Contemporary trends in population ageing and urbanisation in the developing world imply that the extensive out-migration of young people from rural areas coincides with, and is likely to exacerbate, a rise in the older share of the rural population. This paper examines the impact of migration on vulnerability at older ages by drawing on the results of anthropological and demographic field studies in three Indonesian communities. The methodology for identifying vulnerable older people has a progressively sharper focus, beginning first with important differences between the communities, then examining variations by socio-economic strata, and finally the variability of older people's family networks. Comparative analysis indicates considerable heterogeneity in past and present migration patterns, both within and between villages. The migrants' contributions are a normal and important component of older people's support, often in combination with those of local family members. Higher status families are commonly able to reinforce their position by making better use of migration opportunities than the less advantaged. Although family networks in the poorer strata may effect some redistribution of the children's incomes, their social networks are smaller and insufficient to overcome their marked disadvantages. Vulnerability thus arises where several factors, including migration histories, result in unusually small networks, and when the migrations are within rural areas.

Background/Aims: Cell culture studies suggest that phytoestrogens, abundant in soy products such as tempe and tofu, could protect against cognitive decline. Paradoxically, the Honolulu Asia Aging Study reported an increased risk for cognitive impairment and other dementia markers with high tofu (soybean curd) intake. Methods: A cross-sectional study was carried out in 2 rural sites (Borobudur and Sumedang) and 1 urban site (Jakarta) among mainly Javanese and Sundanese elderly (n = 719, 52–98 years of age). Memory was measured using a word learning test sensitive to dementia and soy consumption was assessed using Food Frequency Questionnaire items. Results: High tofu consumption was associated with worse memory (β = –0.18, p < 0.01, 95% CI = –0.34 to –0.06), while high tempe consumption (a fermented whole soybean product) was independently related to better memory (β = 0.12, p < 0.05, 95% CI = 0.00–0.28), particularly in participants over 68 years of age. Fruit consumption also had an independent positive association. The analyses were controlled for age, sex, education, site and intake of other foods. Conclusion: The results for tofu consumption as a risk factor for low memory function may tie in with the Honolulu Asia Aging Study data. It is unclear whether these negative associations could be attributed to potential toxins or to its phytoestrogen levels. Estrogen (through which receptors phytoestrogens can exert effects) was found to increase dementia risk in women over 65 years of age. Tempe contains high levels of phytoestrogens, but (due to fermentation) also exhibits high folate levels which may exert protective effects. Future studies should validate these findings and investigate potential mechanisms.

The demographic study of child supply has long concentrated on the implications of reproductive excess, rather than a lack of children. In recent years attention to population aging has begun to redress this emphasis, but even in aging research the comparative study of childlessness has a low profile. Data collected as part of anthropological and demographic research on aging in Indonesia are used to question current assumptions and to introduce issues and concepts that shed new light on current levels and experiences of childlessness. In our East Javanese study community 25 percent of the elderly have no living children, and another 15 percent have one child. Provincial and national data indicate that these findings are part of a wider pattern, corroborated by historical evidence from Indonesia, Europe, and populations elsewhere in the world. Analysis of the East Javanese data shows that childlessness is a composite category. Demographic childlessness occurs where a combination of proximate determinants (nuptiality, mortality, primary and pathological sterility) leads to no childbearing and child survival. De facto childlessness arises where there is a lack of support from any children. Actual childlessness aggregates demographic and de facto childlessness, net of adoption or remarriage where these provide alternative access to children. Analysis also takes into account the practices of patronage, charity, and kin support to assess the implications of childlessness in old age where state support is lacking.

Family networks are widely assumed to be a key source of support for older people in Indonesia and Southeast Asia more generally, although empirical study of their composition and functioning is in its infancy. This paper draws on ethnographic and survey data collected in longitudinal research of ageing in three rural Indonesian communities, in order to identify demographic and social factors limiting the size of elders' networks. Gaps in networks commonly emerge as a result of childlessness, migration and alienation, but their implications for older people's vulnerability are shaped by socio-economic status, reputation and cultural norms.