in urban areas. Proportions of older males living alone in both areas are insignificant, but still more common in rural than urban areas (see Table 3). On the contrary, the percentage of older men living only with a spouse significantly exceeds that for older women living only with a spouse in both urban and rural areas, but particularly in rural areas (28.7% in rural areas compared with 19.1% in urban areas). Regional differences in the case of living only with children are not significant, but gender differences within areas are remarkable, indicating a higher percentage of older women living only with children compared to male counterparts who also live only with children. The explanation may be the same as for those living alone; that is, widowhood.

As noted above, the majority of older people in this research live in MGHs. How does this differ between rural and urban areas, given that socio-economic conditions and culture differ? Variation between the two areas is not significant; however, in terms of gender, the difference is clear. Older women living in MGHs are reported more in urban areas than in rural areas (54.3% and 44.6%). This is similar in the case of older men (47.5% vs 38.6% respectively). In other patterns of living arrangement, the variation is not significant.

Differences in resources and vulnerabilities regarding urban and rural areas are significant in some aspects, especially financial security. 79% of rural elderly living alone do not have sufficient income; this is around 46% for urban counterparts. The gap is even more significant between rural and urban areas regarding those who live with a spouse and others; in rural areas, the majority of this group (72.2%) do not have sufficient incomes, while the figure for insufficient income is only 23.5% in urban areas. Differences between rural and urban areas were reported in other living patterns as well, suggesting that rural older people are more vulnerable than their urban counterparts in terms of income adequacy. Pensions are another source of income for older people, but in the case we did not find evidence of variations of living arrangements by region, related to pension status, except for the case of living with others, in which the elderly with no pension were found more in urban than in rural areas, accounting for 60% and 35.4% respectively.

Home ownership is another resource for older people; however, this also varies between urban and rural areas with different housing constraints and conditions. In urban areas, the percentage of elderly living alone who own a house is lower than for those who live in rural areas. This reality is reversed among those who live in MGHs, implying that the percentage of older people who own a house and live in MGHs in urban areas is much higher than in rural areas. These results are evidence of the difficulties in securing independent housing in urban areas, which contribute to the living arrangements of older people.

5. Determinants of Older People’s Living Arrangements

It is crucial to identify how older people choose their living arrangements, as this is closely connected to their circumstances and well-being. This section identifies significant determinants of Vietnamese older people’s living arrangements using binary regression. Dependent variables include five types of living arrangement: living alone, living only with a spouse, living only with children, living with a spouse and children and living in MGHs. In addition to social and demographic characteristics, whether older people have a son, children living nearby and are independent also have impacts on living arrangements.

Before implementing the regression analysis, all variables were tested for correlations. One of the interesting findings is that older people’s independence is only correlated with two patterns of living arrangements: living only with children and living with a spouse and children. However, after controlling for other variables, this appears not to be a significant influencing factor on living arrangements, suggesting a more powerful effect from other factors and/or cultural traditions in which older people expect to live with their children and grandchildren as they age. Traditional norms and values stress the role and responsibility of adult children in taking care of older parents and these norms and values are still strong in Vietnamese society. Thus, it does not matter how independent older people are in relation to their living arrangements; rather, other factors, including marital status, health, area of residence, home ownership, work status, and having a son or grandchildren are important. As a result, the variable capturing independence is excluded from the logistic regression models to ensure correct analyses. Instead, three single variables: adequate income, home ownership and health status are included in the models. Table 4 provides results of the binary regression analyses investigating significant determinants of living arrangement patterns of older people.

Multigenerational households

As mentioned earlier, the traditional pattern of living arrangements among older people in Vietnam is MGHs. In this research, having grandchildren, large household size and better health were found to be positive factors facilitating living in MGHs, among which having a grandchild is the most important. This suggests the critical role of older people in providing care for their grandchildren.

At the same time, age, marital status, home ownership, work status and location of non-co-resident children are negatively associated with this type of living arrangement. This means that older people who are at a more advanced age, no longer married or never married, who do not own their home, are no longer working and who have no children living nearby are less likely to live in MGHs. The number of children a person has turns out not to be a significant factor influencing older people’s living arrangement, contrary to the assumption that having more children would increase the likelihood of older people co-residing with children, including in MGHs (see Table 4).

Living with a spouse and children

Living with a spouse and children is a type of nuclear household where the older couple live with their adult children regardless of the children’s marital status. In this pattern of living, marital status and number of children were found to be significant factors. The results suggest that older people who are currently married and have more children are more likely to live with a spouse and children than are their counterparts (OR = 84.4 and 2.3 respectively). Urban older people also tend to live in this arrangement more than rural older people.

Consistent with the case of living in MGHs, the age of older people is negatively associated with living with a spouse and children. Older people of more advanced ages are less likely to live in this pattern (OR = 0.7; 95% CI = 0.57–0.86). However, the results show that household size and having grandchildren negatively influence this pattern of living. The larger the household size, the less likely older people are to live with a spouse and children. Instead, they tend to live in MGHs.

Results show that older women are less likely to live with a spouse and children than are older men (OR = 0.65; 95% CI = 0.49–0.86) and those who have higher education levels are also less likely to live in this pattern of living arrangement (OR = 0.75; 95% CI = 0.6–0.95).

Living only with a spouse

Living only with a spouse only encompasses older people who are currently married. Therefore, the marital status variable was excluded from the analysis. The results indicate that three factors significantly influence this pattern of living arrangement: the age of older people, residential area and their number of children.

Contrary to the case of those older people living in MGHs, older people who are at a more advanced age are more likely to live only with a spouse than those who are in early old age. Living only with a spouse is also more likely to be seen among rural older people than urban older people (OR = 0.44; 95% CI = 0.32–0.62). Older people with more children are less likely to live in this pattern as they tend to co-reside with children.

Living only with children

This living arrangement pattern is closely related to older people’s marital status, as discussed earlier, with married older people less likely to live in this living arrangement than their counterparts (OR = 0.8; 95% CI = 0.05–0.14). This suggests a change in living arrangements as married older people are highly likely to live with a spouse and children while those who are not married are more likely to live in MGHs, only with children or alone. Household size also negatively influences this pattern of living arrangement as the larger the household size, the less likely older people are to live only with children (OR = 0.02; 95% CI = 0.01–0.06) and the more likely they are to live in MGHs.

The most significant factor that determines this living arrangement is number of children. As shown in Table 4, the more children older people have, the more likely they are to live only with children (OR = 4.39; 95% CI = 2.38–8.1). Older women are more likely to live only with children than older men, which makes gender the second most significant factor (OR = 2.01; 95% CI = 1.30–3.30), following by residential area and age. Urban older people and especially those at advanced ages tend to live more in this pattern of living arrangement. These findings suggest the influence of the higher life expectancy of older women and the differences in structural context on living arrangements in later life.

Living alone

Results from the analysis of this living pattern contradict those regarding MGHs. Older people who own a home are more likely to live alone (OR = 6.49; 95% CI = 3.80–11.1) and those who have children living nearby, are still working, and are more advanced in age have a higher likelihood of living alone. This suggests that people who live alone may still have to work for their living, and may lack support from their family and resources. Those who live in MGHs can be supported by their family, and maintaining employment is a choice and no longer due to concerns about making a living.

Never married or no longer married older people are more likely to live alone than married counterparts. Those who live in rural areas have a higher tendency to live alone than older people who live in urban areas.

Adequate income or pension receipt and having a son have no significant impacts on patterns of older people’s living arrangements.

From the above findings, generally, factors that influence older people’s living arrangements in this analysis encompass demographic information (age, gender and marital status), structure context (rural/urban), resources and vulnerabilities (except for income or pension), family structure (household size, number of children and grandchildren), and the location of the children. The following section discusses these findings in the context of social change in Vietnam.

6. Summary and Discussion

Family is still the major institution that provides support for older people in Vietnam and other developing countries in the region (Bonggaarts & Zimmer, 2002), and multi-generational households, the traditional living pattern, are the most supportive for older people (Be, 2005), facilitating support exchange between generations. Results from this paper indicate that the majority of older people living in MGHs are the older elderly, females and no longer married (divorced, separated, or widowed).

It is crucial to understand the living patterns of older people to assess whether they choose to live with children or whether their circumstances force them to live with their offspring because this is connected to the quality of older people’s relationships with adult children and the parents’ well-being in the later life. When older people age, they become more vulnerable and face more challenges such as with income, difficulties in daily activities and especially health problems. They become more dependent on other people in the family; in most cases in Vietnam, they depend on their offspring. If only age is taken into account, the results from the descriptive analysis would suggest that the majority of older people in more advanced ages are living in MGHs, which is partly to fulfil their need for care in that period of life. However, findings from the regression analysis suggest a converse trend, where older people who are in their early old age are more likely to live in MGHs. The most plausible explanation for this finding is that the adult children of early old age elderly are, to a certain extent, still dependent on their parents (for accommodation, care for grandchildren, and even financial support); in the case of those at more advanced ages, their children may be able to afford to move out. There are also many cases of older people who are ‘independent’ deciding to live with their children. In this case, it is cultural effects or their children’s need of their support (e.g., grandparenting) that lead to this pattern of living arrangement, and the older people are more likely to play the role of support providers rather than receivers to their adult children.

Further, Vietnamese older people tend to live with their sons if they have one, which reflects the patrilineality, patrilocality and patriarchy in kinship systems embedded deeply not only in Vietnam but in other countries in Asia. Son preference has been well documented in previous studies in Vietnam (Bélanger, 2002; Guilmoto, 2012; Haughton & Haughton, 1995; Truong et al., 1997; UNFPA, 2011b)as well as in other countries such as China, India, Japan and Korea. This relates to several cultural and socio-economic aspects. For example, in terms of culture, a son is considered the person in charge of continuing the family line and practising ancestor worship. In regards to socio-economic aspects, they offer old age support and inherit assets. Thus, much research has found that the majority of Vietnamese older people co-reside with a son (UNFPA, 2011b), and when this son gets married and has his own children, older people live in MGHs. This is the effect of life course on older people’s living arrangement patterns (Pfau & Giang, 2007), which may contribute to explaining the high proportion of older people living in MGHs in Vietnam.

Results from the descriptive analysis show that older people who have a son are reported more in MGHs or with a spouse and children than in other living arrangements; those who have no son are seen more living alone. There are many reasons that older people decide to co-reside with a son over a daughter, but in general, these may be condensed in the following points. First, it is the effect of cultural expectations regarding sons’ roles. Second, it is related to gender discrimination or gender prejudice rooted in feudal culture, which considers daughters ‘other people’s children’ when they marry and not always available to provide support for older parents when needed. This perception is opposed to Western culture with a popular proverb saying that ‘my daughter’s my daughter all the days of her life’ (Ray, 1670, p. 52 cited in Speake & Simpson, 2009). Third, social welfare in Vietnam does not meet the demands of older people, which ‘force people to give birth to a son’ (UNFPA, 2011b, p. 27)so that they can rely on them in their advanced age. However, it is also interesting to note that some studies do mention the significance of daughters’ support for their older parents, even when they get married and have their own families; this would be more significant to older people who have no male offspring.

Contrary to these findings on the importance of having a son, results from the regression analysis do not show any significant impact of having a son on examined patterns of living arrangement. This partially suggests a change in perception regarding sons and implies other underlying determinants of older people’s living arrangements. In another words, having a son is not as important as assumed in regards to older people’s choice of living arrangement, although the traditional son preference still exists in society. This finding partly confirms the idea that the Vietnamese family system is a bilateral kinship system, which ‘predominate in South-East Asia and southern India’ (Mason, 1992 cited in Truong, 1997, p. 8).

In terms of gender, results from the analysis in this paper partially oppose Pfau and Giang (2007), in that it indicated a higher percentage of older women living with children, but is consistent with the finding that a higher portion of older women are living alone compared with older men. As discussed in Pfau and Giang (2007), this is because of differences in life expectancies and marital status between males and females. The implication of this gap is closely related to marital status, as it implies more Vietnamese women will become widows as they age than men will, and this leads to a higher possibility of older women living with their children, in MGHs or even alone.

Working status was found significant associated with living in MGHs and living alone. Older people who are no longer working were found more in MGHs. In contrast, those who are still actively working are more likely to live alone than in other living arrangement patterns. A lack of resources may be one of the main reasons that elderly living alone have to work, as the majority do not have adequate income. Previous research has also found that the portion of Vietnamese older people with sufficient income (for example, from savings and pensions) is relatively low, and thus, many still participate in economic activities (GSO, 2010), and this has been reported as increasing over time (Ngo, 2013). To some extent, older people participating in economic activities should not be seen as a negative issue or the result of insufficient social welfare support; rather, it should be considered a contribution older people make to their family’s income (or at least, their attempt to reduce the burden on the family of providing support for them) and society.

Regional variations were reported in this analysis, with variations mostly related to housing constraints in urban areas that lead to a higher percentage of older people living in MGHs in cities. At the same time, more older people live alone or only with their spouse in rural than in urban areas. The differences in living conditions between the two regions may also contribute to different patterns of living arrangements. For instance, family members tend to live in closer proximity in rural areas, which allows older people to live by themselves but still ensure that they have immediate support from their children who live nearby.

The living arrangements of older people comprise an extremely complex aspect because it relates to several layers of their lives, not only at the individual level but at the social level. At the individual level, the decisions reached about living arrangements incorporate the socio-economic characteristics of the families and the elderly themselves, the relationships and exchange of resources between generations and long-term quality of life and well-being. At the social level, such decisions involve providing an adequate and appropriate system of social welfare for the elderly that may allow them to choose whether to live independently or rely on their offspring’s support in their advanced age.

________________________

References

Arokiasamy, J. T. (1997). Social problems and care of the elderly. Medical Journal of Malaysia, 52(3), 232-237.

Be, Q. N. (2005). Living strategies of rural elderly in family transition. Vietnamese Sociological Review, 1(89), 65-72.

Bélanger, D. (2002). Son reference in a rural village in North Vietnam. Studies in Family Planning, 33(4), 321-334.

Bonggaarts, J., & Zimmer, Z. (2002). Living arrangements of older adults in the developing world: An analysis of demographic and health survey household surveys. Journal of Gerontology, 57B(3), S145-157.

Browne, K. (2011). An introduction to Sociology. Cambridge, United Kingdom: Polity Press.

GSO. (2010). The 2009 Vietnam population and housing census: Completed results. Retrieved from http://www.gso.gov.vn/default_en.aspx?tabid= 515&idmid=5&ItemID=10799

GSO. (2011). Vietnam Population and Housing Census 2009: Age-sex structure and marital status of the population in Vietnam. Retrieved from https://www.gso.gov.vn/Modules/Doc_Download.aspx?DocID=13286

Guilmoto, C. Z. (2012). Son preference, sex selection, and kinship in Vietnam. Population and Development Review, 38(1), 31-54.

Hashimoto, A. (1991). Living arrangements of the aged in seven developing countries: A preliminary analysis. J Cross Cult Gerontol, 6(4), 359-381.

Haughton, J., & Haughton, D. (1995). Son preference in Vietnam. Studies in Family Planning, 26(6), 325-337.

Hirschman, C., & Nguyen, H. M. (2002). Tradition and change in Vietnamese family structure in the Red River Delta. Journal of Marriage and Family, 64(4), 1063-1079.

Hirschman, C., & Vu, M. L. (1996). Family and household structure in Vietnam: Some glimpses from a recent survey. Pacific Affairs, 69(2), 229-249.

Knodel, J., & Chayovan, N. (2008). Population ageing and the well-being of older persons in Thailand: Past trends, current situation and future challenges. Thailand.

Knodel, J., Friedman, J. E. D., Truong, S. A., & Bui, T. C. (2000). Intergenerational exchanges in Vietnam: Family size, sex composition, and the location of children. Population Studies, 54(1), 89 - 104. doi:10.1080/713779067

Lam, N. N. T. (2008). Industrialization and changes in Vietnamese families. Paper presented at the 3rd International Conference on Vietnam Studies, Hanoi, Vietnam.

Mason, K. O. (1992). Family change and support for the elderly in Asia: What do we know? Asian-Pacific Population Journal, 7(3), 13-32.

Ngo, T. T. D. (2013). Elderly people in Vietnam: Some issues of concern from a family and gender perspective. Paper presented at the Ageing in Asia Pacific: Balancing State and Family, Cebu, Philippines. 3-4 April, 2013.

Palloni, A. (2001). Living arrangements of older people. In United Nations (Ed.), Ageing and Living Arrangements of Older persons: Critical Issues and Policy Responses (Vol. Special Issues). New York: United Nations.

Pfau, W. D., & Giang, T. L. (2007). Patterns and determinants of living arrangements for the elderly in Vietnam. In T. L. Giang (Ed.), Social issues under economic transformation and integration in Vietnam (Vol. 2, pp. 147 - 176). Hanoi, Vietnam: The Publishing House of Social Labour.

Ray, J. (1670). A complete collection of English proverbs (3rd ed.). London: Printed by J. Hughs, J. Torbuck, O. Payne, T. Woodman.

Tran, T. M. T. (2016). Life arrangement and care provision of lef-behind elderly in Vietnam. Vietnam Journal of Social Sciences, 5(175), 36-50.

Truong, S. A., Bui, T. C., Goodkind, D., & Knodel, J. E. (1997). Living arrangements, patrilineality and sources of support among elderly Vietnamese. Asia-Pacific Population Journal, 12(4), 37-49.

UNFPA. (2011a). The ageing population in Vietnam: Current status, prognosis, and possible policy responses. Retrieved from http://vietnam.unfpa.org/ sites/asiapacific/files/pub-pdf/Ageing%20report_ENG_FINAL_27.07.pdf. Retrieved from

UNFPA. (2011b). Son preference in Vietnam: Acient desires, advancing technologies. Retrieved from http://vietnam.unfpa.org/en/publications/son-preference-viet-nam-ancient-desires-advancing-technologies.

UNFPA. (2011c). Young people in Vietnam: Selected indicators. Retrieved from http://vietnam.unfpa.org/en/publications/young-people-viet-nam.

United Nations. (2005). Living arrangements of older persons around the world, 119. Retrieved from http://www.un.org/esa/population/publications/ livingarrangement/report.htm

Velkoff, V. A. (2001). Living arrangements and well-being of the older population: Future research directions. In United Nations (Ed.), Ageing and Living Arrangements of Older Persons: Critical Issues and Policy Responses (pp. 376-385). New York, USA: United Nations.

Vietnam Women's Union. (2011). Vietnam national ageing survey (VNAS) 2011: Key findings. Retrieved from https://www.researchgate.net/publication/ 280728578_VIET_NAM_AGING_SURVEY_VNAS_2011_KEY_FINDINGS.

* Ph.D., Institute for Family and Gender Studies, Vietnam Academy of Social Sciences.

[1] This paper was derived from the author’s PhD Thesis “Ageing and Intergenerational Relationships in Vietnam”. Accessible at https://openresearch-repository.anu.edu.au/handle/1885/151965

[2] Data retrieved from HelpAge International’s website, accessed at http://ageingasia.org/ageing-population-vietnam/ on 14 April 2016.

Sources cited: JOURNAL OF Family and Gender Studies. Vol. 15, No. 1, p. 38-64