2 Extremely Rapid Process of Population Aging
2.1 Fastest Increase in Proportion of Elderly Persons
Previous studies (e.g. , , , , , , ) show that, although the proportion of elderly age 65 and above of the Chinese population is not very high now (5.6% in 1990), the speed of population aging will be extremely fast in the first half of the next century. Under medium fertility [Note 1] and medium mortality [Note 2] assumptions, the Chinese elderly age 65 and older will account for 15.8 and 23.1 percent of the total population by 2030 and 2050, respectively (, ). The 1998 projections by the United Nations under medium fertility and medium mortality assumptions show that the Chinese elderly age 65 and older would comprise for 15.7 and 22.6 percent of the total population in 2030 and 2050, respectively [29, p273]. The medium variant of the U.N 1998 projection is almost the same as our medium projection 10 years ago, performed independently and based on substantially different methodologies and base populations in different years [Note 3]. Such consistency confirms that extremely rapid population aging in China in the first half of next century is definite - the proportion of elderly population age 65+ in 2050 will be more than four times as high as in 1990. The annual rate of increase in the proportion of the elderly population between 1990 and 2050 is 2.3 percent.
In European societies, the aging transition has been spread over one century or more. In China, however, this change will take place within a few decades and reach more or less the same level of population aging as in most of the developed countries by the middle of next century. The proportion of the elderly in China will increase much faster than in almost all other countries in the world. It will take about 20 years for the elderly population to increase from 10 percent to 20 percent in China (2017-2037), compared to 23 years in Japan (1984-2007), 61 years in Germany (1951-2012), 64 years in Sweden (1947-2011), and 57 years in the United States (1971-2028) . Japan is regarded as a country with very rapid population aging, but the aging process of the Chinese population will be even faster than that for Japan . Table 1 presents the percents of elderly persons age 65 and above in 1990, 2030, and 2050 in selected countries. Figure 1 shows the average annual growth rates of the proportions of elderly between 1990 and 2050 in China and selected developing and developed countries with large population sizes. By the middle of the next century, the proportion of elderly persons in China will be higher than in the U.S. by 0.9 percentage points, and the average annual increase between 1970 and 2050 in China is 2.6 times as high as in the United States. The anticipated proportion of elderly population in China in 2050 is somewhat lower than in Canada, France, and U.K., and substantially lower than in Germany, Italy and Japan. But the annual increase of the percent of elderly population between 1990 and 2050 in China will be much higher than in those European countries, and 44 percent higher than in Japan.
It is interesting to note that China is not alone with respect to extremely rapid population aging among developing countries. The proportion of elderly in Korea will climb to a higher level with a larger annual increase rate than in China. Mexico and India, two developing countries with large population sizes, also will undergo very rapid population aging at annual increase rates of 2.6 and 2.1 percent, although their proportions of elderly in 2050 will be substantially lower than in China. The annual increases in the proportion of the elderly between 1990 and 2050 in China, India, Korea and Mexico are all much higher than in European and North American countries. This fact deserves serious attention, not only in those developing countries, but also from international organisations and developed countries as well (, , ).
2.2 Huge Numbers of Elderly Persons
The very large size of the elderly population is another unique characteristic of population aging in China. In 1990, there were 63 million elderly persons age 65 and over. By the years 2030 and 2050, there will be 232 million and 331 million elderly people in China respectively under the medium mortality assumption, based on our projection (, ). The most recent revision of the United Nations population projection forecasted that there will be 234.5 million and 333.6 million elders in China in 2030 and 2050, respectively, under their medium mortality assumption [29, p273]. Again, the surprising consistency of the projected total numbers of elderly in China in the next century, produced independently by different scholars at different times, 10 years apart, following substantially different approaches, confirms the anticipated huge quantity of elderly persons in China in the next century.
Table 1 also presents the numbers of elderly persons in other selected countries projected by the Population Division of the United Nations (, ) under the medium variant. Under the medium mortality assumption, China's elderly population will be fairly close to the total population size of the United States, and 4.4 times as large as the U.S. elderly population by the middle of the next century. China's elderly population will outnumber India's by 103 million, while the Chinese total population size will be smaller than India by 51 million in 2050.
2.3 Even More Extremely Rapid Increase of "Oldest Old" Persons after 2020
Most younger elderly persons (less than 80 years old) are relatively healthy, but the oldest old usually need help. The oldest old persons consume amounts of services, benefits, and transfers far out of proportion to their numbers. For example, about a quarter of medicare payments to hospitals were on behalf of the oldest old patients in 1988 in New York City . According to a German study, 1.7, 3.2, 6.2, 10.7, and 26.3 percent of the elderly age 65-69, 70-74, 75-79, 80-84 and 85+, respectively, regularly need health care services . Obviously, the oldest old persons are most likely to need help. However, in China and in almost all other developing countries, very little is known about the oldest old, and almost all published statistics, based on census data, are truncated at age 65 or so. The Population Division of the United Nations has taken a ground breaking step forward by revising the U.N. population estimates and projections to extend the open-ended age interval from 80+ to100+ . We examine here the projected oldest old population in China in the next century based on our study and compare them to the most recent projections by U.N. Population Division.
Table 2 provides the projected numbers and percentage distributions by various age groups for the elderly population in China. There were about 8 million "oldest old" (age 80 and over) in 1990. As compared with the increase of all elderly persons age 65 and above, the number of the oldest old will climb much faster to about 13, 32, 76, and 114 millions in the years 2000, 2020, 2040 and 2050, respectively, under the medium mortality assumption. The average annual increase rate of the oldest old between 1990 and 2050 will be 4.2 percent! The percent share of the oldest old among the elderly population will be nearly tripled from 1990 to 2050. From 1990 to 2040, the share increases by approximately 2.5 percentage points per 10 years. But in the 10 years from 2040 to 2050, the share increases by 10.6 percentage points! The main reason why the number of "oldest old" will climb so quickly after year 2040 is that China's baby booms in the 1950's and the 1960's will fall into the category of "oldest old" at that time.
As shown in Table 2, the numbers of elderly persons age 65+ projected by Zeng and Vaupel  are fairly close to the ones projected by the U.N. Population Division . However, the U.N.'s projected number of oldest old in 2050 (99.6 million) is considerably smaller than that projected by Zeng and Vaupel (114.4 million), and the projected age distributions of the elderly population differ (see Table 2). This discrepancy is mainly due to the different approaches for interpolating age-specific death rates in future years. We believe that our projected larger numbers and higher proportion of the oldest old may be closer to the reality of the future trend, although subject to a lot of uncertainties (see Appendix for a technical explanation). Despite the discrepancy and uncertainties in accurately forecasting the oldest old population, it is certain that the oldest old will increase tremendously in the next century in China (also see: Mayer et al. 1992: 81-82). The middle of the next century will be a hard time for the country due to the serious problems of population aging.
2.4 Extraordinarily Rapid Population Aging under the Low Mortality Scenario
The extremely rapid population aging discussed above are all based on the medium mortality assumed by the U.N. Population Division  and by us. The underlying assumption of the medium mortality variant is that there will be slow progress in reducing mortality in China during the next century - from a life expectancy of 68.4 years for both sexes combined in 1990 to 78.8 in 2050. This is quite conservative, given the fact that life expectancy in Japan in 1995 was already 80 years. Some recent research indicates that there might be a significant improvement in mortality in the next century, because of biomedical advances and breakthroughs, and better personal health practices, such as healthy diets, not-smoking and exercise etc. We, therefore, made another optimistic scenario, namely, life expectancy for both sexes combined was assumed to approach 84.9 years by 2050 , a level that is about 4.5 years higher than in Japan today. This low mortality scenario is subject to uncertainty, but we believe that it is not impossible. For example, male and female life expectancies in Japan in 1950 were 7.7 and 10 years lower than in the United States, respectively, but the difference disappeared in 1960s [25, p32]. Rapid socio-economic development plus the Eastern Asian style of healthy diets and habits in China may narrow the gap between Chinese and Japanese mortality levels in the next century. Some scholars thought that a life expectancy of 85 years represented the limit of human life expectancy (e.g. , ). However, most scholars now think that human beings can, on average, live much longer than 85 years (e.g. , , ). Despite uncertainty, the medium and the low mortality scenarios bracket an informative range of possibilities in China during the first half of the next century.
Under the low mortality scenario, the elderly will comprise 17.4 and 26.5 percent of the total Chinese population in 2030 and 2050, respectively. The annual increase rate of the proportion of the elderly population age 65+ between 1990 and 2050 is 2.6 percent. By the year 2020, 2030, 2040 and 2050, there will be 187, 264, 370 and 407 million elderly people, respectively, in China. Under the low mortality scenario, the oldest old will number 38, 58, 100 and 161 million in the years 2020, 2030, 2040 and 2050, respectively (, , [Note 4]). Obviously, if the biomedical breakthroughs and improved health practices make the low mortality scenario become reality, population aging problems in China will be much more serious in the next century.
2.5 More Serious Aging Problems in Rural Areas than in Urban Areas [Note 5]
Although fertility in rural areas in China is much higher than in urban areas, aging problems will be more serious in rural areas because of the continuing massive rural to urban migration of which the large majority is young people. Under the medium fertility and medium mortality assumptions, the proportion of the elderly will be 26 and 22 percent in rural and urban areas respectively by the middle of the next century. The proportions will be 31 percent in rural area in contrast to 26 percent in urban areas under the medium fertility but low mortality assumptions .
While the percent of elderly in rural areas in the next century will be substantially higher than in urban areas, the rural elderly are much less able to obtain necessary social support and services. According to a survey of the elderly conducted in 1992 in 12 provinces (Beijing, Tianjing, Shanghai, Zhejiang, Jiangsu, Heilongjiang, Shanxi, Shananxi, Shichun, Guangxi, Guizhou, and Hubei), only 5.9 percent of the rural elderly age 60 and over were pension recipients, in contrast to 73.7 percent in the urban areas. 66.6 percent of the urban elderly had their medical expenses paid entirely or partially by the government or collective enterprises in 1991. However, this figure was only 9.5 percent for the rural elderly. In another survey conducted in 1987, a relatively small proportion (32.5 percent) of the elderly in urban areas reported that they had difficulties in obtaining medical care, while a big majority (94.8 percent) of the rural elderly had medical difficulties. About 21.3 percent of the urban elderly reported that their nutrition status was bad, and this percent was as high as 53.3 for rural elderly (PRI of CASS 1988).
It is also important to note that the extremely rapid and large scale population aging in China is accompanied by a per capita GNP that is considerably lower than many other developing countries, especially in rural areas. Thus, resources for addressing the serious problems caused by rapid population aging are very limited.
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