• 2. Family Support For The Elderly In Urban China: An Intergenerational Relations Approach
• Chapter
• University of Michigan Press
• pp. 33-59
• View Citation
• Additional Information

2

FAMILY SUPPORT FOR THE ELDERLY IN URBAN CHINA: AN INTERGENERATIONAL RELATIONS APPROACH

Yuan Fang and Martin King Whyte

One of the primary objectives of the Baoding survey project, as noted in chapter 1, was to determine how China’s multiple revolutions have affected the well-being of the elderly in that country, and in particular whether they have weakened the traditional obligation of grown children to provide support and assistance to their aging parents. The current chapter presents a descriptive overview of the situation of the elderly and the nature of intergenerational relations in Baoding in 1994, based upon a wide range of measures included in our parent survey. This overview will help us examine whether and how China’s revolutions have affected the urban elderly in that country. We are, of course, dealing with a “moving target,” since rapid economic and social changes have continued since the time of our survey and will likely characterize Chinese society for many decades to come. Our survey data also help to establish a baseline against which these continuing and future changes can be judged.¹ Many aspects of the well-being of China’s urban elderly reviewed in these pages will be subjected to more detailed examination in later chapters of the present volume.

Basic Demographics and Living Arrangements

As noted in chapter 1, the design for the Baoding survey called for a probability sample of residents of that city aged fifty and older. The final sample of 1002 older Baoding residents ranged in age from fifty to ninety-one. Given our lower limit of age fifty, our sample tends to be concentrated in the “young old” range—493 respondents, or 49.2%, were between ages fifty and fifty-nine. Another 371 individuals, or 37%, were between ages sixty and sixty-nine, and the remaining 138 individuals, or 13.8%, can be considered the “old old”—those aged seventy and above. Since most other studies of the elderly focus on those over age sixty, many of the statistics presented in this chapter will be broken down into these three broad age-groupings so that any distinctive features attributable to including those between ages fifty and fifty-nine will be visible.

As noted in chapter 1, an excess of males in the sampled neighborhoods in Baoding led to our sample having a higher proportion of males than most studies of older populations. Overall 52.5% of our parent respondents were men. As the figures in Table 2.1 reveal, it is particularly in the oldest ages that men are most overrepresented, constituting 64.5% of those over age seventy in our sample (see the first panel of Table 2.1). The figures presented in the lower half of Table 2.1 reveal that an overwhelming 88.4% of our respondents were married at the time of our survey, again an unusually high figure for an older population. A portion of this high level can be attributed to the fact that in our fifty to fifty-nine age group, 96.6% of the respondents are currently married and only 3% widowed. However, even among those respondents over age seventy, more than two-thirds are currently married and less than one-third have been widowed.² The figures in the bottom half of Table 2.1 show two other striking features characteristic of older Chinese populations. First, there are not any individuals who have never been married, although never-married individuals were eligible for inclusion in our sample. Traditionally, high marriage rates in China went hand in hand with very low divorce rates, and that characteristic is also apparent in our figures. Only four individuals in the entire sample gave their current marital status as divorced. A separate question determined that another thirty-five individuals had earlier marriages that ended in divorce, with most of those individuals subsequently remarrying. Even taking the latter cases into account produces a total of less than 4% of Baoding respondents over age fifty who ever experienced a divorce.

These figures on marital status produce a distinctive context of aging. The majority of even the “old old” in our sample are currently married, and the possibility of relying on a spouse may make the need to rely on grown children less pressing. Of course, it is also possible that having a spouse who due to health problems requires extensive care may add to the burdens on the partner, thus increasing the need for assistance from children.

During the 1970s state policy in China was aimed at enforcing a two child limit on urban families, and, after the launching of the “one child family” policy in 1979, that limit was reduced to one. However, as noted in chapter 1, most Baoding respondents had begun, and many of them had already completed, their fertility before these limits went fully into effect. The current number of children of our respondents ranged from zero (only 3 cases) to eight. The average number of living children for the whole sample was 3.24, and 71% of our respondents had three or more living children.³ This figure, it should be noted, is substantially below the “traditional” levels of fertility for urban China, which were in the range of five children or higher.⁴ However, despite some reduction in fertility compared to the historical norm, most of our respondents have several living children whom they might be able to rely on for support in old age.

Are these children available to provide family support? Availability would require at a minimum that a grown child lives in Baoding, although children located elsewhere could send funds to parents and visit occasionally. It is notable that of the total of 3249 living children of our Baoding respondents, fully 88% currently also live in that city. This figure signifies that most Baoding parents have more than one child available within the Baoding area (the precise average being 2.85 per interviewed parent). The bureaucratic system of job assignments and the absence of a labor market for most of the past 40 years have produced an unusual degree of immobility of grown children.⁵

How likely is it that Baoding parents will live with a child versus on their own? The figures computed from our survey differ markedly, depending upon whether we consider coresidence with any child versus with any married child. Overall, 64% of parents interviewed live with one or more child, but only 35% live with one or more married child. These figures suggest an underlying pattern in which most unmarried children continue to reside in the parental home until married, but most married children do not remain in joint residence with parents for prolonged periods. Furthermore, there does not appear to exist a strong obligation currently for at least one grown child to remain coresident with aging parents. This conclusion is reinforced by the figures presented in Table 2.2. That table shows the percentages of interviewed parents in our three age groups who coreside with at least one grown child (i.e., age eighteen and up), and with at least one married child. For the “young old,” those between ages fifty and fifty-nine, the large majority have at least one child coresiding (78%), but only a minority (32%) have a married child in the home. When we get to those parents over age seventy the picture changes. Predictably, most grown children have gone off to live elsewhere, and only 47% of those in this age group have any grown child at home. But in most of these cases (41% out of 47%) that child is a married child. These figures also indicate that a slight majority of the “old old”—those over age seventy—do not have any child living in their household.⁶

Our data yielded several other statistics of interest in regard to the living arrangements of older Baoding residents. First, the average number of members in the households of parents is 3.6, with the range from one to twelve. In our sample there were thirty-three older Baoding residents interviewed, or 3.3% of the sample, who live by themselves—mostly widowed individuals living alone. Another 28.1% of the sample involved two person households, and in the great majority of cases these are older married couples living by themselves. More than two-thirds of all households include others besides the respondent and spouse, and most often these others are children. Looked at in terms of the conventional categories of family structure, 3.3% of our respondents live alone, 61.9% live in nuclear families, another 32.3% in stem families (with one married child coresident), and only 2.5% in the traditionally favored joint families.⁷ Our results coincide with much previous research that indicates that joint families are now fairly rare, particularly in urban China.⁸ What is more novel are the further indications in our data that there is no longer a strong obligation or expectation for Chinese parents to end their days living with a married child. Living on their own seems to be the norm for aging parents in Baoding these days, by a modest margin.⁹

Is the percentage of aging Chinese parents in Baoding low or high, compared to other Chinese populations? In order to answer this question we recompute coresidence figures for parents over age sixty, since that is the age conventionally used in such computations. Of the Baoding parents who were sixty or older, 50% were living with a grown child, but only 38% were living with one or more married child. A recent compilation yields estimates from Taiwan ranging from 69 to 85% of those over sixty living with a grown child, and 56–71% living with a married child. The same compilation cites earlier surveys from urban areas of the PRC with ranges from 56–71% of individuals over age sixty living with a child, and 38–52% living with a married child.¹⁰ The 1989 survey data for Taiwan used later in the current volume yield figures of 50% of urbanites over age sixty living with a married child, and 63% among the majority Taiwanese population, in contrast to the 38% for Baoding (see Table 9.2). Thus our Baoding figures show a tendency toward slightly lower coresidence with children than found in earlier PRC surveys and levels substantially below those found in surveys on Taiwan.¹¹ The Baoding figures are also substantially lower than figures derived from rural surveys in the PRC. For example, one recent study of a village in Heilongjiang Province yielded a figure of 70% of those over age sixty living with a grown child, and 64% living with a married child.¹²

In the past when parents lived with one or more married child, that child was most likely to be a son. The strong patrilineal bias of the Chinese kinship system dictated a pattern in which daughters married out and joined other families, and only a son or sons remained. In our Baoding sample there are 253 parents (25.2%) who live with one or more married son, but only seventy-nine parents (7.9%) who live with one or more married daughter.¹³ Thus in Baoding it is still about three times as common to live with a married son as with a married daughter. Most of the parents we interviewed have several grown children, as indicated earlier, and fewer than 10% of them have no sons at all. Since we have also noted that there are very few older Baoding residents living in joint families, it turns out that having a particular constellation of grown sons and daughters available makes little difference in whether parents reside with a grown child or not. For example, while the overall percentage of parents living separately from all married children is 65%, for those with only one married son, only one married daughter, and two or more married children of one sex and at least one married child of the opposite sex, the comparable percentages are 64%, 65%, and 64%.14 It is not the case, then, that the fairly high proportion (for a Chinese population) of parents living separately from children can be attributed to not having children of the appropriate age and sex available for coresidence.

The figures reviewed so far do not tell us how willingly or unwillingly parents entered their current residential arrangements. To get at this issue, we asked parents who do not live with a grown child an open-ended question about why they reside separately, and about 18% gave answers that indicated some concern for potential conflicts with children. However, the most common reasons cited for residing separately were housing space considerations (either insufficient space in the parent’s home or the ability of children to obtain better housing on their own) or convenience factors (e.g., workplaces of children being too far away), with these reasons accounting for about 58% of the cases of separate residence. On a similar note, the most common reasons cited for residing with a grown child involve constraints on housing availability and having children not yet married, rather than any strong preference for tradition or need for care from a child. Furthermore, most parents told us that the decision about residing separately or together with one or more child was made either by the parents themselves (38.1%) or jointly with the children (46.9%), with children dictating the decision in only 7.3% of the cases (others outside the family dictated the decision in the other 7.6% of the cases). In most cases (87%) the housing that parents lived in was owned by or allocated to them or their spouse, rather than to a child.¹⁵ It might also be noted that 96% of the parents we interviewed perceived themselves or their spouses to be the head of the household. Furthermore, of those who live with one or more grown child, about 76% said that household money is managed by themselves or their spouse and 90% are satisfied or very satisfied with the amount of voice they have in family decisions.

In general the scattered pieces of evidence reviewed so far paint a fairly consistent picture. Most Baoding parents have children living with them until those children marry, and many have had one or more married child residing with them in the past.¹⁶ However, over time most coresiding married children obtain separate housing elsewhere and move out, and it appears that the decision to do so is usually a mutual one based upon pragmatic considerations, rather than the product of tensions between the generations. Similarly, those parents who do reside with one or more grown child usually do so for practical reasons, and as a result of a decision made by the parents or jointly. Usually such coresiding parents do not become highly dependent, but instead continue to play a strong role in family management. Additional evidence reviewed in chapter 4 indicates little difference between parents and children in views toward the advantages and disadvantages of coresidence. We find no sign that substantial proportions of Baoding parents who live separately would prefer to reside with a grown child but are prevented from doing so by family conflicts and the unwillingness of their children. At the same time, coresidence does not usually indicate a high level of dependence upon the grown child.

Employment and Income

The extent to which older Baoding residents need to rely on assistance from their grown children depends partly on whether they retain independent sources of income. The most common sources of continuing income for the elderly are wages from current employment and pensions. While the patterns of continued employment of older Baoding residents are analyzed in some detail in chapter 3, it is useful here to preview the general situation. Overall, 34% of our parent respondents are currently employed. Of these, 25% are still working at their original workposts and another 9% had retired but then started working for pay again. As the figures in Table 2.3 make clear, being in the paid labor force is more likely if the respondent is male and in the younger age groups. Overall, 55% of respondents under age 60 are still employed, in contrast to only 7% of those over age 70; for female respondents the corresponding figures fall to 30% and 0%.¹⁷

The other primary source of personal income for older Baoding residents is pensions, and overall 60.2% of our respondents have pension incomes. This level of pension eligibility is substantially higher than in urban Taiwan, where mostly only retirees from government agencies, the military, and large corporations qualify.¹⁸ Those receiving pensions tend, not surprisingly, to be older and predominantly male, for those over age 60. The figures in the lower panel of Table 2.3 show that for males the percentage receiving a pension increases from 23% for those under age 60, to 96% for those over age 70; for women, on the other hand, the percentage for pension receipt actually falls from 63% to 29% in the same age groups. From these figures it is clear that the group receiving pensions least often is women in the oldest age group who were primarily housewives and never or only irregularly in the labor force.¹⁹ Taking these two sources of earnings into account, overall about 15% of our Baoding respondents have neither wage nor pension income, 25% have only wage income, 51% have only pension income, and 9% have both wage and pension income. Once again, age and sex differences affect which category a respondent is likely to belong to, with the gender contrast most clear in the over seventy age group. Among the “old old” in our survey, only 5% of males have no income, while 84% have pensions and 11% have both pensions and wages; for females in this age group, fully 71% have neither wages nor pensions, and only 29% are receiving pensions (with the other categories vacant). Any conclusion about older women living in poverty must be qualified, however, by recalling that most of even these older women are still married, rather than widowed. Thus most benefit from the pensions and perhaps additional earnings enjoyed by their husbands.

The general conclusion that the great majority of Baoding parents have some earnings of their own to rely on is reinforced by examining data on cash incomes from our survey. The average personal income of parents was 323 yuan per month in 1993, based upon respondent recall. This figure is slightly lower than the average personal earnings reported by the children we interviewed, which was 363 yuan per month. (If we consider only those parents who had some income, the average parental income was 372 yuan per month, slightly higher than the average child income.) The longstanding stress on seniority in the Chinese wage system means that, with pensions generally calibrated at 70–80% of pre-retirement pay, a retired parent may still have a higher personal income than many of their children do.²⁰ There is no clear tendency for older Baoding residents to be in a weak and dependent position financially. A final statistic further reinforces this conclusion: on average the personal income of each older Baoding respondent we interviewed constituted almost half (46%) of the total income of their families.²¹

Although the economic problems of those older Baoding residents who have no personal income at all should not be minimized, the overall context revealed by these figures is a positive one. Most aging parents in Baoding have some earnings of their own, and they are not in the position of depending entirely upon financial assistance from their children (or others). Insofar as such assistance from children is rendered, it is often supplementary, rather than constituting the primary means of subsistence of the parent. Furthermore, in many cases the parent may still earn more than some or all of their children do, and in such cases the tables may be turned, with children turning to parents for financial assistance. (The extent of this reversal of dependency will be documented later in this chapter.)

As a broad generalization, the living arrangements and financial situation of the elderly in Baoding differ from their counterparts both in China’s rural villages and in Taiwan. As we have seen, fewer older Baoding residents live with a married child than is the case in these other Chinese locales. However, it is less vital financially that they do so, and coresidence is a less contentious issue, since unlike older residents of China’s villages or Taiwan, most receive pensions and some have other sources of income. The relatively low level of coresidence with grown and married children is not perceived as a problem by aging parents in Baoding.²²

Health Status and Medical Problems

We noted at the outset of this chapter that familial support for the elderly involves much more than simply financial support from children. We will discuss later a variety of other kinds of assistance that children may provide, but first it is important to consider the health status of our Baoding parent respondents. If parents are financially secure and healthy, usually they can continue to function on their own. If either economic needs or health problems become serious, they may have to turn to other sources for assistance. The traditional source of assistance with the physical and health problems of the elderly in China was, once again, grown children. As noted in chapter 1, throughout the centuries Chinese children were raised with stories of heroic efforts by children to provide comfort to parents and to cater to their daily needs and infirmities. What sort of physical shape are older Baoding residents in today, and to what extent do they need assistance due to health problems?

We asked our sample of older Baoding residents a number of questions about their health status and medical problems, and their answers can be used to present a general picture of their situation. When asked to list any ailments they were currently suffering from, almost half (49.3%) of Baoding parents did not list any problems at all. Another 10.8% said that they had one or more ailments, but none that caused them any daily inconvenience. Then there were 26.5% who said they suffered from some inconvenience as a result of health problems, and 13.3% who reported substantial inconvenience. Looked at in a different way, 26.7% of Baoding parents reported one current ailment, 15.6% reported two current ailments, and 8.4% reported three or more such ailments. We also asked our respondents what level of difficulty they might have in performing eight simple physical chores: shopping, climbing two or three flights of stairs, walking 200–300 meters, lifting and carrying a ten-kilogram object, opening by hand a tightly sealed bottle, standing in place for two hours, riding a bicycle for five kilometers, and getting on a public bus without assistance. Just over half of our respondents (56.3%) said that they could perform all of these tasks without difficulty. Another 18.1% said they would experience some difficulty performing one or more of these chores, but no major difficulty with any of them, while 15.5% said they would experience major difficulty or be unable to perform one or two of these tasks, and 10.2% said they would be unable to perform three or more of these tasks. The subjective self-evaluations of our respondents coincide fairly well with these figures. Overall, 50.4% of Baoding parents rated their health as good or very good, 33.9% as so-so, 13.4% as not so good, and 2.3% as poor. It might also be noted that 9.9% of Baoding parents reported that they had been hospitalized within the last six months.

As expected, these health indicators vary across age groups, with the “young old” in better shape than the “old old.” As the figures in Table 2.4 show, as we move from those under age sixty to those over age seventy, the percentage reporting no health problems declines from 56% to 38%, while the proportion reporting serious inconvenience as a result of health problems rises from 7% to 19%. Looked at in terms of the number of current ailments, the proportion suffering from three or more health problems rises from 4% among those under sixty to 15% among those over seventy. In terms of our index of difficulties in performing eight physical tasks, when we go from those under sixty to those over seventy, the percentages who say they can perform all the indicated tasks without difficulty falls from 69% to 34%, while the number who would have considerable difficulty performing three or more of these tasks rises from 3% to 28%. There is a weaker contrast in the same direction visible in the subjective perceptions our respondents hold about their health, with the percentage reporting good or very good health falling from 54% among those under age sixty to 50% among those over age seventy, and those reporting not so good or poor health rising from 12% to 17%. Hospitalization within the past six months also varies modestly by age group, with 8.5% of those under age sixty having been hospitalized, in contrast to 15.2% of those over age seventy.²³

Without comparable figures from other times and places, it is difficult to know what to make of these statistics. However, on balance the picture they yield of the health status of older residents of Baoding seems fairly positive. Depending upon which indicator we use, overall only 8–16% of Baoding parents report health problems serious enough to cause them major difficulties in life. Even among those over age seventy, the proportion experiencing serious health problems rises to only 15–28%. Although there is a sizable minority of Baoding parents, particularly among those over age seventy, who have health problems and infirmities that may make the care and assistance of grown children important, most seem to be healthy enough not to require regular care of this type.²⁴

We also attempted in a rough way to judge the psychological health of Baoding parents and the prevalence of signs of depression. Obviously the measures one can include in a questionnaire are less satisfactory than what could be obtained in a clinical examination. We presented each respondent with a series of statements about moods and emotional states and asked whether they often, sometimes, rarely, or never had each specific feeling. Responses to thirteen of these questions were closely related to one another, and these items were used to compute a mean depression index for each respondent.”²⁵ Judged by these individual measures and the summary scale, relatively few Baoding parents are experiencing serious levels of depression. Generally only 5–10% reported experiencing often any of the negative emotional states listed, and another 10–25% sometimes. The mean score on this depression index for all Baoding parents was 3.3 (on a scale from 1=often experienced these problems to 4=never), which indicates a predominance of “rarely” and “never” answers. It is also interesting to note that there was no overall tendency for the “old old” to report more depression symptoms than the “young old.” The correlation between age and the depression index score was -.06, which is not statistically significant. However, there was a clear gender difference, with females on average reporting more depression symptoms than males.26 Generally about 15% of Baoding mothers interviewed often experienced the various negative emotional states we used, in comparison with about 5% of fathers; at the other end of the scale, generally about 35% of the mothers reported they never experienced the various negative emotions, in comparison to about 45% of the fathers. The correlation between gender and our summary depression index was statistically significant (r=-.18, p<=.001), and among mothers there is also a statistically significant tendency for older women to be more depressed (r=-.13, p<=.01). In sum, most Baoding parents report themselves to be in quite good emotional as well as physical condition. However, women report more symptoms of depression than men, and older women are the most likely to experience emotional problems

Our parent survey contained questions about the extent to which parents currently receive four specific types of support from any source, and whether they provide others with these and one additional type. Overall, only 3.8% of parents were receiving physical assistance with such things as bathing, getting dressed, and going to the bathroom; 31.6% were receiving help with domestic chores; 24.6% were receiving financial support; and 34% were receiving material gifts—i.e., things to wear or eat. Overwhelmingly, these forms of assistance were being provided by a spouse (particularly for physical care and help with chores) or by children. Although in China the government encourages work units, neighborhood committees, and other organizations to provide assistance to the elderly, only 7.4% of Baoding parents report receiving any regular assistance from outside of the family.27 Those parents who were not receiving any of these kinds of assistance were asked who they would turn to if they needed them, and again, overwhelmingly, they mentioned various children, but rarely a work unit or other outside agency.

Family exchanges are not always in one direction, and many parents are regularly providing support to others, and particularly to their children. Again, looking at overall figures, 8.4% of parents reported helping others with physical care, 9.1% with domestic chores, 24.5% with money, 18.5% with gifts of food or clothing, and 35.3% with child care (the added category). In Table 2.5 the exchanges in both directions are examined for the three separate age groups. In general the figures in the table reveal an expected pattern. Younger parents tend to provide assistance somewhat more often, and to receive assistance somewhat less often, while with older parents these patterns are reversed. However, the differentials are fairly modest, and a majority of “old old” parents are not receiving the specific kinds of assistance inquired about, and significant minorities of those over age seventy are still providing assistance to others.

In general it appears from these figures that assistance from grown children to parents is somewhat more common than assistance from parents to those children, although childcare is an obvious exception to this generalization. These figures appear to contrast with the patterns of intergenerational exchanges found in contemporary Western societies in several ways. In general in the West parents are less likely to regularly receive assistance from grown children. And usually the balance of intergenerational exchanges is reversed, with parents helping grown children more often than they receive assistance from their children.²⁸ Although the figures for Baoding support a picture of mutual assistance between generations, the filial “tilt” is nonetheless visible, making obligations on grown children generally heavier than on aging parents.

We can examine the pattern of financial dependency versus independence for parents in greater detail by utilizing the results of follow-up questions about the average monthly amounts of money received from others and provided to others. Following the scheme proposed by Wang Mei and her colleagues, we can categorize the monetary flows involving parents into four categories.²⁹ “Independence” refers to cases in which cash is neither given nor received, “supported” involves cases where parents receive cash and do not provide any, “fostering” refers to parents who provide cash but do not receive any, and “mutual benefit” refers to instances in which cash is both received and provided. By these definitions, 56.7%) of Baoding parents are classified as independent, 19.8% as supported, 19.4% as fostering, and 4% as mutual benefit. If exact amounts of cash flowing in both directions are compared, then 22.2% of parents are net recipients, 57.1% are neither net recipients nor providers, and 20.7% are net providers of cash support. Once again, these flows vary across age groups in a predictable way, with older parents more likely to be net recipients of cash support, and less likely to be net providers of such support, than younger parents.30 These results reinforce conclusions reached earlier in this chapter. The most common situation is for Baoding parents to be relatively independent or to be involved in balanced exchanges with their children, rather than to be highly dependent. With advancing age the balance tips toward greater dependency, but only modestly.

There are other ways in which parents provide assistance and support to their grown children besides those already described. We asked Baoding respondents about the extent to which each individual child had relied on family assistance in a number of important past transitions. On average parents said 25–40%) of their children had received some or a good deal of family help with their homework, 8–13%) in getting into the desired school, 35–40% with getting their first job, 30–35% with changing jobs, 15–35% in obtaining housing, and 25–30% with finding their spouse.³¹

For those parents who live with one or more married child, we also inquired about the pattern of division of chores across generations. Most parents reported that they or their spouse do most or all of the grocery shopping, cooking, and cleaning, with children shouldering primary responsibility for these chores in only a small minority of families.³² Care for grandchildren was also said to be more often done by the grandparent/respondent and his or her spouse than by the parental generation. Laundering clothes was the one chore we inquired about for which there was a rough balance, often with each generation or even each individual washing clothes separately. The only domestic chore we asked about that children performed more than their parents was purchasing and carrying home heavy objects, such as propane gas canisters and sacks of grain. Although the particular set of questions we asked may skew the picture, in general these results suggest that grown children often benefit more than they are burdened in domestic chore exchanges when they live together with an aging parent or parents. The relationship between parental assistance to children and the provision of support from children to parents is explored further in chapter 8 in this volume.

Support from children to aging parents, as noted at the outset, takes many forms. Even if the provision of financial and physical assistance are the most essential forms for those in need, the provision of emotional and social support is also vital to parental well-being. Evidence from our survey indicates that most Baoding parents are doing quite well in terms of these kinds of support as well. The context here needs to be kept in mind: most interviewed parents have several adult children living in Baoding, even if only a minority are actually living with a married child.

Contact with non-coresident children appears to be frequent. We asked about the frequency of contacts with each grown child, and generally about 30% claimed daily or almost daily visits, and another 30% said several times a week. In only 6–8% of the cases did parents report infrequent contacts (once a year or less), and these mostly involved children who lived far away from Baoding. For the sake of comparison, we found one study that reported that 8% of non-coresident adult children in Japan saw their parents on a daily basis, and similar data for the United States showing the same was true for 16% for sons and 24% for daughters.³³ Since the average Baoding parent has about three adult children residing nearby, these figures mean that most have multiple children they see on a daily or almost-daily basis. Most parents reported visits with grown children as more frequent than with any other kind of relative or with neighbors or friends. For example, 41% of the parents interviewed said that there was no single friend or neighbor that they got together with as often as once a week.

We also asked parents a variety of questions about their relations with each child when that child was growing up and at present. The responses to these questions also indicate parent-child relationships that are primarily positive. For example, generally about 75% of parents report that arguments with any particular child were rare or never occurred, although 45–50% said that when their children had opinions that differed from those of their parents they would sometimes or usually express them.³⁴ When disputes with a child did occur, parents claimed that about 65–70% of the time the parent would prevail in the argument. Parents reported that 25–30% of their children were very obedient when young, and another 65–70% fairly obedient; fewer than 5% were characterized as even somewhat disobedient. On a similar note, 25–30% of grown children were described as currently very filial, with another 65–70% fairly filial, with under 5% characterized as somewhat or very unfilial. Elsewhere in the questionnaire, only 6.7% of parents said they did not get sufficient respect from their grown children. In contrast, 86.3% of parents felt that the respect they received was sufficient, while 7.1% said they got too much respect! These responses contrast with parental evaluations of what is happening in Chinese society generally. Two-thirds of parents claim that respect for elders has diminished in China compared with twenty years earlier, but few appear to see signs of such a deterioration in relations with their own grown children.³⁵

A variety of more specific questions about family relations yield a similarly positive picture. Parents report that they are often (64%) or sometimes (24%) consulted about important family decisions; seldom (29%) or never (43%) criticized by their grown children; always (50%) or sometimes (38%) asked for advice by children facing important decisions; and that they often (24%) or sometimes (48%) offer unsolicited advice to their children and that often (64%) or sometimes (33%) their children pay attention to such parental advice. Furthermore, 44% of the parents reported that if a married child and his or her spouse were having conflicts, the respondent might be asked to mediate. In Chinese extended families of the past, conflict between the mother-in-law and daughter-in-law was an endemic source of family tension, but only 6% of the parents whose households contain this dyad claim that this kind of conflict occurs with any regularity.³⁶ Summing things up, the overwhelming majority of Baoding parents describe themselves as either very satisfied (15.1%) or satisfied (81.5%) with the emotional support they receive from their children, with less than 5% expressing any degree of dissatisfaction.

In Table 2.6, age group variations in responses to general questions about relations with grown children are displayed. Some of the contrasts across age groups are statistically significant, while others are not. Those over age seventy are significantly less likely than younger parents to report that they are consulted about important family decisions (line one), asked for advice by their adult children (line three), or that they offer children unsolicited advice (line four), and older parents are slightly less likely to be asked to mediate the marital disputes of their children (line six). However, at the same time these older parents are significantly more likely to report that they are never criticized by their children (line two). These responses are consistent with an interpretation of parents becoming more removed from intense involvement in family decisions at older ages, but in the process being treated even more deferentially. Overall there is no significant difference across the age groups in satisfaction with the emotional support received from children (line seven), and parents over age seventy are actually slightly more likely than others to feel that they receive “too much” respect from their children (line eight).

Perhaps there is a degree of bias built into these answers, with Baoding respondents not willing to reveal the extent of family problems and dissatisfactions to outsiders—our interviewers. We have no way to independently judge the state of relationships in Baoding families.³⁷ However, there is a degree of congruency in the various kinds of information collected in our interviews. The figures presented on parental financial and health status, on parents giving and receiving various kinds of support, on the frequency of contacts with non-coresident children, and on the quality of parent-child relations, all seem consistent with one another.

Although this chapter has relied almost entirely on data from interviews with parents, rather than from our child interviews, there is a certain plausibility in this positive picture from the standpoint of grown children as well. The modal situation for such a child is to have benefited in the past and to be benefiting currently from parental assistance, to live separately but not too far away from the parents, and to have parents in fairly good physical and economic health. Such children do not face a one-sided burden of providing constant support to needy and infirm parents; instead they can provide modest supplementary support in a variety of ways while continuing to benefit from parental assistance. At a later stage in the life of the parent or for the minority of cases involving a destitute or disabled parent, the burden of support may be much greater. However, in such cases there will usually be a legacy of positive parent-child relations in the past that is likely to induce children to bear this burden. (See further discussion of this point in chapter 8.)

Conclusions

Although subsequent chapters in this volume will explore in greater detail many of the issues raised here, some preliminary conclusions are in order. In some ways the statistics reviewed here reveal sharp differences from the patterns of China’s past. In particular, older Baoding residents are more likely to live independently now, rather than with one or more grown children. However, this trend cannot be interpreted as a sign of breakdown of traditional obligations for family support. Indeed, the Baoding data provide very few signs of such a breakdown, or of any impending crisis in support for the elderly. Rather, our data provide evidence of a successful but partial transition away from the traditional system of familial support. Pensions and other public benefits have bolstered the position of the elderly without apparently weakening the obligation of children to assist their aging parents.

Of course, our survey deals with only one medium-sized city, not urban China generally. We have also noted that the situation in China’s rural areas, where about 70% of the population still resides, is different in important ways. However, in general the elderly in Baoding appear to be doing fairly well by relying on a combination of public and familial sources of support.

Most of the Baoding parents we interviewed have some source of independent income and only minor health problems. The great majority are still married. Most also have several adult children living close at hand, or at least within the city boundaries of Baoding. The average parent has frequent contacts with those grown children who are not coresident, receives a variety of forms of assistance from children, and reports quite harmonious relations with them. No exact comparisons are possible due to lack of comparable data, but it appears that Baoding parents are involved in much more extensive and frequent exchanges with their grown children than would typically be the case in Western societies. Even in selected comparisons with other Asian societies, Baoding parents seem well-connected to their grown children. Although proportionally fewer of them coreside with a married child than is the case in Taiwan, more of their grown children live nearby, and the level of contacts with such non-coresident children appears higher than in Japan. The result of all of these features is that most respondents express considerable satisfaction with their lives and families. Generally only 5–15% of Baoding parents say they face situations involving hardships—of poverty, physical incapacity, and inattentive or disrespectful grown children.

The one partial exception to this generally positive situation concerns older women. In certain respects women over 70 in our sample are more vulnerable than others—few have any independent source of income, and more of them are suffering from serious physical and emotional problems. However, in other respects these older women share the benefits of their male and younger counterparts, with frequent visits and exchanges with grown children, and with a somewhat higher likelihood of living with a married child.

Although we lack comparable surveys for Chinese cities in the past, the many revolutions China has faced in the twentieth century do not appear to have substantially undermined family support for the elderly, at least in Baoding. Indeed, we have suggested that some parts of those revolutions may have enhanced the situation for China’s elderly—for example, by keeping children close at hand and providing improved health care and pensions for most older urbanites.³⁸

Although the situation of the Baoding elderly as of the mid-1990s appears quite positive, it remains unclear what the future will hold. Several elements that bolster the current positive picture are likely to be threatened in the future. In particular, as indicated in chapter 1, in the future very few elderly in China’s large cities will have three or more grown children to turn to for support, and most will have only one. Furthermore, fewer grown children are likely to remain close at hand, as economic reforms and revived labor markets open up possibilities for jobs and residence elsewhere. The impact of the reforms on the economic situation of the urban elderly is also uncertain. Continued rapid growth may improve living standards, increase the richness of leisure activities for the elderly, and make it easier to cope with daily chores. However, state work units that have provided the bulk of urban pensions are under great pressure to “smash the iron rice bowl” of benefits to their employees, with no substitute society-wide source of such benefits yet in place. Since the mid-1990s, in many Chinese cities pensioners have complained publicly that the pensions to which they were entitled were not being paid due to the financial difficulties of their former enterprises. It seems that elderly urbanites are becoming more economically vulnerable just when the number of children available to provide assistance is declining sharply. However, it is also possible that as reforms proceed further, the need of both parents and children to rely on special help from their families may decrease, as more needs can be met through market exchanges.

For these and other reasons, the role that various forms of familial support will play in the lives of China’s elderly urbanites in the future remains unclear. However, familial support of China’s elderly has remained central in the face of much turmoil and many challenges over the past century. Therefore, we should not be too hasty in concluding that in the future this support will weaken dramatically. For the foreseeable future, grown children are likely to retain a central role in the lives of China’s elderly, a role that is much greater than has generally been the case in recent times in Western societies.

NOTES

¹ In many recent analyses, the changes in China since Mao’s death are seen as ushering in a more “normal” phase in which economic development and societal modernization common to many developing societies have an impact on family relationships. If this scenario is accurate, then debates about whether such global changes necessarily undermine the status of the elderly and require the state to increasingly replace the family as the basis for support of the elderly are clearly relevant to the Chinese case. For one effort to examine such arguments critically, see Yuan Fang, “The Role of the Chinese Elderly in Family and Society,” Beijing University Journal (Philosophy and Social Sciences), no. 4, (1987) (in Chinese). However, before analysts can judge the applicability of these debates about global changes to the Chinese case, it is necessary to know what impact the unique and tumultuous changes that China experienced in the twentieth century have had on the elderly and intergenerational relations.

² For comparison purposes we note that in the five largest cities included in the 1989 Taiwan survey that will be used in chapters 9–11 in the current volume, 61% of those over age sixty were married, compared to 81% of the same age group in our parent sample in Baoding.

³ Separate questions were asked about the number of children ever born to respondents and any spouses they had had. Those figures were more variable, ranging from zero (twenty-six cases) to 25 (two cases)! The mean number of children ever born to respondents and spouses was 3.52. Twenty-five respondents had adopted children from others, and ten respondents acknowledged giving up a child for adoption by others.

⁴ Even among those over age sixty in Baoding, fertility is slightly lower than in urban Taiwan. For this age group the mean number of living children was 3.7, compared to 4.3 for their counterparts in Taiwan’s five largest cities, and 4.8 for Taiwanese in those cities (see the figures in Table 9.2 in the current volume).

⁵ It is difficult to find exactly comparable figures for other societies. However, the results of one survey conducted in the Albany, N.Y., area in 1988 provide some context for our Baoding findings. In that survey roughly half of the adult children of parent respondents were reported by the parents to live within a half hour travel time away, a percentage much higher than the authors had expected to find. See John Logan and Glenna Spitze, Family Ties (Philadelphia: Temple University Press, 1996), 99. Nonetheless, it is a level much lower than the almost 90% of grown children of our Baoding parents who live within a roughly comparable distance (if we can assume that children living anywhere in Baoding can travel to their parents’ home in thirty minutes or less). More comparable perhaps are figures from Taiwan, where the 1989 survey used in later chapters of the current volume revealed that 70%) of the grown children of parents over age sixty living in the five largest cities on that island were living “nearby,” compared with 84% of the grown children of Baoding parents in this age range (see Table 9.2 in the present volume).

⁶ Widowhood increases the likelihood of living with at least one married child, particularly for women. About 53%) of widowed parents live with one or more married child. If the parent is widowed and over age seventy, the figure rises to 61%, and for “old old” widowed females the figure is 65% (as compared to 56% of males in this situation). These differentials correspond in a rough way to differences in expressed attitudes. Overall, 71% of parents agreed or strongly agreed with a statement that if they remained healthy, an old person should remain independent, rather than relying on grown children. However, at the same time 75% agreed with a statement that a widowed older person should not continue living independently. Note, however, that even among the oldest widowed females, the actual level of coresidence with a married child (65%) is below the level of general support for coresidence in this situation (75%).

7 Some of the nuclear families are incomplete nuclear units, which refers to married couples with no coresiding unmarried child, a widowed parent living with an unmarried child, etc. A joint family refers to a parent or parents living with two or more married children. At the extreme end of the range, there were three parents in our sample who were living in a joint arrangement with three or more married children.

⁸ See, for example, Martin K. Whyte and William L. Parish, Urban Life in Contemporary China (Chicago: University of Chicago Press, 1984). There is some dispute in the Western literature about how common or rare joint families were in earlier times. Many studies have argued that although the joint family was the traditional ideal, only a few relatively rich families could hope to keep such a complex unit together, and that most people even in the past lived in nuclear or stem families. However, Arthur Wolf has provided evidence that under some circumstances, joint families could be widespread among the ordinary population. See Arthur Wolf, “Chinese Family Size: A Myth Revitalized,” in The Chinese Family and its Ritual Behavior, ed. Hsieh Jih-chang and Chuang Ying-chang (Taipei: Academia Sinica, 1985).

⁹ There is only a weak relationship between the distribution of family structures and parental age group. The percentage living in nuclear or subnuclear units goes from 68 to 64 and then to 59 as we move from those under age sixty to those seventy and older, while the percentage of stem families goes from thirty to thirty-three and then to thirty-eight in the same comparison (with joint families constituting 2%, 3%, and 3% in the three age groups).

¹⁰ See John R. Logan, Fuqin Bian, and Yanjie Bian, “Tradition and Change in the Urban Chinese Family: The Case of Living Arrangements,” unpublished paper, Table 1.

¹¹ The levels of coresidence between parents and grown children shown in all such Chinese surveys are considerably higher than found in contemporary Western societies. For example, the 1988 survey in the Albany area cited earlier yielded a figure of 23% of parents over age sixty who had an adult child coresiding, with national surveys in the United States yielding figures in the same range (well under 30%, compared to the 50% in our Baoding survey). See Logan and Spitze, Family Ties, 42–43, 205. Still, these figures from surveys on Chinese populations help to dispel the stereotype that virtually all elderly Chinese live with a grown child.

¹² Figures computed from Yunxiang Yan, “Elderly Support and the Crisis of Filial Piety,” chapter 7 from Private Life under Socialism: Individuality and Family Change in a Chinese Village, 1949–1999, Table 9 (Stanford: Stanford University Press, forthcoming).

¹³ These figures involve a small amount of double-counting, since there were seven parents who live in unusual arrangements with both one or more married son and one or more married daughter.

¹⁴ Only those parents with no children at all and with two daughters but no sons were substantially more likely to live alone (100% and 81%, respectively). It should be noted, once again, that a good portion of those parents who do not live with any married child still have one or more unmarried child at home.

¹⁵ It is important to understand that there is still little in the way of a private housing market visible in Baoding. Overall, only 6% of parents live in privately owned housing (5%) owned by themselves and 1% owned by others), with the remainder living in public housing allocated by work units (78%) or by city housing offices or others (16%). A slightly higher proportion of the children we interviewed—16%—live in privately owned housing.

¹⁶ Forty-five percent of our parent respondents told us they had lived with another married child in the past, and some parents had lived with several. A not unusual pattern is for an older son to marry and temporarily reside with the parents and then move out when independent housing becomes available or when a second son is about to marry. However, our data indicate that in a good many instances the sequence does not end with the parents residing with the youngest married son, but instead living independently of any children.

¹⁷ In the 1950s the Chinese government, following the Soviet precedent, enshrined a gender differential in retirement from state units by mandating sixty as the retirement age for men, fifty-five for women in mental labor jobs, and fifty for women in manual labor jobs.

¹⁸ Among Baoding parents over age sixty, 77%) had pension income, compared to 27% among urbanites in Taiwan in 1989, and only 13% for Taiwanese, who were much less likely than Mainlanders to retire from covered employers (see Table 9. 1 in the present volume). In more recent years pension reform efforts in Taiwan have spread pension eligibility somewhat more broadly.

¹⁹ Since the end of the 1950s most women under age fifty have been regularly employed for wages, but the older women in our sample established their families prior to this mobilization. Even when employed, women are more likely than men to work in smaller firms in the collective sector, rather than in state firms. Small collective-sector firms less often have pension systems, while pensions have long been mandatory in state firms.

²⁰ Adjustments have been made in the pensions of many work units to take inflation into account. As a result, the cash value of the average pension among our respondents was about 2 1/2 times the (unadjusted) level of their final pay prior to retirement (again, with all figures based on respondent recall). However, see chapter 3 for a discussion of the problem of income “slippage” faced by retirees.

²¹ More specific comparisons between parent and child incomes are possible using the Baoding survey data. For the 731 pairs of parents and children interviewed in the survey, the recollected parent monthly income in 1993 was higher than the interviewed child in 47.7%) of the cases, exactly the same in 6.5% of the cases, and lower than the child in 45.8% of the cases. On a more subjective question to parents about their economic wellbeing compared to each one of their grown children, parents reported that they had a better living standard than 39% of their children, about the same in comparison with 32.9% of their children, and a worse standard than 28.1% of their children. In general, all of the figures available convey a story of parents on average doing about as well as their children, or slightly better.

22 For more on the comparison with Taiwan, and particularly on the scarcity of pensions for the elderly on that island, consult chapter 9. The struggles and tensions surrounding the issue of whether the rural elderly will live with their grown children are vividly described in Yunxiang Yan, “Elderly Support and the Crisis of Filial Piety,” op. cit. See also Haiou Yang and David Chandler, “Intergenerational Relations: Grievances of the Elderly in Rural China,” Journal of Comparative Family Studies 23 (1992): 431–53; Guo Yuhua, “The Aged and the Young: A Case Study of Disputes about Supporting the Aged in a Village in North China,” unpublished paper based upon fieldwork in a village in Hebei; Qin Zhaoxiong, “The Process of Change in the Chinese Rural Family: Government Policy and Confucianism,” unpublished paper based upon fieldwork in a village in Hubei.

²³ Particularly at the older ages, the figures in Table 2.4 conceal significant gender differences, with women on average reporting more health problems than men. For example, 46.1%) of men over age seventy reported no current ailments, but this was the case for only 24.5% of women over seventy. Similarly, 10.1% of men over age seventy reported three or more current ailments, while this was the case for 24.5% of women over seventy. Not surprisingly, an even clearer difference emerged in regard to the difficulty of performing physical tasks, with 49% of men over age seventy reporting that they could perform all eight indicated tasks without difficulty, but only 6% of women making the same claim. Along the same lines, about 9% of men over age seventy reported their health status as not so good or poor, but this was the case for more than 30% of women in that age group. Curiously, despite these differences, women in all age groups had slightly lower rates than men of reported hospitalization within the last six months.

²⁴ Again these generalizations apply less well to elderly women in Baoding. Among women over seventy as judged by the same standards used in the text, 20–40% have health problems or infirmities serious enough to interfere with their lives in major ways.

²⁵ The thirteen statements were: I worry about little things; I have no appetite; I can’t concentrate; I feel happy (reversed); I feel lonely; I feel depressed; I enjoy life (reversed); I feel my life is a failure; I feel disliked by others; I feel full of energy (reversed); I don’t sleep well; I can’t find the energy to get things done; and I feel everything is an effort. The reliability coefficient alpha for this scale was a very robust 0.85. Given the wording of these questions, a high score actually indicates the relative absence of depression symptoms.

²⁶ Most research in Western societies has also found higher reported rates of psychological distress for women than for men, with substantial debate about the reasons for this disparity. See, for example, the discussion in R. Kessler and J. McLeod, “Sex Differences in Vulnerability to Undesirable Life Events,” American Sociological Review 49 (1984): 620–31; Walter Gove, “Gender Differences in Mental and Physical Illness: The Effects of Fixed Roles and Nurturant Roles,” Social Science and Medicine 19 (1984): 77–91.

²⁷ The reader should keep in mind, however, that most respondents are receiving pensions from their former work units, as well as other public benefits not listed here, such as medical insurance. Obviously these forms of support for the elderly are not being considered by respondents in answering our questions about receiving support from others. In Part III of this volume, and particularly in chapter 5, intergenerational exchanges are analyzed in great detail, including a comparison of the reports of children and their parents for the same exchanges.

²⁸ For example, in the study of intergenerational relations in Albany cited earlier, the authors conclude that parents receive help from children “somewhat rarely,” with only 15% or less of parents reporting specific kinds of assistance from any child, and only 27% reporting that they received any of several kinds of assistance. On the other hand, 58% of parents reported providing some assistance to one or more children. See Logan and Spitze, Family Ties, 32–33.

²⁹ Wang Mei and Xia Chuanling, “Some Tentative Analysis on Family Support of the Elderly in Contemporary China,” China Population Science, No. 4, 1994 (in Chinese).

³⁰ To be specific, among those parents under age sixty, 25.3% provide cash support to others (almost always children) and 13.2% receive such support; among those over age seventy, 11.5% provide cash support and 29.8% receive it.

³¹ These figures include a range of percentages because our parent questionnaire had a section asking separately about exchanges with each of their children, with the responses differing somewhat from child to child. Other research reinforces the picture of parents and other relatives playing an important role in the employment of their children in both the socialist and reform eras. See Yanjie Bian, “Guanxi and the Allocation of Urban Jobs in China,” China Quarterly, No. 140, (1994) 971–99.

³² Specifically, for grocery shopping 74.8% of parents reported that they or their spouse is primarily or solely responsible, while the children are primarily or solely responsible in only 13.7% of the families. For cooking the comparable percentages are 71.4% and 13.2%, and for cleaning 62.8% and 16.1%. (These percentages do not add to 100 because the category “each generation equally” is omitted, and because in some cases other family members perform these chores.)

³³ See Larry L. Bumpass, “A Comparative Analysis of Coresidence and Contact with Parents in Japan and the United States,” in Tradition and Change in the Asian Family, ed. Lee-Jay Cho and Moto Yada (Honolulu: East-West Center, 1994), 229. If we include adult children who coreside with their parents then the Japanese figures for sons are much higher than those for the United States—48% of sons in daily contact (and 22% of daughters).

³⁴ The format of these questions involved asking a respondent a series of questions about their relations with child one, then repeating the same questions with regard to child two, and so forth through all of their grown children. The figures presented in the text are rough averages of the percentage totals for the first five children. (The small Ns at higher parities make the percentages unstable.)

³⁵ Note that the reference point for this subjective comparison is the year 1974, which is toward the end of the “Cultural Revolution decade.” That period is usually seen as characterized by low respect for all authority, including parental authority. Nonetheless, most parents (as well as most of their children, who were asked a parallel question) perceive a decline in respect for the elderly since that time.

36 Traditional family morality dictated that in the event of such a conflict, the son was supposed to side with his mother and against his wife. Although such conflicts are described as uncommon today, when they do occur our respondents claim that the most common approach is for the son to remain neutral (58%), with siding with the mother only slightly more common than siding with the wife (25% versus 17%).

³⁷ We do have responses in the child questionnaires to many of the same questions—responses not summarized here. Those responses in general yield a similarly positive picture about family relationships. However, since the children we interviewed might also be affected by a desire to minimize the extent of family problems in front of strangers, this corroboration cannot be viewed as an independent check.

³⁸ The contention that post-1949 changes have on balance benefited more than undermined China’s elderly is not new. The most extensive earlier Western analysis of this topic came to similar conclusions. See Deborah Davis-Friedmann, Long Lives: Chinese Elderly and the Communist Revolution, rev. ed. (Stanford: Stanford University Press, 1991).

Several other measures were also included in our analyses as control variables. Much theorizing and research have suggested that the needs of the elderly, or their lack of material resources or ability to maintain a normal life, are major determinants of old-age support.²² By including variables measuring the physical and financial status of the parents, we will be able to test whether the effect of parental investment on old-age support persists, even after variables representing parental needs are controlled. The age of the parent interviewed was included. A scale measuring the functional status of parent interviewed was also used to measure the physical frailty of parents directly. That measure was derived from eight questions in the parent questionnaire that describe daily activities that require a moderate amount of physical strength. The coding scheme is as follows: 0—if the parent-interviewed had no difficulty in doing any of these eight physical activities; 1—if the parent did some of these activities with a little or some difficulty; 2—if the parent reported it was “very difficult” or “can’t do it” for one or two such activities; and 3—if the respondent reported “very difficult” or “can’t do it” for three or more such activities.”²³ (See the age distribution of our parent sample on this scale in Table 2.4 in chapter 2.) Thus, the higher the value, the frailer the parent.

One additional potential threat to filial obligations involves China’s increasingly strict family planning policy. As 43 Therefore, very few of the parents we interviewed had only one child. However, if present trends continue urban China will eventually consist almost entirely of parents with only one child. The potential implications of this rapid and dramatic shift are unclear but potentially monumental. In barely a generation, China has gone from a situation in which aging parents (such as those in the Baoding survey—see the discussion in 44 In sum, recent changes in demography and childrearing practices may be undermining filial obligations even more than the political and cultural turbulence described above. However, any examination of the impact of the one child policy on filial obligations is of necessity a task for future research.