Southern African Regional Poverty Network (SARPN) SARPN thematic photo
Country analysis > South Africa Last update: 2020-11-27  

 Related documents

International Food Policy Research Institute (IFPRI)

HIV/AIDS, land-based livelihoods, and land reform in South Africa

International Food Policy Research Institute (IFPRI) and RENEWAL

January 2006

SARPN acknowledges the International Food Policy Research Institute (IFPRI) as the source of this document:
[Download complete version - 550Kb ~ 3 min (80 pages)]     [ Share with a friend  ]

Executive Summary


In 1994, South Africa’s Department of Land Affairs initiated a land reform programme. Land reform was – and still is – intended to redress the racial imbalance in landholding, develop the agricultural sector and improve the livelihoods of the poor. These far-reaching objectives were derived from an understanding that land reform has the potential to make a direct impact on poverty through targeted resource transfers, while simultaneously addressing the economic and social injustices caused by colonial and apartheid dispossession. However, there is now a growing concern that the pro-poor objectives of land reform may be under threat from the HIV/AIDS epidemic. However, thus far there has been little tangible action aimed at either better understanding the relationship between land reform and the HIV/AIDS epidemic, or in respect of adjusting land reform policies and practice in reaction to the epidemic.

Although the present study is, as far as we are aware, the first and only research effort thus far of the relationship between HIV/AIDS and land reform in South Africa, it is not the first treatment of the relationship between HIV/AIDS and land-based livelihoods, either in South Africa or elsewhere in the region. However, these prior exercises had some notable shortcomings for which the present study seeks to correct. The first major shortcoming of earlier efforts – including the HSRC’s own research – was focusing only on HIV/AIDSaffected households, which meant that the actual impact of HIV/AIDS could not be distinguished from other possible influences. The present study therefore deliberately seeks to interview both affected and non-affected households. The second major shortcoming was that most of the earlier studies were merely snapshots in time. Given the dynamic manner in which HIV/AIDS affects households, as well as the dynamic nature of rural livelihoods generally, it is preferable to trace patterns of change over time. (In addition, as of yet there are no longitudinal studies of land reform, whether of projects or particular beneficiaries, thus our understanding of land reform is itself more one-dimensional than is desirable.) Accordingly, the study is conceptualised as a longitudinal study covering three years. The study is presently concluding its first year, which has focused on 10 sites in three provinces. The idea is to return to the same sites repeatedly over the three years, to bear witness to the relationship between HIV/AIDS and land reform and/or land-based livelihoods as it unfolds over time. Thus the information collected for the first ‘wave’ is considered baseline data. It was not expected at this time that a full understanding of the relationship would have been established. Indeed, much of the analysis that follows pertains to the nature of land reform projects and land-based livelihoods, and infers the connection to HIV/AIDS rather than observes it directly.

The ultimate aim of the research is to generate actionable policy recommendations and programme responses, first of all by answering basic questions such as to what extent and in what way the HIV/AIDS epidemic poses a threat to South Africa’s land reform programme, and secondly by identifying specific ways in which land reform policy and practice should be adjusted. However, it should be clarified that the study is not solely about the impact of HIV/AIDS on land reform and rural livelihoods, but equally about whether and how land reform can serve as an intervention to mitigate the impact of HIV/AIDS on affected households.

Study description

The study is designed as a collection of mutually illuminating case studies, covering a range of project type and geography. Because resources were limited, the study was confined to three provinces – Limpopo, Eastern Cape, and KwaZulu-Natal – which were chosen largely on account of their large rural, poor populations that stand to benefit from land reform, as well as the fact that they are the three provinces with the largest numbers of black households engaged in agriculture. In each province, three or four sites were identified, including sites consisting or redistribution project, restitution projects, and communal sites, i.e. communities that are not part of the land reform programme. The inclusion of communal sites serves two purposes. First, communal areas are areas in which the Communal Land Rights Act (CLRA) will eventually be implemented, thus the research ultimately aims to generate insights that will be useful to the development of implementation systems for CLRA. Second, communal areas serve as a sort of ‘control’ against which the data from redistribution and restitution projects can be interpreted.

In each site, various research activities were conducted, including project visits, key informant interviews and household interviews. Around and apart from the sites, in each of the three provinces the research team undertook a scoping exercise to determine the extent to which local civil society institutions were aware of rural people being displaced from their homes, and held a workshop with provincial officials of the Department of Land Affairs together, where possible, with other provincial stakeholders.


As with similar studies, the present study was handicapped by the fact that it did not have hard information as to respondents’ ‘HIV/AIDS-affectedness status,’ i.e. whether or not it is a household in which a household member has died from an AIDS-related condition, is presently ill with an AIDS-related condition, or is aware or suspected of being HIV positive. The study relied therefore on respondents’ direct revelations, as well as a few other imperfect but necessary indicators. The fact that the research team’s interpretation of which households are and are not affected is not wholly erroneous, is supported by the correlation between imputed affectedness status and other variables, e.g. household welfare. However, it remains a concern that more is left to guesswork than is desirable.

Based on this approach, the estimated ‘prevalence’ of HIV/AIDS-affectedness varies enormously across sites, with communal sites tending to be on the high side (near 50%), and redistribution and restitution projects varying from 5% to over 40%.

Octoplus Information Solutions Top of page | Home | Contact SARPN | Disclaimer