3. THE RELEVANCE OF HIV/AIDS FOR MoAs

The relationship between the HIV epidemic and MoAs is bi-directional:

• the HIV epidemic affects MoAs and their work (their clients, staff, services, programmes, etc.);

• MoAs can affect the spread of HIV infection and the impact of AIDS through their policies, strategies and programmes.

This section examines the impact of HIV/AIDS on MoAs and their work, identifies key issues and selected MoA coping mechanisms, and provides examples of response measures. Four areas of HIV/AIDS impact are reviewed:

a) MoA staff vulnerability to HIV infection and AIDS impact;

b) the disruption of MoA operations and the erosion of capacity to respond to the challenges being posed by the HIV epidemic (see Box 4);

c) the increased vulnerability of MoA clients to food and livelihood insecurity; and;

d) the continued relevance of certain MoA policies, strategies and programmes, in view of the conditions being created by HIV/AIDS (see Box 4).

The findings of this section are summarized in the table on the inside back cover.

3.1 MoA staff vulnerability to HIV infection and AIDS impact

MoA staff and their families are directly affected by the HIV epidemic through HIV/AIDS-related morbidity and mortality. Levels of HIV/AIDS prevalence among MoA staff are likely to be at least as high as national average estimates. This would translate into prevalence rates of nearly 36% within the Ministry of Agriculture in Botswana, 25% in Swaziland, 23% in Lesotho, and just under 20% in Zambia and Namibia. Preliminary evidence shows that in Kenya's MoA, 58% of all deaths in the last five years have been AIDS-related.^[14] In Malawi, at least 16% of the staff of the Ministry of Agriculture and Irrigation (MoAI) are living with HIV/AIDS, 76% have lost at least one colleague and 60% have lost at least one close relative to AIDS.^[15]

Relatively little is known about the impact of the epidemic on MoA professional and support staff and on how these are coping. According to the MAAIF in Uganda, "support staff has fared even worse than professional staff [in terms of HIV/AIDS impact]: their low income and need to supplement their earnings by seeking favours, which are sometimes paid back through unprotected sex, has made them particularly vulnerable to HIV infection."^[16]

3.1.1 MoA staff knowledge and awareness of HIV/AIDS

It is commonly assumed that government staff in countries heavily affected by the epidemic has adequate knowledge of HIV prevention, care and support. However, such assumptions may be misleading, as HIV/AIDS awareness-building exercises are often one-off events that may only target some of the staff. In particular, support staff may not be included in such exercises. Moreover, more emphasis is often placed on HIV prevention rather than on care and support and on coping with AIDS impact.

A case in point is Malawi where an impact assessment survey found that knowledge of HIV/AIDS among MoAI staff was far from adequate: While 66% of survey respondents knew that HIV/AIDS was preventable, 9% believed this was not the case while about 25% were not certain. The survey also found that 70% of the respondents felt relatively safe from HIV infection. Among female technical employees in particular, more than 80% felt that HIV infection was not a serious risk. However, among female industrial employees, 42% felt they were at considerable risk.^[17] These findings show that: a) the perceived self risk of HIV infection was very low in the ministry; b) awareness and knowledge of the epidemic may not be sufficiently internalised; and c) certain categories of employees may be more vulnerable to HIV infection than others. Lastly, it is likely that certain categories of MoA staff (such as drivers, messengers and other support staff) are not being reached by information, education and communication (IEC) campaigns.

3.1.2 Attitudes toward HIV/AIDS within MoAs: stigmatisation and discrimination

In a number of countries, MoA staff attitudes toward HIV/AIDS may not be characterised by tolerance, acceptance and supportiveness as commonly assumed. In particular, AIDS stigmatisation may be widely prevalent. This is partly reflected in the fact that AIDS is rarely acknowledged as a cause of death even in countries that are "open" about the epidemic, such as Uganda. While the degree of stigmatisation may vary widely from country to country, AIDS stigma is a key constraint in confronting the epidemic: in particular, it may deeply affect working relationships, staff performance and morale and may undermine efforts to mitigate its effects.

Negative attitudes toward MoA staff living with HIV/AIDS and their families have important implications: asymptomatic staff may be reluctant to disclose their status early in fear of losing their jobs; staff living with AIDS may be discriminated against in the workplace and may be forced to retire at a time when they need the income most; and staff with family members living with HIV/AIDS may live in fear of the consequences of being "found out." Anecdotal evidence on discrimination in the workplace is abundant but no systematic analysis has been undertaken of discrimination in MoAs.

3.1.3 MoA staff exposed to high risk situations

There are certain categories of MoA staff that may be particularly vulnerable to HIV infection. These would include mobile professional and support staff who need to travel in order to carry out their duties: agricultural extension workers, high level professionals who frequently attend seminars, conferences and in-service training as well as drivers. These groups often have to spend extended periods away from their homes and families.

An impact assessment conducted by the Ministry of Agriculture and Irrigation of Malawi found that among MoA male staff, drivers, supervisors, middle and top managers were most vulnerable to HIV infection. Among female staff, messengers and secretaries were perceived to be most vulnerable. Reasons given for this increased vulnerability included: i) the fact that these jobs required frequent travel to the field, which separated employees from their spo uses for prolonged periods of time; ii) better-off male staff were more likely to have more than one sexual partners; iii) worse-off female staff were more likely to offer sex for money.^[18]

3.1.4 Response Measures

Integrated HIV/AIDS workplace programmes

Few MoAs have HIV/AIDS workplace programmes that extend beyond HIV prevention. Issues relating to care and support of persons living with HIV/AIDS and/or their families tend to be dealt with on an ad hoc basis while working conditions continue to be the same. Integrated AIDS workplace programmes are needed to:

a) institutionalise IEC prevention initiatives and ensure that awareness-building campaigns in particular target both professional and support staff;

b) create a supportive working environment by eliminating the stigma surrounding HIV/AIDS. It cannot be emphasised enough that as long as HIV/AIDS stigma is present, it is unlikely that an enabling environment can be put in place to address the vulnerability of MoA staff to HIV infection and AIDS impact;

c) prevent discrimination of employees living with HIV/AIDS and/or their families through appropriate policies and adjustment of benefits and procedures that take into account AIDS impact;

d) modify working conditions of employees exposed to high risk situations which render them vulnerable to HIV infection (i.e. align duty station and home bases so that MoA staff does not work in one area and live in another; limit the number of overnight stays required of MoA staff during duty travel, etc.). A concerted effort should be made not to stigmatise these employees by singling them out but to address the conditions which expose them to an increased risk of HIV infection.

e) help staff members and their families cope with AIDS impact and plan for the future through counseling, legal advice, loans, etc.

There are a number of guidelines for developing workplace policy and programmes on HIV/AIDS, including one developed by the Community Agency for Social Enquiry (CASE) in South Africa.^[19] Such guidelines can assist MoAs in defining the concerns of their employees; identifying the responsibilities of managers, employees and supervisors; and prioritising key legal, personnel and policy development issues related to HIV/AIDS.

3.2 Disruption of MoA operations and erosion of capacity

HIV/AIDS disrupts MoA operations by severing key linkages in the service delivery chain between MoAs and their clients, for instance, through its impact on the agricultural extension service. Agricultural extension workers give farm households access to improved agricultural practices, new technologies, improved seeds, etc. In addition, they also provide technical advice on credit, marketing and farm management. In many rural areas, agricultural extension workers are the only contact farmers have with support services (see Box 6). When they fall sick or die, rural communities lose access to extension advice and services when they need them most.^[20]

Further, extension workers are often responsible for collecting data for district MoA information systems. Prolonged illness and death among extension workers may thus result in gaps in MoA district data collection systems and in data bases on the basis of which agricultural policies, strategies and programmes are designed.

HIV/AIDS also impacts on MoAs at the organizational level by claiming the lives of highly qualified staff who may be difficult to replace. Many such civil servants have been trained abroad, have a long record of professional experience and may have specialized in areas that are not easy to fill in. Their demise is more than just a loss in staff: it can create a vacuum in the structural organization of an MoA. According to the Ministry of Agriculture, Animal Industry and Fisheries (MAAIF) in Uganda, the loss of senior administrators has often left significant gaps in the structure of the Ministry.^[21] HIV/AIDS has also contributed to the elimination of the post of County Extension Coordinator (CECs are officers-in-charge of counties and district Subject Matter Specialists who provide technical back-up and support supervision to extension workers) because in a number of counties CEC positions were vacant for a prolonged period of time. Given that the skills of supervisors are often derived from many years of experience, the loss in output due to HIV/AIDS is likely to be much greater than that measured by their wage.^[22]

For MoAs, as for other Ministries and rural institutions, erosion of capacity translates into a diminished capability to deliver services, to cope with crises (inclusive of HIV/AIDS), and to function as organizations. In other words, the impact of the epidemic makes it even more difficult for MoAs to address their mandate, let alone the challenges posed by HIV/AIDS.

In addition, addressing the effects of the HIV epidemic requires skills on the part of MoA staff which may not be part of their formal education, training and professional experience. Thus, even though MoA professionals may be confronted with HIV on a daily basis, they may be unable to cope with the technical challenges posed by the epidemic. In Zambia, it has been reported that increased adult morbidity and mortality among senior MoA staff has "definitely affected the planning and administrative capacity of the MoA to implement agricultural programmes", according to the Ministry of Agriculture, Food and Forestry (MAFF).^[23]

Few countries in sub-Saharan Africa have systematically assessed the impact of HIV/AIDS on their Ministries of Agriculture and their work. In Zambia, a UNAIDS/UNDP initiative in the MAFF has, since 1999, been collecting data on increased MAFF absenteeism, mortality among staff members, funeral and associated costs, staff turnover and causes of death and absenteeism.^[24] A similar exercise has been undertaken in Malawi with support from UNAIDS and the World Bank.^[25] Some of the findings of these assessments are presented below.

3.2.1 The determinants of HIV/AIDS impact

This section reviews the main determinants of HIV/AIDS impact on MoA operations and capacity.

a) Reduced staff productivity

i) Loss in human resources

Many MoAs have not systematically analysed the toll of HIV/AIDS-related morbidity and mortality and the extent of disruption of MoA operations due to HIV/AIDS. However, data from the Ministries of Agriculture in Kenya and Malawi as seen above show that the epidemic is exacting a heavy toll.

Responses to HIV/AIDS-related loss of human resources appear to be ad hoc rather than a result of pro-active policies. For example, MoA staff at headquarters level are usually not being replaced. In Uganda, the Public Service Commission (PSC), which is responsible for recruiting staff for all Ministries, does not replace staff with new recruits, in view of the government's restructuring exercise at Ministry headquarters. This policy has, according to the MAAIF, "covered" staff losses experienced to date.^[26] In Tanzania, the Ministry of Agriculture and Cooperatives (MoAaC) similarly indicated that the restructuring exercise within the Ministry has "masked" the problem of staff loss due to HIV/AIDS.^[27]

At the field level, the MAAIF in Uganda has responded to the loss of professional staff by re-deploying university graduates at sub-county level. This strategy is reported to pose great challenges in terms of workload and technical proficiency. In Malawi, in order to cope with the loss of staff, the MoAI has contracted 42 retired Field Assistants, 7 veterinary assistants, and 4 Farm Home Assistants to fill in the vacant posts.^[28] The effectiveness of this strategy has yet to be determined.

ii) Absenteeism due to morbidity and funeral attendance

In the mid-1990s, one FAO study found that up to half of agricultural extension staff time in one district in Uganda was lost due to HIV/AIDS. Staff members were frequently absent from work caring for sick relatives or attending funerals. In addition, some staff members had fallen sick themselves.^[29] Today, the MAAIF reports that increased and prolonged morbidity of focal point officers renders the "implementation of certain key activities impossible."^[30] In Malawi, there have been reports of "fisheries field [extension] staff [being] absent to attend funerals half or three quarters of the working days per month".^[31] In Namibia, the Ministry of Agriculture, Water and Rural Development (MAWRD) argues: "...[the] increasing absence from duty (with leave) by staff members attending funerals of relatives [is making it] difficult to have a meeting with all staff present".^[32] The MAWRD in Namibia reports that, frequently, training or field day exercises organized by the Ministry are being postponed due to funerals. "This trend is increasing, and, we, in extension, will have [to have] flexible programmes to accommodate cancellations at short notice, and then reschedule activities also at short notice".^[33]

iii) Morbidity-related on-the-job fatigue

While there is no hard information on productivity loss due to on-the-job fatigue related to AIDS morbidity, this could well be significant. According to the MAWRD in Namibia, "during the later stages of the disease [AIDS], the ability [of staff members] to work decreases dramatically. This affects work performance ..."^[34]

iv) Staff demoralisation

The distress generated by young adult morbidity and mortality in the workplace should not be underestimated as an important factor in reduced staff productivity, operational efficiency and quality of output.

b) Increase in Ministerial expenditures^[35]

i) Costs related to HIV/AIDS absenteeism

HIV/AIDS absenteeism includes the time spent seeking medical treatment by sick staff members, sick leave (exemption from duties on medical grounds), unofficial leave and caring for sick family members. In Swaziland, a government employee may have up to six months sick leave at full pay and then another six months at half pay before becoming retired on medical grounds.^[36] In Zambia, the MAFF provides for continuous absenteeism from work up to 90 working days with full salary. Thereafter, the employee is put on half salary for another six months before being asked to retire on medical grounds. However, unlike in the private sector, public sector regulations on these provisions are not strict, so a staff member may be given more time to recover, at full salary, at the discretion of the head of department.^[37]

ii) Medical costs

In Uganda, as in other countries, the MAAIF has no provision for health funds or insurance for its personnel. Health care costs are treated by the Ministry on a case-by-case basis. As a result, Ministerial resources are re-allocated to provide basic health care support to HIV-infected persons "through humanitarian considerations and cost-benefit considerations by the management".^[38] In Zambia, successive collective agreements between the MAFF and the unions have resulted in a medical allowance for MAFF staff and their immediate families under special cases (recurring and complicated illnesses, such as tuberculosis, and conditions requiring specialist treatment). According to this agreement, which does not set a ceiling on the amount to be spent, the MAFF is expected to meet the medical costs as prescribed by the physician. In some cases, this could even include the cost of treatment and upkeep abroad of senior MoA staff members.^[39]

iii) Burial costs

Burial costs usually include the purchase of the coffin, the funeral grant, transport costs, subsistence allowances and miscellaneous costs, and represent a significant, unplanned, expense for many MoAs.^[40] In Uganda, the deceased are transported to their place of birth with all the attendant expenses, which are usually assumed by the MoA. Similarly, in Malawi, the MoAI is responsible for covering all funeral expenses of deceased staff members, including transporting the body and personal belongings of the deceased. A funeral grant is also given to the family of the deceased.^[41] Conversely, in Namibia, staff funeral costs are not assumed by the MoA.

Preliminary findings from Zambia indicate that the total recorded deaths of 936 MAFF staff members from 1990 to 1998 would have cost the Ministry about 2.8 billion Kwacha (K), at an average of K300 000 per death, at the 1999 rate of exchange.^[42]

Funeral grants are an important component of funeral costs. In Zambia, the MAFF allocates a fixed funeral grant for the death of a serving staff member and his/her nuclear family. In 1999, the grant for a serving staff member amounted to K250 000 while the grant for the death of a member of the nuclear family amounted to K200 000.^[43] Transport and fuel costs (see Box 7) need to be taken into account for the entire duration of the funeral, and particularly for the period leading to the burial.

iv) Recruitment and replacement costs/productivity loss after training

There is little hard information on the costs incurred by MoAs to recruit and replace staff members lost to disease. In Zambia, the MAFF reports: "it is important to note that replacement costs of [certain] officers are very high as it is expensive to offer the specialized training involved and usually this training takes a long time".^[44] It has been argued that, in Zambia, the MAFF is among the Ministries with the highest trained professionals.^[45] For instance, a Master's degree is required for heads of department while a diploma is the minimum requirement for other professional staff. MAFF professionals spend more time in university (five years for a Bachelor of Sciences in agriculture and six years for a Bachelor's in Veterinary Medicine) than professionals in other fields. The duration of these programmes aside, the science programmes that MAFF professionals pursue are among the most costly: not less than US$2500 per year, minus salary, to train an extension officer and US$3000 per year to train a veterinary doctor. Postgraduate studies undertaken abroad may cost the MAFF between US$15 000 and $20 000 a year and between US$40 000 and $50 000 for PhDs.^[46]

In addition, the cost of holding interviews (hotel costs, sitting allowances for the interview panel, etc.) can be substantial. In Zambia, interview panelists alone earn as much as K90 000 per day per person. If the panel consists of between five and ten individuals who may interview candidates over a period of up to ten days (depending on the number of candidates), then recruitment-related costs can be prohibitively expensive. The recruitment of new staff entails costs toward orientation and retraining, and settling allowances which are equal to one month's salary.^[47]

v) Terminal benefits

The increase in young adult morbidity and mortality is likely to exacerbate financial pressures on MoAs through the payment of a growing number of terminal benefits to the families of deceased staff.

vi) Costs incurred to protect the rights of staff members living with HIV/AIDS at the workplace

The MAAIF in Uganda assumes the cost of transferring staff affected by the epidemic to areas of their choice or of convenience (usually closer to their homes).^[48]

Given the significant medical and other expenditures incurred by MoAs as a result of increased young adult morbidity and mortality, certain human resource policies and employee benefits and procedures may be inadequate and in need of review (current sick leave provisions, procedures for processing terminal benefits and emergency advances for the terminally ill, etc.). The MAFF in Zambia has argued that even though it is fully aware of the disruption that HIV/AIDS causes the Ministry, "these concerns have not yet been put in a format to provide policy guidelines for decision-making with respect to HIV/AIDS".^[49] This points to the need to quantify the various AIDS-related costs incurred by MoAs and make them accessible to policy- and decision-makers.

c) Increase in staff turnover

The extent of staff turnover due to increased young adult morbidity and mortality is difficult to assess given the re-structuring exercises undertaken in many Ministries. Nevertheless, anecdotal evidence suggests that staff turnover is high among a number of MoAs in Eastern and Southern Africa.

d) Increase in the workload of MoA staff

Given that many staff members who die are not subsequently replaced, the immediate impact of young adult mortality is an increase in the technical workload of staff members at headquarters and fields level alike. For instance, the workload of agricultural extension workers in a number of countries has increased to a point where many are unable to work effectively. While HIV/AIDS is not the only contributing factor, it has certainly exacerbated this trend. Equally important is the increase in the administrative and management workload of MoA staff.

e) Loss of knowledge, skills and experience

While not easily quantifiable, many of the persons who have succumbed to AIDS have been highly educated, experienced individuals whose knowledge, skills and experience cannot easily be replaced. According to the MAFF of Zambia, "the loss of knowledge, skills and experience has been significant before replacements are found".^[50] In other words, HIV/AIDS is eroding MoA capacity not only through losses in human resources but also through the loss of vital technical, administrative and managerial skills. As seen above, coping mechanisms such as the deployment of junior (and thus often less experienced) professionals or retired professionals (a sound short-term strategy which is, however, not sustainable in the long term) do not fully replace the skills and institutional knowledge base that has been lost.

The loss of staff and the corresponding loss of knowledge, expertise and experience depletes the pool of highly specialized MoA personnel and affects the quality and continuity of MoA services.

3.2.2 Examples of responses

a) Human capacity development^[51]

While this section deals with capacity development directly related to HIV/AIDS, it cannot be emphasized strongly enough that overall capacity development in MoAs is essential if HIV/AIDS concerns are to be addressed. If MoA capacity is severely limited in the first place, it is unlikely that HIV/AIDS-related capacity development initiatives will be effective.

i) Sensitisation

A number of MoAs in sub-Saharan Africa have undertaken sensitisation of senior administrators, district agricultural officers and district subject-matter specialists on HIV prevention, on caring for persons living with HIV/AIDS, and on the effects of the epidemic on the agricultural sector. However, it appears that sensitisation tends to take place as a one-off event rather than as an on-going process. As such, it ends up being a goal in itself rather than a means to an end. Once sensitisation is over, there are usually no follow-up activities to build upon the skills and information imparted, such as concrete initiatives on how to integrate HIV/AIDS into divisional/departmental or district level workplans and into MoA budgets. Lack of funding for follow-up activities is a major constraint.

ii) Training

Training on HIV/AIDS has been undertaken in a number of MoAs in Southern and Eastern Africa. However, as with sensitisation, training has been largely health-oriented (HIV prevention, etc.) rather than agriculture-specific, and fairly limited in scope. Moreover, training has primarily focused on assisting households directly affected by HIV/AIDS, and particularly on those living with HIV/AIDS, rather than on the survivors and on households indirectly affected by the epidemic. In Uganda, three training booklets have been prepared by the MAAIF: HIV/AIDS and Nutrition, Feeding Guidelines for People Living with HIV/AIDS; and a Training Guide on HIV/AIDS for agricultural extension workers. Due to a lack of resources, however, subsequent training of MAAIF staff has been limited. Only one training-of-trainers workshop was held for 17 Agricultural Officers, Assistant Agricultural Officers, Veterinary Officers and Fisheries Officers, and three district level training workshops for a total of 83 field extension workers under a UNDP-funded project in 1997 (see below).^[52]

The MoAaC in Tanzania has identified training to strengthen the analytical capability of decision-makers to plan for the socio-economic impacts of the epidemic on rural households and communities as a key priority.^[53] However, even where ministries have identified priority training areas, these often do not materialise due to lack of funding and follow-up.

b) Mainstreaming HIV/AIDS in the work of MoAs

Multi-sectoral responses to the HIV epidemic in the early 1990s involved primarily the appointment of AIDS Focal Points to mainstream HIV-related activities within MoAs. In the case of Zambia, the MAFF established a focal point on HIV/AIDS to liaise with key officers of various departments, including the Agricultural Information Service, Human Resource and Administration, and Planning and Cooperatives. This team constitutes the Ministry's HIV/AIDS Committee. At provincial level, formally trained provincial focal point persons on HIV/AIDS coordinate activities in three provinces (Copperbelt, Lusaka and North-Western). They, in turn, liaise with District Focal Point persons who are in charge of executing MAFF initiatives on HIV/AIDS at district level.

HIV/AIDS mainstreaming exercises have had a mixed record. In the case of Zambia, it has been argued that HIV/AIDS focal points in MoAs have had a limited impact in mitigating capacity loss.^[54] Furthermore, their technical know-how on HIV/AIDS was reported to be in need of improvement and their activities in need of evaluation and adjustment.^[55]

Key constraints encountered by the MAFF AIDS Control Programme include the following:

• the absence of a mandate on HIV/AIDS;

• the absence of political backing for the HIV/AIDS Focal Points and for the HIV/AIDS Committee;

• the fact that the HIV/AIDS Committee is composed of staff from various departments within the MAFF who have other responsibilities and therefore limited time to devote to HIV/AIDS; and

• lack of resources, and particularly a budget for HIV/AIDS initiatives.^[56]

UNDP has been supporting HIV/AIDS mainstreaming exercises in line Ministries (including MoAs) in a number of countries (Botswana, Zambia and Uganda among others). In Uganda, the MAAIF embarked on an HIV/AIDS mainstreaming exercise in the mid-1990s. The Family Life Education (FLE) unit of the MAAIF assumed the role of integrating the Ministry in an expanded national response to the epidemic. One of its initiatives was to integrate HIV prevention and care for affected families into the core activities of the Ministry. According to the MAAIF, mainstreaming has been largely successful due to the following reasons: a) the manager of the FLE unit has been actively involved in the national AIDS Control Programme since 1994; b) a number of senior staff members of the MAAIF have also been involved in the development of the Ministry's strategic plan for the expanded national response to HIV/AIDS; and c) each MAAIF department has a focal point officer responsible for the HIV/AIDS programme.

According to the MAAIF, key constraints to mainstreaming HIV in the ministry include the following: the absence of commitment and support from some heads of departments; the absence of funds; and the brevity of the mainstreaming project (limited to about a year). At present, according to the MAAIF, there is "great need for financial assistance to undertake training, workshops and research".^[57] In fact, one of the challenges facing the AIDS Control Programme of the MAAIF when the UNDP project ended was how it would be able to use the training materials it produced once the funding ceased.^[58] In effect, as soon as the project ended, the MAAIF had problems training its Field Extension Workers.^[59] According to the appraisal report of this mainstreaming project, the perception of the programme being 100% funded by donors inadvertently eroded the ability of the Ministry to sustain the activities in place.^[60]

While focal points for AIDS may be useful instruments for mainstreaming, there has been a tendency to situate these within "soft" units, such as the Family Life Education unit in the case of Uganda's MAAIF, rather than within "hard" units (livestock, crop production, fisheries, agricultural extension, etc.). This identification of the HIV/AIDS focal point with "soft" units can render the mainstreaming of HIV/AIDS within the core areas of MoA work more difficult. In fact, in the case of Uganda, the HIV/AIDS mainstreaming exercise was perceived to be an added-on "project" rather than an attempt to integrate HIV in MoA programmes.

c) Adjusting MoA budgets

Most countries in Eastern and Southern Africa have so far not included HIV/AIDS in their budgets due primarily to financial constraints. Even without HIV/AIDS, many MoAs face acute financial problems. The MAFF in Zambia reports that: "...it is important to point out from the outset that the current existing resources, especially finances, in the Ministry are so erratic and inadequate that the implementation of most of the agricultural programme activities has virtually stalled. It is difficult, therefore, to perceive how under the current funding arrangements within the Ministry, HIV/AIDS could be effectively addressed".

Financial pressures are particularly acute at the district level, "where meager resources are thinly distributed over a wide range of competing needs", according to the study on the impact of HIV/AIDS on Zambia's MAFF.^[61] The study concludes: "An increase in the incidence of illness and/or death means a corresponding increase in the demand for financial assistance to cover funeral and other related costs. Given the culturally sanctioned reverence for funeral and burial rites among African communities, preference would be given to funeral costs over other financial demands".^[62] In other words, HIV/AIDS depletes MoA funds earmarked for agricultural service provision and may deplete funds allocated for agricultural investment.

This raises the following dilemma: given the scarcity of resources in MoAs, funding is needed to initiate and follow-up on pilot HIV/AIDS activities. Yet, HIV/AIDS mainstreaming exercises that have been fully externally funded have ended up being perceived as added-on "projects" with little ownership at the end of the pilot phase. Experience with mainstreaming HIV/AIDS in development programmes also shows that once such projects come to an end, activities virtually cease.

For these reasons, it is important to adjust MoA budgets to reflect the direct and indirect costs of HIV/AIDS and the need for response measures to the impact of HIV/AIDS. Uganda's MAAIF is perhaps the first MoA to introduce HIV/AIDS into its 2001 budget. This indicates a major shift in approach as HIV/AIDS becomes a factor to be reckoned with at the budgetary level. For, unless HIV/AIDS features in MoA budgets, it is unlikely that measures to address the HIV epidemic will be introduced in MoA divisional and district level workplans and thus in core agricultural policies, strategies and programmes.

3.3 Increased vulnerability of MoA clients to food and livelihood insecurity

The socio-economic impact of HIV/AIDS on rural households and smallholder agriculture, the vulnerability of smallholder farmers to its effects, and the coping mechanisms of households and communities have been explored in some depth over the last decade.^[63] For this reason, they are not dealt with in detail here. Suffice it to say that a number of studies have shown that the impact of HIV/AIDS is most severe on smallholder agriculture-the primary economic sector and engine of growth of many sub-Saharan African countries-through its effects at the household level. Smallholder agriculture in sub-Saharan Africa relies almost exclusively on family labour-often the only productive resource poor people have.

3.3.1 The impact of HIV/AIDS on food and livelihood security

In terms of household food security, HIV/AIDS impacts on the productive capacity of farm households, thus influencing availability, access and utilization of food in the following ways:

a) Adverse effects on land/labour productivity and on agricultural production

Household labour quality and quantity are reduced, first in terms of productivity, when HIV-infected persons fall sick, and later when the supply of household labour declines because of patient care (this burden falls mostly on the women who are also the main food producers in sub-Saharan Africa) and death (see Box 8). For example, in one village in Tanzania, in households with an AIDS patient, nearly 30% of household labour was spent on AIDS-related matters (including care of the patient and funeral duties). If two people were devoted to nursing the patient, as was the case in 66% of recorded cases, the total labour loss was 43% on average.^[64]

The impact of HIV/AIDS morbidity and mortality not only affects labour inputs to farm production, but, more significantly, it disrupts the household production-domestic labour interface^[65] by diverting women's labour from regular caring activities to caring for persons living with HIV/AIDS. This can have severe repercussions not only on food production but also on food and livelihood security, health, education (children are often taken out of school), nutrition and family welfare more generally (see Box 9). The illness and death of a young adult woman can have a particularly dramatic effect on the household, given that women provide up to 80-90% of labour and managerial services for smallholdings in rural areas and are the primary care givers.^[66]

In the case of crop production,^[67] FAO research in Eastern and Western Africa in the early 1990s shows that the impact of HIV/AIDS resulted in one or more of the following consequences:

• a reduction in the area of land under cultivation;

• changes in cropping patterns/shift from cash crops to subsistence production;

• a decline in the range of crops cultivated;

• a decline in the ability to control pests;

• loss of soil fertility;

• a decline in crop yields; and

• loss of crop and farm management skills.

Recent evidence from MoAs in Eastern Africa shows that these early trends continue to prevail. According to the MAAIF of Uganda, in severely affected districts like Rakai and Masaka, up to 25% of households are cultivating less land as a result of the HIV epidemic.^[68] A decline in cash crop production, and particularly coffee, which is labour intensive, is also being observed.^[69]

More importantly perhaps, Uganda's MAAIF recently reported that AIDS is contributing to food scarcity in areas hitherto known for food availability and surplus.^[70] In Mansa District, Luapula Province, Zambia, one study revealed that malnutrition was perceived to be a major risk in 60% of families affected by AIDS.^[71] Another study in Zambia found that the combined effect of drought and AIDS made it difficult for farms to recover from the 1992 drought. This was largely due to the fact that key agricultural tasks, such as planting and weeding, usually undertaken by women, were neglected when women had to care for persons with HIV.^[72]

b) Decline in on- and off-farm disposable household income

HIV/AIDS greatly increases household expenditures and adversely affects on- and off-farm income, and especially the availability of disposable cash which largely determines the amount and quality of food that can be purchased. Household income declines due to:

• increased expenditures for special foods, medication and treatment, transport to and from health care facilities and funerals (see Box 10). According to a simulation study in Kenya, AIDS costs represented 78% of farm household income during the first year of AIDS impact (with one adult death) and 167% the second year;^[73]

• the loss of the income of the HIV patient (either in the form of labour or remittances);

• HIV/AIDS stigmatisation that may prevent persons with HIV from continuing to exercise their trade; etc.

In the rural Rakai district of Uganda, for instance, households can spend up to a third of their annual cash income on monthly medical care or on a single funeral. Family assets, such as livestock, land and property, may also be sold.^[74] It is worth noting that the drastic reduction in income and productivity often occurs at a time when expenses related to treating the infirm increase exponentially.

c) Erosion of farm household resource and asset base

The MAAIF in Uganda reports that many affected households sell their food crops in order to cover hospital expenses. Some households even sell off their land to raise money for medical care. In fact, it has been argued that it is becoming increasingly common for some hospitals and clinics to encourage terminally ill patients to surrender land titles as security for medical bills.^[75] A World Bank study found that asset ownership declined when an HIV positive household member died, but remained stable when the death was not related to HIV/AIDS.^[76] This erosion of the household resource base deprives families of the essential means to sustain themselves. Surviving widows and their children often have great difficulties in retaining family land and other assets which tend to revert to the late husband's family.

d) Erosion of knowledge base and skills for agricultural production

Agricultural work is becoming increasingly less attractive in rural areas even among youths who have been brought up in farm households. Conversely, urban centres have been identified with physically less demanding work as well as with higher wages. Given that rural youths spend most of their time in school, their participation in farm work is limited to the peak season. A combination of these factors is contributing to a situation whereby youths are less inclined to make a living off the land and are losing essential skills needed for agricultural production. This is further compounded by HIV/AIDS which has left behind more than 12 million orphans to date in sub-Saharan Africa.

The death of one or both parents to HIV/AIDS often means that younger members of the family may not have the necessary knowledge, experience and management skills to run the farm household. Similarly, if one parent dies, it may be that the surviving parent does not have the skills in farming and/or marketing certain crops.

A study of orphan-headed households in two districts of Kenya that have been heavily impacted by HIV/AIDS recently found that when asked if they knew where to get information on food production, a staggering 82% of the orphans replied in the negative.^[77] About 80% of orphan-headed households expressed the need for agricultural assistance and improved practices. Only about 7% of orphan-headed households had adequate knowledge of agricultural production.^[78]

3.3.2 Examples of responses

A number of response measures may help to mitigate the impact of HIV/AIDS on farm households, such as:

a) The introduction of labour- and capital-saving agricultural and household technologies and practices

These may include:

• early maturing, disease-resistant crop varieties that are easily threshed and pounded and thus require less labour; one example is a new strain of cassava, recently introduced in Uganda by IFAD, which is resistant to mould disease;^[79]

• inter-cropping to reduce weeding time;

• integrated pest management to reduce the need for costly chemical inputs;

• the promotion of draught power and of improved technologies for animal husbandry;

• zero or minimum tillage to reduce the need for high-cost ploughs and oxen;

• the introduction of farm equipment that can be used by women, the elderly and youths (i.e. lighter ploughs and planters with modified hoes);

• appropriate technologies for food preparation (i.e. efficient stoves) and for water and fuel collection (water points, etc.); and

• labour-sharing arrangements.^[80]

One example of a project promoting labour- and capital-saving technologies is the Zambezi Valley Organic Cotton Project, supported by the Zimbabwean non-governmental organization African Farmers' Organic Research and Training (AFFOREST). Many AIDS widows have joined the project because organic cotton has few, if any, external input costs and a lower labour requirement than conventional farming.^[81] An analysis of input costs during the 1997-98 season in the Zambezi Valley showed that organic cotton farmers could save more than Z$200 per acre, compared with conventional cotton growers. In addition, while conventional cotton farmers spent more than 15 hours per week on operations connected with pesticide use, including purchase from the supplier, organic farmers spent 1-2 hours per week scouting for pests and predators.^[82]

Small differences in gender roles and in resources among households and communities can influence how effectively households respond to the epidemic. For example, the burden of caring for HIV/AIDS patients usually falls on women and children, who otherwise would be engaged in farming or other productive work or be attending school. An FAO study found that whether women are allowed to ride bicycles and whether bicycles are available can be an important determinant of the marketing capacity of an affected household or community. Gender roles also influence the continuation or adoption of labour-saving responses, such as the use of oxen or access to land and/or credit.^[83]

b) The enhancement of household income-generating capacity

Enhancing income-generating capacity is critical in mitigating the impact of HIV/AIDS on rural households. It can make important contributions to household survival by helping to maintain expenditure patterns. Income-generating capacity can be enhanced in various ways, such as through micro-credit. In Uganda, one such project gave micro-grants of about US$100 to 30 families or to a group of people living with HIV/AIDS to finance low-input income-generating activities, including gardening and fishmongering.^[84] Another way to enhance income-generating capacity is to build the asset buffer of households by expanding their opportunities to own livestock and by protecting existing herds through veterinary care.^[85]

c) The promotion of women's and children's rights to land and other property

As seen above, land ownership becomes critically important, particularly when male heads of household die after long and costly illnesses, often leaving their wives and children without resources. Securing land ownership rights for women and children, in collaboration with other Ministries and institutions, can help to ensure that vulnerable households are able to support themselves. In view of the complexity of land tenure systems in many parts of Africa, this issue needs to be reviewed in depth if viable solutions are to be found.

d) Apprenticeship schemes and agricultural skills training for adolescents

In view of the loss of agricultural knowledge, skills and practices among orphan children and adolescents described above, it is essential that apprenticeship schemes and training in farming skills are organized to enable the younger generation of survivors cope with AIDS impact in the long-term. Through such programmes, orphans can enhance their livelihood options and acquire skills that will enable them to support themselves as well as their siblings. The NGO Uganda Women's Effort to Save Orphans has developed apprenticeship programmes for out-of-school adolescents with considerable success. Orphans receive on-the-job training and earn as they learn. After the training, they are introduced to business enterprise management and subsequently loaned money to start their own businesses. Such efforts need to be expanded and replicated.

3.4 Are MoA policies, strategies and programmes still relevant?

3.4.1 The implications of HIV/AIDS for MoA policies, strategies and programmes

"Any development programme that does not deliberately address HIV/AIDS is bound to fail as the benefits that may be perceived in the programme could potentially be overwhelmed by the negative impact of HIV/AIDS", reports the MAFF of Zambia.^[86] The MAFF indicates that in the process of implementing the Agricultural Sector Investment Programme (ASIP), it has been recognized that HIV/AIDS is a cross-cutting problem that needs to be integrated in the overall programme. However, there is no mention of the impact of HIV/AIDS on sector policies in MoA policy documents.

This section shows how the relevance of some MoA policies and strategies can be called into question by the conditions being created by the HIV epidemic.

a) The limitations of a production-oriented approach

The impact of HIV/AIDS on rural livelihoods can be such that it may neutralize household efforts to boost agricultural production-a key objective of MoAs. According to Uganda's MAAIF, the focus of its agricultural policies is on generating technologies and providing services to ensure improved quality and quantity of agricultural produce and products for domestic consumption, food security and export. The status and living conditions of the producers per se, however, have not been addressed. The MAAIF recognizes that it needs to formulate policies and develop programmes that address the human factor of production, i.e. the quality of life of the producers, inclusive of the impact of HIV/AIDS thereupon.^[87] This will necessitate a paradigm shift from a production- to a client-based approach in which MoA policies, programmes and strategies reflect the economic and social changes in the rural environment (including market liberalization, privatization, HIV/AIDS and other shocks, etc.) as well as the evolving needs, constraints and living conditions of their clients, including their health and nutritional status. While the need for a client-based approach goes beyond AIDS impact, the epidemic does lend more urgency for such a shift in approach.

b) Farm household labour: an abundant resource?

A number of farming systems being promoted across sub-Saharan Africa are based on the premise that there is an abundant, near inexhaustible supply of labour (and particularly female labour). Given the scale of HIV/AIDS, however, this can no longer be assumed to be the case. In addition, the value of so-called "unskilled" labour in agriculture has been greatly underestimated. What is commonly thought of as "unskilled labour" has accumulated location- and task-specific skills that can be hard to replace.^[88] As the rural exodus is a major concern of many MoAs, the additional loss of labour due to HIV warrants some attention: urban labour shortages can be replaced with migrant labour, but labour shortages in many rural areas may be more difficult to compensate for.

c) Changing farm household typologies

Current farm household typologies upon which agricultural policies and programmes are designed may no longer be valid. The parameters of vulnerability of rural households, farming systems and rural livelihoods are changing as a result of HIV/AIDS-induced young adult morbidity and mortality and the subsequent shifts in household demographic structure. Moreover, given the changes in composition of MoA clienteles (with increasing numbers of elderly, youth and women-headed households), existing extension strategies may not correspond to field realities.

3.4.2 Multi-sectoral responses to HIV/AIDS

Multi-sectoral responses initiated in the 1990s were intended to address the cross-sectoral impact of the HIV epidemic and, in the case of agriculture and rural development, to help rural institutions and other organizations cope with the impact of the epidemic on their work. While a number of countries in Eastern and Southern Africa adopted multi-sectoral responses that included MoAs, response measures to HIV/AIDS within MoAs have been largely health-dominated. This is partly due to the fact that in practice, HIV/AIDS is still primarily situated within a health-dominated paradigm and is perceived to be far removed from the core work of Ministries of Agriculture.

For example, in Uganda, the AIDS Control Programme of the MAAIF aims to:

• strengthen HIV/AIDS education in order to halt its spread among the staff and the farmers and fishermen they serve;

• help infected and affected persons cope with HIV/AIDS; and

• mitigate the adverse socio-economic effects of HIV/AIDS through the promotion of profitable agro-enterprises which are labour-, energy- and time-saving.^[89]

It appears, however, that more emphasis has been placed to date on the first two objectives, and particularly on information, education and communication (IEC) initiatives, and less attention has been paid on mitigating the adverse effects of HIV/AIDS through core agricultural initiatives.

In Tanzania, the MoAaC, with support from UNAIDS, has recently finished implementing a one-year pilot project in four regions (Iringa, Mbeya, Rukwa and Ruvuma) to help address the impact of HIV/AIDS on the Ministry and its work.^[90] This project has similarly focused primarily on health-oriented IEC activities.

Multi-sectoral responses in MoAs have had limited success in going beyond a health-dominated focus to the HIV epidemic due to a number of reasons, including the following:

a) Many MoAs have not assessed for themselves the effects of the HIV epidemic on their work and/or on the agricultural sector. For this reason, they have been unable to pinpoint how their work has been affected and how agricultural and rural development policies need to be adjusted. MoAs may be aware of existing HIV/AIDS impact assessments on agriculture/rural development. However, as the MoAaC in Tanzania indicated, "The Ministry was not involved [in such studies]". The question of ownership of impact assessments and of multi-sectoral responses to the HIV epidemic is critical but all too often neglected. This is because impact assessments undertaken by bilateral agencies, UN agencies and NGOs have usually not actively involved MoAs. In turn, MoAs do not make use of these studies. As a result, according to the MoAaC in Tanzania: "the Ministry does not have sufficient evidence (facts) to justify the review of current [agricultural] policy".^[91]

b) The absence of MoAs from multi-sectoral HIV/AIDS initiatives. For example, in March 2000, Zambia established a National Council and Secretariat on HIV/AIDS/Sexually Transmitted Diseases and Tuberculosis - a body designed to advocate effective multi-sectoral approaches for the prevention of HIV transmission, care and social support, as well as for impact mitigation. The Cabinet Committee of this body includes the Ministers of Mines and Minerals Development, Health, Education, Communications and Transport, Presidential Affairs and Information and Broadcasting Services.^[92] The MAFF appears to be absent from this multi-sectoral effort.

Adopting a multi-sectoral response to HIV/AIDS does not merely entail the introduction of HIV/AIDS focal points and HIV prevention/IEC activities in Ministries of Agriculture and other Ministries. Nor does it only mean adding HIV/AIDS - specific initiatives, or, more generally, public health initiatives on to existing programmes. It also entails incorporating the developmental implications of HIV/AIDS into core agricultural policies, strategies and programmes. For this reason, a shift is required toward a developmental paradigm of response to the epidemic that complements health-based initiatives with core agricultural initiatives.

Multi-sectoral responses to HIV/AIDS have not yielded concrete and tangible results to date. This is partly because they involve lengthy processes that require considerable financial and human resources, political commitment at the highest level, extensive networking and collaboration between Ministries. However, this does not necessarily mean that the approach as such is conceptually flawed or dated. Rather, it may mean that more work is required in this area along with a recognition that changes are needed in the way development policy and practice are conducted.

It has been argued that the multi-sectoral response to the epidemic is currently being superseded by a focused treatment and prevention response premised on "proven approaches".^[93] According to this view, "this may mean less emphasis on the multi-sectoral approach and greater emphasis on the most promising prevention interventions. The latest phase also includes a sharper focus on the ethical and resource issues associated with new treatment and prevention options, such as anti-retroviral therapy and prevention of maternal-to-child transmission".^[94] Such a focused treatment and prevention approach, however, would have little, if any, role to play within the core work of MoAs.

Figure 2 below depicts the various approaches of response to HIV/AIDS, inclusive of the developmental approach discussed above. In a developmental paradigm of response to HIV/AIDS, the focused treatment and prevention response forms part of the multi-sectoral and public health responses. The issue of the relevance of agricultural policies, strategies and programmes to the conditions being created by the epidemic would be placed into sharp focus in a developmental approach to multi-sectoral responses to HIV/AIDS.

Figure 2: Conceptual framework of responses to HIV/AIDS