Evolution in intergenerational exchanges between elderly people and their grandchildren in Taiwan; data from a multiple round cross-sectional study from 1993 to 2007

Together with low fertility rates and a longer life expectancy, the rapid ageing of populations is a global phenomenon which requires a research and policy response concerning the potential consequences of this important demographic evolution [1, 2]. The "generational contract" is the most important and contentious dimension of contemporary welfare systems for an ageing society. The key issue presented by population ageing is "to protect the old and invest in the young while keeping a balance between financial sustainability and the principles of social justice and fairness" [3]. However, most of the discussion focuses on its public (state and market provided) dimension, namely retirement pension systems and health and long-term care policies [4]. The importance of the private (family) dimension, which refers to the transfer of money and help between generations in the family, as well as the interaction between the public and private dimensions for an intergenerational contract have not been discussed sufficiently.

The social evolution and changing patterns in intergenerational transfers are multi-dimensional phenomena resulting from complicated factors. Instead of describing all the changes in intergenerational transfer, we focused on a single indicator, namely "child care provided by the grandparent" as the core component demonstrating the evolution in the intergenerational contract. Child care provided by their grandparents covers various contexts, both physical and mental. Some clear examples of physical care include taking grandchildren to school and picking them up again, or cooking and doing the laundry for them; mental support includes taking an active interest in their life and reading books for them. Typical and common arrangements for child care provided by grandparents are baby-sitting over the weekend or during the evening, looking after children when their parents are at work, or taking care of grandchildren under other circumstances on a regular or irregular basis [5]. Studies in the United States focusing on the intensity of grandchild care provided by grandparents during the child's first three years found that grandparents provided child care more sporadically when mothers were relatively young and worked nonstandard hours but provided extended full-time grandparent care for mothers with more extensive full-time employment [6]. These both reflect the parents' need for their own parents' assistance in taking care of grandchildren in the changing society.

The role of grandparent was pointed out to be central to the model of intergenerational solidarity, and taking care of their grandchildren has been identified as a particularly important form of multigenerational family support [7, 8]. Theoretically, exemplary factors affecting intergenerational family transfer include structural factors such as demographic structure of families and labor-force structure, institutional factors like family and social security policies as well as cultural factors such as family and gender value [9]. It appears that younger, healthier elders with sufficient resources and time are more willing to take care of their grandchildren [3, 10]. Further, previous studies not only suggested that grandparents' provision of grandchild care was driven by the availability and willingness of grandparents and the needs of parents and grandchildren [11], but also described the influence of grandparents' socio-demographic characteristic and availability, including partnership status and co-residence with their children, on elders' behavior in taking care of grandchildren [12‐15]. In addition, studies comparing European countries have found the institutional social welfare system has a significant impact on grandparents' provision of grandchild care [3, 16]. However, the changing pattern and the interaction between social welfare systems (public dimension) and intergenerational transfer concerning grandparents' provision of child care (private dimension) has not yet been evaluated in Asian countries.

Taiwan is a typical Chinese society which is strongly influenced by the traditional value of filial piety. Individuals are brought up to take the responsibility of extending the family generations and to have frequent intergenerational transfers. The culture not only emphasizes the responsibility of adult children to take care of their elders, but also addresses the old people's wish and willingness to have and to take care of grandchildren [17, 18]. With a decreasing fertility rate (from 1.81 in 1990 to 1.68 in 2000 and then to 1.1 children per woman in 2006) and increasing life expectancy (from 71.6 years for men and 77.6 years for women in 1993 to 75.09 years for men and 81.90 years for women in 2007), ageing is an emerging issue - in the near future (2026) in Taiwan the old-age population will be over 20% [19]. Combined with the social evolution brought about by rapid industrialization and urbanization during the 20th century, the pattern of intergenerational transfer in Taiwan is changing. For example, fewer and fewer elders live with their children and more elders hold the idea that they should rely on themselves for their future life [19]. Moreover, the popular requirement and the recognition of the government's responsibility to offer welfare systems, including universal health coverage for the people and minority groups, also became apparent during this period [20]. However, the pattern of elders' behavior in taking care of grandchildren was poorly documented concerning these social changes. The possible effect of the social welfare system on it was also lacking. The aim of this study is to investigate the changing pattern of grandchild care provided by grandparents in Taiwan. Our hypothesis is that society has evolved over time in that elderly adults take more care of their grandchildren than before. To do so we used the data from the Study of Health and Living Status of the Middle-Aged and Elderly in Taiwan.

Our study results show the changing pattern of elders' behavior in taking care of grandchildren from 1993 to 2007. Over the given period the provision of grandchild care increased, especially for elders in their 60s and 70s. In detail, in 1999 elders in their sixties provided much more child care than those in 1993, and elders in their seventies provided much more child care in 2007 than before. Elders' age, gender, partnership status, co-residence with their children, employment status and self-reported health status were significantly related to their provision of grandchild care.

With industrialization and urbanization, the female labor force participation rate in Taiwan increased from 36.4% in 1960 to 49.6% in 2010 [22]. In the mean time, illiterate elders in Taiwan gradually became less frequent and more old people lived in urban than in rural areas. Moreover, more elders lived without partners and fewer and fewer lived with their children. Traditionally, the ideal realization of filial piety for the elders was to live in a three-generational household or live with adult children. However, the emerging nuclearization of both young and old generations is producing complicated intergenerational relations. With urbanization, more people worked and lived in cities. They might remain in the city after their retirement for its familiar living environment, better accessibility to medical care, and to live closer to their children so as to have closer contact between generations, including providing care for the children of their children. It might also be the reflection of elders' awareness of their future needs for care and help. Although the government will take the responsibility of the care of old people when they are completely incapable for taking care of themselves and lack sufficient resources, family members are the first priority for being care-givers at home. In order to receive help from their children in return, elders provided grandchild care even though it was more difficult to live with their children.

Furthermore, older adults without a job provided more grandchild care than elders with a full time job in our study result. However, more older adults had part time job in 1999 than elders in 1993 and 2007. This might result from the better health of the older adults with a part time job than those with no job in 1999 following Taiwan's implementation of National Health Insurance (NHI) in 1995.

The health care system is one of the important social welfare policies, especially for elderly people. Health care in Taiwan is administrated by the Department of Health, which introduced National Health Insurance on 1st March, 1995. Before NHI, there were several health insurance systems such as health insurance for government officers, farmers and laborers, but only 57% of the population was covered in 1994 [23]. With strong political backing, NHI was introduced in 1995 and the coverage rate then rapidly reached 97.62% for females and 94.11% for males in 1999 and then 99% by the end of 2004 [24]. As a single-payer compulsory social insurance plan, mainly financed through premiums based on a payroll tax, the insurance fee of NHI was controlled by the government and remained relatively low priced. As a compulsory insurance with a low insurance premium, NHI provided a possibility of an affordable non-employer based health insurance and a safe health environment for elders' future. This was supported by the old people's great satisfaction and feeling of security toward NHI (Table 1). With previous findings that older adults were more likely to work less or retire after receiving access to non-employer based insurance [24, 25], the implementation of NHI and its comprehensive coverage rate might give the elderly more leeway to make their retirement decision and work-family balance. Therefore, elders might tend to retire to improve their family life, including taking care of their grandchildren. The significantly decreased elderly employment rate from 1993 to 1999 might also be supporting evidence for this explanation. Also, the result that mainlanders provided significantly more grandchild care might also support the argument. The term 'mainlanders' refers those who migrated with the Kuomintang (KMT) to the island of Taiwan from mainland China between 1945 and 1950, following the Chinese civil war. Because these elders had comparatively less family support in Taiwan, they had to work more for their future life security. Therefore, they might have felt less necessity to work after the implementation of NHI and so have a greater chance to devote more time to their family.

There was an interesting finding that the self-reported health of elders became worse after the implementation of NHI in 1999 and then became better in 2007. A possible explanation is that in order to save money elders might not have had a health examination before the implementation of NHI, but elderly people knew their real health status after the NHI's provision of a health check-up. Therefore, elders' self-reported health status in 1999 was worse than 1993. But then with the increasing understanding of healthier life styles and the improvement in medical techniques, the elder's self-reported health status became better in 2007.

There were some limitations in this study. First, our finding can only be considered an association rather than a cause because the information on elders' individual characteristics and their behavior in taking care of grandchildren was cross-sectional. Second, the age of the grandchildren was not specifically considered due to the lack of information in this study. Third, we believe we underestimated elders' provision of grandchild care in Taiwan for the high "reporting threshold"; those that stay with grandchildren for only part of a day might not be considered in Chinese culture as taking care of grandchildren. Fourth, the causal relationship between work and elder's behavior of taking care of their grandchildren is not clear.

In conclusion, our study result illustrates the increasing prevalence of grandchild care provided by grandparents over time in Taiwan. Our finding does, nonetheless, suggest a complex interaction between governmental welfare systems and intergenerational family support in shaping the work-family life balance for older adults. We suggest that the interaction between population policies and those of social welfare, including health care and childcare policies, should be carefully evaluated. We would also encourage policy makers to widen their thinking concerning any particular social welfare policy, linking its specific purpose with other, surrounding, functions. Further studies are needed to evaluate the change of social welfare policies and its effect on intergenerational transfer in an ageing society.

Feng-jen Tsai holds a PhD in public health and an LLM degree in law. She is a lawyer in Taiwan and a postdoctoral fellow in the Institute of Social and Preventive Medicine University of Geneva.

Sandrine Motamed is MD, MPH. She is lecturer at the University of Geneva (Medical Faculty), chief resident at the University Hospital of Geneva and adjunct associate professor of University of Hokkaido, Japan.

Nadia Elia is MD, MSc Epidemiology from the Institute of Social and Preventive Medicine, University of Geneva.

Andre C Rougemont, MD, MPH is full professor and, since 1993, the Director of the Institute of Social and Preventive Medicine, University of Geneva.