Who Gets and Who Doesn T

WHO GETS AND WHO DOESN’T? REFLECTIONS ON THE UP-TAKE OF THE CHILD SUPPORT GRANT IN BHAMBAYI, SOUTH AFRICA

Dr Barbara Simpson and Dr Tanusha Raniga

Loyola Journal of Social Sciences, Vol XXIV, no. 2, July-Dec 2010.

INTRODUCTION

The decisions regarding choices and access to social grants and services in both the developed and developing countries differ considerably and are ultimately dependent on the particular political and ideological positions of that country (Pelman, 2007). One of the choices to be made is whether social security programmes should be universal or targeted. Universal programmes allocate benefits to all irrespective of income, whereas targeted programmes are used to select beneficiaries according predetermined criteria. Targeting is thus meant to achieve the widest coverage of the target group by concentrating benefits on them (Gulan, 1994, Cuenca, 2007).

South Africa has a number of social security benefits all of which are targeted to specific vulnerable groups of people. The child support grant is one such benefit and is a means-tested, cash transfer paid to the primary caregiver of the child. It was introduced in 1998 and was specifically intended to address child poverty. Initially, children up to the age of seven received the child support grant of R100.00 per month. Eligibility for the child support grant has been progressively expanded, and as of 2010, children up to the age of 18 receive an enhanced grant of R250.00.

This article investigates the up–take of the child support grant in Bhambayi, a mixed formal/informal settlement near Durban, South Africa. The study took place in 2007, when the child support grant was R220.00 and children up to the age of 14 were eligible to receive the grant. The article is divided into five sections. In the first section, we discuss the targeted nature of the child support grant in South Africa. We then provide background information regarding the research site and the rationale for undertaking a survey in that area. In the next section, we describe the research study. The results are then presented and we focus specifically on coverage of the child-support grant and exclusion and inclusion errors. The article ends with recommendations and conclusions.

THE SOUTH AFRICAN CHILD SUPPORT GRANT: A TARGETED SOCIAL SECURITY PROGRAMME

The debate about whether targeted or universal programmes are better suited to alleviating child poverty is, according to Notten and Gassman (2007), a long standing one. The argument is that the more accurate the targeting is in reaching the poor, the less wastage there will be, and the less will it cost to achieve the desired effect of reducing poverty (Sen, 1995). In countries where there is pressure to reduce social spending, improved targeting is seen as a way of reducing poverty with less expenditure (Besley and Kabur, 1990).

There is indeed some evidence to suggest that targeting is more efficient. Using mathematical modeling, Besley (1990) showed that poverty is reduced more effectively and efficiently when benefits are targeted to the poor based on a means test as opposed to universal benefits. Similarly, Edmonds (2005) found that profileration to non poor households was lower in the means-tested programme than the universal programme in Slovenia but Notten and Gassman (2007) showed that while using a means test improved coverage rates to children living in poor households in Russia, universal coverage would achieve greater levels of poverty reduction.

However, these and other authors (such as Besley and Kanbur, 1990) also point out that means tested programmes are more efficient are tentative and that factors such as administrative costs, individual responses and incentive effects and political economy considerations are difficult to quantify and may affect the efficiency of means tested programmes and their ability to impact on poverty reduction.

South Africa’s choice of targeted social security can be attributed to a number of reasons. Firstly, South Africa has a long history of targeted social security that began in 1928 with the introduction of a state old age pension (for poor people who were classified white or coloured). Over the years, foster care grants and disability grants as well as grants for single women (state maintenance grant) were introduced. Although these were gradually extended to all race groups, people who were not white received lower amounts. Since 1994 and the abolition of apartheid and its discriminatory policies, social security has been equalized and extended to people of all races. The child support grant was introduced in 1998 to replace the state maintenance grant which had excluded African children and was available to ‘White’, ‘Coloured’ and ‘Indian’ children (Lund 2008).

Secondly the choice of targeted programmes is underscored by the economic policy of the country. In 1996, South Africa adopted a neo-liberal capitalist macro-economic policy called GEAR (Growth, Employment and Redistribution) and in 2006, this was replaced by ASGISA (Accelerated and Shared Growth Initiative of South Africa). These policies are intended to stimulate economic growth through increased levels of local and foreign investment and create employment, thus improving the lives of the poor. However, proponents of social democratic thinking have criticised these macro economic policies as reducing the state expenditure on social security and reducing the responsibility of the state to care for the poor (Bond 2007; Desai 2007: Frye 2007). Universal grants in this policy context are thus not realistic.

Thirdly, as much as economic policies influence the choice of targeted programmes, there remains a strong ideological reluctance to implement universal social security programmes. There is a general public perception that dependency on state grants encourages laziness and erodes “ubuntu” which in South Africa refers to helping and caring without reward (Holscher, Kasiram and Sathiparsad 2009).

Despite such negative perceptions, one of the positive developments is that South Africa’s Constitution (RSA, 1996) provides for the personal, political and socio-economic rights of both adults and children. Section 28 of the Bill of Rights in the Constitution outlines the ideal environment for children’s care and protection such as the right to family care, basic nutrition, shelter, health care, and social services where children may thrive and meet their full potential. This constitutional imperative to protect children and ensure that their human rights are upheld, has guided the development of legislation and policy designed to protect children. The new Children’s Act (RSA, 2005) and the Children’s Amendment Act (RSA, 2007) provide the legislative framework for the implementation of a wide range of services that will protect children from exploitation and abuse (Jamieson, Proudlock and Waterhouse, 2007/2008). These Acts oblige national, provincial and local governments to provide adequate budgets and human resources for the provision of equal access to social services for children 18 years and younger, including those with disabilities and chronic illnesses.

In addition, social security specifically focusing on children has also been introduced to ensure that children’s basic needs are met. The child support grant is now the state’s largest social assistance programme in terms of numbers of beneficiaries and is the key poverty alleviation programme targeting children (UNICEF, 2008). In 2009, just over nine million children are received the child support grant (Hall, 2010).

BACKGROUND TO THE RESEARCH

The research took place in an area called Bhambayi which is situated in Inanda, approximately 25 kilometers from the Durban city centre, a large city of just over 3.4 million residents (Statistics South Africa, 2007) on the eastern seaboard of South Africa. The name “Bhambayi” derives from an Africanisation of the name “Bombay”; the area was so named because of its location on the site of the Phoenix Settlement, the original homestead of Gandhi, when he lived in South Africa. Until the mid 1980s, Gandhi’s descendants lived on this homestead, which comprised of Gandhi’s original house, a primary school, a museum and the original printing press of the newspaper, ‘The Indian Opinion” which Gandhi had established in the early 1900s.

In August 1985, political unrest resulted in widespread violence in the Inanda area, and the Indian residents of the Phoenix settlement were forcibly evicted and the Gandhi homestead destroyed. A large scale invasion of the land took place and an informal (squatters) settlement developed. Very soon, Bhambayi was home to an estimated 30 000 people, which increased to 50 000 by 1992 (Hindson and McCarthy, 1994). Violence continued to rack Bhambayi which was divided into two factions, each owing allegiance to a rival political party. Further upheavals in 1992/1993 resulted in many people fleeing the area and by 1996, there were 2395 informal structures and 13 594 people living in Bhambayi (Simpson, 2001).

It was in this context that the School of Social Work at the University of Natal (now the University of KwaZulu- Natal) first met with some members of the Bhambayi community as part of a community course that introduced students to the realities of poverty in South Africa. As a result of this meeting, the community leaders asked the School of Social Work to consider offering social work services in the area as none existed and there was clearly a need for them. In 1992, the first group of students together with a social work lecturer began working in Bhambayi. During the first two years very little progress was made – regular outbursts of violence put the students at risk and for weeks at a time, they were unable to enter the community. At the end of 1993, it was decided to close the student unit, as fears of increased political intolerance and violence in the run up to the first general elections increased. This decision was met with dismay by the community, and as a result of several discussions, the School of Social Work promised to visit Bhambayi after the elections, and to re-assess the situation.

As is now history, the first general elections were peaceful and the new democratic government was welcomed by the majority of South Africans. At the same time, efforts towards establishing peace in Bhambayi were underway, with the establishment of the committee comprising members of the two factions, and the building of an office on the site of the dividing line between the two factions. This positive development meant that the School of Social work could once again work in the community, and since 1995, students and staff of the UKZN have had a presence in Bhambayi.

Today, Bhambayi is home to about 10 000residents, and consists of 900 formal house,s and about 1800 informal dwellings or shacks. The homes have access to electricity, but there are no domestic water connections and the residents have to use communal taps. Roads have been built and several welfare organisations now offer social welfare services in the area. The Phoenix settlement has since been restored and at the time of writing is a tourist attraction. The school has reopened and serves about 850 children from the area.

The School of Social Work and Community Development at the University of KwaZulu- Natal continues to send social work students to Bhambayi for practice experience. Over the past few years, more and more people who approached students for help, were battling to cope with the combined challenges of HIV/ADS and poverty. The need to understand the extent of the problem became obvious, and in consultation with the community leaders it was decided to undertake a community survey. This article focuses on one aspect of the survey, that of, child support grants.

THE RESEARCH STUDY

The purpose of the study broadly was to explore the impact of poverty on households in Bhambayi. The study was multi-phased with the first phase being a quantitative household survey, and the second and third phases being qualitative studies of the experiences of poverty of residents and service providers with poverty alleviation in Bhambayi. This article derives from phase one and focuses on one sub-set of data relating to households where there were children age 14 and under, the eligible age of receipt of the child support grant at the time of research (2007).

Data for the quantitative household survey were collected by means of a personally administered structured questionnaire. This questionnaire was developed in consultation with the community development committee and revised several times. During the initial meeting with the community development committee, the types of information that the researchers and the community leaders and representatives considered useful was clarified, and a draft questionnaire was presented at a follow up meeting. The community development committee also provided feedback on the advisability of types of questions, nd the questionnaire was again revised and discussed at a further workshop.

It was pilot tested using a total of 18 households (nine from the informal and nine from the formal section of the area) in order to pretest the questionnaire and enhance the reliability and validity of the data.

The total number of households in Bhambayi is 2720, of which 920 are formal houses and 1800 are informal houses. A systematic random sampling strategy was adopted to select 180 households from each section. We acknowledge that this methodology skewedthe results in favour of the formal section. The vast geographic area that constitutes the informal sections, research budget constraints and limited human resources, all contributed to our inability to target more than 10% of the informal households.

A team of six social work students and three community members conducted personal interviews with household heads. If the household head was not present at the time of the visit, a follow up visit was done. Interviews lasted approximately one hour. However, in some cases, interviewees appreciated the opportunity to talk and share their stories and interviews ended up lasting longer. In total, 169 households from the formal section and 180 households from the informal section were surveyed. Questionnaires were checked and data cleaned by the senior researchers. The statistical package, SPSS (12.0) was used to collate and analyse the data. The sub-set of data in respect of households where there were children age 14 and under was then extracted.

The research questions which guided the analysis of this sub-set of data were:

• What is the coverage of the child support grant in Bhambayi?
• What are the characteristics of the child support grant households in terms of school attendance and hunger?

RESULTS AND DISCUSSION

Population profile

A total of 1789 people formed part of the survey. Of these, 813 (45.4%) were adults, 534 (29.8%) were young people aged 15-18 and 442 (24.7%) were aged 14 and below. These figures are lower than the 2008 mid year South African population estimates which put the number of children age 14 and under at 15.6 million or approximately 32% of the total population (Statistics South Africa, 2008).

Coverage of the child support grant

PLACE TABLE ONE HERE

Of the 442 children aged 0-14, 297 (67%) received the child support grant. Among those who received the grant, 159(53.5%) were girls and 138 (46.5%) were boys.

The child support grant is specifically intended to target poor children and is subject to a means test that mandates the caregiver’s income be less than R1100.00 per month for people residing in an informal settlement. Using this means test, the results were further analysed to determine whether or not the children receiving the grant were eligible. In considering these figures a note of caution is however necessary and the reliability of the income figures may be suspect. It is possible, for example, that the respondents under-reported monthly income. Also, in an area such as Bhambayi, income is not stable as people engage in part time and/or temporary employment. Despite this limitation, it was particularly useful to examine the figures with a view to further service delivery.

PLACE TABLE TWO HERE

The results indicate that of the children who receive the grant, 235 (79.1%) qualified for the grant while 62 (20.9%) children did not qualify. A further 112 (25.3% of the total number of children in Bhambayi) qualified for the grant but did not receive it. In total, there were only 33 (0.7%) children who did not qualify for the grant and who did not receive it.

Errors of exclusion

Errors of exclusion occur when children do not receive the benefits intended for them. In Bhambayi, a total of 347 children were eligible for the grant and 67.7% of these children were receiving the grant. These figures are somewhat lower than national figures derived from a household survey in 2007, that estimated that approximately 86% of eligible children were receiving the child support grant (Hall, 2008). Other studies have also estimated exclusion errors ranging from 20% to up to one third of all eligible children. (Hall, 2007; UNICEF, 2008)

There may be a number of reasons why children in Bhambayi who are eligible for the child support grant are not receiving it; a major one is the lack of birth certificates, which are necessary before an application for the grant can be made. Bhambayi is a resource poor area and the household based survey findings revealed that 67% of the economically active people are unemployed (Raniga and Simpson, forthcoming). Moreover, it is simply too expensive to travel to Government departments to apply for identity documents and to access services.