Issue 23 - Article 7

'New variant famine': hypothesis, evidence and implications

May 29, 2003
Alex de Waal

Under the combined pressures of HIV/AIDS, poverty and drought, southern Africa faces a new type of famine, calling for a new type of response.

In southern Africa, an HIV/AIDS epidemic, severe poverty in the agrarian sector and external shocks such as drought threaten to create a famine with a distinct vulnerability profile and a new trajectory of impoverishment and coping: a ‘new variant famine’. This concept is drawn from existing models of food insecurity and rural livelihoods, revised in accordance with what we know about household responses to HIV/AIDS and reasonable assumptions about future outcomes should rural agrarian communities come under severe compound stresses.

The impact of HIV/AIDS

The HIV/AIDS epidemic forms the central component of the ‘new variant famine’ hypothesis. HIV/AIDS undermines the sustainability of the livelihoods of affected households, primarily through the costs of the care and treatment of sick household members, the labour costs associated with supporting the sick and orphans, funeral costs and through the shortage of adult labour. This in turn compels adjustments to farming practices, rendering them less productive and more sensitive to external shocks. Household-level studies beginning in the late 1980s confirm this trend across southern Africa. The tendency of the private sector to shift the burden of AIDS by withdrawing sickness benefits increases the burden on rural households. Urban households may also send children orphaned by AIDS to the home village for care. As a result, the rural economy is bearing a disproportionate share of the costs of the HIV/AIDS epidemic, while also being less able to call upon support from the urban economy in times of distress.

In aggregate terms, what we are witnessing is the emergence of a new category of poor people: the ‘AIDS-poor’. The basic vulnerability feature of households directly afflicted by HIV/AIDS is the scarcity of labour relative to the number of household members (in other words, they have an adverse dependency ratio). These people cannot be defined geographically or in terms of occupational status: they are scattered throughout rural society. Current aggregate statistical measures for poverty (both average income and headcounts of those living on less than a dollar a day), and population-level dependency ratios do not pick up this category of the AIDS-poor. The tools are too blunt and there are too many confounding factors. There is a need for new methodological tools to measure and monitor this new component of poverty.

Other households may suffer indirectly: their AIDS-afflicted neighbours may no longer hire labour, or they may assume the burden of upkeep of orphaned children. Some may, also, benefit, by acquiring the land of the distressed household. Again, this area needs more research.

In times of acute communal food insecurity, AIDS-afflicted households are particularly vulnerable. They are likely to find that famine coping strategies are less available and less viable. They may find that they are unable to call upon the support of their neighbours and kin because these networks are already saturated by the demands of caring for orphans or supporting the sick. They may be unable to undertake labour-intensive coping strategies such as gathering wild foods and finding casual employment. Households without mature adults may lack the experience and skills that they need to cope in this way. Hence, they are likely to resort more rapidly to responses such as selling essential assets, crime and sex work.

Screen Shot 2012-11-02 at 2.18.54 PM

‘New variant famine’: the implications for aid

The implications of this hypothesis are clear: we are not witnessing a short-term episode of acute food insecurity in southern Africa. There will be no rapid bounce-back to normality should the rains return, as we are accustomed to seeing after drought-famines in various parts of Africa. Rather, the crisis is likely to leave a very substantial proportion of the population in a situation of chronic extreme food insecurity, without sustainable livelihoods and most probably reliant for their survival on a long-term programme of international social welfare. This large, destitute population is likely to suffer from high rates of HIV and AIDS.

If the ‘new variant famine’ hypothesis is even partly correct, it has a range of implications for famine early warning and prevention, and for relief interventions. Early-warning systems will need to develop new indicators for identifying AIDS-related food insecurity. Because of the stigma, denial and discrimination that is still widely associated with HIV and AIDS, this cannot be done by identifying those who are living with HIV. Rather, it is better to identify proxies that are more directly related to the livelihood outcomes of AIDS-afflicted households.

The most important indicator is the availability of labour relative to the number of household members. The concept of ‘effective dependency ratio’ (EDR) is a potential indicator. The EDR ratio is adjusted to include chronically sick adults (whether suffering from AIDS or any other chronic disease) in the dependent category. Other indicators may include the production of low-labour distress crops such as cassava, rather than cereals.

Means of preserving viable livelihoods despite the impact of HIV/AIDS will need to be identified and developed. Most research indicates that there is in fact no such thing as a sustainable livelihood for a household in which one or more mature adults are sick with AIDS. The challenge is therefore one of blunting the impact, for example by providing direct assistance, skills training or micro-credit to healthy family members. Food aid has been used in this way, blunting the effects of HIV/AIDS on affected households. Other responses include increasing incomes (for example through better prices for farm produce); increasing productivity; and (perhaps the most viable) providing assistance to meet some of the additional burden of care. One assistance option is direct financial aid to families caring for orphans.

We must face the distinct possibility that we can no longer talk about food aid and other forms of welfare assistance as short-term measures until ‘normal’ development is ‘resumed’. Where there is a generalised epidemic of HIV/AIDS, the task will be to prevent the further impoverishment of poor households, and this will require a scaled up welfare programme. Although anathema to developmental orthodoxies, such programmes of vulnerability reduction will surely be preferable to waiting until afflicted households have become wholly destitute.

Food-assistance targeting systems will need to take into account the nature of the AIDS-poor. The nature of the impoverishment and vulnerability brought about by the HIV/AIDS epidemic means that geographical targeting will be inexact. The best means of targeting in these circumstances will probably involve communities themselves identifying the vulnerable.

Food-for-work and other labour-based relief programmes will meet some of the needs of affected communities. If targeted particularly at young women, these may be a means of preventing many from resorting to commercial sex work. However, labour-scarce households may be unable to take advantage of such schemes. Other means of stopping households from descending rapidly through the stages of destitution will also need to be sought, such as cash and food payments for those caring for children.

The vicious interaction between HIV and malnutrition poses particular challenges. Customarily, relief interventions have neglected adults other than nursing and expectant mothers, on the grounds that malnutrition would not seriously endanger their health. This assumption no longer holds. We will need to consider including nutritional supplements to adults in food assistance. Given that only a minority of adults who are living with HIV know they are doing so, or are ready to admit it, this means targeting the entire adult population in an affected area. Because malnutrition is a risk factor in mother-to-child transmission of HIV, expectant and nursing mothers should be a priority.

A supplementary set of challenges arises in the context of programmes for the scaled-up provision of anti-retroviral (ARV) therapy. For ARVs to be effective, those who take them must not only be well-nourished, but must consume a high-protein diet and eat several times a day. The challenges of a scaled-up programme of administering ARVs for a poorly nourished population engaged in a desperate struggle to find food will be insurmountable. In fact, stabilising the livelihoods and nutrition of a population stricken by a generalised HIV/AIDS epidemic is a prerequisite for effective ARV provision.

The worst-case scenario is that current food insecurity combined with the HIV/AIDS epidemic creates a wide, severe and intractable famine. A less severe outlook is for the continuing impoverishment of a wide section of rural southern Africans and the continued high prevalence of HIV. National and international mobilisation of resources, capacity, programmes and political will can prevent this. Unfortunately, time is not on our side: the crisis of ‘new variant famine’ may well be upon us before we fully understand what it is, or have the tools and means to respond.

Alex de Waal is Director of Justice Africa and senior advisor on HIV/AIDS and governance to the UN Economic Commission for Africa.

References and further reading

T. Barnett and P. Blaikie, AIDS in Africa: Its Present and Future Impact (London: John Wiley, 1992).

C. Baylies, ‘The Impact of AIDS on Rural Households in Africa: A Shock Like Any Other?’, Development and Change, vol. 33, no. 4, 2002, pp. 611–32.

S. Rosen and J. Simon, ‘Shifting the Burden of HIV/AIDS’, Boston University Center for International Health, February 2002,

‘Household Food Economy and HIV/AIDS: Exploring the Linkages’, Food Economy Group, 2002.

T. Barnett and A. Whiteside, ‘HIV/AIDS and Development: Case Studies and a Conceptual Framework’, European Journal of Development Research, vol. 11, no. 2, 1999, pp. 200–34.

G. Rugalema, ‘Coping or Struggling? A Journey into the Impact of HIV/AIDS in Southern Africa’, Review of African Political Economy, 2000, 26.86, pp. 537–45.

D. Webb and S. Paquette, ‘The Potential Role of Food Aid in Mitigating the Impacts of HIV/AIDS: The Case of Zambia’, Development in Practice, vol. 10, no. 5, 2000, pp. 694–700.


Comments are available for logged in members only.