Social assistance, gender, and the aged in South Africa

Abstract

This paper reviews the history of the noncontributory social pension in South Africa, as well as recent work on the distributional and poverty-alleviating effects of this program. The pension has a strong gender dimension, reaching three times as many women as men, and has an unambiguous impact on reducing household poverty, particularly among Black South African households. The existing literature also suggests that the pension reaches unintended beneficiaries within households and that strongly gender-differentiated patterns emerge both in the sharing of pension incomes by pensioners and in the behavioral responses of other household members to pension receipt.

Keywords:

• Social pensions
• welfare and poverty
• income pooling

Notes

JEL Codes: J18, I30, J16

The elderly are defined as people who are 60 years or older. For simplicity, this convention is followed throughout this paper despite the differences in pension-age eligibility rules by gender.

Racial categorization was institutionalized by the apartheid state. The Population Registration Act of 1950 required that all South Africans be racially classified, and this resulted in the legal construction of four main race groups: White (European descent), Black (African), Indian (predominantly descendants of South Indian indentured laborers brought to work on British-owned sugar plantations on the southeast coast of South Africa) or Colored (of mixed-race descent). Classification into these categories was based on appearance, social acceptance, and descent, and was vigorously enforced by the apartheid state.

See note 2 for discussion of race categorizations.

Take-up rates are the rates of enrollment for a program.

The remaining age-eligible Black South Africans were either in richer households with a private pension or were the very poor who had difficulty accessing pensions (Anne Case and Angus Deaton 1998).

This is approximately US$132 in PPP adjusted terms.

Primary healthcare is free for all South Africans. With respect to secondary healthcare, the elderly in receipt of a pension are eligible to receive free secondary healthcare services at public hospitals.

The housing subsidy scheme makes special provision for the elderly, and the value of the subsidy for the elderly exceeds the maximum value of subsidies for other beneficiaries.

The elderly account for 8 percent of the total population in developing countries, 5 percent in other African countries, and 19 percent in developed nations (Department of Social Development 2002a).

These estimates, however, assume the absence of any crowding out of informal assistance or remittances as the result of pension receipt. Robert Jensen (2003) presents evidence to suggest a 20 – 30 percent reduction in remittances for every rand of pension income received by South Africa's elderly, implying that the poverty-reducing effects of the pension may be lower than previously thought. However, Dori Posel (2001) argues that remittances are more strongly influenced by the characteristics of the remitting household than by the pension status of the recipient.

Headship is a self-identified concept reported by the interview respondent.

This differential age-eligibility rule has been contested in post-apartheid South Africa. In the latest challenge, the Communicare Residents Association has taken a class action suit before one of South Africa's provincial high courts, arguing that this law embodies discrimination against men (Cape Argus June 20, 2002).

In the case of a single individual, the means test is calculated by taking the sum of the individual's income and an income value assigned to the assets that the individual owns. For married individuals, the means test is calculated by taking the mean of the pooled income (actual income and the income value of owned assets) of the pensioner and their spouse. The means test does not take the income of other household members into account, thus removing any incentive for households to rearrange themselves in order to become eligible for pension receipt. If the individual's income exceeds the means test, the size of the pension payment is reduced on a rand-for-rand basis, in line with their income status.

According to a National Food Consumption Survey (Department of Social Welfare 2002a), 40 percent of households are now headed by elderly persons as a direct result of the HIV/AIDS epidemic. These households would be characterized as skip-generation households.

Significant at 1 percent level.

Significant at 1 percent level.

Significant at 1 percent level.

Significant at 1 percent level.

Additional information

Notes on contributors

Justine Burns

JEL Codes: J18, I30, J16