The experience of cash transfers in alleviating childhood poverty in South Africa: mothers' experiences of the Child Support Grant.

Cash transfer (CT) programmes are increasingly being used as policy instruments to address child poverty and child health outcomes in developing countries. As the largest cash-transfer programme in Africa, the South African Child Support Grant (CSG) provides an important opportunity to further understand how a CT of its kind works in a developing country context. We explored the experiences and views of CSG recipients and non-recipients from four diverse settings in South Africa. Four major themes emerged from the data: barriers to accessing the CSG; how the CSG is utilised and the ways in which it makes a difference; the mechanisms for supplementing the CSG; and the impact of not receiving the grant. Findings show that administrative factors continue to be the greatest barrier to CSG receipt, pointing to the need for further improvements in managing queues, waiting times and coordination between departments for applicants trying to submit their applications. Many recipients, especially those where the grant was the only source of income, acknowledged the importance of the CSG, while also emphasising its inadequacy. To maximise their impact, CT programmes such as the CSG need to be fully funded and form part of a broader basket of poverty alleviation strategies.

Introduction

In recent years, cash transfer (CT) programmes have been increasingly recognised for their important role in social protection in developing countries, and in contributing to broader global development goals as expressed in the Millennium Development Goals (MDGs), such as those relating to health, education and hunger (Basset, 2008; Devereux & Sabates-Wheeler, 2007; DFID, 2005b; Ellis, Devereux, & White, 2009; Henlon, Barrientos, & Hulme, 2010; Robertson et al., 2013; Samson, van Niekerk, & Mac Quene, 2006; Samson et al., 2008; World Bank Policy Report, 2008). Many advocates of CT programmes consider them to be a cornerstone of poverty alleviation. In the context of low- and middle-income countries outside Latin American and Caribbean Countries, South Africa is unique in implementing one of the largest unconditional CT programmes for children. Established in 1998, the Child Support Grant (CSG) was intended as an instrument to alleviate child poverty. It has evolved from a CT to children aged 0–7 from poor households via their primary caregivers at R100 (US$12) a month to widened eligibility criteria by 2011 that included children up to 18 years of age with a CT of R320 (US$32) per month.

With regard to the original assumptions underlying the design of the CSG, in particular the value of the grant, Lund, one of the original architects, suggests that the original benefit level was determined by fiscal and political constraints. Crucially, the need to sell the very idea of a child grant to politicians meant that it had to start at modest levels. Thus, while initially the intention was to link the value of the CSG to the cost of raising a child, in the end the Lund Committee had to restrict its value to food costs only. This early decision was a fiercely contested one with many in the committee believing that the amount was so negligible (then, US$6) that it was hardly worth the trouble of applying for it (ACESS, 2002; Lund, 2007). Nevertheless, the committee tried to justify the benefit level by aligning the value of the CSG to some objective measure of need, which at the time was the apartheid-era Household Subsistence Level (HSL) designed especially for Black Africans and thus very minimalist. It was hoped however that once the CSG was secured (accepted and implemented), the amount of the CT would eventually rise, reflecting a more realistic, higher measure of need (Lund, 2007). Although the CSG has generated significant interest in social policy circles as a large CT programme in an African country, it has not been possible, until recently, to conduct extensive rigorous impact studies on the CSG (Case, Hosegood, & Lund, 2004). Many of the early studies focused on the implementation and reach of the CSG. These studies reported challenges with take-up rates of the CSG relating to lack of awareness and knowledge about the requirements for applying for the grant, turnaround times for application processing (Kola, Braehmer, Kanyane, Morake, & Kimmie, 2000; Sogaula et al., 2001) and the application process itself (Zembe-Mkabile et al., 2012). Since then some of the barriers to accessing the CSG, such as knowledge about application requirements, and turnaround times for processing applications have improved (Zembe-Mkabile, Doherty, Sanders, & Jackson, 2014; Zembe-Mkabile et al., 2012). However, many children, especially infants and young children below the age of 2 have continued to have lower take-up rates than older children [Zembe-Mkabile et al., 2012; Department of Social Development (DSD), The South African Social Security Agency (SASSA), & The United Nations Children's Fund (UNICEF), 2011].

Early evidence on the impact of the CSG suggested that there is an association between the receipt of the grant and improved child nutritional status (Aguero, Carter, & Woolard, 2006), higher school enrolment (Samson et al., 2006) and reduced child hunger (Samson et al., 2008). More recent attention has focused on concerns about its potential to produce unintended effects, such as decreasing the motivation to seek and hold on to employment and increase in teenage pregnancies (Biyase, 2005; Makiwane, 2010; Makiwane & Udjo, 2007; Richter, 2009; Steele, 2006; Surender, Noble, Wright, & Ntshongwana, 2010). However, research has thus far shown no links between the CSG and lower rates of participation in the labour market (Noble & Ntshongwana, 2008; Noble, Ntshongwana, & Surender, 2008; Surender et al., 2010) with new evidence (Eyal & Woolard, 2011) suggesting that receiving the CSG is associated with increased participation. Studies examining the link between CSG receipt and teenage pregnancy have shown no such association (Makiwane, 2010; Makiwane & Udjo, 2007; Mokoma, 2008).

Whilst these studies are important in initially indicating the impact the CSG might be having on children and households, we still have limited understanding about the precise mechanisms involved or how mothers experience the CT in their daily lives. There are other questions about whether the benefit is enough for recipients to cope with the current food price inflation without a corresponding increase in its cash value. These questions are best suited to qualitative inquiry, of which there has been so far very little in South Africa (Hunter & Adato, 2007; Liziwe & Kongolo, 2011).

This paper seeks to address these questions and contribute to the evidence on the CSG in particular and CT programs in general. We report findings from an in-depth qualitative exploration of the experiences of mothers in receipt of the grant, as well as the experiences of those who, though eligible for the CSG, have not been able to access it. The study offers a comparative analysis, which allows for a more distilled understanding of how the grant is experienced by recipients.

Methods

This qualitative study is a sub-study of a large cluster randomised intervention trial (PROMISE EBF) undertaken between 2005 and 2008. For details of the trial see Tylleskar et al.'s (2011) study.

Qualitative sub-study

Analysing complex and sensitive dynamics such as the ones in this research poses several challenges. A qualitative research design was therefore chosen to produce data that are useful for both explanatory and evaluative purposes, since it emphasises the in-depth subjective views and experiences of those directly affected by the programme (Denzin & Lincoln, 2005). As with all qualitative researches, the concern was with identifying the contextual conditions favourable to examine how CTs might be effective rather than establishing causality between variables (Mahoney & Goertz, 2006). The sampling criteria were purposive rather than random with no attempt to achieve a statistically representative sample. However, efforts were made to achieve a balanced mix of respondents in terms of geographical location in order to give voice to as wide a range of experiences and views as possible.

The assumptions underlying this study are based on a framework adapted from Leroy, Ruel, and Verhofstadt's (2009) study (Figure 1). These assumptions are threefold, as follows. First, that giving cash to women leads to an increase in household income and increases women's agency – this leads to household food security and improvements in the quality and quantity of food that is available for children to eat, which leads to improvements in child nutrition and other child health outcomes. Second, in the hands of women, CTs will facilitate the use of health services, which will in turn contribute to improved child health. Third, that CTs should improve school enrolment and attendance as they enable mothers to pay for the hidden costs of accessing education (transport, uniforms, school lunch), and that they will reduce the opportunity cost of sending children to school when they could be contributing to the household income, which will lead to improved educational outcomes.

Figure 1.

Mechanisms by which CTs might affect child outcomes. Source: Adapted from Leroy et al. (2009).

Sampling

Forty-one in-depth interviews were conducted with CSG receivers and non-receivers, (who were eligible recipients) in the three trial sites and an additional fourth site in the Western Cape (Khayelitsha). Khayelitsha was added as a fourth site both to increase the sample of CSG non-receivers and to provide a pure informal settlement, unlike the other two mixed urban sites (Paarl and Umlazi). A subset of households was selected from each of the three sites for qualitative interviews. CSG receivers and non-receivers were identified from the quantitative study data and the families were contacted to invite them to participate in follow-up in-depth interviews. More than 30 families were contacted in each site and 10–15 families were available to be interviewed. In Khayelitsha, CSG non-recipients were selected through convenience sampling with the help of a Community Health Workers working in the area. Recruitment continued until we determined that theoretical saturation had occurred (i.e. that no new information was emerging from the interviews; Glaser & Strauss, 1967).

A CSG receiver was defined as a mother receiving the CSG for the ‘index’ child (the child being followed from the quantitative study), regardless of other children in the household or their CSG recipient status. A CSG non-receiver was defined by the index child failing to access the grant, regardless of other children in the household who may have been in receipt of the CSG. In total there were 21 CSG receivers and 20 non-receivers, which included one focus group interview in Umlazi Township.

The mothers interviewed ranged in age from 17 to 42 years old. Marital status rates differed by site – in the urban sites of Paarl, Khayelitsha and Umlazi, all participants were single, whilst in the more rural Rietvlei site most respondents were married. In Paarl and Rietvlei, the majority of participants had more than one child, while in Umlazi all but one participant had one child. All respondents across all sites were unemployed (except for ad hoc casual work or what is locally referred to as ‘piece jobs’).

Data collection and analysis

Interview topic guides were developed and piloted in 2011 and conformed to best practice in qualitative research methods literature (Mays & Pope, 1995). All interviews were undertaken by the lead author in the local Xhosa/Zulu language of the participants. In all cases, the child's mother was the interviewee.

Data were transcribed and translated into English and checked against the original recording to ensure accuracy. Following each interview, field notes were written to capture the context, home environment and non-verbal communication. These were analysed after each interview and used to guide further interviews where appropriate. All data were analysed using Graneheim and Lundman's (2004) manifest and latent content analysis methods. Together with some of the co-authors, the lead author read through each of the transcripts, noted initial thoughts and began manifest coding of the data. Initial codes were grouped together into categories that were then further transformed into major themes. All authors discussed this first phase of the analysis, and once there was consensus, the codes, categories and themes generated from this analysis were then used in the analysis of the remaining transcripts.

Results

Respondents were asked about their experiences of raising their children with or without the CSG. Benefit recipients were asked further probing questions about how they used the CSG, their views on its ability to meet the needs of their children and the CSG in the context of other sources of income. Four major themes emerged from the data: (1) barriers to accessing the CSG; (2) life without the CSG; (3) how the CSG is utilised and the extent to which it makes a difference; and (4) the mechanisms recipients used to supplement the CSG.

Barriers to accessing the CSG

Two types of CSG non-receivers were interviewed: the first, mothers who were not in receipt of the CSG for the index child and had no other children in receipt of the CSG; the other, mothers who were not in receipt of the CSG for the index child but in receipt of it for other children in the household. Unsurprisingly, the mothers who were not in receipt of any CSGs felt the absence of the grant more keenly than mothers who were in receipt of some CSG benefits.

Reasons for non-receipt ranged from administrative factors such as not possessing a birth certificate, or an identity document (ID), to institutional capacity including long queues, delays and errors at the social grants administration offices:Participants identified several barriers in the process of applying for the grant, which delayed, and sometimes prevented, access to the CSG. Stories of having to rise in the early hours of the morning to queue at local grants offices and then having to wait for up to six hours for the offices to be opened, with no guarantee that they would be allowed to submit their application, were common. One participant described the experience of making such repeated trips at least five times before she was eventually allowed to submit her application. Making these trips meant that she had to depart her house with her husband leaving their small children sleeping alone:Similar experiences were recounted in Paarl and Umlazi: Respondents also reported incidents of corruption by administrators of the CSG, and how this becomes a further barrier to access:

The [hospital] card got lost … when I went to the hospital, they said it is difficult to make one, you are supposed to have three affidavits, now then it got lost with his father, now when it is made by him, it is not possible because his surname is not the same as mine, and because of that she does not get the grant. (RT007, CSG Non-receiver, Rietvlei KZN)

I have had to go there to try and apply for him every week from when he was born … [we get up at] two in the morning, if you fail to get up at 2[am] and wake up at 4[am] maybe, then you might not be served that day, even now I was served because I got up at 2am. We go on foot, we walk to (section) V … you get there and you wait outside the yard, we light a fire, you wait until they open around 8 am. (KLT001.CSG Non-Receiver, Khayelitsha Western Cape)

Imagine waking up at four and queuing at the hall outside, then when you get there they tell you that your thing is wrong, go back and start afresh … so you become demotivated, you quit [trying]. (PA005, CSG Non-Receiver, Paarl Western Cape)

[T]hey get there early because the queues are always very long … people are always fighting, fighting about queuing, like people who jump the queue because they want to get this money. People wake up early to go there, at 3am, I know people who go there very early, because there is no food at home. (M1, FGD, Umlazi, KZN)

One of my siblings has even given up trying to get the grant because each time she goes there they say ‘no you need to come back with this, you didn't bring that,’ and she says ‘but how is that because I have all the documents that are needed’ … The problem is that those people [officials] want money, that is why they are sending her on the run-around … if you do not have money you will never get that grant. (M2, FGD, Umlazi, KZN)

My grandmother said you have to pay R150 to some officials there who help you get grants. That is how it is in my community, for instance with me … I went there for … three months without receiving it. (M4, FGD, Umlazi, KZN)

Life without the CSG

Non-recipients of the CSG experienced the absence of the grant to different degrees depending on their access to other sources of income from which they could draw. Naturally, mothers would also use the CSG(s) of other children in the household to meet the needs of the children not receiving the grant:The absence of the CSG was of course more keenly felt in households where there was no other source of income. In one household, a mother talked about the experience of living without the CSG, which she had been receiving for her three children, after it was suspended:Non-recipients experienced regular shortages and relied on social support networks. One young mother related a story about seeking help from friends when her food runs out:A young mother who was receiving a disability grant (DG) for herself but no CSG for her child expressed the sense of panic she experiences towards the end of the month when she runs out of supplies for her baby:

I buy for the one who does not get the grant … I pay his money for transport, R80, then I also buy the things for his lunch box … he is the one who is schooling. (RT001, CSG Non-Receiver, Rietvlei, KZN)

There will be a big difference [when she gets her own grant] because all this time she has been surviving on the other children's money. I used this money to buy her formula milk at the time when she was still feeding on milk … I took it from this money. (RT007, CSG Non-Receiver, Rietvlei, KZN)

But this month they go without food, they go to school. They leave this house with no food, they wash with whatever they can wash with, and they leave for school because their future has to be bright … with no soap to bath with, no washing powder for their school uniforms when they get back from school, they come back wanting to wash their school uniforms, no powder (RT010, CSG Receiver, Rietvlei, KZN)

Sometimes I take [baby] to her father's place … but there is no difference, still she comes back with nothing, and then I ask my friend [for help] and she gives me. So … I have to pay her back (crying). (UM007, CSG Non-receiver, Umlazi, KZN)

When I can see that the formula milk and diapers are running out, I think where am I going get them from … and what if she gets sick now and I need to take her to the clinic? (UM003, CSG Non-receiver, Umlazi, KZN)

How the CSG is utilised and the extent to which it makes a difference

Those in receipt of the CSG reported that it is mostly used to purchase basic food items. However, most reported that the CSG was only enough to buy staples and that expensive food items such as meat were only purchased on ‘pay-out day’:The importance or role of the CSG was typically discussed in the context of meeting important basic needs for food; but sometimes its significance was apparent in meeting seemingly non-basic needs, such as treats for children:It was striking to find that similar to non-recipients, many recipient mothers also experienced frequent food shortages. Strategies for coping with food shortages included attempts to find ad hoc work, and more usually, borrowing money from friends and neighbours to survive until the end of the month:The next main item besides food, on which the CSG was spent, was school-related needs. Mothers used the CSG to buy school uniforms, food for lunch boxes and to pay for school transport, activities and trips. Mothers acknowledged that it would be a struggle to meet these needs without the CSG:What also came across clearly was that recipients’ mothers prioritised payment of school-related expenses over any other need in the household:Another mother talked about spending most of the first CSG back pay that she received on school- or crèche-related expenses for her child:Interviewees were asked about what they did when a child was ill in the household. Mothers’ responses reflected different strategies of coping with illness, depending on their distance from health facilities, the ability to put aside a portion of the CSG for savings for health care and access to credit.

I buy groceries, mielie-meal, rice, flour, sugar, five litre oil, I buy potatoes and meat … but we don't buy much meat, just something for that moment which finishes up fast … and salt and soups. (RT009, CSG Receiver, Rietvlei, KZN)

I am not able to buy them treats, you just buy food, you buy mealie-meal … potatoes … rice … sugar … oil, fish oil (cooking oil), salt, soups, by then there is no more money, you buy matches and candles so that you can have light at home … I can buy meat on that first day, but by the second day I cannot. (RT008, CSG Receiver, Rietvlei, KZN)

One of the kids had their birthday recently and I had to buy her cake, I had saved a R100 [from the CSG] and put it away for quite some time so that I could buy her a cake Monday … I thought that I should save it for her because it's her money. (PA005, CSG Receiver, Paarl Western Cape)

Yes the food is enough but then by month end we are pulling hard. When you have just received the money food can be purchased … but towards the month end the food runs out, there is none of it … [then] we hustle with piece jobs that we get in [people's] houses … and then we manage to eat until month end. (RT008, CSG Recipient, Rietvlei, KZN)

Yes it [food] does get finished before the month ends, and I have to borrow from someone, and say, ‘just give me something, maybe R20 so I can buy some mielie-meal here at Shoprite.’ (PA005, CSG Receiver, Paarl Western Cape)

I also buy the uniform, the firstborn; I buy his uniform with the grant money. (RT002, CSG Receiver, Paarl Western Cape)

We buy clothes in January and December. The only time they get them (clothes), they get them in January for school and in December for Christmas…in order to buy food we go to look for piece jobs just because we have to buy uniform with their money. (RT008, CSG Receiver, Rietlvei, KZN)

You see, when it [first] came … it was about R500 because it's for three months. I paid her crèche … It's R60 per month. I paid it for two months [because] I owed them R120. I bought ‘carry’ things (lunch boxes) with the remainder. (PA007, CSG Receiver, Paarl, Western Cape)

Those who lived far away from clinics talked about their attempts to put aside some of the CSG money each month in case their child needed to go to the clinic:However, many acknowledged that it was often difficult to put money aside. In such cases, when a child fell ill and there was no money to take them to the clinic mothers would borrow money to pay for these expenses. Health needs were also frequently left unmet as relayed by one mother:Most participants complained about the cash value of the CSG asserting that it was very small at R260 per month (then US$35). Some mothers had very strong negative feelings about the amount of the grant and refused to identify anything positive about it.In most interviews, mothers struggled with their ambivalent feelings about the usefulness of the CSG in raising their children. On the one hand, they acknowledged the vital role it played in their households and how it was often the only source of income, while simultaneously lamenting its inadequacy in meeting the needs of their children:

I try to save it so that if a child is sick, I have it. (RT011, CSG Receiver, Rietvlei, KZN)

… she had a toothache and was sent back from school, she was asked why she doesn't tell her mother to take out the tooth, she told them she hasn't received her grant for the month yet. She had this tooth ache even the night before … and I knew I couldn't go and get it taken out then. (PA005, CSG Receiver, Paarl Western Cape)

Oh, I don't really count it (the child support grant) … it (the CSG) doesn't really help me … That R260 is nothing, it is nothing, when you get it, it gets finished quickly on children's things. (PA006, CSG Receiver, Paarl Western Cape)

So we use R500 for a whole month in this house, but I thank the government even there, for the way he helps us with it, because otherwise we wouldn't even know what to mix with what, but we would like to ask the government to please increase this grant amount, because the children use this same money for school, for food, when they are sick, all on this same money. (RT010, CSG Receiver, Rietvlei, KZN)

I manage to bring them up with it although this money is too little, it is not a lot of money at all, but they are able to grow. (RT002, CSG Receiver, Rietvlei rural village, KZN)

The role of additional grants in supplementing the CSG

Despite its relatively modest value, the CSG was reported to have a useful effect in contexts where it was combined with other grants either when there were several CSGs being received in a household or and especially when there was a mix of CSGs and one other ‘big’ grant such as a DG or an old age pension (OAP) (each with a cash value of about R1370 (US$137) a month). In these households, the CSG seems to have a multiplier effect and an accumulative impact of the pooled income. In particular where the CSG was present with a DG or an OAP, mothers seemed to feel less financial pressure in meeting the needs of their children. One mother who is receiving a DG and three CSGs said:Another mother who is receiving three CSGs and a Child Dependency Grant (equivalent to a disability grant) added:Even in families where there was no other source of income or other big grants, multiple CSGs within a household (i.e. more than one child receiving the CSG) appeared to be enough to stave off the worst levels of destitution:

No we meet the needs of our family, I cannot say we do not. No I manage. There isn't anything that I think causes me stress. (KLT001, CSG Receiver, Khayelitsha, Western Cape)

… there is a huge difference now, even whilst still earning the children's one, only it was helping me, but now that I am getting grants for all of them, it helps me a lot. Especially since the older one's dependency grant has increased. (RT002, CSG Receiver, Rietvlei rural village)

We live on this children's grant … the father is unemployed, he just holds piece jobs and I am also unemployed. They are being raised by this grant…the main source of financial support is this grant. (RT009, CSG Receiver, Rietvlei, KZN)

… they (CSGs) have made a huge difference because they eat, I clothe them, and things that are wanted at school, they come and say and it then gets done for them. (KLT001, CSG Receiver, Khayelitsha, Western Cape)

The role of CSG in social support networks, access to credit and informal social insurance

On the whole, mothers confirmed that without regular financial support and assistance from relatives or from the fathers of their children, it was difficult if not impossible to manage.Systems of reciprocity and informal social support networks were crucial both at times of unforeseen expenses such as medical bills or more routinely when these mothers experienced food shortages:It was interesting to note therefore how the CSG intersected with these traditional and well-established systems. Many mothers reported that the existence of the grant allowed them greater capacity and leverage to borrow because having a CSG made them credit-worthy in their communities as there was a guarantee that they would pay back their loans. In particular, at the end of the month when the CSG ran out they were able to borrow money in the knowledge they would soon have incoming income again:Some however viewed the enhanced credit-worthiness that came with receiving a CSG negatively. Specifically, some mothers described how loan sharks wait for CSG receivers who owe them money at the hall where the grant is paid out, and because they have exorbitant interest rates, they take the entire amount before recipients are able to purchase anything:Finally, some CSG recipients were able to participate in more organised (though still informal) insurance schemes such as burial societies, and others were able to participate in stokvels (savings clubs). The CSG recipients who reported participation in a savings club papered to mostly utilise them in order to be able to purchase clothes for their children:The ability to save a portion of the CSG in local savings schemes was not the same for all of the mothers. Thus for every mother who described their ability to save some of the CSG in stokvels, there were many mothers who talked about the difficulties of saving any money from the CSG each month:

… it's better when you have a partner, [because] there is someone who is helping you … When your partner is not present, it becomes very difficult … perhaps at times you are not working you are dependent on this R260. (PA007, CSG Receiver, Paarl Western Cape)

[T]hat sort of thing [running out of food and supplies] we cannot avoid, it does happen, it truly becomes like that and you realise that now times are tough, I am struggling…but my sister gets paid every week so she manages to cover whatever we run out of. (PA005, CSG Receiver, Paarl Western Cape)

Yes, food gets finished and you find yourself having to figure out what to do, maybe for those few days you have to borrow from the stokvel club [savings clubs]. (RT001, CSG Receiver, Rietvlei, KZN)

If there is an illness you know that you can go and borrow money because you will get the grant at the end of the month, like even loan sharks in our areas need proof that there is some money that you are getting before they give you money…leave them your grant card so they keep it until pay-day where you get the grant and pay them back their money. (M6, CSG Receiver, FGD, Umlazi)

They wait for them there at the hall. I was watching this other time, there was a girl who was fighting there outside. They said this money is for the children, how can you take all the children's money? It's the loan sharks that I really see as a problem in this grant money. But it is because we are hungry … (PA002, CSG Receiver, Paarl Western Cape)

Yes when I think I'm in a tight spot I borrow. But it's not right because this is not my money, [but my children's] but I borrow because this money is spent on them. (RT002, CSG Receiver, Rietvlei, KZN)

I also participate in a stokvel, R150 every month … at the end of the year [with the money] I first buy clothes for my children. I then buy groceries with the remaining money. (RT003, CSG Receiver, Rietvlei, KZN)

It's impossible to save right now, you use it and you have R100 left for an example. With the R100 you can maybe buy bread, bread is R10, that's R10 a day, so in the second week the money is finished … so you can't have a lot of money left over from that [the child grant], it's too little. (PA006, CSG Receiver, Paarl, Western Cape)

Discussion

It is important to recognise that CTs are only one of a number of anti-poverty strategies available to governments in the global South. In particular, recent research is beginning to demonstrate that in contrast to South Africa, others of the so-called ‘BRICS’ countries (especially China) appear to be guided by a social development approach which establishes stronger links between the economic and social dimensions of development rather than traditional donor organisations (Urbina-Ferretjans & Surender, 2012, 2013).

Nevertheless, it is equally true that in contrast to current BRIC orthodoxy, since the turn of the twenty-first century, there has been a steep ‘and rapid rise of social protection up the policy agendas of developing countries’ (Devereux & Sabates-Wheeler, 2007) including those of sub-Saharan Africa. Previous scepticism (dominant in the era of structural adjustment programmes) about the affordability and appropriateness of social expenditure and transfers in development contexts has, it seems, given way to a discourse of ‘pro-poor’ and ‘transformative’ social protection (Andrews, 2009; DFID, 2005a, 2008; ILO, 2008). Rather than being viewed as an obstacle, this social dimension is now increasingly viewed as an important prerequisite for successful development (Surender & Walker, 2013).

Thus, support for CTs and social protection in general is growing within the region as evidenced by the 2006 Livingstone Call for Action Summit, the 2008 African Union's Social Policy framework and the 2012 World Bank's 10-year strategy for social protection in Africa which notes that ‘cash transfers are an important tool for addressing the regions’ development, poverty alleviation and human rights aspirations’ (World Bank, 2011). Given these developments, in particular in the context of the post-2015 MDG agenda, it is an important juncture to examine whether the emphasis on CTs, prominent in South Africa, has proved in practice to be an effective strategy for the delivery of welfare for the poorest.

In this context, this qualitative study, conducted across diverse settings, which explores families’ experiences of receiving or not receiving the CSG in South Africa shows mixed results. For many of the families, the CSG was the main source of income, without which they would have no means of meeting their children's needs. Further, the findings show that the CSG is an important source of financial support for all families, irrespective of geographical location or household composition, providing for basic needs such as food, schooling and health care. The grant expands the roles of recipients as consumers, opening up otherwise unavailable opportunities to feed their children, mitigate financial crises and participate in local savings schemes. Further, the grant allows access to reciprocal exchange networks such as informal credit, to alleviate their poverty and manage vulnerability.

The central role the grant plays in ensuring survival for many poor families is further underscored by the poignant experiences of destitution related by those who, though eligible for the CSG, were not in receipt of it. While major improvements have been made in the turnaround time for processing applications, our findings show that negotiating the queues and long waits to submit an application are still major problems. Harrowing experiences of leaving their children unattended in the early hours of the morning to travel back and forth to social services offices and enduring long queues to try and submit applications were common.

However, in the context of overwhelming poverty as experienced in the study communities, it is clear that the CSG is not adequate on its own to meet even the most basic needs of its beneficiaries (i.e. food costs), for which it was established. Most early studies highlighted the barriers to accessing grants and the subsequent lack of take up as one of the main factors explaining why the CSG has had such a minimal impact on poverty (Goldblatt, Rosa, & Hall, 2006; Lund, 2011). However, our study shows that even when the programme is being implemented as intended and families access the grant, the contribution of the CSG to childhood poverty alleviation is small.

The results show that the CSG is mostly used to buy staples, and is insufficient for greater dietary diversity, with many mothers frequently running out of food before the end of the month. These findings concur with a recent study that reported worrying levels of food insecurity among CSG recipients (Patel, Hochfeld, Moodley, & Mutwali, 2012). In order to manage, families pool resources such as additional social grants, utilise social support networks to help meet unexpected expenses and access informal credit through savings clubs and burial schemes.

It is now well documented that the poor are particularly vulnerable to crises and ‘shocks’ which reproduce and aggravate the effects of poverty in ways that undermine whatever little income they may have (Devereux, Marshall, Macaskill, & Pelham, 2005; Ellis et al., 2009). As such, poor households suffer most during times of crisis since they generally do not have sufficient savings or insurance to cushion them against unexpected expenses such as burial costs and health-related costs (Ellis et al., 2009; Kongolo, 2007). In this study, mothers reported that when unexpected expenses occurred, such as an illness or school-related costs, they were forced to borrow or to adjust the household budget by reducing food. Most respondents in this study reported that they had to borrow each month to make ends meet, and this rendered families vulnerable to loan sharks.

Several factors seem to explain, at least in part, the diluted impact of the CSG in the lives of its recipients. First, in this study, the CSG was typically introduced in a context of widespread unemployment, where the grant became the only predictable source of income, even though this was not the intention of the CT programme. Previous research shows that in South Africa in the presence of high unemployment rates, the poor cluster around social grants, particularly the CSG and the OAP, where these grants take care of not only the needs of direct recipients, but entire households (du Toit & Neves, 2009b; DSD et al., 2011). Proponents of a universal income grant for working-age adults have argued that the clustering of the poor around social grants is a result of generalised poverty and high unemployment rates in South Africa, and that a basic income grant would reduce the current pressure on social grants and would result in CTs benefitting the specific groups they were designed for (Frye, 2005; Noble et al., 2008).

Second, while the architects of the CSG planned to link it to the cost of raising a child, in the end budgetary constraints limited its goals to food costs only (Lund, 2007). However, as these findings suggest, maintaining the grant at R280 (US$30) per month per child means that whilst reaching more children, it often only partially covers the food costs of its beneficiaries, in part because the cash value of the CSG has not increased at the same rate as inflation and food prices (Lund, Noble, Barnes, & Wright, 2009). These findings support the argument that to maximise the potential effectiveness of CTs, their value should be linked to food price movements and the cost of essential non-food items, and adjusted for household size (Devereux et al., 2005).

Third, it has been suggested that the effectiveness of child CT programmes in developing countries depends, in large part, on their linkage to other anti-poverty measures. In this paper, although CSG recipients had access to free primary health care services, those who lived far from such facilities had to use the grant to pay for transport. Similarly, while education is free, CSG recipients still had to pay for other indirect school-related costs such as school lunches and school uniforms. This is a classic feature of developing country contexts: services may be free but the associated costs of accessing them still have to be borne by the poor. In South Africa, the CSG was originally intended to be ‘one in a basket’ of poverty alleviation strategies (Lund, 2007), and while other interventions and services aimed at supplementing and complementing the CSG do exist (e.g. the school feeding scheme, fees exemption for CSG recipients, free school transport in rural areas), their implementation is patchy and uncoordinated. This is also true for other poverty alleviation strategies that have been implemented as major alternatives to social grants such as the Expanded Public Works Programme (EPWP). The EPWP was established in 2004 as the largest development programme in South Africa, tasked with reducing poverty, unemployment and skills shortage (Mccord, 2003; McCutcheon & Parkins, 2009). However, the EPWP has largely failed to achieve its objectives and has been dogged by issues of low remuneration and very transient, unreliable employment only for a few (McCutcheon & Parkins, 2009). Ultimately, the findings of this study point to a number of lessons that inform our understanding of the CSG and CTs more broadly. One of the key lessons is that even though small CTs make a difference in the context of abject poverty, their adequacy should not be overstated. Second, choices made at the design phase of a policy programme continue to affect its shape, form and impact long after its implementation, and such choices are difficult to reverse. This is especially true for the CSG benefit level. Although the CSG was designed with the hope that its value would increase significantly with time, this has not occurred in terms of its real value, and the fact that it started so low has determined its value over the years even though it is clear that this is too small to make a significant dent on child poverty.

Fourth, our findings are consistent with and help explain why the policy efforts of the post-apartheid government have not been able to lower, in any significant way, the very high levels of poverty and inequality overall (Leibbrandt, Woolard, Finn, & Argent, 2010), despite its strong commitment to upholding the constitutional goals of social rights and justice. Poverty levels have decreased only slightly since 1994, and after nearly a decade of social assistance programmes, the proportion of the population in poverty (using any measure) has remained virtually static. Overall income inequality actually increased from 1994 to 2008 (Gini coefficient increase from 0.66 to 0.70 during this period) and continues to persist at acute levels for Black African and Coloured racial groups (Gardin, 2011; Hall & Woolard, 2012; Leibbrandt et al., 2010).

Thus, while the CSG is extremely important in the composition of household income for the very poorest (two-thirds of the income for the bottom quintile comes from social assistance), and it succeeds in reducing the poverty gap for this group, the findings from this study help to illuminate why its effect on poverty remains diluted.

At first glance, our findings may seem contradictory as they, on the one hand, highlight the importance of the grant in poor households, but on the other, reveal its inadequacy in meeting the needs of recipients. We argue that such seeming contradictions demonstrate the complex nature of small CT programmes trying to address big problems. In contexts of extreme poverty, any small amount of money is welcomed and will be experienced as crucial in ameliorating the worst effects of poverty, but this does not negate the reality that small CTs are limited in their ability to significantly contribute to poverty alleviation. Indeed, other scholars have pointed out that the importance of the CSG:

should not give the impression that these minimal forms of protection are sufficient for addressing chronic and structural poverty … social protection can ameliorate poverty for some, but it is not a solution on its own to the massive inequality and continuing poverty imposed by the skewed and relatively unredistributive nature of the South African economy … left to itself, social protection can only offer a kind of ‘sustainable poverty. (du Toit & Neves, 2009a, 24)

Conclusion

The CSG is clearly a crucial source of income for poor families. Eligible children who are not in receipt of this grant have no cushion against the worst effects of poverty and destitution, thus the factors that continue to prevent access to this important form of social protection must be addressed. However, even as the CSG contributes to childhood poverty alleviation in South Africa, it is by no means a magic bullet. As such, it makes more sense to align expectations of the CSG's performance with the original intention with which it was established, that is, as one in a basket of poverty alleviation strategies targeting poor children (Lund, 2007). Viewed with this lens, the CSG is playing an important role in the lives of many poor children and their families in South Africa. However, to lift children out of poverty and subsequently improve child health outcomes, the ‘basket’ needs to be filled, and established strategies such as nutritional support, free school transport, free access to health care, the EPWP, broader infrastructural development in remote rural areas, and other livelihood options need to be better synergised. Further, a more comprehensive social security system for unemployed adults is needed.