References Tables


1. The medium fertility assumes that the rural total fertility rate would gradually decrease from 2.65 observed in late 1980s to 2.23 in 2000, and reach the replacement level of 2.15 by the middle of the next century. The urban sector would have a total fertility rate of 1.64 in year 2000 and become 1.70 in 2050 [47]. These assumptions made by us 10 years ago seem quite plausible based on today's information. The latest official figure of Chinese TFR in 1998 is 1.84 (People's Daily, Oct. 12, 1999). But we believe that it contains substantial underestimation. Our estimates are that current TFR in rural areas is about 2.2, and the urban TFR is about 1.6 (see [53], for a detailed discussion). We estimated that the weighted average of TFR for whole China is about 2.0 today (with an estimated 60% and 40% rural and urban population respectively). As stated above, we assumed that TFR in rural and urban areas of China in 2050 would be 2.15 and 1.7 respectively. The weighted average for the whole country would be 1.8 in 2050 (with assumed weights of 20% and 80% of rural and urban sectors respectively).
2. These projected figures are all under the medium fertility assumption, which assumes that rural fertility will gradually reach the level of replacement by the middle of next century, and the urban sector will have a total fertility rate of 1.7 in year 2000 and remain constant afterwards.

Life expectancies at birth under the medium and low mortality assumptions are listed below:

Year   Medium mortality Low mortality
Male female Both sexes Male female Both sexes
2000 Rural 69.0 72.0 70.5 71.0 74.0 72.5
  Urban 72.0 76.0 74.0 74.0 78.0 76.0
  Total 70.5 74.0 72.3 72.5 76.0 74.3
2050 Rural 75.0 77.5 76.3 79.4 85.6 82.5
  Urban 78.0 81.0 79.5 81.9 89.1 85.5
  Total 77.4 80.3 78.8 81.4 88.4 84.9


3. Given the evidence that urbanization is underway and the likelihood that substantial rural-urban fertility differences will persist, the model developed by Zeng and Vaupel [47] incorporates the dis-aggregation of the population according to the rural-urban dichotomy and a net migration flow from rural to urban areas by single year of age. The dynamics of the model are based on the calculation procedures of multiregional population projection ([31], [32], [34], [15]). The U.N. 1998 revision follows a classical approach of population projection with 5-year age classification without consideration of rural-urban differentials and dynamics ([28], [29]).
4. The United Nations 1998 projections do not include low mortality variant for China and any other countries, so that no comparisons under the low mortality assumption between our and UN's projections for China and between China and other countries are available.
5. The U.N. Population Division did not make population projections classified by rural and urban sectors. We, therefore, are not able to make comparison to the U.N. projection for the rural-urban differentials of population aging.
6. The 1998 healthy longevity survey was conducted in 22 provinces: Liaoning, Jilin, Heilongjiang, Hebei, Beijing, Tianjing, Shanxi, Shaanxi, Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Shangdong, Henan, Hubei, Hunan, Guangdong, Guangxi, Sichuan, Chongqing, covering 985 million people, 85.3 percent of the total population in China. The survey interviewed 9,073 oldest old individuals age 80 and above (see [58], for a more detailed description of the survey and the data quality evaluation).
7. The Chinese censuses define the nursing homes or other kind of long-term care institutions for elderly, as well as other households consisting of young or middle age persons, who have no familial relationships, as "Ji Ti Hu". "Ji Ti Hu" can be translated into English as "collective households" directly from the Chinese words, or "institutional households" based on its substantive meaning for elderly. We use the term of "institutional households" in this paper dealing with elderly.
8. It is almost impossible to compare the rural and urban differentials between 1990 and 1982, due to large changes in the rural-urban administrative boundaries in 1990, which included substantial portion of rural people into the urban boundary and did not reflect the truth of urbanization [20].
9. Crude divorce rates were more than doubled between 1982 and 1995, and there is no doubt that divorce rates will continue to increase in China [57]. Assuming that the urban Chinese divorce propensity would increase from 10 percent in 1990 to 40 percent in 2050 was an extrapolation based on the observed trend between 1982 and 1995, but it may not necessarily be true in the future since the rapid increase in divorce rates observed in 1980s and 1990s in China may not always keep the same momentum. Nevertheless, we present this assumption here as a "scenario". Furthermore, a sensitivity analysis [55, p77] has demonstrated that different assumptions about the future increase of Chinese divorce propensity by 100, 200, 300, 400 percent from 1990 to 2050 are not sensitive to (only slightly increase) proportion of elderly living alone, which is the major focus here.


References Tables

logo70.gif (2450 bytes)

Family Dynamics of 63 Million (in 1990) to more than 330 Million (in 2050) Elders in China
Zeng Yi, Linda George
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871