The consequences of HIV/AIDS impact on small farmers

As discussed in the previous chapter, the impact of HIV/AIDS is being felt, be it directly or indirectly, on almost all aspects of the farming systems studied. In some cases there is a tangible though not necessarily quantifiable effect on farming systems, especially in Uganda, where the epidemic is seen at an advanced stage of its development, and thus its impact is more directly evident.

In both Tanzania and Zambia, the impact is less evident but the vulnerability of the farming systems and the trends in agricultural development generally, may be used to preview, and hopefully prevent, as serious a deterioration of the status quo because of the HIV/AIDS epidemic as might otherwise be the case.

In Chapter 3 consideration was given to the impact of the HIV/AIDS epidemic on different aspects of small farming systems in Eastern Africa, this chapter considers the consequences of the changes that are occurring in the rural communities studied.

The main consequences include: the threat to household (see box below) and community food security; decline in the nutritional and health status of smallholders and their families; decline in educational status as children are forced to leave school; and changes in the social system as households adapt to the impact of HIV/AIDS. These latter may involve the break-up of families, a growing incidence of female-headed households, and changes in attitudes to indiscriminate sex. The impact of HIV/AIDS is also likely to be most severe among already vulnerable groups such as the malnourished and the food insecure.

More positively, individuals and groups have begun to develop survival strategies in the light of the changes which are occurring as a result of the impact of HIV/AIDS on their lives. Despite what are, in some cases overwhelming mental and physical hardships, with spouses losing not only their partners but also one or more children and experiencing increasing difficulties maintaining and making a daily living, households have continued to function.

Box 10: The Zambian farmer who only just copes

In the village of Mpongwe, amongst the more productive farmers, those whose production is now being affected by illness, or is likely to be in: the near future one: of the farmers identified three ways in which HIV/AIDS had and was affecting his life. These were through; (i) the impact on his cattle, with draught oxen having died for lack of care; (ii) his six children from his first marriage had gone to live with their maternal grandmother on the death of their mother his wife; (iii) the shortage of labour meant that his maize crop which be said had declined sharply, was affecting his own staple food security.

In some of the case studies described in Chapter 3, farmers have expressed interest in supplementing their incomes through small income-earning activities either on or off-farm. Such activities are extremely diversified and include bee keeping, beer brewing, tailoring, pig keeping, etc. Smallholders have also shown a growing interest in receiving extension advice, skills training or education in areas such as basic literacy, numeracy and health.

4.1 HIV/AIDS: a threat to food security?

4.1.1 The situation in Zambia
4.1.2 The situation in Tanzania
4.1.3 The situation in Uganda

In considering the consequences of the general decline in the standard of living and in household productivity as a result of HIV/ AIDS, the question arises as to whether food security is at risk.

Food security is defined by FAO (1994) as "enough nutritious and safe food being available and accessible for a healthy and active life by all people at all times". At the household level it is defined as "the capacity of a household to procure a stable and sustainable basket of adequate food" (IFAD 1993). This food must be adequate in terms of quantity and quality to compose a diet which meets both the nutritional needs of the household members as well as being culturally acceptable. Stable access is assured through various mechanisms that enable the household to procure food supplies across seasonal and transitory shortages. Sustainable access requires that the means of food procurement in the long run are consistent with:

- sustainable resource use and management;
- maintenance of productive assets;
- self reliance and human dignity; and,
- overall livelihood needs (IFAD 1992).

The nature of food access depends on the particular conditions prevalent in a given farming system. Food may be derived from home production, through purchase, through gathering (e.g. fishing), or through barter arrangements, gifts or some other means of transfer.

4.1.1 The situation in Zambia

In early 1993, up to 2.6 million people (33 percent of Zambia's total population) were reported as being vulnerable to food insecurity. Of these, over one million were in the marginal rural areas (IFAD, 1993).

In the rural areas, food insecurity is due to low smallholder productivity associated with maize monocropping, as well as limited access to agricultural services such as input supplies and other resources, as well as past market distortions (and consequent crop mix distortions) associated with Government price subsidization policies for maize. In towns and cities, food insecurity and malnutrition are related to the deterioration of purchasing power as a result of the long term stagnation of the economy, the scarcity of income-earning activities, inflation and the effects of a severe structural adjustment package. Urban earnings are linked to rural food insecurity through remittances (see Chapter 3), which are particularly important in certain farming systems. Such systems are likely to be adversely affected by the HIV/AIDS epidemic in the urban areas as well.

In Zambia, the highest levels of malnutrition are in the agriculturally-marginal rural areas, the pert-urban localities and the shanty towns of major urban centres. In general, malnutrition appears to be more serious and widespread in the northern, north western, Luapula, western and eastern provinces. Zambia also has one of the highest levels (54%) of childhood growth stunting in Africa (International Fund for Agricultural Development (IFAD) /World Food Programme (WFP) 1993). Stunting is an indicator of long-term deprivation.

Analysis of labour and labour management in the Chipese area of Mpongwe, showed that the more food secure clusters had men and women (a husband and wife) involved in production, and the most efficient were those in which men organised timely ploughing arrangements, either through tractor or oxen hire, or through labour-pooling arrangements. This situation arises through the need to plant as large an area of maize as possible to ensure a surplus, and the fact that maize tends to be a crop controlled by men - they have greater access to credit and usually control the marketing arrangements.

In the area of Teta, however, the situation is different. The area planted to maize and other crops grown in ploughed fields is much smaller than in Mpongwe. It is therefore easier for women themselves to manage this operation, and to hire oxen to cultivate this smaller area.

Diversity in the production system gives women more independence. The cultivation of local maize and beans are largely women's activities; it is only with the advent of Irish potatoes, an expanding cash crop, that women have become involved with the techniques of wetland production. The construction of mounds and planting of sweet potatoes, beans, cassava and millet also tend to be activities controlled by women. As a result, a larger proportion of the production output, including some cash income, and the marketing of such output is controlled by women. Maize production itself is often split between men and women, with both husbands and wives tending their own fields. During a series of meetings with villagers, of the 25 women in the women's meeting, only three were not tending their own fields. Accordingly, it would appear that women are going some way towards ensuring their own staple food security.

The net result is that amongst the vulnerable group (see Chapter 2) from which farmers were interviewed in Teta, four of the nine were headed by women. These women were often involved in hiring oxen or piecework labour. With a range of crops to sell, and access to millet for beer brewing, it is possible for them to raise enough income to purchase such resources. Because of the quantity of millet that is grown, beer brewing amongst these more food secure farmers is a regular not an occasional income-earning activity.

It would also appear that livelihood security is better guaranteed because of the greater range of options in the system. The different production types and crops mean that generalised crop and income failure is unlikely - risk being spread more widely.

The resource poor producers (cluster type 4) in Teta were also mostly women. Several of them were divorced or widowed, or if married, receive less assistance from their husbands. Thus less labour was available. The households of these women producers are probably more food secure than their Mpongwe counterparts because of the diversity of staples, relish crops, and income sources available. During the drought, the women resorted to beer brewing and the selling of crops such as vegetables, beans, sweet and Irish potatoes, rather than undertaking piecework, as was the case with the women in Mpongwe. This meant that less labour was diverted from their own fields.

In this community, the poorest producers tend to be reliant primarily on citemene production (see Chapter 3) of millet for their staple food security; they grow little maize and do not use fertiliser. Consequently, although beer brewing is a survival strategy, there will be more conflict over the use of millet for food or beer. These households run out of staple food earlier than other households. They also tend to have fewer relish crops.

In Chipese district, it was estimated that by the time the HIV/AIDS epidemic reaches the stage of increased mortality, probably within two to three years, the impact on agricultural production and livelihood security would have increased substantially. It was also clear that as sickness and death due to HIV/AIDS rises, the effect will be to exacerbate the already significant vulnerability and food insecurity of large numbers of women and children. This is due to the fact that it is common amongst the matrilineal peoples (prevalent in many parts of rural Zambia) for the death of a parent, as with the already common event of divorce, to lead to the break up of the nuclear family itself. Death like divorce causes social dislocation. Women, with their children, move back to the villages of their own mothers, or other matrilineal kin, leading in these villages to an increasing number of single parent producers and a growing dependency ratio.

4.1.2 The situation in Tanzania

An estimated 90 percent of the population of Tanzania are reputedly dependent upon their own subsistence (FAO; 1992). Research in two areas - Mtwara and Morogoro - indicates that 74 percent and 60 percent, respectively of the households surveyed had insufficient food. In such circumstances, the impact of illness or death on the household labour economy and thus on nutrition may be very dramatic indeed. For example, Semali and Ainsworth (1993) report that in Kagera the restriction of normal activities due to illness or injury (not necessarily related to HIV/AIDS) was resulting in a very considerable loss of working time in agriculture in 1992.

However, in an examination of indicators of food and energy adequacy for Tanzania, it was found that those regions which are most exposed to the HIV/AIDS epidemic in the medium-term are not, with certain exceptions', the regions which also have a current energy deficit. This is illustrated in Table 3 below which compares exposure to the HIV/AIDS epidemic and food insecurity as measured by dependence on emergency food aid.

The point however is that even in relatively robust systems such as those of the Kagera and Mbeya regions in Tanzania (and similar systems in Uganda), ten years of HIV/AIDS exposure do, nevertheless, result in dramatic declines in the standard of living, particularly among poorer households, and in a marked decline in the energy balance as indicated by the decline in the production of matooke and the rise in the production of sweet potatoes, cassava and yams.

Table 3: Emergency food aid and epidemic exposure
Region Population N. in Target Group Target Group as % of population Target Group Rank Epidemic Impact Ranking
Arusha 1390000 81833 5.89 9 11
Coast 656000 83289 12.70 3 5
Dodoma 1272000 82508 6.49 6 12
Iringa 1243000 50000 4.02 11 2
Kagera 1363000 73600 5.40 10 1
Kilimanjaro 1140000 96903 8.50 4 9
Lindi 665000 131949 19.84 2 18
Mara 998000 378891 37.97 1 7
Mtwara 914000 76518 8.37 5 6
Morogoro 1257000 79785 6.35 7 4
Mwanza 1931000 30000 1.55 14 3
Singida 824000 20000 2.43 12 10
Tabora 1065000 64890 6.09 8 9
Tanga 1320000 31564 2.39 13 3

4.1.3 The situation in Uganda

A small-scale but detailed study of nutrition in rural Masaka carried out in 1993 (Seeley) showed very low levels of current nutrition and food security in households. According to the field work undertaken for the present study much the same situation appears to be true at present in Gwanda and in other similar communities in Uganda. Moreover, a community wealth evaluation estimated that about 45 percent of people in Gwanda are poor (according to the community' own standards). But only ten years ago Gwanda was considered to be a wealthy community, by local rural standards.

With the advent of the HIV/AIDS epidemic, the community has seen a measurable decrease in its own standard of living. In the Table 4 below which summarizes the coping responses of 14 households in Gwanda of which case study analyses were made, some 36 percent (of the 14 families surveyed) reported food shortages. A similar study carried out in the community of Nakyerira found that 40 percent of the ten households interviewed suffered food shortages. These figures may be compared with those for the third area studied in Uganda, the Ndaiga communities on which HIV/AIDS had had, (at the time of the study), little if any impact. In Ndaiga of the 16 households surveyed, only two (or 12 percent) had reported food shortages.

In Gwanda as noted in other communities the severity of epidemic impact is unequally experienced as between differently-resourced households. The richer the household, the longer it can maintain adequate food supplies; the poorer the initial endowment of a household, the more difficult this will be. Given that the community itself considered that at present 45 percent of households were poor in terms of its own standards of evaluation, then it must be concluded that for 45 percent of the people in the community, any further stresses, whether related to the epidemic or for some other reason, can only have dire results.

Table 4: Summary of coping responses in Gwanda
Household Experience Number Percentage
HIV/AIDS affected 8 57
Decline of cultivated area in the past ten years 7 50
Decreased range of crops 10 71
Labour shortage 12 86
Change in cultural practice 8 57
Food shortage 5 36
Land shortage 2 14
Decline in off-farm activity 4 28
Increase in off-farm activity 2 14

Note: these data were collected from a village transect and are not the result of a random sample. They are illustrative only. Total number of cases 14, percentages rounded up or down to nearest whole number.

4.2 The declining status of nutrition and health standards in epidemic-affected communities

In Chapter 3 it was observed that a recurring pattern in the communities studied is the change in the volume and kinds of crops produced in the farming systems. Partly as a result of this, and other factors, e.g. the process of impoverishment induced or exacerbated by HIV/AIDS affliction, it would appear that levels of nutrition are falling. The reliance on starchy staples like cassava and sweet potatoes, compared with other more nutritious traditional crops like matooke, and or protein from animal products, is partly to blame. In addition there is the problem of the lack of understanding of the nutritional value of other produce, fruits and vegetables which though they may be produced are not favoured in the diet (see Chapter 3).

Lower levels of nutrition result in the increased vulnerability of the individual to disease and thus to a decline in health. Increasing morbidity in turn has a negative affect on household and individual productivity, resulting in long hours and hard labour for the already weak, and creating additional burdens on able-bodied family members.

One of the exercises that was carried out during the visit of the survey team to the Chipese area of Zambia was a health time line. Some of the older men participated especially, in identifying the type and severity of diseases that had been experienced in different decades since the 1920s. What the exercise showed was that by the l940s, several of the early twentieth century diseases such as smallpox and leprosy had disappeared, whilst others such as venereal disease and pneumonia had become much less prevalent. The 1940s through to the 1970s were relatively free of major diseases or epidemics. In the 1980s, however, health trends took a significant turn for the worse again. Many diseases became more virulent and widespread than they had been for decades - sexually transmitted diseases, scabies, pneumonia and tuberculosis. Malaria became much worse and is now the most commonly registered illness for admission to both the Mpongwe and Ibenga hospitals. Cholera also reared its head again in the late 1980s. And then there is AIDS.

During case study interviews, information was sought on major illnesses experienced within the community and deaths that had occurred within the last three years. Within most villages, illnesses such as malaria (fever), coughing and diarrhoea are common, the latter especially among children. Pneumonia is relatively common too, and with children malnutrition is an increasing problem. Older women and men suffer from backache, aggravated in some cases by the need to produce more food because of an accumulation of mouths to feed. Children who are ill during the weaning age of about 8 to 18 months can have a serious effect on women's agricultural production. In Kapenda village, a young woman reported having spent three weeks in hospital in January 1993 with a baby that had fever and diarrhoea. January is the peak weeding month. Consequently her 0.6 hectares of maize produced enough to eat as green maize only.

By contrast, another young woman also spent two-and-a-half months in hospital at the end of 1991 with a child suffering from diarrhoea and malnutrition. In 1993 the couple were still able to produce 40 bags of maize from 2 hectares. Her husband had assumed responsibility for staple food security and did the work when she was ill.

It appears clear from the general health status of this population that it is already profoundly vulnerable - especially the women and children. If the impact of the HIV/AIDS epidemic is considered in addition, the situation may have critical consequences.

4.3 HIV/AIDS: education and declining school attendance

One of the consequences of impoverishment and contracting household incomes in communities like Gwanda and Nakyerira in Uganda, is that with limited cash incomes, families can barely afford to send their children to school. On the one hand, extra hands may be required in the household to undertake farm or domestic chores. On the other hand, and judging from the case histories where several or all the children in a family were attending school, rather than helping with the farm work' education also appears to be a priority. In some families resources were provided by NGOs (see Chapter 3) to help with children's education. Certainly, if they are not provided with external assistance, families that are already at or close to subsistence level will have to forego educating their children simply to meet basic survival needs.

In Ndaiga and Nakyerimira it was estimated that 60 percent of all children of primary school-age children (6-14 years) were not attending schools. The reasons advanced for this phenomenon were lack of money to pay for their education.

It was also noted that female children were less likely to be in schools than were their male counterparts. Of all the children in school only about 10% were females. Some of the reasons given were the early marriage of girls and their leaving school because of pregnancy. This in turn has discouraged parents from paying for their female children.

4.4 The problem of orphans

The situation is still more serious with respect to orphans. With increasing numbers of children left orphaned by the HIV/AIDS-related deaths of their parents, the education of such children is a special problem. If orphans are taken over by other relatives, the cost of educating such children may well strain the resources of the receiving households. The children themselves may not have the time, let alone access to money, or even the intellectual wherewithal to consider education as a priority. Again, in areas with a high incidence of HIV/AIDS and large numbers of orphaned children, NGOs have so far been making efforts to fill the gap.

In the Mbeya region of Tanzania, the local hospital at Igogwe, which runs a home for orphans, reported that although attempts were made by relatives to care for orphans, they find it impossible to shoulder the burden of their own children and the orphans together. Most orphans are adversely affected in terms of their access to essential services, particularly education. They cannot pay school fees or buy uniforms and other school necessities such as exercise books and text books.

4.5 HIV/AIDS and changes in the social system

It is difficult to state absolutely that HIV/AIDS has been directly responsible for marked changes in the social organization of rural families. It is certain, though, that the epidemic impact has often exaggerated pre-existing problems. In Zambia, for example, where divorce was reported by focus group respondents to be a serious and growing phenomenon, it seems likely that the current economic circumstances have exacerbated the divorce rate, although there is no direct evidence for this. Previously marriages used to be arranged and a bride price was paid. These days almost no bride price is paid. Moreover, marriages frequently occur due to the pregnancy of the girl. Her parents may force the pair to marry but such a marriage may not last long.

In matrilineal social groups in Zambia, another practice which exacerbates the break up of marriages and has a negative impact on agricultural production is that of sons-in-law going to live in the village of their spouse's parents. The son-in-law has an inferior status and is rarely motivated to work particularly hard.

Rapid Rural Appraisal studies in the Central and Copperbelt provinces of Zambia have found that the most vulnerable and resource-poor farmers are always from indigenous matrilineal social groups. In this system, the increasing incidence of divorce results in social dislocation for both women and children. When a wife is forced to leave the house she goes back to her parents with her children. Fathers thereafter are likely to take only limited responsibility, if any, for their children.

In cases of divorce, which seems to be a growing phenomenon in the Zambian communities studied, a woman moves back to the village of her own mother, usually with most or all of her children. In the event of her death, it is possible the children will be subsequently split up (see Chapter 3). This is very common in the case of urban deaths. But in an area such as the Chipese community the net result will be an accumulation of children in the households where there are middle-aged or older women.

4.6 A change in attitudes to social customs?

During a focus group meeting at the village of Mpatisha in Zambia the women were asked about the responsibility of a husband for his children if his wife should die. Their first response was that if the wife of their son dies the children should go to relatives of the wife. 'Those children [of our sons] are not our relatives'. But asked a question which challenged the morality of this response and it provoked a very different reaction (see Box inset overleaf):

This exchange gives an insight into many of the now extremely negative 'customs' that are associated with matrilineal culture. The normative basis of the custom or tradition has disappeared, and now practice is driven only by expectation. The women showed through their general agreement with the remark that change was up to them and that if they do so others may follow their example -thus much of "tradition' is just example. Practices are followed because others do so, even if when challenged most will agree that the practice has no sound moral basis and has negative consequences.

The social problem is how to change undesirable 'customs', which have long evolved away from any ethical basis, and now merely undermine social solidarity and exacerbate the vulnerability and trauma experienced by women and children.

4.7 The growing reliance on income-generating activities

About 40 percent of farm households in Gwanda reported that they supplemented their incomes from off-farm activities. These range from petty trade, fishing, and beer brewing to small-scale handicraft production. For the majority of Gwanda households, off-farm income-earning activities do not feature in their livelihood strategy because they: do not have the small amounts of capital necessary to start this kind of work, they choose to limit their consumption, and they are able to sustain an adequate standard of living from their farms.

The case material for Gwanda suggests that there are two clear categories of household with major off-farm components to their livelihood strategy. The first category is composed of wealthier households which, even if they have been affected or afflicted by the epidemic, are able to maintain a high standard of living because they were well-off before they were affected by the epidemic and this has acted as a buffer to its impact. For these households, who are in the minority, farming was only one among a fairly large number of income-generating activities in which the household enterprise was involved.

By contrast, there is another category of households which have taken-up off-farm work as part of their coping response to the depredations of the epidemic. In such cases, which often seem to be female-headed households, these activities will be on a very small scale indeed, perhaps a little beer brewing, mat making or bark cloth production.

Whether or not the decision to enter into off-farm activity is taken may also have to do with the cultural expectations of a particular household. It may be no coincidence that the widow of a high salary earner for whom the epidemic has meant a dramatic drop in level and expectations of life should choose an off-farm activity as an effort to retain a lost standard of living. However, for those in the community who form the vast majority, the middle and the poor, there is perhaps no cultural expectation of anything much more than the difficulties that they currently face. Additionally, they may not have the social support networks which would provide access to the small amounts of capital required to start up an off-farm enterprise.

Box 11: Focus group meeting (Mpatisha, Zambia)

Question:"lf you have both a son and a daughter and they both marry and have children, who are the grandchildren, those from the daughter or from the son?"

Answer:"They are both our grandchildren. But it is a tradition we follow that only the children of our daughters are our grandchildren, but it is not a good tradition."

Question: "Can a chief change a tradition?"

Answer, first respondent: "We can't follow the chief unless we ourselves believe it. An individual we can copy, but not the chief."

Answer, second respondent:"lt is up to me to change. Now we have been discussing we can see that it is good that we keep the grandchildren of sons as well as daughters. We can change and others who are not here will stay the same. But they can see what we do."

In Nakyerira, as in Gwanda, farm production cannot provide for all the nutritional, social and economic needs of the majority of households. Even for those able to produce a surplus for sale, marketing remains a major bottleneck, and this is particularly the case in households where a person who has died as a result of HIV/AIDS or from another cause was instrumental in marketing farm produce.

Many of the young men and women, form part of the migrant, often semi-skilled, population seeking employment in Kampala, Mityana or elsewhere, sending remittances home. Most single women heads of households engage in making and selling local beer. In contrast to Gwanda, fewer women are involved in weaving mats and baskets or making other traditional handicrafts for sale. Apparently there is little local demand for these products. However, those who do undertake such activities were more likely to be in the middle income households.

Young unmarried adults often combine several occupations into a livelihood strategy because, as they say, "agriculture does not pay enough to sustain a person". Although the predominant occupation of young men seems to be cultivation of crops and raising of small stock, approximately 85 percent are also engaged in casual labour.

The main off-farm activity for the few adult male householders is trading in agricultural produce.

Except for the middle-aged single women heads of households who are often engaged in the brewing and sale of beer, women are not involved in many off-farm activities. However, some of the young single mothers seek employment from better-off families and perform such tasks as land preparation and weeding.

In contrast to the opportunities which appear to exist for supplementing incomes from activities other than those traditionally or normally relied upon, the situation in Ndaiga is also relevant. Here, although there is little HIV/AIDS impact on the communities, given the livelihood strategies characteristic of the area, there seem to be many reasons for the epidemic to spread in the next few years. In that case it is possible to speculate that the dominance of maize monocropping for the market, the presence of a relatively large landless population and the absence of off-farm income-earning opportunities and market isolation, would indicate that this farming system and its connected livelihood strategies would be less resilient than for example those in Gwanda.