Southern African Regional Poverty Network (SARPN) SARPN thematic photo
Country analysis > South Africa Last update: 2020-11-27  

 Related documents

Getting by…
Benefits of non-contributory pension income for older South African households

Valerie MС€ller, Institute of Social and Economic Research, Rhodes University

Monica Ferreira, The Albertina and Walter Sisulu, Institute of Ageing in Africa, University of Cape Town

Grahamstown and Cape Town


November 2003

В© Institute of Ageing in Africa, University of Cape Town, 2003
This report was posted with permission of Professor Monica Ferreira, director of the Institute of Ageing at the University of Cape Town.
This report formed part of the Non-Contributory Pensions and Poverty Study (NCPPS) studies in Brazil and South Africa;
more details of the NCPPS can be obtained from
[Download complete version - 400Kb ~ 2 min (87 pages)]     [ Share with a friend  ]

Executive summary

Scant attention has been given in the literature on poverty to the role of non-contributory pension programmes in poverty reduction in developing countries. Very few developing countries indeed have a non-contributory pension programme. Brazil and South Africa, both middle-income countries with a vast poor population and ethnically diverse populations, have well-established, similar programmes in place. Comparative surveys were carried out by a multi-national team of investigators in the two countries in 2002/03 to investigate the benefits of social pension income for older households - specifically, how pension income augments other household income and alleviates household poverty, and thereby enhances household members' participation in the development process. An older household was defined as including at least one person aged 55 years and over. Only the South African survey and its findings are reported here.

The South African study examined the general living conditions, and financial and health situations of 1 111 older households. Survey sites were selected in two provinces: the Cape Town metropole in the Western Cape and rural areas of the Eastern Cape. A multi-stage cluster sampling design was applied to select realised subsamples of 374 rural black (RB) households in the Eastern Cape, and 324 urban black (UB) households and 413 urban coloured (UC) households in the Western Cape. Data collection was contracted to a commercial research firm. A questionnaire was specially constructed by the multi-national investigators for use in both the South African and Brazilian surveys, which was administered to heads of the sampled households.

Topics covered in the survey instrument included the household's material living conditions, household composition, economic activities, income and expenditure and the servicing of debts, health and care, and perceived quality of life. A separate module was administered to all older household members (55 years and over), and covered topics relating to pension and unemployment issues, intra-household dynamics and the employment of older persons.

The survey data are tabulated in appendices to the report and the results of an analysis of the household data are presented and discussed in the report. Separate profiles are given of household members aged 55 years and over and of old-age pensioners resident in the households. The data and the results are extensive, and only selected findings are given in this summary.

Among the key findings were: In RB households pensioners were often the sole income earners. The majority of older persons lived with children or grandchildren; 5% of the RB subsample and 8% of the UC subsample lived alone. The study confirmed that older black households are the poorest in South Africa. Old-age pension income competed with wage earnings as the most important source of income in the surveyed households, both in terms of access and size of earnings. However, older RB households were virtually dependent on social pension income and to a far lesser degree on other government transfers. Although the grandchild generation accounted for 39% of RB household members, only 3.5% of these households benefited from a child support grant; in UB households, grandchildren comprised a quarter of household members and 18% received the grant.

RB households were on average the largest households; a quarter comprised eight or nine persons. Some 15% of these households received remittances from persons outside the households; between 18 and 21% of RB pensioners received money from children. A large majority of black social pensioners pooled their pension money with other household income. Few households were able to save or invest in a small business. Less than 4% of older household members were self-employed or employed others. On average, only 2% of earners in all households derived income from odd jobs, hawking or other informal sector work; informal work appeared to be non-existent in the RB households. In RB households, in particular, pension income was expended on the education of children in the household.

Other findings included pensioner households being creditworthy; membership of burial societies and church groups being widespread; RB households in particular borrowing from micro lenders in times of financial difficulty; and few pensioners reporting problems accessing their pension money and less than 7% of surveyed beneficiaries accessing their pension through a bank.

Overall, a gradient of disadvantage across the subsamples was noted in the survey findings, with RB households being the most disadvantaged in all domains surveyed, UC households being the least so, and UB households falling inbetween. The relative disadvantage and comparative advantage, and the differential benefits of pension income for poor older households in the three subsamples are highlighted and discussed in the report.

Conclusions drawn in the report include the importance of non-contributory pension income to the welfare of poor older households; the comparatively heavier health and child care burden shouldered by RB households; the widespread custom of pension-sharing in RB households in particular; and the irony that few black pensioners are able to use their pension income for themselves but must share the income to meet the survival needs of their family. By contrast, UC older households benefit from other government transfers, such as the child support grant, to a greater extent and have a lower debt burden than the other two subpopulations. A greater number of UC pensioners than other pensioners are able to use their pension money for themselves. It is proposed in the report that inequalities in the positive impact, or benefits of pension income for poor older households across South Africa's subpopulations may be strengthened through greater access of black households, particularly in rural areas, to free education, free health care, basic free water and electricity, and the child support grant.

Octoplus Information Solutions Top of page | Home | Contact SARPN | Disclaimer