The full report can be accessed (see 'Download complete version' above);
this summary has been extracted by SARPN from the report.
This paper is intended to respond to the need to better understand the implications of the
AIDS pandemic for government ministries responsible for agricultural and rural development
in eastern and southern Africa. There is now widespread recognition that HIV/AIDS is not
simply a health issue. A coordinated approach will be necessary to effectively combat the
epidemic and its consequences, one that cuts across many sectors of the economy and many
government ministries. While many in the agricultural sector embrace the idea of playing a
role to combat HIV/AIDS, there has been very little analysis by agricultural policy analysts to
guide them. Moreover, despite the fact that the pandemic is now in its third decade in Africa,
available analysis to date provides a very murky picture as to how HIV/AIDS is affecting the
agricultural sector ' its structure, cropping systems, relative costs of inputs and factors of
production, technological and institutional changes, and levels of production and marketed
surplus. Until these issues are clarified, policy makers will be inadequately prepared to
forecast anticipated changes to the agricultural sector and respond proactively. For reasons
presented later in the paper, anticipation of future impacts and proactive policy responses are
likely to make a critical difference in averting future crises and chronic poverty among the
countries hardest-hit by HIV/AIDS.
Scope of the Paper
The paper was prepared as a background document for an International Workshop on
Agricultural Policy and HIV/AIDS: Addressing the Linkages, organised by FAO's Gender
and Population and Agricultural and Development Economics Divisions in Maputo,
Mozambique, in November 2003. In Section 2, it reviews the empirical evidence concerning
the effects of AIDS on the agricultural and rural sectors, with a focus on those factors that
influence agricultural production, such as land, labour, capital, and technology and discusses
potential policy implications. Its geographic focus is placed on the hard-hit countries of
eastern and southern Africa, where HIV prevalence rates among the general population
exceed 10 percent.1 It concludes that the AIDS epidemic will have such fundamental effects
on agriculture and rural livelihoods in the next several decades among the seven to ten
hardest-hit countries of eastern and southern Africa that some rural development strategies
previously considered uneconomic or inappropriate for other reasons may deserve serious
Section 3 presents some elements of a framework for governments and donors to assess and
potentially modify existing agricultural programmes, policies, and investment strategies for
achieving their agricultural and rural development objectives after factoring in the likely
impacts of HIV/AIDS on the rural socio-economy. In addition, some broad steps for
mainstreaming HIV/AIDS concerns into agricultural policy and programming are outlined.
The chapter closes with illustrative practical examples of how HIV/AIDS could be integrated
into specific agricultural policy considerations.
It is important to note that the paper does not attempt to cover all aspects relevant to
mainstreaming HIV/AIDS concerns into agricultural and food security policy. Particularly
measures for ensuring food access of both urban and rural population groups, such as safety
nets and social protection, including linkages between food aid and agricultural policy are not
discussed in detail as these are dealt with elsewhere in the literature2. Similarly, while some
impacts of HIV/AIDS on institutions at meso-level and macro-level are highlighted, the focus
of the paper is placed on those policy implications which can be drawn directly from the
analysis of AIDS' impact on factors of production.
Multiple Objectives of Agricultural Policy
Agricultural policy has always been designed to meet a number of objectives. In eastern and
southern Africa, for example, agricultural policy for the past several decades has been aimed
at raising incomes for smallholder farmers, producing enough basic foodstuffs to meet
national consumption requirements, placating politically important interest groups, raising
foreign currency through cash crop exports, and reducing the cost of delivering basic food
items to consumers' tables in order to enhance household food security, all within the budget
constraints imposed by scarce treasury resources (e.g., Government of Uganda, 1998;
Government of Kenya, 2002; Government of Mozambique, 2001; Government of Zambia,
2001; Bates, 1981; Jayne and Jones, 1997).
The onset of the AIDS pandemic has added yet another dimension to agricultural policy. It is
widely accepted that HIV/AIDS will affect many aspects of the rural economy in the hardesthit
countries of Africa, although the particular pathways, impacts, and magnitudes remain
unclear and continue to be debated. The challenge for analysts, agricultural policy makers,
and donors is to understand with greater precision how the rural socio-economy is being
affected by the disease, and consequently how agricultural and rural development policy
should be modified to better achieve national agricultural sector objectives.
Effects of AIDS on Future Demographic Changes
In most of eastern and southern Africa, where HIV prevalence rates generally exceed 10
percent, there will be many fewer adults in the coming decades compared to a "no-AIDS"
scenario (US Census Bureau, 2002).3 By the year 2010, five countries in the region will be
experiencing negative population growth rates: Botswana (-2.1 percent per year),
Mozambique (-0.2 percent), Lesotho (-0.2 percent), Swaziland (-0.4 percent), and South
Africa (-1.4 percent) (US Census Bureau, as reported in Way, 2003, p. 5). By 2020, AIDS
mortality will produce population pyramids in these countries never been seen before (Figure
By 2025, summing across the seven countries where HIV prevalence exceeds 20 percent,
there will be roughly 20 million men in the working age years between 20 and 59 years as
opposed to 31.5 million if AIDS had not existed. By contrast, there will be only 18 million
women in the 20 to 59 year age range as opposed to 32 million in the "no-AIDS" case. And
because of the early death of so many adults of reproductive age, there will also be many
fewer children born, also indicated in Figure 1. Population pyramids in 5-6 other countries
will have similar shapes, though less extreme than those shown in Figure 1.4
Figure 1a. Population in the Medium Variant ("with AIDS") and in the No-AIDS
Scenario ("without AIDS"), by sex and Age Group, 7 Most Highly Affected Countries,
Figure 1b. Projected Population in the Medium Variant ("with AIDS") and in the No-AIDS
Scenario ("without AIDS"), by Sex and Age Group, 7 Most Highly Affected Countries, 2025.
However, it is also important to compare future projected population to current population.
Notwithstanding the catastrophic death toll that is projected to occur over time in these
countries, the absolute numbers of adults projected to be alive in 2025 is roughly similar to
what it is today. This is because the momentum of population growth would have produced
much greater population sizes in the coming decades. While AIDS is projected to erode
population growth to roughly zero in the seven hardest-hit countries, the net result is a
roughly stable number of working age adults over time. Table 1 presents population figures
in 2000 for selected age/sex categories and compares this to projected population estimates in
* Botswana, Lesotho, Namibia, South Africa, Swaziland, Zambia, and Zimbabwe
Source: US Census Bureau, 2003.
According to these demographic projections, there will be a slight increase in the number of
men between 20 and 59 years of age between 2000 and 2025, and virtually no change in the
number of women. The projections indicate a decline in the number of males and females
below 20 years of age by 2025. Because AIDS will particularly influence the number of
people under 20, both through the impact of increased child mortality and fewer adults living
long enough to have children, dependency ratios may actually become slightly more
favourable over time.
These projections are consistent with those of the United Nations (2003). According to its
projections, countries with HIV prevalence above 20 percent will register annual population
growth rates of roughly +0.2 percent between 2000 and 2025. Countries with HIV prevalence
between 10 and 20 percent (Cameroon, Central African Republic, Kenya, Malawi, and
Mozambique) will have population growth rates of +1.33 percent per year.
However, not reflected in these figures are the losses of labour availability due to poor health
prior to death, care-giving for those afflicted with the disease, and mourning periods after a
death. Moreover, it is quite different to achieve stability in population sizes due to avoiding
unwanted births by curbing fertility, than through the loss of adults. Curbing population
growth through the death of adults creates myriad social disruptions, for example through the
long-term difficulties faced by children of deceased parents (e.g., Gertler et al., 2003; Yamano
and Jayne, 2003). People who survive into adulthood in poor food scarce societies, have
already received substantial social investments (education, skills, food production). Their
death translates into a loss to society of existing knowledge and skills as well as the transfer
of knowledge to succeeding generations.
Overall, the picture that emerges is that the domestic labour force is unlikely to grow in the
hardest-hit countries over the next several decades, but it will not shrink either. The quality of
the labour force is likely to be adversely affected by AIDS. In the event that labour shortages
do arise, it may be envisioned that migration outflows may increase in areas where HIV
prevalence is relatively low and where population pressures are already intense (parts of east
Africa such as Burundi and Rwanda).
Future mortality rates in these countries are driven by current HIV prevalence rates. Thus
there may be limits to what can be done to alter mortality rates from AIDS in the near future.
But over the longer run, adult mortality and the shape of population pyramids can be
influenced by policy and other initiatives, such as new treatments and success in behaviour
change. This is likely to offer the most effective approach to addressing many of the looming
problems facing the agricultural sector in the long run (15-20 years and greater).
Seven countries in the eastern and southern Africa region have an estimated (15-49 year) HIV prevalence
exceeding 20 percent: Botswana, Lesotho, Namibia, South Africa, Swaziland, Zambia, and Zimbabwe (US
Census Bureau, 2002). Five other countries, Cameroon, Central African Republic, Kenya, Malawi, and
Mozambique, have HIV prevalence rates exceeding 10 percent. For shorthand, we hereafter refer to these
countries as the "hardest hit" countries.
see for example: Holzman, Robert and Steen JС€rgensen (2000): Social Risk Management: A new conceptual
framework for social protection, and beyond. Social Protection Paper No. 6, The World Bank;
Botswana Institute for Development Policy Analysis (2000): Impact of HIV/AIDS on Poverty and Income
Inequality in Botswana; Kaddiyala, Suneetha and Stuart Gillespie (2003): Rethinking Food Aid to Fight AIDS;
WFP (2002): Food Security, Food Aid and HIV/AIDS: A Five Country Case Study. Internal Draft Report
prepared by Tango International.
While it is not our goal here to explain why HIV prevalence rates are so high in these countries compared to
elsewhere in Africa, we note that these countries share unique structural features of their economies that
exacerbated the spread of the disease. Migration, mostly by men, from their rural farms to urban areas, mines,
and commercial farms for employment has been a fundamental economic and social feature of most of these
countries. The separation of husband and wife, coupled with the concentration of men in housing complexes for
long periods of time gave rise to social and sexual risk behavioral responses that have contributed to a much
more rapid spread of the disease in these countries than elsewhere (Epstein, 2003). See also Chin (2003) for
related epidemiological-based explanations of regional differences in the spread of the disease.
This assumes that current projections by UNAIDS and US Census Bureau are correct. Bennel (2003) argues
that in some cases, official HIV prevalence is probably overstated and that advocacy is getting in the way of