|
Summary Report
The social and economic impact of South Africa's social security system
Commissioned by the Directorate: Finance and Economics
Department of Social Development
Dr. Michael Samson, Ms. Una Lee, Mr. Asanda Ndlebe, Mr. Kenneth Mac Quene,
Ms. Ingrid van Niekerk, Mr. Viral Gandhi,
Ms. Tomoko Harigaya, Ms. Celeste Abrahams
|
|
|
[Download complete version  - 114Kb ~ 1 min (27 pages)]
[ Share with a friend 
]
|
Executive Summary
Social grants in South Africa play a critical role in reducing poverty and promoting social development. This study evaluates the social and economic impact of State Old Age Pensions (SOAP), Disability Grants (DG), Child Support Grants (CSG), Care Dependency Grants (CDG), Foster Care Grants (FCG) and Grants-in-Aid (GIA). The analysis evaluates the role of social assistance in reducing poverty and promoting household development, examining effects on health, education, housing and vital services. In addition, the study assesses the impact of social grants on labour market participation and labour productivity, providing an analysis of both the supply and demand sides of the labour market. The study also quantifies the macro-economic impact of social assistance grants, evaluating their impact on savings, consumption and the composition of aggregate demand. Most of the statistical analysis focuses on the CSG, SOAP and DG since sample sizes are sufficiently large for these grants to support significant inferences.
South Africa’s system of social security successfully reduces poverty, regardless of which methodology is used to quantify the impact measure or identify the poverty line. Nevertheless, the quantitative measure of poverty reduction is sensitive to the methodological choices. For instance, the measured impact is consistently greatest when employing the total rand poverty gap as an indicator. The poverty headcount measure, however, consistently yields the smallest results. Likewise, the choice of poverty line heavily influences the measurement of the quantitative impact. The current social security system is most successful when measured against destitution, and the impact is smallest when poverty lines ignore economies of scale and adult equivalence issues. For instance, South Africa’s social grants reduce the poverty headcount measure by 4.3%, as measured against the Committee of Inquiry’s expenditure poverty line (with no scales). The social security
system, however, reduces 45% of the total rand destitution gap—an impact more than ten times greater.
Using the Committee of Inquiry expenditure poverty line (without scales), a 10% increase in take-up of the SOAP reduces the poverty gap by only 1.2%, and full take-up by only 2.5%. The take-up rate for the SOAP is already very high, and many of the eligible elderly not already receiving the SOAP are not among the poorest South Africans. As a result, further extensions of the SOAP have limited potential in reducing poverty. Extensions of the Disability Grant offer greater promise, although at substantially greater expense. A 50% increase in DG take-up reduces the total rand poverty gap by 1.7%, and full take-up generates a 5.1% reduction. The greatest poverty reducing potential lies with the progressive extension of the Child Support Grant. Extending the eligibility age to 14 reduces the poverty gap by 16.6%, and a further extension to age 18 reduces the gap by 21.4%. Increasing the real grant payment (as the government did in 2003) generates an even greater impact. The extension to age 14 yields a 22% poverty gap reduction, while the extension to age 18 reduces the poverty gap by 28.3%. Combining the higher CSG extended to age 14 with the full take-up of the SOAP and the DG yields a reduction in the total rand poverty gap of 29%.
The magnitudes of these effects, of course, depend critically on the poverty line by which the impacts of the reforms are measured. For instance, the 29% reduction in the total rand poverty gap measured using the unscaled Committee of Inquiry expenditure poverty line is less than half the magnitude of the reduction in destitution, which amounts to a 66.6% reduction. Likewise, the impacts of the scaled Committee of Inquiry income and expenditure poverty lines are substantially greater than for the unscaled poverty lines. The impact of the “all grants” package measured with the scaled Committee of Inquiry income poverty line reflects a 47.4% reduction, and with the expenditure poverty line, a comparable 47.5% reduction. As this makes apparent, the distinction between income and expenditure poverty has not generated material differences in this analysis. Likewise, the impact using the unscaled Committee of Inquiry income poverty line (a 28.9% reduction) is virtually the same
as that using the unscaled Committee of Inquiry expenditure poverty line (a 29.0% reduction). For almost every simulation, the HSL poverty line generates very close results to those yielded by the scaled Committee of Inquiry income and expenditure poverty lines, in spite of the substantial methodological differences distinguishing the HSL measure. The relative poverty line yields results that are not closely comparable to any of the other poverty line measures, with the results generally falling in between the results of the Committee of Inquiry scaled and unscaled poverty line measures.
The evidence in this report documents the substantial impact of South Africa’s social security system in reducing poverty and destitution. The magnitudes of the results are sensitive to methodological issues. It matters whether the poverty line is relative or absolute, whether it is scaled for household composition and economies of scale or not, and to a small extent whether it measures income or expenditure. Likewise, it matters how the poverty
impact is measured—using poverty headcount or variants on the poverty gap. Nevertheless, the qualitative results, and the answers to critical policy questions, are robust to different methodological approaches. South Africa’s system of social security substantially reduces deprivation, and the progressive extension of the magnitude, scope and reach of social grants holds the potential to dramatically diminish the prevalence of poverty in South Africa.
The results of this study provide evidence that the household impacts of South Africa’s social grants are developmental in nature. These findings are consistent with international lessons of experience, as well as with previous studies of South Africa’s system of social security. Social security programmes in Brazil, Argentina, Namibia and Botswana yield positive impacts in terms of reducing poverty, promoting job search and increasing school attendance. Past studies of social security in South Africa have focused on the State Old Age Pension, identifying important positive effects in terms of broadly reducing
household poverty as well as improving health and nutrition.
Poverty and its associated consequences erode the opportunities for children and youth to attend school, fomenting a vicious cycle of destitution by undermining the household’s capacity to accumulate the human capital necessary to break the poverty trap. The statistical evidence from this research documents the extent to which poverty exerts a negative impact on school enrolment rates. Many poor children cannot attend school due to the costs associated with education, including the necessity to work to supplement family income. In addition, communities that are resource-constrained provide lower quality educational services, which negatively affects enrolment rates. Social security grants counter these negative effects by providing households with more resources to finance education. New findings from this study demonstrate that children in households that receive social grants are more likely to attend school, even when controlling for the effect of income. The positive effects of social security on education are greater for girls than for
boys, helping to remedy gender disparities. But both the State Old Age Pension and the Child Support Grant are statistically significantly associated with improvements in school attendance, and the magnitudes of these impacts are substantial. This analysis only measures the direct and static link between social security and education. To the extent that social grants promote school attendance, they contribute to a virtuous cycle with long term dynamic benefits that are not easily measured by statistical analysis.
Nationally, nearly one in five households experienced hunger during the year studied (2000). The highest income provinces—Gauteng and the Western Cape—have the lowest prevalence rates of hunger. The prevalence rate of hunger is highest in one of South Africa’s poorest provinces—nearly one in three households in the Eastern Cape experiences hunger. However, another of the poorest provinces—Limpopo—has the third lowest hunger prevalence rate in the country. Meanwhile, Mpumalanga—with a poverty rate below the national average—has the second highest hunger prevalence rate in the country. Social
grants are effective in addressing this problem of hunger, as well as basic needs in general. Spending in households that receive social grants focuses more on basics like food, fuel, housing and household operations, and less is spent on tobacco and debt. All major social grants—the State Old Age Pension, the Child Support Grant and the Disability Grant—are significantly and positively associated with a greater share of household expenditure on
food. This increased spending on food is associated with better nutritional outcomes. Households that receive social grants have lower prevalence rates of hunger for young children as well as older children and adults, even compared to those households with comparable income levels.
Receipt of social grants is associated with lower spending on health care, perhaps because social grants are associated with other positive outcomes that reduce the need for medical care. For instance, the World Bank identifies the important link between improved education and stemming the spread of HIV/AIDS. Likewise, social grants are associated with greater household access to piped water. The evidence in this chapter underscores the importance of moving beyond measures of income poverty in the assessment of social deprivation. In case after case in this study, household outcomes conflicted with the simple
implications of monetary income rankings. While many measures of well-being are
correlated with aggregate income and expenditure, the exceptions affect large numbers of people and require careful policy analysis. The interaction between social security and household well-being is complex, and further research continues to explore these interactions. In particular, the broad measures of household well-being analysed in this chapter exert profound effects on labour productivity and the ability of workers to find jobs. Employment in turn provides access to resources that promote improved education, nutrition, health and other outcomes.
Conventional economic theory suggests that social grants may undermine labour force participation by reducing the opportunity cost of not working. Models developed for industrialised countries and applied broadly to South African data sometimes corroborate this hypothesis. However, when models are developed that reflect the labour market behaviour of South Africans who receive social grants, the results contradict this hypothesis. The response of very low income South Africans to a marginal increase in their income is
significantly different from the response of median income South Africans.
To the extent that social grants create adverse labour market effects, the adverse consequences stem from distortions in social security targeting mechanisms. For instance, to the extent that the State Old Age Pensions are employed to target the non-pensioner poor, then the grants may encourage a household formation response that impedes job search. These types of problems can be addressed by broadening the base of the social security programmes. A more comprehensive system of social security generates fewer distortions from the incentive effects created by the social grants.
This study explicitly examines the impact of social grants on the labour market participation, employment success and realised wages of South Africans in households receiving social grants. While statistical analysis cannot prove causation, the empirical results are consistent with the hypotheses that:
- Social grants provide potential labour market participants with the resources and economic security necessary to invest in high-risk/high-reward job search.
- Living in a household receiving social grants is correlated with a higher success rate in finding employment.
- Workers in households receiving social grants are better able to improve their productivity and as a result earn higher wage increases.
The empirical evidence discussed in this chapter demonstrates that people in households receiving social grants have increased both their labour force participation and employment rates faster than those who live in households that do not receive social grants. In addition, workers in households receiving social grants have realised more rapid wage increases. These findings are consistent with the hypothesis that South Africa’s social grants increase both the supply and demand for labour. This evidence does not support the hypothesis that South Africa’s system of social grants negatively affects employment
creation.
At the macro-economic level, South Africa’s system of social development grants tends to increase domestic employment while promoting a more equal distribution of income. The effects of grants on national savings and the trade balance are ambiguous, since grants have two competing effects on the national savings—one through private domestic savings, and the other through the trade deficit. Depending on the magnitude of the effects, grants could improve or worsen national savings and the trade balance. Initial analysis suggests
that the impact on savings may be negative, while that on the trade balance may be positive. However, since much of the savings of upper income groups are offshore, the negative impact is unlikely to be significant, particularly given the small share of private savings in the national savings rate. The impact on inflation may also be ambiguous. The increase in overall demand in the economy may generate some inflationary pressure. However, the relatively low rate of capacity utilisation may enable the economy to meet this demand without significant increases in inflation. Likewise, the positive trade balance effects may lead to an appreciation of the rand, tending to dampen imported inflation. On balance, the macro-economic impact of South Africa’s social security system is largely positive. These positive macroeconomic effects support higher rates of economic growth, which are reinforced by the social security system’s positive effects on income distribution and education.
|
|
Related documents