Annie Barber *

Introduction[1]

This research is based firmly in the context of the HIV/AIDS epidemic in South Africa. The purpose is to investigate the role of Savings and Credit Groups (SCGs) in helping poor, rural households affected by HIV/AIDS to achieve consumption smoothing and thus improve coping strategies as well as contribute to women�s empowerment within and outside the household.

Until recently, the AIDS epidemic in South Africa faced political neglect and denial. The South African government refused to make antiretroviral medication (ARVs) publicly available until 2008 - the end of former President Mbeki�s rule. This neglect fed an epidemic that has reached disastrous proportions: today more than 5.6 million South Africans are HIV positive[2]. This is more than any other country in the world. Every day around 1000 of HIV/AIDS affected people die, while at the same time, more than 1200 people become newly infected (Gow, 2010). Within this, the virus has been found to disproportionately affect women. The prevalence of HIV among young women in particular is three times higher than it is for men of the same age in South Africa.

The result of this neglect is not only epic proportions of people living and dying with HIV/AIDS. It has also shaped and distorted the country�s response to the epidemic. With the absence of medical treatment available to the public, the focus of all South African AIDS activism over the last decade has been ARV roll out (ibidem). With most stakeholders looking to the health sector for solutions, there is significant neglect of other HIV/AIDS strategies. Despite common consensus around the links between poverty and HIV (Donahue, 2000), there is a lack of practical livelihood and coping strategies for people to weather the lingering and debilitating shocks caused by the disease.

The principal policy implication of this paper is that the role of SCG membership for the coping mechanisms of women affected by HIV/AIDS is not well recognised, and is underutilised in the policy arena. It aims to demonstrate how SCGs offer an important strategy to mitigate the effects of HIV/AIDS on heavily affected communities and households. This case study documents the experiences of a successful programme, demonstrating that SCG membership is allowing people, and particularly women, affected by HIV/AIDS (PABHA) to employ stronger coping mechanisms in times of crisis and avoid strong reliance on family, neighbours or high-interest debts. Credit received is used for consumption smoothing as well as for productive activities, suggesting an increase in economic stability.

SaveAct SGSs members, Eastern Cape

The model

Founded in 2005 in South Africa, SaveAct provides support for the formation of SCGs to almost 10,000 members across two provinces (KwaZulu-Natal and the Eastern Cape). The model was developed through working with women, with the intention of providing a potential platform for women to change gender relations. Women�s empowerment, as well as enabling support for vulnerable groups, such as PLWHA is at the heart of this work. Each SCG consists of up to twenty-five members with every member receiving basic financial education before beginning their first savings cycle. SaveAct also offers an enterprise course for members aiming to develop an income-generating activity (IGA).

The model employed by SaveAct is savings-led and involves no external credit[3].Embedded in local knowledge, it builds on the old traditional stokvel practice in rural areas but involves strict adherence to rules and procedures with tight monitoring and mentorship from SaveAct Field Officers. Each SCG is self-forming. Where there is a strong groundswell of demand, each self-selecting group is supported by the SaveAct field officer to elect its own Management Committee, including a chairperson, bookkeeper and key-holders. Every SCG formalises its own individual Group Constitution outlining the basic rules of practice. To save, members purchase shares in the group - the price of a share (normally R50 � R100) is agreed in the Constitution. By providing a formal savings mechanism for poor people this model challenges the myth that �poor people can�t save�. In fact, poor people save more frequently than any other group, precisely because they must (Collins et al, 2009). Members can purchase between one and five shares per meeting but are not obliged to purchase shares every month. A record of share-purchases is kept by both the bookkeeper, Field Officer and is recorded in each member�s individual �Share Book�. Each savings cycle lasts for twelve months with members meeting on monthly. The capital accumulated through the purchasing of shares is distributed between members through small loans on request and approval of the group throughout the year. A rate of interest decided in the Group Constitution (normally 10% per month) is charged for each loan. Loans can be paid back over a three - four month period. The total accumulated capital, including the profits made on loan interest is shared between members in an annual lump-sum according to the amount of shares held by each member. Average annual profits made on savings in SaveAct groups are 40% but can be as high as 80%. Member equity exceeds $200 on average, with annual capital share-outs reaching up to $600.

Each SCG also builds an Emergency Social Fund (ESF) by contributing a small monthly amount agreed in the Constitution (usually R10.00, equivalent to 1.2 USD). The ESF is used to assist group members in times of crisis or hardship, the exact terms of which are specified in the Group Constitution. The ESF is given as a grant and requires no repayment. At the end of a savings cycle, any remaining money from the ESF may be shared out or kept for the next cycle depending on the wishes of SCG members.

The model and HIV/AIDS

In recognition that the lives of their members unfold in a complex and multifaceted context and that AIDS is a part of this context, SaveAct made the decision to partner with local HIV/AIDS-oriented Community Based Organisations (CBOs) to specifically target PABHA. All of SaveAct�s Eastern Cape implementing partners have HIV/AIDS expertise; however all HIV/AIDS-linked SCGs are treated like any other and are given no HIV/AIDS-specific support from SaveAct.[4]

HIV/AIDS-affected households are more likely to face recurring and debilitating financial shocks that can push households deeper into a poverty trap. Recurring bouts of illness and death erode the financial resources of immediate and extended families as they pay for multiple hospital trips, medical expenses, and funeral costs (Donahue, 2000). The death or sickness of a principal breadwinner affects the wellbeing of the whole household as well as any connected supporting households.

In the event of an adverse shock such as death or sickness, households with no effective safety-net are likely to employ one or more of the following strategies:
- Reduce food consumption (e.g. by skipping meals and reducing quality of food);
- Cut back on non-emergency health costs;
- Remove children from school or delay payment of school fees or related costs;
- Sell valuable assets;
- Burden kinship, extended family and community members with requests for assistance;
- Borrow from mashonisas (local informal money lenders) at usuriously high interest rates

SCG membership can help HIV/AIDS-affected households to achieve consumption smoothing and thus avoid employing any of the negative and potentially irreversible coping strategies outlined above. The goal is that households build up their asset and savings base so that transitory income shocks do not affect consumption patterns (Murdoch in Zeller, 1999). SCG members have cheap and immediate access to credit so that they avoid resorting to the sale of assets and borrowing at exorbitant rates, a strategy which could destroy future productive capacity (Donahue, 2000).

Local context

This study is based on SaveAct�s Eastern Cape activities targeting PABHA. Based in the remote rural town of Matatiele, the Eastern Cape office serves a total of over 3,500 SCG members with one Programme Coordinator, two Field Officers and five small-scale implementing partners, monitored and supported by SaveAct staff.

SaveAct targeted PABHA through strategic partnership with local HIV/AIDS[7] CBOs. SCGs were formed on top of existing Home-based Carer (HBC) and Support Groups (SG). These SCGs then grew organically, allowing some new members to join. SaveAct supports four SGs and six HBCs, serving ten groups with a total of 158 members . Three groups were formed in 2008, five formed in 2009 two in 2010. Average group membership is fifteen people, with a gender ratio of 88% female: 12% male. SCGs are in six local sub-village level areas in the Alfred Nzo and Ukhahlamba Districts of the Eastern Cape including, Matatiele, Xaxazana, Khoapa, Mpharane, Khubetsoana and Dikamoreng. These areas fall under the Matatiele and Elundini Local Municipalities.

With 74% of its population living below the poverty line, the Eastern Cape has the highest rate of extreme poverty in South Africa (Alfred Nzo District, 2011). In the Alfred Nzo district, 82.3% of the population live in poverty. Poverty levels are found to be rising in the Eastern Cape, from a provincial average of 71% in 2004 to 74% in 2011 (ibidem). The root causes of high poverty rates in the Eastern Cape, a former designated �homeland�[6], are linked to the policies and practices of the apartheid regime. Even in the post-liberation era, rural areas in the Eastern Cape show the effects of apartheid planning and are characterised by economic marginalisation and a distinct lack of services and infrastructure. This is also in large part due to the sheer isolation and generally lower economic potential of the area.

With regards to the employment pattern, entrepreneurial levels are low and most households rely on seasonal migrant work and remittances (ibidem). With South Africa�s high general unemployment rate, the only viable livelihood option for the poor in Alfred Nzo and Ukhahlamba is self-employment. This includes mainly seasonal petty IGAs and small-scale agriculture. Many households in Alfred Nzo and Ukhahlamba experience severe economic instability, with periodic shocks that lead to debilitating spirals of debt and hunger. Financial literacy levels are low and many fall prey to the economically crippling interest rates of loan sharks.

The area experiences high levels of HIV/AIDS and associated opportunistic infections such as Tuberculosis, resulting in low life expectancy levels. Access to quality healthcare is scarce and is made difficult by the vastness of terrain, poor access roads and lack of public transport. HIV/AIDS support has been particularly poor. There are now steps in place to try to redress the situation, however a large portion of those who should be on ARVs are not currently on treatment. The HIV/AIDS infection risk is exacerbated for women by high levels of gender inequality and gender-based violence (GBV). GBV and gender inequality makes women largely powerless to ensure safe sex and condom usage, potentially accelerating the spread of HIV. There are also high levels of female-headed households in the study area (FHHs) (58%), leaving women and children to carry a significant burden of the effects of the epidemic.

A significant number of women in the area head households because their partner has taken a job as a migrant worker. During the study, many women told stories of their battle for financial support from these partners. While some men remitted money regularly, other women reported that financial support was sporadic and unreliable. Many young women in the study were single parents, receiving no financial or other support from the fathers of their children.

Methodology

Of the 158 members[5] served, a one-third sample size of fifty-three SCG members was selected. Of the participants, 88% were female and 12% were male, reflecting the overall gender dimensions of all relevant groups. All the participants belong to households that are poor or very poor and are reliant on a combination of social grants, remittances, sporadic piece jobs and small IGAs. All are affected by HIV/AIDS and are vulnerable to shocks related but not limited to the epidemic. Among the participants, 33% were openly HIV-positive themselves. 14% were supporting somebody with HIV internally, and 52% were supporting one or multiple people in external households. 47% of households were either fully or partially supporting children orphaned by AIDS.

Research participant, Xaxazana

Research techniques follow a mainly qualitative approach. Four principal research tools were employed with all fifty-three participants: seasonality analysis[9], crisis trend analysis[10], loan usage ranking[11] and general questionnaires. A further twenty-one of these participants were randomly selected for semi-structured individual interviews (SSIs). Each tool uses Participatory Rural Appraisal (PRA) methods to uncover the complexity and messiness behind the coping strategies of PABHA. The two complementary tools, seasonality analysis and crisis trend analysis, were used based on a study of consumption smoothing by H�gman (2009)[12] and formed the basis around which the other methods were chosen. These tools were combined with field observations by the researcher and in-depth discussions and interviews with key staff members and Field Officers.

Results

A seasonality analysis of income, expenditure, savings, and credit
Income flows vary quite substantially throughout the year. Through SSIs and discussions, most participants identified monthly social grant payments as income sources, topped up by irregular money from remittances, local piece jobs and (seasonal and sporadic) IGAs. Some members relied on a small salary from work as a home-based carer, which is paid irregularly and erratically by the Department of Social Development. Income is highest in March, June, September / October and December. Between March and May some vegetable and home-grown crop harvests occur contributing to a slight rise in income. Many informal stokvel groups have an annual pay-out between October and December. Many SCG share-outs also occur around this time.

Expenditure patterns are very volatile with dramatic rises in December and January and significant dips between March - May and August - September. According to information from SSIs, expenditure is high around year-end because of the festive period involving Christmas and New Year, as well as the Xhosa �Ukolusa� Coming-of-Age Ceremony[13], which requires significant investment. In January, most participants pay the highest amount of school-fees and related costs such as new uniforms, sports clubs and excursions; this was cited as the single most difficult time of year in SSIs. Expenditure then dips between March and May because of crop harvests meaning that participants spend less money on food over these months. They also reported that this is an easier time of the year because of mild weather. Expenditure rises from May - July as members prepare for the harsh winter months, purchasing warmer clothes and fuel. November is the ploughing season for many and so money is spent on purchasing / hiring equipment leading up to this month.

It is interesting to note that while savings levels could be expected to closely track income, the data line is the least volatile of all, following a relatively stable pattern throughout the year. This gives an indication of the commitment of the target group to their savings activities. It is not clear from this research whether this is in any way connected to HIV/AIDS linkages and thus the potentially increased need to develop an effective safety-net. Savings levels can be seen to rise gradually towards the year-end. Many participants explained that they begin saving at this time in anticipation of high expenditure in January and as expected, this is the month with the lowest levels of savings. Another explanation for this is the possibility that, as reported in some SSIs, members save a small proportion of loans in case of an emergency or difficulty in repaying thus topping up savings levels even when income levels are low.

Credit levels are highly volatile, rising substantially in January and then dipping to relatively lower levels throughout most of the following months. Credit levels could be anticipated to be high around year-end; however this is not the case. Levels actually drop around this time. In SSIs, many members explained a desire not to take loans during the festive period in anticipation of high levels of expenditure in January. Many predict difficulty in repayment at this time and so refrain from taking large loans around December, saving money from share-outs and loans earlier in the year instead. This is advice that is reiterated in SaveAct�s financial education to help members avoid repayment difficulties. As expected, credit levels rise dramatically in January to cover school fees and related costs. They also rise slightly just before and during the winter months when SSI respondents report spending extra on medical costs, warm clothes, fuel and food.

Figure 1. Seasonality analysis of income, expenditure, savings, and credit

Source: Author�s own data including all totals

A significant result is that expenditure levels do not closely follow income flows. In fact during low income periods, expenditure is at its highest (January) and spending levels dip when income levels are relatively high from March � May. This is highly indicative of consumption smoothing, suggesting that participants are in some way protected from the volatility of income levels and strong financial pressures.

This evidence is supplemented by the recurring theme in SSIs of �control�. The ability to plan and save was cited as one of the three most important benefits of SCG membership by 29% of participants. As Nomsa Didi (female SCG member, Xaxazana) states: "I can now plan and budget for the future so have more control over my life". This was raised by Didi during her SSI, she continued: "I feel that my household status has improved as a result of being able to plan for future needs .... My relationship with my husband has improved because I now contribute more to household costs and can make some of the decisions". While the utility of credit was frequently underlined, the advantages of saving appeared to be equally appreciated. Members described the coping strategies they are able to employ to prepare for routinely difficult times. As Zandile Malgas commented, "there is no more uncertainty at the end of each month because I know I will have enough money". When asked about frequency of loan use, an average figure of 5.2 loans per member was given. 14% of participants reported having difficulty repaying credit. These figures suggest that loans are being taken on a regular, but not overwhelming basis.

Trend analysis of credit use - crisis mitigation
Five years ago family was the most frequently used source of funds in times of crisis. This is followed by use of informal groups such as stokvels and burial societies. When asked about interest rates on money borrowed from stokvels, responses ranged from 30% - 40% - a figure as high as some loan sharks or �Mashonisas�. Mashonisas were the third most frequently used financial instrument in dealing with adverse shocks, with interest rates cited as around 50% - 60% per month. Fourth most frequently used was own savings, this backs up the theory that poor people can and do frequently save. Formal bank loan services are fifth at 13%. When discussed, it became clear that participants included local �microcredit� companies in this sector, many of which charge interest rates of up to 40%. SaveAct services were not used as this is before its existence in the area.

In results regarding last year, the most frequently used financial service in times of crisis is SaveAct Credit. This is followed by the SaveAct Emergency Social Fund at 17%, then family at 16%. Usage of informal savings groups such as stokvels remains relatively high, falling from 21% to 12% between this year and five years ago. The use of Mashonisas has dropped dramatically to just 1%, as has use of formal bank loans (2%). Use levels of own savings remain similar and sale of assets has crept up from 0% to 2%.

Results from �this year� and �last year� are very similar. At 38%, SaveAct loans are the most frequently used financial instrument in times of financial crisis: this time they are, however, followed by family at 15%. The SaveAct Social Fund usage appears to drop to 13%. When discussed in SSIs it became clear that this is because the Fund is only paid out for certain crises stipulated in the Group Constitution and with interviews taking place in the middle of a savings cycle, circumstances qualifying members as fund recipients may not have occurred. Use of Mashonisas remains low at 1%, as does that of formal bank loans - 2%. Usage of other informal savings groups and own savings remains stable.

The above results indicate that crisis coping strategies have changed significantly among participants over five years, with SaveAct SCG services becoming the most dominant. It is noteworthy that use levels of own savings have remained relatively constant, suggesting that these households may save both within and outside of their SCGs. Another significant trend was a more intelligent use of Mashonisas. Members use Mashonisas or other high-interest money lenders in the event of a crisis occurring in between SCG meetings. They then immediately repay with a SaveAct loan from their next meeting. In this way they are taking advantage of loan shark convenience but avoiding payment of usurious interest rates.

This data suggests that SCG membership is allowing PLWHA to employ stronger coping mechanisms in times of crisis and avoid strong reliance on family, neighbours or high-interest debts. SSI discussions reveal that it is also easing the burden on households supporting one or more people living with HIV/AIDS, or dealing with the death of a relative from an AIDS-related illness. As SaveAct services have become dominant and represent the only apparent major change in the target group�s access to financial services, it can be deducted that these changes in behaviour can be largely attributed to SCG membership.

Loan usage
The most frequent use of credit is for investments in home improvements. The next most frequent purpose is to cover school-fees and related costs. This is followed closely by expenditure on health-related costs and is especially significant for PLWHA who are likely to have high outgoings related to health. Also important to note is that repayment of previous debts constitutes the fourth most frequent use of credit. Transport also constitutes a significant proportion of credit usage. When discussed in SSIs, participants explained that this largely covered transport to hospitals in an emergency, monthly transport to collect ARV medication and transport to schools. The total of credit spent on IGAs, both new and existing, is joint fifth. Livestock and food are the least frequent categories of credit usage.

This data shows that credit is used to cover basic needs such as food, transport, schooling and health costs. The need for credit to meet these basic needs indicates the high vulnerability levels of clients. It also indicates that the consumption smoothing evident in the above seasonality analysis results is significantly related to the access to cheap and immediate SCG credit. Results do show signs of more productive activities taking place, such as investments in home improvements, new and existing IGAs, livestock and repayment of previous debts, suggesting increases in economic stability.

Gender dynamics and inter-household relationships
Questions five and eight of the general questionnaire regarding self-confidence and household status were designed to capture a sense of the gender dynamics among participants, as well as potentially reflecting on issues of HIV/AIDS stigma. 100% of participants responded positively to both questions. These were the only two questions to receive such overwhelmingly positive responses suggesting large impact in these domains. Recurrent themes in SSIs with both men and women centred on increases in self-confidence and self-esteem, with many attributing these changes to increasing self-sufficiency[17]. Nombeko Duna is HIV-positive herself, and was abandoned by the father of her children. She talked about a greater feeling of self-worth and confidence: "I feel that my household status has improved significantly because I have been able to make a lot of home improvements... I feel more able to provide for my family and more respected... I have more hope for the future as a result of joining SaveAct." Similarly, Nonkosikhona Sibizo feels more confident; "I now participate in making more household decisions in conjunction with my husband. Before, he made all the decisions as he was often the only earner, now I bring money too."

The majority of participants reported increased levels of respect. Just under half of the women in the SSI sample were unmarried and reported being principal household decision-makers. All married women reported either sharing this role with husbands or playing a limited part in the process. Some female participants relayed stories of increased participation in household decision making. Nonkosikhona (female SCG member, Xaxazana) sums up an example of this as she states: "My status in the household has improved and so has my relationship with my husband because I can now contribute. Before he was often the only bread-winner so he was the person who made all the decisions. Now I can make some decisions". They attributed this change to the fact that husbands noticed concrete changes in household living standards stemming from SCG activities. Since women are now holding resources and building assets their bargaining power is stronger and their partners must share decision making responsibilities.

Overall, the experiences of different participants, both male and female, are diverse and highly dynamic. One common impact of SCG membership on all participants however, is a sense of increased self-worth and self-confidence. For some participants this has the potential to translate into concrete changes in relationships while for others this is a more difficult process.

Noamen Nongenzi with her 6 children and sister in their new kitchen, Khoapa

Extra-household relationships
During SSIs, participants were asked about status in the community. Ninety-five of participants reported an improved status as a result of SCG activities. For many participants, this centred around no longer being dependent on others and being able to make visible improvements to housing without incurring large debts. "My neighbours now say that my household is �rich�!" reported Sibongile (female SCG member Xaxazana), who is the head of a female-headed household of five, receiving no financial support from the father of her own two daughters (one who is HIV-positive[8]), nor from her daughter�s nine-month old son. Similarly, Fundiswa (female SCG member, Matatiele), an HIV positive single mother of two with no support from the father reports; "I no longer have to borrow money from my family; I now sometimes even lend money to them so I feel more respected in my community".

The SCGs are setting an example for improved gender equality in the study areas. In August 2009 in Xaxazana, the male partner of an SCG member demanded to be given her entire share out amount. He sent a relative to the SCG in question to get the money. In this instance the case was taken by the savings group members to the area chief. All SGC members asserted the rights of this SCG member over her own savings. The chief ruled against the member�s partner, leaving her in control of her won savings and setting an example for other groups in the area. The SCGs are also providing a powerful platform for collective action in the areas included in the study. In the case of a rape of a young girl in Xaxazana, SCG members actively organised themselves to support the family of the victim. They arranged community meetings to present their discomfort to the wider community. They challenged the courts not to grant bail to the alleged rapist. This was achieved. From these examples we can conclude that women are feeling emboldened by their participation in SCGs to take up sensitive and challenging community issues that may otherwise have been avoided.

Conclusions

The situations of poor rural households affected by HIV/AIDS are extremely complex and difficult to grasp at a single point in time. In spite of this, this study presents important evidence of consumption smoothing and improved coping mechanisms among PABHA. With results from the trend analysis coupled with the seasonality analysis and other data, it is reasonable to assume that the changes were due in a significant measure to SCG activities.

Results clearly show that levels of expenditure do not closely track income flows. There is also a stable demand for saving and cheap credit amongst participants. Usage of loans sharks and other high interest-charging money lenders has decreased dramatically, as has the incidence of borrowing from family members. In turn, use of SCG loans in times of crisis has increased substantially, becoming the dominant financial instrument for participants. These results suggest that risk reduction activities among PABHA have been strengthened. The target group are avoiding irreversible coping strategies and are gradually engaging in more productive activities. Stable levels of consumption suggest that they are more protected from insecure and volatile flows of income and are thus more able to meet basic needs. Results indicate that the impact of adverse shocks has been cushioned and coping strategies strengthened, thus overall vulnerability to poverty has decreased among participants.

Consumption smoothing and strengthened risk reduction activities have resulted in positive effects on the health of participants as well as tangible improvements in living conditions. In addition, improvements in wellbeing are evident, with qualitative evidence that PABHA of both genders have been empowered through this experience. There is also further qualitative evidence that gender relations are shifting slightly as a result of SCG membership, with women experiencing more respect and more frequently taking part in household decision-making processes.

A significant indication of project success is the large and constantly increasing demand for SaveAct services among participants and the fact that 98% of all PABHA have recommended the programme to a relative, friend or neighbour. Demand in the target area is currently too high for SaveAct to meet and plans to expand and spread the project to new areas are underway. With this target group being largely typical of rural populations affected by HIV/AIDS in South Africa, it is likely that the same impact can be achieved on a much larger scale and there is thus great scope for project replication in other areas.

References Alfred Nzo District Municipality (2011) Annual Report : http://www.andm.gov.za/site/index.php/2011/05/annual-report/ Collins D., J. Morduch, S. Rutherford and O. Ruthven (2009) Portfolios of the Poor: How the World's Poor Live on $2 a Day. Princeton: Princeton University Press.
de Aghion A., J. Morduch (2005), The Economics of Microfinance. Cambridge: The MIT Press
Donahue J. (2000) �Microfinance and HIV/AIDS: It�s Time to Talk�. USAID http://www.microfinancegateway.org/gm/document-1.9.29149/2740_file_02740.pdf
Gow J. (2010) �The HIV/AIDS Epidemic in Africa: Implications for U.S Policy�, Health Affairs: http://content.healthaffairs.org/content/21/3/57.full
H�gman A. (2009) Microsavings Effects on Consumption Smoothing: A Case Study of South Africa. Thesis Work D, Uppsala University, Department of Economics.
KHANA (2011) �Evaluation of Microfinance Pilot Project for HIV-affected Families�. USAID: http://www.aidsalliance.org/includes/Publication/Microfinance_Pilot_Project.pdf
Zeller M. (1999) The Role of Micro-Finance for Income and Consumption Smoothing. International Food Policy Research Institute (IFPRI): http://www.microfinancegateway.org/gm/document-1.9.25376/20683_20683.pdf