Study Paper No. 6
THE IMPLICATIONS OF HIV/AIDS FOR RURAL DEVELOPMENT POLICY AND PROGRAMMING:¹ Focus on Sub-Saharan Africa

3. Conceptual framework on the implications of HIV/AIDS
for rural development policies and programmes

Rural development does not merely consist of the total of various isolated sub-sectors (infrastructure, employment, education, health, etc). It is a dynamic, integrated and interdependent system of productive and other components, operating through a network of inter-related sub-sectors, institutions and rural households with linkages at every level of activity. The efficiency and effectiveness of each sub-sector, institution, household, etc. depends, to a large extent, on the capacity in other parts of the system. If this capacity is eroded through HIV, then the system's ability to function will be diminished.

Thus, HIV/AIDS does not merely impact on certain rural development sub-sectoral components leaving others unaffected. If one component of the system is affected, it is likely that others will also be affected either directly or indirectly. In other words, the impact of HIV is not only cross-sectoral, but, more importantly, systemic. If the linkages between sub-sectors, institutions and households are not identified and addressed as such, then the analysis of the impact of HIV/AIDS will be incomplete (given that the full picture is more than just the total of its parts) and programme responses will be inadequate.⁴¹

If rural society is seen as a living organism, the systemic impact of HIV/AIDS can be depicted as a series of attacks on its immune system, leading to a host of chain reactions as the society and its various organs try to fight back and adjust. Without wanting to take this analogy too far, there are valid parallels at the physiological and social levels: the institutional response to HIV/AIDS is essentially the social immune system attempting to strengthen itself to fight back without being able to cope well enough.⁴²

The systemic impact of AIDS occurs within an inter-sectoral environment-rural development. However, most formal rural institutions are organised and operate sectorally. As HIV/AIDS also cuts across sectors, addressing the epidemic sectorally may not be sufficient. The question is: how can existing structures and institutions be modified to enable cross-sectoral collaboration so as to deal with problems that cross existing boundaries. And by extension, how can multi-sectoral responses be operationalised at central, district and village levels?

For the purpose of analysis, the conceptual framework presented below focuses on selected rural development focus areas (poverty alleviation, food and livelihood security, and the empowerment of rural women) and sub-sectors (labour and infrastructure) which are intended to serve as examples. The policy focus areas and sub-sectors are all inter-related. For instance, poverty alleviation is essential for food security and vice-versa. This accounts in part for the resulting overlap in terms of the co-factors of susceptibility and vulnerability to HIV. More importantly, however, the systemic effects of the epidemic and the inter-sectoral nature of rural development need to be kept in mind when examining the inter-relationships between sub-sectors and institutions, as these are critical to the formulation of policy and programming responses.

This framework represents only one of many different ways that the implications of HIV for rural development policies and programmes can be analysed. As such, it has a number of constraints. For example, an important dimension which is missing from the framework is how policy-makers and programme planners (including community, divisional and district councils) can take the HIV epidemic into account in their budgets, planning and programme of work. It will be the task of the forthcoming case studies to further develop, refine and test this framework, adjust it according to their consultations and findings, and generate policy and programme responses.

The conceptual framework presented below focuses on the following issues:

• How are rural development policy focus areas and sub-sectors susceptible and vulnerable to HIV/AIDS and vice-versa?
• What is the systemic impact of HIV/AIDS on rural development policies and programmes and vice-versa?
• What are the implications of HIV/AIDS-related susceptibility, vulnerability and impact for rural development policies and programmes?
• What planning mechanisms can be put in place to help generate policy and programme responses that will arrest the spread and mitigate the effects of the epidemic?

3.1 Key cross-cutting issues

The following key points, which cross-cut the proposed conceptual framework are integral to any analysis of the inter-relationships between rural development and HIV/AIDS and to the generation of policy and programming responses:

1. The HIV epidemic tends to exacerbate existing problems of rural development (poverty, food insecurity, etc.) through its catalytic effects and systemic impact.

For example, food insecurity is exacerbated by HIV/AIDS. However, food-insecure households are affected not only by HIV but, more commonly, by one or more of the following factors:

• drought (which may lead to a decline in food production);
• migration (which may lead to farm labour shortages and a decline in food and export crop production);
• gender inequalities (women who produce most of the food in Africa have limited access to productive resources, such as land and water, and to support services); and
• poverty and livelihood insecurity (falling household incomes, limited off-farm employment opportunities, etc.).

Similarly, HIV/AIDS exacerbates labour and employment-related problems. However, the issue of discrimination in the workplace, for example, is not one for HIV/AIDS alone but is part of the wider problem of disciminatory practices in workplaces which have adverse effects on many more people than just those with HIV/AIDS. The issue of absenteeism is also not exclusive to HIV/AIDS but is related to a host of other structural, political and organisational problems in the workplace.

2. In areas heavily affected by HIV/AIDS, the catalytic effects and systemic impact of the epidemic on rural development may:
a) amplify existing development problems to such an extent as to trigger structural changes; and/or
b) create new problems and challenges for rural development.

For instance, the pattern of adult mortality is changing dramatically in countries with mature HIV epidemics. It is not merely a case of mortality decline being reversed. In the past, declines in mortality occurred mostly at the infant and child levels and resulted in a marked increase in life expectancy. These days, as a result of HIV/AIDS, life expectancy is declining dramatically in many countries, mostly due to young adult mortality. For instance, in Zambia, without AIDS, life expectancy in 1996 would have been approximately 60 years; due to AIDS, it is now estimated at about 35 years. Projected data on life expectancy for 2010 paint an even bleaker picture: in Zimbabwe, for example, life expectancy without AIDS would have reached about 70 years, but as a result of the epidemic it is expected to decline to about 32 years.⁴³

Vertical mother to child transmission of HIV is resulting in an increase in infant and child mortality which is transforming the root causes of infant and child mortality. Such changes in young adult and infant mortality, as well as in life expectancy, are likely to have serious repercussions on the structure and composition of population pyramids across countries with mature epidemics, and more generally, on human development. For instance, UNDP's Human Development Index (which combines life expectancy, a measure of educational attainment and GDP per capita) is declining sharply in many sub-Saharan countries as a result of HIV/AIDS.⁴⁴

The emergence of child-headed households and the breakdown of certain informal rural institutions are just two examples of new problems being created in areas heavily affected by HIV/AIDS. Both have significant implications for rural as well as national development but have largely not been addressed as such to date.

The uneven and heterogeneous distribution of HIV/AIDS has important implications for rural development policies and programmes: a) nation-wide average HIV prevalence rates, policies and programmes are likely to be misleading; b) responses need to be tailored to the needs and socio-economic conditions of particular areas, according to the extent and nature of the impact of the epidemic; and c) the structural changes and new problems brought about by the impacts of the epidemic (old and new) need to be taken into account.

3. Given that many problems arising from the epidemic are not specific to HIV/AIDS, policy and programme responses need not be HIV/AIDS-specific but must address the root causes and consequences of the wider challenges to rural development. In other words, a developmental rather than an AIDS-specific focus is critical to tackling the multi-sectoral complexity of the epidemic and its systemic impact and to ensuring the sustainability of both HIV/AIDS responses and rural development efforts.

Given the systemic impact of the epidemic, the focus of analysis and response need not lie in AIDS-specific policies and programmes to what are often perceived to be AIDS-specific problems. For instance, as we have seen, food insecurity is not specific to HIV/AIDS; the impoverishment of female-headed households is not specific to HIV/AIDS impact; and restricted access to social and support services is not limited to families affected by HIV/AIDS. Thus,

• appropriate technology is a key issue not only for HIV/AIDS-affected households, but for all poor and vulnerable households suffering from shocks inflicted by drought (crop failure), war/civil unrest, migration, etc. Labour-saving technology development for smallholder agriculture is a response not only to labour shortages related to HIV/AIDS-affected households in particular, but to other vulnerable, labour-deficient households as well.

• similarly, it is unlikely that discriminatory practices against employees with HIV or AIDS will cease unless discrimination in the workplace is tackled in its own right. In other words, the issue is not merely discrimination of men and women with HIV and AIDS in the workplace, but discrimination of male and female employees in the workplace in general, regardless of the reason. Responses to the effects of the epidemic on labour and employment need to address the broader issues of discriminatory practices, absenteeism, losses in production, etc. to which AIDS is but one contributing factor, albeit an increasingly prominent one.

A shift in approach to HIV/AIDS is needed, premised on the recognition that poverty-, gender-, equity-based rural development policies and programmes are the fundamental elements of a multi-sectoral response to the epidemic that will facilitate the integration of HIV across sectors, including rural development.

4. The policy environment plays a key role in defining the parameters of susceptibility/vulnerability to HIV/AIDS and of the impact of the epidemic.

Biased policies and institutional arrangements, such as urban-oriented investment, have traditionally deprived many rural areas of resources, excluded them from the benefits of development and accentuated the impact of other poverty processes (drought, war/civil strife, environmental degradation, migration and HIV/AIDS).⁴⁵ The need to review such policies and re-orient resources toward rural development is important not only for rural development per se, but also for addressing the epidemic.

For instance, institutional arrangements and socio-economic conditions that have perpetuated rural poverty, according to the International Fund for Agricultural Development (IFAD), include:

• the lack of access to land;
• inequitable sharecropping and tenancy arrangements;
• poor markets;
• limited access to credit, inputs and technology; and
• ineffective agricultural extension services.⁴⁶

Other constraints include the lack of training facilities; the lack of grassroots institutions needed to foster people's participation; and inadequate research and technology related to smallholder farming systems (see section 3.2)

Addressing the rural policy environment, and in particular the constraints to rural poverty alleviation, food and livelihood security, gender inequalities, etc. is critical to creating an enabling environment for the generation of appropriate responses to HIV/AIDS.

5. Gender, age and marital/family status play as decisive a role in determining susceptibility/vulnerability to HIV/AIDS and the potential impact of the epidemic as do economic and cultural conditions. For this reason, the interplay between these factors needs to be considered at each stage of policy and programme development.

Susceptibility and vulnerability to HIV/AIDS needs to be differentiated by gender, age and marital/family status for many reasons: women are biologically more prone to HIV infection; youths tend to be more sexually active than mature adults; single people may have more sexual partners than married people; and so on. Similarly, the impact of HIV/AIDS is likely to vary depending on the interplay between these three factors.

Below are some examples of the interface between rural development, gender and/or age and/or marital status and HIV/AIDS:

The critical link between food insecurity, gender and HIV, though scarcely documented, is important. FAO research in Uganda⁴⁷ (see Box 4) and in West Africa shows that the most immediate problem for many AIDS-affected female-headed households is not medical treatment and drugs but food and malnutrition. It appears that when a household has been affected by male adult mortality, surviving widows and their families often have few, if any, assets to dispose of in their time of need. Thus, food security coping strategies may disintegrate quite soon after male adult death and food consumption may decline sharply. For this reason, food security and nutrition should be key elements in the response to HIV.

It has been argued that for many widow-headed households, the main constraint following the death of the spouse is not labour shortage (possibly due to changes in cropping patterns) but cash income (given that men are often the main cash income earners). In one forthcoming study, the most immediate need reported by widows in Tanzania was credit to establish small projects that could be combined with farm and domestic work.⁴⁸

Another example of the interface between gender, rural development and HIV is the impact of HIV/AIDS on gender roles in agricultural production at the household level, and in particular the constraints that the epidemic poses on the ability of affected families to cope. This is partly due to constraints inherent in the traditional division of labour and partly due to the fact that women and children may lack the requisite skills and experience for certain farming tasks. The latter constraint is exacerbated by HIV and AIDS because farming, marketing and managerial skills are not being passed fast enough by men on to women and their children (or by women on to men and their children). Issues that need to be addressed are: a) how to gain acceptance of new gender roles and tasks; and b) how to devise formal and informal ways of ensuring that requisite skills are passed on to women and children.49

Susceptibility and vulnerability to HIV/AIDS and the extent of the impact of the epidemic varies not only by gender but also by age group (infants, children, young adults, mature adults and the elderly). Given that the majority of Africa's population is under 15 years of age and that this largely corresponds to the "window of hope" (age that is relatively HIV/AIDS-free), children and adolescents are a particularly important target group. More generally, the felt needs and constraints of different age groups need to be considered in policy and programme responses:

• children (health, education, socialisation, and skills/knowledge that were not passed on to them from their parents, etc.);
• youth (skills, market information and ability to earn a living). Youth is increasingly becoming less interested in subsistence agriculture and more enticed by activities that yield quick cash incomes such as trade and high value horticultural crops. For this reason, they need different skills, inputs and market information. The implications for agricultural research, training and extension are significant as these institutions need to re-orient their focus and approach.⁵⁰
• the elderly as main family providers (how to enhance their food and livelihood security, etc.). This also has implications for research, training and extension, credit, etc.

Marital and family status also contribute to susceptibility and vulnerability to HIV/AIDS as well as to the extent of the impact of the epidemic. Differentiating between the following groups may be significant in policy and programme analysis and formulation:

• Single men and women
• Married men and women (monogamous versus polygamous unions)
• Divorced men and women
• Separated men and women
• Widows/widowers
• Cohabiting men and women
• Number of children/dependents

For example, the inter-relationships between polygamy and HIV/AIDS may be important for the spread of HIV and the generation of appropriate responses.⁵¹ While polygamy is widely practiced in Western Africa and the Sahel, it is also prevalent in Eastern and Southern Africa: for instance, in Tanzania and Uganda, about a third of rural married women are in polygamous unions; in Malawi, Zambia and Sudan, the rate is between 21-23%.⁵³

Issues to be investigated include the following:

• How and to what extent does polygamy facilitate the transmission of HIV?
• Given that polygamy is more prevalent in rural than in urban areas, is it contributing to bridging the gap between rural and urban HIV rates?
• Are co-wives (and their children) equal before the risk of HIV infection and AIDS impact? What are the inter-relationships between informal rural institutions, polygamous households and HIV/AIDS?

Difference in age between partners is a key variable of the epidemic and of polygamy (which tends to involve considerable differences in age between spouses) and thus all wives may not be affected equally from the impact of an ailing husband. It is likely that the risk is greater for younger wives. This has important implications for the spread of HIV as younger wives will have a greater chance of infecting others and of bearing infected children.

Another dimension of polygamy that needs to be considered is the heterogeneity of coping strategies. Household coping strategies of monogamous and polygamous unions are likely to differ, according to spouse rank or status. For instance, a first wife may receive or be able to muster more support from the extended family than a second or third wife; in addition, she may have older children who might get preferential treatment in terms of food or schooling within the family. Coping strategies are also likely to differ within polygamous unions: for example, some polygamous families live under the same roof while others live in different dwellings; some wives are fully supported by their husbands while others may have to fend for themselves and their children.

6. The policy and strategy recommendations put forth by the World Conference on Agrarian Reform and Rural Development and by the World Food Summit provide a springboard from which to mainstream HIV/AIDS in rural development policies and programmes. In particular, WCARRD's focus on poverty alleviation and participation by rural people in the institutions that govern their lives as a basic human right, and the World Food Summit emphasis on food security and sustainable human development are not only prerequisites for the revitalisation of the rural economy, but also for effective responses to HIV/AIDS.

For instance, the World Food Summit pledged "to ensure gender equality and empowerment of women.⁵⁴" To this end, it recommended the following actions:

• Promote women's full and equal participation in the economy, and for this purpose introduce and enforce gender-sensitive legislation providing women with secure and equal access to and control over productive resources, including credit, land and water;
• Ensure that institutions provide equal access for women;
• Provide equal gender opportunities for education and training in food production, processing and marketing;
• Tailor extension and technical services to women producers and increase the number of women advisors and agents; etc.⁵⁵

These recommendations provide an entry point and springboard for mainstreaming HIV/AIDS in rural development.

Similarly, the World Food Summit calls for the promotion and implementation of "agricultural and rural development schemes targeted at increasing on-farm and off-farm employment." Governments and civil society are urged to "adopt policies that create conditions which encourage stable employment, especially in rural areas, including off-farm jobs, so as to provide sufficient earnings to facilitate the purchase of basic necessities, as well as encourage labour-intensive technologies where appropriate.⁵⁶" Stable rural employment is likely to enhance livelihood security, slow down rural-to-urban migration, and increase rural incomes, all of which may, in turn, contribute to arresting the spread of HIV and mitigating the impact of the epidemic.

Thus, policy and programme responses to HIV/AIDS will benefit considerably from using the recommendations and follow-up activities to the two world fora (as well as other major world conferences mentioned above) as a platform for action.

7. Rural development policies and programmes in support of poverty alleviation, food and livelihood security, the empowerment of rural women, etc. are, in effect, also HIV prevention and AIDS mitigation measures and vice-versa.

For instance, any attempt to arrest the spread and impact of the epidemic must focus on rural poverty in general and vulnerable groups of rural poor in particular. Policies and programmes that reduce rural poverty will ease the economic constraints faced by the poor (and particularly women) in paying for essential HIV prevention services, such as treatment of STDs and condoms, and will therefore indirectly promote behavioural and lifestyle changes among rural men and women.⁵⁷

"There is potentially important synergy between AIDS mitigation and anti-poverty programs," has argued the World Bank. "For example, the finding that poor households are more vulnerable to the impact of an AIDS death implies that general anti-poverty policies can also be AIDS mitigation policies.⁵⁸" Thus, income-generating programmes for vulnerable groups that raise funds for social services, such as health dispensaries, can enable poor rural men and women to improve their health status as well as treat STDs, both of which will help minimise the risk of HIV transmission.

Other examples include the following:

• As seen in section 2.1, low potential agricultural areas are susceptible and vulnerable to HIV/AIDS, as young men tend to leave their villages in search of better employment opportunities in the towns. Thus, enhancing employment opportunities for rural youth and improving living conditions in rural areas are not only policies to stem rural exodus but are also AIDS prevention and mitigation measures targeting the group most at risk of HIV infection.⁵⁹
• Rural development policies and programmes aimed at improving rural women's access to sustainable livelihoods, and enhancing their living conditions and inferior socio-economic status are likely to reduce the spread of HIV transmission and lessen the impact of the epidemic. For example:

  - access to clean water is likely to have a marked effect on the amount of time women have for other productive activities and for childcare; it will also help improve domestic hygiene and health, enhance childcare as well as crop and/or animal care, and facilitate household maintenance;

- improving rural women's income earning opportunities by including them in income-generating schemes will enhance their livelihood security in times of stress (drought, HIV/AIDS, etc.); and

- household labour-saving appropriate technologies, such as fuel-efficient stoves, food-grinding machines, etc. will increase the amount of time women have for productive and childcare activities.

The World Bank finding that adult death depresses per capita food consumption in the poorest households by 15% implies that AIDS deaths occurring in poor households exacerbate poverty. Therefore, "when AIDS mitigation policies are targeted to households that were poor before the AIDS death, they are likely to prevent the affected household from slipping further into misery as a result of the death. In this case, AIDS mitigation could be effective in limiting the depth, if not the extent, of poverty.⁶⁰" In other words, AIDS mitigation measures can be effective poverty allevation measures as well.

8. While rural development programmes can be integrated with HIV/AIDS prevention and mitigation programmes, HIV/AIDS-specific policies and programmes have an important complementary role to play.

As HIV/AIDS-related issues extend across sectors, and given the current financial and human resource constraints, they must first and foremost be mainstreamed in existing development priorities. However, this is not to argue that HIV/AIDS-specific policies and measures have no place in a response to the epidemic. On the contrary, as experience from mainstreaming other cross-sectoral issues (such as gender and the environment) has shown, it is vital to draw attention to the significance of HIV/AIDS and to the specificity of its impact, analyse its rural dimension and pinpoint its implications for policies and programmes. Where appropriate, HIV/AIDS-specific IEC components may be introduced, severely affected or high risk population sub-groups may be targeted, and special programmes (such as on youth and HIV/AIDS) may be implemented.

Table 1: THE IMPLICATIONS OF HIV/AIDS FOR RURAL DEVELOPMENT POLICY & PROGRAMMING

N.B.

1. Where appropriate, susceptibility and vulnerability to HIV/AIDS, and the impact and implications of the epidemic for rural development policies and programmes need to be differentiated by gender, age and marital/family status.

2. Where possible, column 3 on the "Impact of HIV/AIDS" should also include the impact of rural development policies and programmes on the spread and impact of HIV/AIDS.

* This column refers to each and every focus area (poverty alleviation, food/livelihood security, etc.)

3.2 Rural development policy and programme focus areas

3.2.1 Poverty alleviation

According to an internationally adjusted standard of absolute poverty, sub-Saharan Africa has four times as many poor people as non-poor people.⁶¹ Economic growth in the region is expected to barely exceed 3.3% per annum, while the minimum growth rate needed to reduce the number of poor is 4.7%. Thus, under current conditions, poverty in sub-Saharan Africa is likely to increase further over the next decade.⁶²

Who are the rural poor in Africa?⁶³

• Smallholder farmers (73% of the rural population)
• Nomadic pastoralists (13% of the rural population)
• Households headed by women (31% of rural households) cross-cut these two groups
• Artisanal fisherfolk (5% of the rural population)
• Displaced people and refugees (about 14 million)

How are the rural poor susceptible and vulnerable to HIV/AIDS? To take but one example, displaced people and refugees (in countries like Angola [see Box 5], Liberia, Rwanda, Somalia, Sudan, etc.) are caught in a vicious circle of war, which exacerbates poverty and either forces them into refugee camps or prompts them to migrate in search of new homes. Social dislocation has been associated with high risk behaviours, such as drug abuse or commercial sex work. Refugee camps are extremely high risk environments for the transmission of HIV. In some countries, like Tanzania, it has been found that 33% of sexually active adults in refugee camps are HIV positive⁶⁴. For those fleeing war to find security in refugee camps, HIV can pose an equally deadly threat.

However, the risk is not limited to refugees inside camps. Often, the interaction between refugee camps and the surrounding communities can enhance the spread of HIV. The former may have more resources than the latter, leading people from nearby communities to enter the camps either in search of food or to exchange services with the refugees (food may be given in return for sex and so on). In both cases, given the high seroprevalence rate among refugees, this may lead to the spread of HIV from the refugee camps to surrounding communities. Given the systemic impact of the epidemic, institutional collaboration is critical: relief agencies should focus on refugee camps while rural institutions focus on poverty alleviation programmes in surrounding communities.

The effects of HIV/AIDS on rural poverty have only been indirectly documented in studies dealing with agricultural production systems in Eastern and Western Africa.⁶⁵ Household coping strategies are still poorly understood (particularly in terms of gender, age and marital status differentiation) and little is known of the long-term systemic impact of the epidemic on affected households (particularly on children and the elderly). For instance, are affected households able to recover from the shocks of young adult morbidity and mortality? Which households in particular (male/female-headed, monogamous/polygamous, etc.) are better able to cope with mortality and morbidity? Are households with infants and young children worse affected than other households and if so, what are the policy and programme implications? What will be the effects of impoverishment on the younger generation? In particular, what will be the effects of inter-generational poverty for rural development?

The following general observations can be made on poverty and HIV:

• Most people with HIV/AIDS are poor. This is why AIDS is characterised as a "disease of poverty" (see Box 6).
• The economic impact of HIV/AIDS is greater on poor households. Given that households which experience adult mortality draw on their assets (human and financial) to cushion the shock, households with fewer assets are likely to have more difficulty coping than households with more assets.
• Coping mechanisms are less effective in poor households. This is evident in the brevity of survival time from initial HIV infection to death in Africa (6-7 years rather than 15 years);⁶⁶ the deterioration of children's nutritional status; and the decline in school enrolment among poor households.⁶⁷

It has been argued that possibly the greatest impact of HIV/AIDS may be felt by those rural people whose activities are not counted by standard measurements of economic performance and productivity. The resources, time and labour of those working in the informal sector, in subsistence agriculture and in rural households (particularly women) are for the most part invisible in quantitative terms.⁶⁸

At the macro level, the impact of HIV on rural poverty may result in any of the following consequences:

• the creation of new, inter-generational poverty (through the impact of HIV on children and on asset depletion; see Box 10), deepening of existing poverty and increased indebtedness;
• increased discrimination and marginalisation of the poor, and especially women (who are often perceived to be responsible for transmitting HIV);
• increasingly unequal asset (and particularly land) distribution; it appears that land ownership is being concentrated through the effects of HIV (i.e. land is being sold to cover medical and funeral costs);⁶⁹
• increase in the prevalence of the feminisation of poverty and agriculture;
• rise in number of female-headed households; and
• increase in rural exodus and migration as a coping strategy for HIV/AIDS- affected households.

At the programme level, formal rural institutions engaged in poverty alleviation need to respond to HIV, as the viability of on-going programmes may be undermined. For instance, rural credit programmes (a key instrument of poverty alleviation) may be at risk as a result of HIV/AIDS for three reasons:

1. increased mortality may raise the number of defaults;
2. AIDS-affected families may be forced to liquidate their assets in order to repay the loans or else have their assets seized, thereby ending up worse off than before they incurred the loan; and
3. AIDS-affected families may have to spend part or all of the credit to finance medical care for family members suffering from AIDS.⁷⁰

Rural credit institutions need to assess whether HIV is affecting credit schemes in a particular area by inquiring if the demand for loans is on the rise and why this is so. Once constraints resulting from HIV have been identified, there are several ways of ensuring that credit schemes remain viable and that families affected by the epidemic benefit from the loans and are thus not forced to sell their assets (land, livestock, etc.) to pay for HIV/AIDS-related expenses. UNDP has been successfully operating micro-credit projects for families affected by the epidemic in rural Uganda, helping to finance income-generating on- and off-farm activities. Women have especially benefited from this credit scheme. Another example of innovative credit programmes for the poor is that of ACORD (see Box 7). Yet, these are isolated efforts. Policy reforms and programmes are needed in the area of credit and income-generation to help AIDS-affected families maintain their productive base.

In responding to the epidemic, formal rural institutions should bear in mind that many disadvantaged households, though not affected by HIV/AIDS, are so poor that their families suffer similar disadvantages. For this reason, the World Bank urges, assistance should be targeted on the basis of both direct poverty indicators and the presence of AIDS in a household, rather than on either indicator alone. One way to do this is to use adult death as a targeting criterion for poverty alleviation programmes (see Box 8).⁷¹ Another way of measuring the same poverty factor would be through household dependency ratios. This would capture households which have acquired additional dependents (i.e. children from AIDS-affected families) as well as those which have directly lost productive adults.

Based on this analysis, the case studies need to address the following questions:

• Are targeting strategies effective in reaching vulnerable groups of poor rural households? Are households affected by HIV/AIDS included among vulnerable groups?
• What are the effects of inter-generational poverty on rural development and what policies and programmes need to be put in place to counter-balance these effects?
• Is there an increase in landlessness in heavily affected areas and, if so, how can this be addressed? Which families are most likely to lose their assets and how can they be assisted most effectively? Who is benefiting from the adverse effects of HIV/AIDS (such as asset liquidation) and at what price and cost in terms of landlessness and levels of future poverty?
• What will be the effects of these poverty-related phenomena (asset/land concentration) for rural development? Are they likely to stimulate rural-to-urban migration? Are they likely to increase the ranks of the landless and dispossessed? If so, what are the policy and programme implications for rural development?
• How can targeted poverty reduction efforts be effectively coordinated with programmes to mitigate the impact of the epidemic?

3.2.2 Food security and sustainable livelihoods

Access to an adequate amount of food is the most basic of human needs and rights. Food security is dependent on four factors: availability, stability and accessibility of food, and good health. To achieve national food security, a country must be able to grow sufficient food or have enough foreign exchange to enable it to import food. Similarly, households must have sufficient income to purchase the food they are unable to grow for themselves. The basic causes of food insecurity are low productivity in agriculture combined with fluctuations in food supply, low incomes and insecure livelihoods. About 44 out of 49 countries in Africa today are classified by FAO as low-income and food-deficit (LIFDCs).⁷²

Food-insecure adult household members are susceptible and vulnerable to HIV/AIDS: malnourishment and poor nutrition contribute to a poor health status, and by extension, to low labour productivity, low income and livelihood insecurity. It is unlikely that people with low incomes will be able to treat STDs or opportunistic infections associated with AIDS. PLWAs are particularly vulnerable as, without a good diet, they cannot prolong good health and live a longer life to provide for their children.

Key factors related to HIV/AIDS that impact on household food insecurity include:

• the loss of prime-age adult on- or off-farm labour; labour shortages lead to a decline in productivity;
• a decline in household income and loss of assets, savings, remittances, etc;
• an increase in household expenditures (medical treatment and transport, special foods for the infirm, etc.) and
• an increase in the number of dependents relying on a smaller number of productive family members.

These factors may result in any of the following scenarios:

• they may trigger food insecurity in previously unaffected households;
• they may increase the frequency and extent of food insecurity among households previously only suffering from seasonal food insecurity;
• they may render vulnerable households chronically food insecure and their members chronically undernourished with severe consequences for infants, young children, pregnant/lactating women and the elderly.

At the programme level, the viability and sustainability of food security programmes may be undermined by the epidemic: for example, the impact of HIV on livestock programmes may mean that:

1. family members have to sell their livestock to finance medical care for AIDS patients;
2. if the person in charge of the livestock dies, family members are often unable to manage due to the loss of skills and relevant experience;
3. in some parts of Uganda, it has been reported that the price of livestock has declined substantially as a result of people selling their animals partly because of livestock diseases and partly because of AIDS.⁷³

Box 9 shows how food production and post-harvest protection programmes may similarly be affected by the epidemic.

Given the labour shortages experienced by many rural households as a result of migration, shifting employment patterns and HIV/AIDS, there may be a need, in some instances, to review labour-intensive food production strategies, upon which food security policies and programmes are often based. In particular, it is important to re-assess labour-intensive food production strategies in areas heavily affected by the epidemic where there may be a need for:

• research, dissemination and promotion of labour-saving technologies and improved farming practices for men and women farmers;
• promotion of drought- or disease-resistant crop varieties;
• adaptation of post-harvest protection measures to account for the loss of knowledge on storage of particular crops (see Box 9); and
• dissemination of knowledge on nutrition, diet and health.

Labour shortages also raise the issue of the sustainability of traditional agricultural production methods. In Kagera, Tanzania, for instance, bananas are not being mulched or replanted in heavily affected areas, resulting in falling yields and reduced soil fertility. Fallowing practices entail clearing new areas every few years, but as labour is no longer readily available, there is overcropping. What will be the impact on yields and on food security in the medium term? How can traditional coping strategies be maintained given the increasing pressures on subsistence agriculture?⁷⁴

Agricultural research programmes need to investigate farmers' supply response to AIDS in terms of output or labour inputs and the special needs of farm households with fewer working adults and higher dependency ratios (i.e. for appropriate technology). To date, agricultural research has, for the most part, remained static and has not focused on the changing felt needs of farmers. According to one researcher, "farmers [in Tanzania] seem to be ahead of the national agricultural research systems (NARS)."⁷⁵ The question remains: how can agricultural research, technology and extension become relevant to those who need it most? And how can be it be re-oriented to focus on rural environments in flux?

Agricultural extension programmes need to ensure that strategies for labour-substitution, technical advice and credit services are made available to enhance food and livelihood security.⁷⁶ They also need to review the impact of HIV/AIDS in terms of increased mortality among agricultural extension staff and of the reduction of the work week as a result of the rise in funeral attendance. In some parts of Uganda, for instance, the six-day work week has been informally reduced as a result of increased mortality and morbidity related to HIV/AIDS.

The link between household labour shortages and livelihood insecurity is another major area of concern. Labour shortages reduce opportunities for labour diversification within households. Such diversification is critical for livelihood security, particularly in low-potential areas or in regions plagued by drought where agricultural incomes are not assured.

Coping mechanisms of informal rural institutions with regard to food insecurity are poorly understood, particularly cooperative production and marketing arrangements. For instance, how does a community copes with labour shortages in an AIDS and non-AIDS scenario. Do traditional labour-sharing arrangements (communal/individual) operate under the impact of HIV/AIDS? How does the principle of reciprocity work in areas heavily affected by HIV/AIDS? What happens when AIDS-affected households cannot contribute labour and thus do not receive assistance when they need it most? How vulnerable are traditional labour-sharing coping mechanisms to HIV and how can these be strengthened? Are there any relief-oriented mechanisms in place (i.e. by church groups) when other reciprocal mechanisms fail?

Labour mobility and migration are equally important factors in the analysis of food security and sustainable livelihoods. Rural-to-urban migration has traditionally acted as a safety valve and mechanism for rural accumulation (through remittances and savings for investments in technology and inputs). The HIV epidemic is eroding the savings capacity of rural households as youth that migrates to urban areas and contract HIV return to their villages when they fall sick. The question is, what effects does the epidemic have on savings flows and other remittances to rural areas and what are the ramifications for rural development? Has the behaviour of rural households using migration as a strategy for accumulation changed in the face of HIV?⁷⁷

Nutrition can be used as an entry point for a more comprehensive understanding of the inter-relationships between HIV/AIDS and food/livelihood insecurity, as well as of the changing dynamics of food and nutrition coping mechanisms at the household level. Given that nutrition assessments are location-specific (being dependent on the agro-ecological, socio-economic and cultural environments), they can also serve as AIDS mitigation measures, where appropriate.

In view of the above, the case studies should consider the following:

• What is the impact of the epidemic on the nutritional status of infants, children, pregnant women, the elderly and adults? Are there signs of gender-based and/or age-based differentiation?
• If food is the most immediate problem for many HIV/AIDS affected households (as seen in section 3), how can food security programmes provide relief when needed and what is the most effective way of delivering it?
• What are the main food security coping strategies adopted by male- versus female-headed households?
• How can livelihood security strategies be integrated in poverty alleviation and food security programmes?
• How can the livelihoods of the poorest be strengthened through self-help mechanisms (savings and funeral societies/groups, etc.)?
• How can research for technology development be targeted to the needs of poor, smallholder farmers?

3.2.3 Empowerment of rural women

Despite their critical contribution to food production, food security and rural development in general, rural women in Africa have often been overlooked by rural institutions. Their limited access to productive resources (land, water, etc.), technology, inputs, support services (agricultural research, training and extension, credit, markets) and social services (education, health, etc.) and low socio-economic status are important in the context of HIV.⁷⁸ "Low-income, income inequality, and low status of women are all fairly highly associated with high levels of HIV infections," argues Martha Ainsworth of the World Bank.⁷⁹

The illness and/or death of a woman is likely to threaten household food security, given that women provide the bulk of the labour for food production, animal tending, crop planting and harvesting across Africa. In fact, it has been argued that female morbidity and mortality has "a particularly dramatic impact on the family."⁸⁰ If women fall ill while their husbands are working in urban areas, the overall socialisation and education of the children and the management of the household may be seriously affected.⁸¹ Moreover, studies have shown that children's nutritional status is more closely related to the mother's work and income than to the father's.⁸²

This brings us to the gender-, age-, and marital status-differentiated effects of HIV on household income and expenditure. Little is known on how household expenditures change under the impact of HIV/AIDS morbidity and mortality. It has been found, however, that women tend to spend their income toward family needs and well-being (particularly on nutrition, health and education for their children), men tend to spend their income toward their personal needs (durable goods, such as radios, alcohol, entertainment). What are the implications of changes in household expenditure patterns for the nutrition, education and general welfare of young children and youth?

"As the status of women improves, as incomes improve and so forth, we can expect that [HIV] infection levels will be lower," Ainsworth argues.⁸³ Thus, women's access to formal rural institutions, such as health and education, is essential for HIV prevention while their access to technology, inputs, credit, etc. is likely to mitigate the impact of AIDS by enhancing women's employment opportunities and incomes.

Policies aimed at improving the socio-economic status of women are likely to enhance some women's ability to negotiate safe sex, although less so in rural areas. Nevertheless, expanding female education and employment opportunities, guaranteeing basic inheritance, property, and child custody rights; and outlawing and severely punishing slavery, rape, wife abuse and child prostitution,⁸⁴ are essential for the protection of vulnerable women from HIV/AIDS.

The support of formal rural development institutions will not be enough to diminish the susceptibility and vulnerability of women to HIV/AIDS. As seen in section 2.3.2 and in Box 11, the social safety net that informal rural institutions have traditionally provided to poor, vulnerable women is now threatening their very lives and families through practices such as wife inheritance, sexual cleansing rites, polygamy, etc.

These practices, in effect, create an impasse for rural women: on the one hand they have limited access to formal rural institutions; on the other hand, the traditional support they have received from the extended family and kinship systems is, under the influence of HIV, becoming a source of further insecurity. If they refuse the support offered by traditional rural institutions they will be ostracised, turned away from their homes and left unable to support their children (given the absence of support from formal rural institutions). If they comply, they may get infected and die, possibly also giving birth to children with HIV or, if they are already infected, they may infect other members of the extended family. Thus, if women merely refuse to give in to traditional practices, they will be faced with the loss of the last vestige of support available to them.

Given the inter-relationships between HIV, gender and rural development, the case studies should address the following questions:

• How can rural women's access to rural institutions be facilitated and their participation in rural institutions strengthened?
• Given that the main constraints to the introduction or enforcement of gender-responsive legislation providing women with secure and equal access to and control over productive resources are customary laws and practices, how can these be overcome? In particular, how can formal and informal rural institutions work together to overcome these constraints?
• How can formal rural institutions facilitate the provision of equal gender opportunities for education and training in food production, processing and marketing and improve women's access to agricultural extension services?
• How can formal and informal rural institutions work together across sectors to help modify traditional social safety mechanisms such as wife inheritance, sexual cleansing rites, etc. so as to minimise the spread of HIV?
• What are the coping mechanisms of monogamous versus polygamous unions? Is there co-wife solidarity, as is often assumed, or are infected wives and their children ostracised?

3.2.4 Labour

A first important point that needs to be made when assessing the impact of HIV on labour is that it is not homogeneous but heterogeneous. For instance:

• Within a rural household, there is a marked difference in impact depending on whether it is the man or woman who is affected by HIV/AIDS. For instance, in many communities, sowing is done by women. The death of a woman essentially deprives the farm of accumulated sowing skills that other household members may not have. The loss of such skills may affect productivity. Another example of the heterogeneity of household labour is seen in Box 9 on the impact of HIV/AIDS on post-harvest activities.
• Within a rural enterprise, there is great difference in impact depending on whether a project manager versus an unskilled or semi-skilled worker is affected by HIVAIDS, as the former will be considerably more difficult to replace than the latter.

Thus, the heterogeneity of labour is important when assessing the impact of HIV, as this should not be merely analysed in terms of numbers of casualties but in terms of:

• who is affected within a household, enterprise, etc.?
• at which point of the production cycle does the loss occur?
• what skills are being lost and at what cost to the household, enterprise and rural economy?
• what are the policy and programme implications of these losses for the formal and informal sectors?

The remainder of this section will focus on the interface between HIV and the formal labour sector in the Southern African Development Community (SADC), where of the formal sector working population (urban and rural) of about 61 million, some 15 million employees are expected to die of AIDS in the next decade.⁸⁵ In particular, the main challenges posed by HIV/AIDS for rural labour and employment will be reviewed and discussed. SADC's Employment and Labour Sector has begun to address three main aspects of the epidemic:

• Human rights, or more specifically employment rights (particularly issues related to the stigma and discrimination associated with HIV/AIDS). A code on AIDS and Employment for SADC is being developed to ensure non-discrimination between individuals with HIV and those without, and between HIV/AIDS and other comparable medical conditions.
• Production and productivity issues.
• Employment and labour market issues, including employee benefits and social security issues.⁸⁶

The following labour policy challenges posed by the epidemic remain to be addressed:

• The role of the workplace in HIV prevention.
• Lost skills and experience and the need to deal with anticipated labour shortages⁸⁷ (through multi-skilling, reviewing training practices and doubling up on task performance capability).
• The substitutability of labour. Various kinds of labour are in scarce supply to rural development (i.e. supervisors in plantations, credit managers in rural banks, etc.)
• Losses in production/sub-optimality of production. The main problem arising from HIV/AIDS impact may not be the loss of unskilled labour that can be relatively easy to replace. It may lie in the obstruction of the smooth operation of production, which depends on forms of capital equipment that need to be maintained and repaired without delay when problems arise. Thus, the loss of skilled labour, management capacity, transport, and other services addressing the needs of farmers has to be taken into account. The question is if HIV obstructs the production process from functioning smoothly, how does the sub-optimality of production (higher costs and lower profit margins) affect the production process, employment, and the revenues of producers?⁸⁸ This is all the more important given the increasing trend toward globalisation.
• Rising payroll costs (depending on the level of benefits) related to medical costs, social security, insurance and employee benefits. The issue at stake is how benefits⁸⁹ (for medical health and life insurance schemes) can be maintained and human rights be protected without discriminating against people with HIV/AIDS, whilst containing costs within acceptable limits. These issues concern employers and employees alike. To respond appropriately, there is a need for accurate data on the health of the labour force and on health-related benefit costs.

Attitudes toward HIV among the private sector have changed over time. One private employer representative from Swaziland recently admitted that initially, most employers expected government departments to respond to HIV/AIDS, viewing it as a medical problem. As the epidemic began affecting the private sector in increasing numbers, however, employers began working together with governments and NGOs, particularly in HIV/AIDS awareness campaigns.⁹⁰

In addition, home ownership, the promotion of family lifestyle, the elimination of discrimination of women in the workplace and the provision of primary health care (to prolong the life of HIV positive workers and to reduce the risk of HIV transmission by offering treatment for STDs) are increasingly being recognised by the private sector as cost-effective strategies that can have a impact on the containment of the epidemic, according to a spokesman from the Mhuluma Sugar Estates of Swaziland (see box 12).⁹¹

Questions that need to be addressed include the following:

• How does HIV-related sub-optimality of production affect the production process, employment, and producer revenues?
• Does the country in question have a code on AIDS and employment? If not, how can such a code be put in place and be effectively enforced?
• What is the role of the workplace in HIV prevention and what lessons can be learned from the private sector?
• How can employers maintain employee benefits without discrimination, whilst containing costs within acceptable limits?
• How can public and private rural institutions cooperate in the effort to resolve some of the challenges posed by HIV on labour and employment?

3.2.5 Infrastructure

Rural physical infrastructure can be either public (roads, ports, dams, etc. as well as schools, health facilities, marketplaces) or private (housing). Infrastructure is essential for improving access to services (health, education, agricultural extension, credit, etc.), marketing, trading and raising rural incomes, and more generally, for strengthening rural-urban linkages. Yet, while infrastructure programmes are central to the revitalisation of the rural economy, they can also help spread HIV. How do investments in infrastructure make rural men and women susceptible and vulnerable to HIV/AIDS?

• Rural workers involved in the construction, repair or maintenance of infrastructure are separated from their families over prolonged periods of time and have disposable incomes to spend at their place of work.
• The community in which the infrastructure programme is being constructed, and particularly the women who come to work in the hotels, bars and restaurants created to service these workers, often end up as casual or semi-permanent sexual partners to these construction workers.
• The wives or sexual partners of the construction workers in rural areas and their families may contract HIV from their migrant husbands/partners.
• When farmers are displaced for the construction of large infrastructure projects, as was the case with the Volta River Dam in Ghana (see below), they become susceptible and vulnerable to HIV/AIDS, as they may lose their only means of supporting their families: their land.
• Operation and maintenance professional staff often commute over long distances to oversee their work and are separated from their families.

The construction of the Volta River Dam in Ghana four decades ago necessitated some 8,500 square kilometres to be cleared for the dam's reservoir. This displaced thousands of farmers, many of them women. Men found jobs on the construction site and later as fishermen on Lake Volta. But many women farmers ended up as service workers in the hotels and bars built to cater the construction workers. Pushed by economic necessity, some women had little choice but to resort to prostitution. Illegitimate children were born and, a generation later, large numbers of young women followed in their mothers' footsteps.⁹² HIV prevalence at the side of the dam was found to be five to ten times higher than in the rest of Ghana.⁹³ In addition, migration of commercial sex workers to bigger communities, such as Accra and Kumasi, created fertile ground for the recent spread of HIV.

In Malawi, road construction has been linked to the spread of HIV, while in Lesotho, the Highland Water Project has led to an increase in sexually transmitted diseases in the remote mountain areas, according to recent studies.⁹⁴

Rural markets constitute another critical infrastructure sub-sector that urgently needs to be addressed, given the linkages that have been established (even if only anecdotally) with HIV. Markets involve the movement of large numbers of people who come together to sell or buy goods or to exchange services. This mobile population of sellers and vendors has disposable incomes that are spent in bars, hotels and truck driver stops that abound in market areas. These factors facilitate HIV transmission. Reviewing certain market rules and regulations in conjunction with town councils, such as market opening and closing times, could have a significant impact on arresting the spread of HIV.

Thus, rural infrastructure programmes may increase susceptibility and vulnerability to HIV unless the socio-economic, socio-cultural and gender environments are taken into account, vulnerable groups are identified and remedial measures are adopted. To this effect, rural infrastructure policies and programmes have to build-in mechanisms to reduce the spread of HIV/AIDS not only during construction but also after project completion. Some donors are already moving in this direction. For instance, the European Community indicates in its Transport Sectoral Guidelines that:

"While new and improved infrastructure brings economic and social benefits, it can also facilitate the spread of disease. Opening up new traffic routes and improving access and personal mobility can contribute to the rapid spread of communicable disease such as AIDS. Extra health measures must be given (sic) by health agencies and contractors during the construction stage when there might be many temporary migrants in a community. When works are complete, health agencies must cope with the greater number of travelers who can both bring and carry away infections".⁹⁵

The World Bank is also addressing the issue of increased vulnerability to HIV and rural infrastructure by incorporating STD/AIDS-specific strategies and activities in its projects (see Box 13).

Conversely, some infrastructure facilities, such as health dispensaries, primary health care centers, schools and youth centers can help to arrest the spread of the epidemic. Health dispensaries availing drugs for STDs, information on HIV prevention, and condoms can help reduce the risk of HIV transmission, while medication for opportunistic infections (tuberculosis, diarrhea, etc.) can help PLWAs live longer. Schools with IEC programmes can raise awareness of HIV transmission, patient care and help eradicate the stigma associated with AIDS among youth not yet sexually active. Rural youth centers can disseminate information about STDs/AIDS and condoms as well as provide entertainment for young people to help minimise the spread of the epidemic among a highly vulnerable group. Thus, strengthening this type of rural infrastructure is also an effective HIV prevention and mitigation strategy.

Construction of infrastructure is only one aspect of the problem. The other acute problem is the impact of HIV/AIDS on the maintenance and operation of existing infrastructure. For instance, Malawi is suffering from the losses of highly qualified water engineers that cannot easily be replaced.⁹⁶ More generally, the issue is how roads, dams, schools, public health care centres, irrigation systems etc. will be maintained and by whom, given the losses in human resources, the expanding needs due to population growth and increasing pressures on funding.

Last but not least, is the issue of housing. The construction, maintenance and repair of housing involves tasks that are usually differentiated by gender. Men are often responsible for construction while women tend to be in charge of maintenance and repair. However, housing may deteriorate as a result of adult illness and death, either through the loss of (gender-differentiated) skills within a household or through the loss of income for repair materials. Poor housing means leaking roofs and a deterioration of children's health, leading to susceptibility to diseases (tuberculosis, bronchitis, etc.). It may also mean less security for rural families from bandits, wild animals, etc. The issue of housing may not have been addressed to date but is critical for affected families. In particular, there is hardly any empirical knowledge on how HIV/AIDS affects traditional gender roles in housing construction, maintenance and repair, and in particular, on the loss of knowledge and skills related to housing and on the resulting implications for family welfare. Similarly, little is known about how affected families cope; if they receive any support from the community and if there are any mechanisms in place at the community level to improve housing conditions for affected families.⁹⁷

Issues that need to be addressed by the case studies include the following:

• Are the working, living and family conditions of migrants engaged in infrastructure programmes taken into account? Is primary health care available to them? Do nearby health centers have information on HIV, trained councilors and testing facilities?
• What can be done to minimise the risk of HIV transmission in rural markets and to raise awareness of HIV/AIDS among men and women who frequent the markets?
• How can the adverse impact of rural infrastructure programmes be counterbalanced through, for example, strengthening health or education infrastructure so as to arrest the spread of HIV?
• How can the maintenance and operation of essential infrastructure be ensured given growing pressures resulting from the loss in human resources, funding constraints and population growth?
• How can the effects of HIV/AIDS on housing be addressed and what mechanisms can be put in place to assist affected families and communities?

  ---

Notes:

41. Personal communication, Desmond Cohen, UNDP, HIV and Development Programme, April 16, 1998.

42. Personal communication, Jacques du Guerny, FAO, Focal Point on AIDS, April 17, 1998.

43. Life Expectancy Rate with and without AIDS in Africa: 1996 and 2010, International Programs Center, Population Division, U.S. Bureau of the Census, Washington, DC.

44. Evidence from high HIV prevalence countries in Africa experiencing mature epidemics shows that growth rates of GDP may be reduced by 0.5% to 1% per annum due to the epidemic. A reduction in educational achievement is also being recorded, especially of girls and young women. See Desmond Cohen, "The HIV Epidemic and Sustainable Human Development," paper prepared for the 4th International Congress on AIDS in Asia and the Pacific, October 1997, pp. 8-12.

45. IFAD, The State of Rural Poverty, cited in "Inside the Poverty Trap," Rural Development, No. 15, January 1994, p. 10.

46. Ibid.

47. D. Topouzis and G. Hemrich. The Socio-Economic Impact of HIV/AIDS on Rural Families in Uganda, op. cit., and A. Michaud, Impact du VIH/SIDA sur les systèmes d'exploitations agricoles en Afrique de l'Ouest, op. cit.

48. Personal communication, Gabriel Rugalema, ISS, 2 May 1998.

49. Personal communication, Desmond Cohen, UNDP, 28 April 1998.

50. Ibid.

51. Personal communication, Jacques du Guerny, April 29, 1998. See also Philippe Antoine and Marc Pilon, "La Polygamie en Afrique: Quoi de neuf?" La chronique du CEPED, Centre français sur la population et le développement, January-March 1998, No. 28, pp. 1-4.

52. In Burkina Faso, Togo, Benin, Senegal and other West African countries, one in two rural married women are in polygamous marriages, see Antoine and Pilon, ibid., p. 1-2.

53. Ibid.

54. World Food Summit Plan of Action, FAO, 1996, p. 11.

55. Ibid. p. 12.

56. Ibid., p. 14.

57. The World Bank, Confronting AIDS, op cit., executive summary, p. 8.

58. Ibid., p. 232.

59. Topouzis & Hemrich, "The Socio-Economic Impact of HIV/AIDS on Rural Families," op. cit.

60. World Bank, Confronting AIDS, op. cit., p. 232.

61. The World Bank, Confronting AIDS, op. cit., p. 208.

62. "Rural Poverty in the 1990s," Rural Development, No. 15, 1994, pp. 8-9.

63. IFAD, The State of World Rural Poverty 1994, cited in ibid.

64. Cooperative for Assistance and Relief Everywhere (CARE) estimates, cited in AIDS Prevention and Mitigation in sub-Saharan Africa: An Updated World Bank Strategy, 1996, p. 1.

65. T. Barnett, "The Effects of HIV/AIDS on Agricultural Production Systems and Rural Livelihoods in Eastern Africa," op. cit., and the Zambia, Uganda and Tanzania country reports on which the above summary is based; A. Michaud, "Impact du VIH/SIDA sur les systèmes de production agricole et l'environnement rural au Burkina Faso et en Côte d'Ivoire," op. cit.; Topouzis & Hemrich, The Socio-Economic Impact of HIV/AIDS on Rural Families in Uganda, op. cit.; The World Bank, Confronting AIDS, op. cit., chapter 4: Coping with the Impact of AIDS.

66. Desmond Cohen, "Poverty and HIV/AIDS in sub-Saharan Africa," op. cit., p. 5.

67. The World Bank, using a combination of theoretical analysis and empirical observation and relying primarily on a single survey in Kagera, Tanzania, has arrived at the following conclusions on the interplay between HIV/AIDS and poverty: Most AIDS-afflicted households in Africa are not destroyed (as originally assumed) but respond to the impact of AIDS (and other shocks) using three main coping strategies: Altering household composition (for example, by sending one or more children to live with relatives, or inviting a relative to join the household in exchange for assistance with farming, household and childrearing tasks). The survey found that most households that have lost a member to AIDS are able to adjust household size and dependency ratios in ways that make them similar to households that did not suffer a death. Drawing down savings or selling assets (durable goods, livestock, etc.). World Bank data also shows a considerable decline in participation of households experiencing an adult death in traditional rotating savings and credit associations (ROSCAs) from Kagera, Tanzania. Utilising assistance from other households and from informal rural institutions. Given the scepticism surrounding the methodology and findings of these studies, the Kagera analysis and conclusions have only been selectively incorporated in this paper. See Confronting AIDS, op. cit.

68.SADC, Conference on HIV/AIDS, op. cit., p. 7.

69. Personal communication, Desmond Cohen, UNDP, 16 April 1998.

70. Daphne Topouzis, "The Implications of HIV/AIDS on Investment Centre Work," FAO, January 1994, p. 24.

71. Ibid. The term "working-age" or "productive" adult may be more appropriate than the World Bank's Aprime-age adult."

72. Personal communication, A. Aviles, Attaché du Cabinet, ODG, FAO, April 29, 1998; see also FAO, The Special Programme for Food Security, September 1996.

73. D. Topouzis, "The Implications of HIV/AIDS on Investment Centre Work," op. cit., p. 17.

74. Personal communication, Desmond Cohen, UNDP, 28 April 1998.

75. Personal communication, Gabriel Rugalema, ISS, 2 May 1998.

76. For a review of the impact of HIV/AIDS on agricultural extension services, see Eric Baier, "The Impact of HIV/AIDS on Rural Households/Communities and the Need for Multi-Sectoral Prevention and Mitigation Strategies to Combat the Epidemic in Rural Areas," FAO, 1997.

77. Personal communication, Desmond Cohen, UNDP, 28 April 1998.

78. See Jacques du Guerny and Elisabeth Sjoberg, "Inter-relationship between Gender Relations and the HIV/AIDS Epidemic: Some Possible Considerations for Policies and Programmes," AIDS, No. 7, 1993, pp. 1027-1034.

79. World Bank Newsletter, November 6, 1997.

80. Steven Forsythe and Bill Rau (eds.), AIDS in Kenya: Socio-Economic Impact and Policy Implications, USAID/AIDSCAP, 1996, p. 29.

81. Ibid.

82. Stephen Devereux and Graham Eele, "The Social and Economic Impact of AIDS in East and Central Africa," International Development Centre, Food Studies Group, May 1991, cited in ibid.

83. Stephen Devereux and Graham Eele, "The Social and Economic Impact of AIDS in East and Central Africa," International Development Centre, Food Studies Group, May 1991, cited in ibid.

84. Ibid.

85. SADC Conference on HIV/AIDS, op.cit., p. 13.

86. ibid., p. 14.

87. ibid, p. 13.

88. Personal communication, Desmond Cohen, UNDP, April 16, 1998.

89. Ibid., p. 14. For an analysis of the insurance perspective and HIV see pp. 16-17.

90. Ibid., p. 14.

91. Ibid., p. 15.

92. J. Decosas, quoted in Gerry Toomey, "HIV and Development: Decosas Advocates Local Solutions," The Daily Progress, newsletter of the XI International Conference on AIDS, Vancouver, 11 July 1996.

93. AIDS Analysis Africa, Vancouver AIDS Conference Special Report, Vol. 6, No. 4, August-September 1996.

94. "AIDS and the Transport Sector," Considering HIV/AIDS in Development Assistance: A Toolkit, EEC, 1997.

95. "HIV and the Transport Sector," Considering HIV/AIDS in Development Assistance: A Toolkit, EEC, 1997.

96. Personal communication, Desmond Cohen, 28 April 1998.

97. Ibid.