«D I R E C T I O N S I N D E V E LO P M E N T Human Development Public Disclosure Authorized The Elderly and Old Age Support in Rural China Challenges ...»
This analysis finds notable variation by gender differences according to whether the individual is among the younger or older elderly. Coefficient estimates from labor supply models of participation and hours worked in productive activity are shown in tables 3.5 and 3.6, respectively. Labor
supply of the elderly is affected by the following factors:
• Increases in actual (or potential) household income from other sources reduce the likelihood of an older person still working. This factor is evident Figure 3.6 Correlation of Employment of Elderly and Migrant Status of Adult Children
in table 3.5 from examining the effects of pension receipt and educational attainment on employment.
For each 1,000 RMB per capita received as pension income, the probability that a man works (whether in his sixties or seventies) falls by 8 percent, and the probability that a woman works falls by about 6 percent. Educational attainment of men in their sixties has no significant effect on employment, but each additional year of education is associated with a 1.6 percent reduction in the probability that a woman in her sixties is employed, and a 1.2 percent to 1.3 percent reduction in the probability that elderly in their seventies are working. An increase in the average years of education of other family members has an even stronger depressing effect on labor force participation: higher average levels of education are associated with higher permanent income of the family, which facilitates retirement.
• Ill health, as represented by increases in the activity of daily living (ADL) z-score,11 also lowers the likelihood that an individual engages in productive activities. Because men are often tasked with more
physically demanding jobs in agriculture, this effect is, not surprisingly, stronger for men in both age categories than for women. A one standard deviation increase in the ADL z-score is associated with
11.3 percent and 7.9 percent decreases in probability of employment for men and women, respectively, in their sixties, and a 10.3 percent and 4.9 percent decrease, respectively, for men and women in their seventies.
• Migration of adult children is weakly associated with increased labor supply. A positive association exists between a migrant child and increased labor supply, but it is not statistically significant for men or for women under 70. Lost farm income and uncertainty about land tenure likely dominate the effects of increased income from remittances when the elderly make decisions about working. Women over 70, however, are
8.1 percent more likely to continue working (most likely in agriculture) if the household has a migrant child.
Evidence on hours worked, presented in table 3.6, suggests that migrant adult children have the strongest effect on the work behavior of women over age 70. Women over 70 with a migrant child work 411 hours more during a year, or the equivalent of 10 40-hour workweeks. For regions of the country that plant two crops a year, that would amount to full-time work during the agricultural busy season. The effects of education on hours worked are of the same direction as in the employment models of table 3.5, but once employed, education itself does not strongly affect the number of hours worked.
As might be expected, the number of hours worked by the rural elderly is also very sensitive to their health status and to income from pensions. An increase in one standard deviation of the ADL index is associated with a 417-hour reduction in hours worked by men in their sixties, and a reduction of 294 hours by men in their seventies. Women in their sixties reduce labor supply by 300 hours with a one standard deviation increase in their ADL index, whereas women in their seventies reduce labor supply by 212 hours in response to a similar worsening in their health. Pension income also has a stronger depressing effect on hours worked by men in their sixties.
The urban elderly are much less likely to work after 60 than are the rural elderly. And the results presented here suggest that any mechanism providing the rural elderly with an income flow in old age will likely reduce both employment and hours worked. As China’s government rolls Sources of Support among the Rural Elderly 63
out a national pension system for the rural elderly, rural residents will likely choose to “retire” rather than work until they are no longer able to do so, depending of course on the benefit level.
Conclusion Whereas the urban elderly receive significant support from pensions, the rural elderly rely primarily on their own labor income and financial support from their children. More men over 60 rely primarily on their own labor income for support, whereas elderly women are somewhat more dependent on transfers from family members. Evidence presented on responsiveness of private transfers to income levels and on the determinants of elderly labor supply suggests that the introduction of a pension system may lead to more security for the elderly without requiring them to continue working into old age.
Private transfers are responsive to low levels of elderly income, but they do not perfectly crowd out income from other sources. At levels of income below the poverty line, transfers from adult children increase as elderly income decreases. When elderly incomes are at and even below 64 The Elderly and Old Age Support in Rural China the poverty line, however, private transfers are not perfectly crowded out by increases in income. A social welfare benefit that raises incomes of the elderly will not crowd out private transfers.
Migrant children continue to provide remittance support to their parents, but the predicted range of transfers suggests growing risk that lowincome elderly may be left in poverty. On average, the predicted transfer from adult children is sufficient to maintain elderly incomes above the poverty line. When the range of potential transfers is considered, however, elderly with migrant children clearly face more risk that private transfers will be insufficient to maintain standards of living above the poverty line.
Elderly with higher incomes are less likely to work, as are elderly who are in poor health. Elderly with nonlabor income from pensions or in households with higher levels of education, and thus more incomeearning potential, are less likely to participate in income-earning activities. If the elderly receive some type of social welfare support, they are more likely to exit the labor force when they are over 60.
Elderly in poor health are less likely to work. Declines in health status associated with decreases in ability to perform daily tasks are associated with lower participation in work activity and fewer working hours for those who are still working. Improving use of the health care system for preventive purposes may keep older workers productive and earning incomes for a longer period of their lives. Further research examining the effects of the New Cooperative Medical System is warranted to determine whether it facilitates improved health and work capacity of older workers.
Migration of adult children may significantly affect the work status of elderly women. Having a migrant child in the family raises the probability that a woman over 70 will still be in the labor force and, conditional on working, will work more hours in a year. For men and women under 70, a migrant child has a positive but statistically insignificant effect on participation in income-earning activities.
1. The data were collected before initiation of the national rural pension pilot program in 2009.
2. This finding is consistent with findings in the World Bank’s China poverty assessment (Chaudhuri and Datt 2009) using other National Bureau of Statistics data.
Sources of Support among the Rural Elderly 65
3. See World Bank (forthcoming) for a detailed discussion of the evolution of rural dibao and other forms of rural social assistance, noting coverage expansion of rural dibao from just over 8 million in 2005 to almost 48 million in 2009.
4. Using other methods, Cameron and Cobb-Clark (2008) find no evidence that transfers to parents respond to low parent income in Indonesia.
5. Selden (1993) concludes that a transition to the nuclear family imposes a heavy price on the rural elderly. Living arrangements are thought to be important for elderly support across East Asia, including Cambodia (Zimmer and Kim 2002), Thailand (Knodel and Chayovan 1997), and Vietnam (Anh and others 1997).
6. Two very different conclusions are consistent with evidence of greater incidence of coresidence with age: the oldest, who are more likely to be infirm, tend to move in with adult children; alternatively, if coresidence has an effect on the quality of care provided, then perhaps only the elderly living with adult children reach old age.
7. Coresidence in rural areas of the four RCRE provinces was also somewhat higher than that observed for rural areas of the CHNS panel.
8. Barro (1974), Becker (1974), and Cox (1987) make important distinctions highlighting different motives for transfers. Much of the empirical research in the United States has suggested that intergenerational inter vivos transfers are driven by exchange motives (for example, Cox and Rank 1992; McGarry
1999) rather than by altruistic motives. However, in the United States, the social security safety net provides substantial insurance against poverty in old age, and thus it is not surprising to find an emphasis on the flow of resources from older to younger generations.
9. Knowing whether a household has family members who have migrated is important for making comparisons across households with migrant children and those with family members living locally. The most appropriate existing data source for the study of transfer behavior is the RCRE household survey, complemented by a supplemental survey conducted in RCRE households in
2004. This book uses the 1995 to 2003 waves of RCRE’s annual household survey from four provinces (Anhui, Henan, Jiangsu, and Shanxi) in which a supplemental survey (Supplemental Rural Household Social Network, Labor Allocation, and Land Use Survey) was carried out in collaboration with Michigan State University in 2004. A description of the RCRE survey and a comparison with other household surveys in China can be found in Benjamin, Brandt, and Giles (2005). An important feature of the survey is that it enumerated characteristics of former household members, including their current location and educational attainment. Among the factors influencing the size of private transfers, the analysis includes the number of children (which captures the effect of the potential transfer network size on 66 The Elderly and Old Age Support in Rural China which rural elderly can rely for private transfers) and average education of children living outside the household (which proxies for the quality of the network). Other regressors included in the analysis are age, years of schooling, marital status, whether any household members are in school, number of household residents, number of working household residents, shares of household residents in different demographic categories, numbers of nonresident family members in different demographic categories, village variables to control for the local economic environment, and province-year variables to control for macroeconomic shocks. The analysis uses the approach introduced by Yatchew (2003) and implemented for analysis of transfers in China by Cai, Giles, and Meng (2006).
10. The information comes from a retrospective work history carried out as part of a supplemental survey conducted with RCRE in 2004. The supplemental survey asked about the work history of current and former household residents and their parents.
11. In the CHNS, activities of daily living (ADLs) are captured by questions asking individuals over age 50 to rate the difficulty from 1 (no difficulty) to 4 (cannot do at all) of the following activities: walking a kilometer; sitting continuously for two hours; standing up after sitting for a long time;
climbing one staircase; lifting or raising a 5 kg bag; squatting down, kneeling down or bending over; bathing yourself; eating by yourself; putting on your clothes; and using the toilet. Average responses are calculated for male and female elderly, and then z-scores are calculated for each individual.
References Anh, Truong, Bui T. Cuong, Daniel Goodkind, and John Knodel. 1997. “Living Arrangements, Patrilinearity and Sources of Support among Elderly Vietnamese.” Asia-Pacific Population Journal 12 (4): 69–88.
Barro, Robert J. 1974. “Are Government Bonds Net Wealth?” Journal of Political Economy 82 (6): 1095–1117.
Becker, Gary. 1974. “A Theory of Social Interactions.” Journal of Political Economy 82 (6): 1063–93.
Benjamin, Dwayne, Loren Brandt, and Jia-Zhueng Fan. 2003. “Ceaseless Toil?
Health and Labor Supply of the Elderly in Rural China.” William Davidson Institute Working Paper 579, Department of Economics, University of Toronto, Canada.
Benjamin, Dwayne, Loren Brandt, and John Giles. 2005. “The Evolution of Income Inequality in Rural China.” Economic Development and Cultural Change 53 (4): 769–824.
Sources of Support among the Rural Elderly 67 Benjamin, Dwayne, Loren Brandt, and Scott Rozelle. 2000. “Aging, Well-Being, and Social Security in Rural North China.” Population and Development Review 26 (suppl.): 89–116.
Cai, Fang, John Giles, and Xin Meng. 2006. “How Well Do Children Insure Parents against Low Retirement Income? An Analysis Using Survey Data from Urban China.” Journal of Public Economics 90 (12): 2229–55.
Cai, Fang, John Giles, and Dewen Wang. 2009. “The Well-Being of China’s Rural Elderly.” Background Paper for East Asia Social Protection Team, World Bank, Washington, DC.
Cameron, Lisa, and Deborah Cobb-Clark. 2008. “Do Coresidency and Financial Transfers from Children Reduce the Need for Elderly Parents to Work in Developing Countries?” Journal of Population Economics 21 (4): 1007–33.
CHNS (China Health and Nutrition Survey). Various years. China Center for Disease Control and Prevention and the Carolina Population Center, University of North Carolina at Chapel Hill. http://www.cpc.unc.edu/projects/china.