What is redistributive land reform for? And who should benefit?

Cherryl Walker
Human Sciences Research Council

Paper presented at the Harold Wolpe Memorial Trust conference “The land question in South Africa: The challenge of transformation and redistribution”, held at the V&A Hotel, Waterfront, Cape Town, 25-27 March 2004


It is easy, at events like this, to roll out a wish list of what land reform should achieve and who it should target: poverty reduction, justice, food security, rural transformation, economic growth, redress; for: the landless, the poor, women, the dispossessed, and the previously disadvantaged.

We probably all agree that these are worthy objectives and worthy beneficiaries. However, working at this level of high-flying, feel-good generality does not sharpen our understanding of what contribution land reform can actually make towards either improving or transforming the (different) positions of these somewhat different social categories of people, nor what scale of land reform the state can actually as opposed to ideally implement, nor how much weight we should attach to land reform compared to other state programmes (job creation, housing, health, education, welfare, basic services, infrastructure, conservation, etc.), nor what the trade-offs and synergies between these programmes could or should be, nor, if the state is to prioritise, where it should concentrate its land reform efforts.

How important is land reform in South Africa today? Politically it has a sometimes latent, currently more prominent, always potent emotional and symbolic appeal in national debate about the future and the past; at the local level it resonates powerfully with those living on the margins, although the narrative of local-level redress regularly involves complications in the form of competing claims on specific pieces of land. Economically and socially the importance of land reform is more difficult to compute – less of a priority than jobs for most South Africans; significant, in the absence of jobs, in underpinning the "multiple livelihood strategies" of the poor; linked to identity and citizenship in complex and shifting, situation-specific ways, and probably more usefully but rather prosaically conceptualised as a host of individual, targeted projects within a broad development programme, rather than a grand storming and seizure of the heights of capitalist, racist, commercial agriculture.

In another paper I have tried to make a case for an unfashionably sober but firmly grounded, hence more rather than less effective, programme that would straddle these different currents (Walker, forthcoming). I am drawing on this paper in what I say here.

There are four preliminary premises on which my analysis rests:

·       South African society and economy is no longer primarily agrarian,

·       The non-agrarian economy is nevertheless failing dismally to absorb people, and this is unlikely to change in the near to medium future,

·       Land reform by itself cannot guarantee incomes, livelihoods or social equality (or, for that matter, rural stability), i.e. land reform is not an end in itself but a possible means to several ends, and

·       Limited state capacity (that blandest of catchphrases) is not a temporary aberration, but an institutional reality that will remain with us for a long time to come.

Accepting that these conditions structure the terrain in which land reform is working, I think the state’s land reform programme can and should prioritise the following objectives:

·         Redress and social justice for the victims of forced removals in the more recent (rather than the more distant, irretrievable colonial) past, through the restitution programme.

Land restitution is enshrined in our Constitution and we need to honour that commitment (although I do not think that the current, presidential target of 2005 is achievable, even with the cutting of lots of corners). The current provision makes what I think is a pragmatic but appropriate cut-off point of 1913. The persistent legacy of colonial dispossession before the Natives Land Act has to be addressed through other strategies and programmes. In my view, only those who deal with our history at the level of sweeping generalisations can imagine that we can apply the methods of restitution to the shifting socio-political dynamics and land maps of the 19th century and before that are layered beneath our contemporary land dispensation. The history of conquest, collusion, alliance, dispossession and migration, as well as of tribal, ethnic, class and national identity formation and of changes in tenure regimes is simply too dense, too complex to be managed through such a programme. Let’s deal with the competing claims and boundary disputes of the present, notably those looming around tenure reform in the former bantustans, before we contemplate addressing - reviving - those of the distant past.

·         Legal recognition of and mechanisms for the enforcement of the customary land rights of families (and individuals within families) and communities living on state-owned land in the former bantustans

This should include provisions for state protection of designated communal resource areas in perpetuity, and for demarcating group and household boundaries and transforming rights into other tenure forms (leasehold, freehold) as the need arises locally.

I know this is a workshop focusing on redistribution, but it is impossible to separate the issues of tenure reform and redistribution in policy and practice, and impossible to divorce a discussion on redistribution from an analysis of land reform policy within the former bantustans. In the region of 35% of the rural population live in these areas, which we all know to be characterized by levels of extreme poverty, inequality and dependence on external earnings and remittances. Yet some of these areas have reasonable, even good, agricultural potential and, paradoxically on the face of it, land in these areas can be under-utilised. We also know that coherent systems of land administration have largely collapsed here and that local development agendas are often driven by powerful systems of patronage and short-term goals.

·         State acquisition of privately owned land for the establishment of a range of different types of rural settlements and projects for poor, landless and land-hungry households and individuals from these (communal) areas as well as from commercial farms.

The type of projects that are possible here include small farmer settlement schemes, agri-villages, small agricultural and non-agricultural production enterprises, land-based hospices, children’s villages – these to be developed on optimally located land, acquired strategically and pro-actively by the state to suit the different requirements of these different types of projects, and made available to targeted groups. The focus should not be on agricultural projects only – land reform should be delinked from the current emphasis on land for agricultural development and aligned more closely with broader development strategies.

·         State acquisition of land for the pro-active establishment of peri-urban land reform projects, including smallholdings and urban agriculture settlements.

Land reform policy development urgently needs to come to terms with the huge land, settlement and livelihood needs and potential opportunities within peri-urban and urban (including non-metropolitan) areas, where the bulk of South Africa’s population now resides. This is a very significant area of need, as well as of potential for developing projects that will allow poor people to straddle urban income and land-based food security opportunities better, while having access to the better services and health and education infrastructure of the urban areas.

·         State protection for and enhancement of women’s social standing and economic opportunities through land reform legislation, policy and implementation practice.

Important issues here concern the advancement of women’s rights in communal, family and household land, as well as the recognition of women’s right to participate on equal terms with men as individuals in land reform projects. It is important that the debate on women’s land rights is not understood in terms of women-headed households only, nor as a policy choice between individual or household land rights and interests for women. Most poor women are living in households dominated or headed by men and their interests and rights in the property resources of these households need attention in land reform policy as well. (I develop these points in Walker, 2003.)

The above is necessarily a rather schematic outline of priorities, but I think all these objectives are attainable and relate to actual needs. My list bears some resemblance to the policies of the past ten years, but there are some significant differences in terms of both principle and emphasis, in particular:

·       The proposal that the state acquires suitable land pro-actively for different target groups (which could mean using its constitutional powers of expropriation, with ‘just and equitable’ compensation, or entering the land market as a buyer);

·       The importance given to peri-urban land reform; and

·       The retention of state ownership of communal lands, pending a more organic and locally driven process of tenure retention and tenure change than that currently envisaged under the Communal Land Rights Bill.

Missing from the above set of priorities for land reform is a strong statement about land reform in support of the transformation – read deracialisation – of commercial agriculture. I think this is a not unimportant objective, insofar as we agree that we need to uncouple the strong association between large-scale land ownership and racialised privilege in the countryside, and retain a healthy (and sustainable) commercial agriculture sector to feed our cities, and, in the short to medium term at least, have enough of a surplus to export to the region as needed.

However, I do not think that this objective should be the major priority of the state – state funds and resources, in my view, should be directed towards those most in need, not those who can secure a loan and private-sector support to enter the commercial agriculture sector. On the other hand, given our history, there is a case to be made for some level of state involvement in facilitating this process, for instance through the provision of soft loans and guarantees, as well as in pressurizing a notoriously conservative sector to adapt and open itself up more emphatically to black farmers (to realise where its own best interests lie). We should, however, expect that a non-racial commercial farming sector is likely to be almost as hostile to a more radical redistributive land reform programme as a previously all-white sector has been.

To illustrate my point about the limitations of a redistributive land reform programme conceptualised primarily and simply in terms of deracialising the commercial farming sector, by overturning the monopoly of white farmers on ownership of the 68% of the country falling into this sector, I think we should consider the following.

There are currently somewhere between 50,000 and 60,000 commercial farmers in South Africa, the great majority (but no longer the totality) of whom are classifiable as white. One reasonable estimate of the level of landlessness in the former bantustans is 675,000 households (Aliber and Mokoena, 2003). Thus it is theoretically possible to replace all white commercial farmers with black commercial farmers – eradicate the white hegemony over non-reserve land of the past 90 years - and (assuming these new black commercial farmers were drawn from the landless) reduce the pool of landless in the Bantustans by no more than 10%. In other words, if the only consideration shaping the transfer of land is that the recipient is classifiable as black, one could have a successful land redistribution programme that makes no or a very limited impact on poverty reduction and the transformation of the agrarian economy.

To make the same point in relation to current land reform targets: In the Northern Cape there are today somewhere in the region of six to seven thousand commercial farmers, who together own just under 30 million hectares. This is the largest block of commercial agriculture land in the country, but probably the least valuable from a farming and livelihoods perspective, certainly the least suitable for intensive settlement and use and the least capitalized. 30 million hectares is 25% of the total land area of South Africa and about 36% of total commercial agricultural land (approximately 82 million hectares). Thus – hypothetically – simply by directing all land reform resources at replacing white farmers in the Northern Cape (only 10% of the national total) with black farmers, it would be possible for the DLA to exceed the current land reform target of 30% of all agricultural land redistributed by 2015. However, it should be obvious that if this transaction were to be no more than a straight colour-coded swap, the impact, while not without social effect in the Northern Cape, would be negligible in terms of making substantial inroads on rural poverty and land hunger in that province, while leaving land issues in the rest of the country essentially untouched. The Northern Cape example, I think, illustrates the limitations of regarding the inadequate shorthand of 87/13% (the conventional but misleading figures for white/black-owned land) as the major indicator of inequality and exclusion in contemporary South Africa.

In the rest of my presentation I want to take a step back and look at the limits on land reform as a goal of state policy. I am certainly not arguing that we don’t need land reform – hopefully I have made the case for it above. I am, however, concerned about overly optimistic - misplaced - expectations of what land reform can achieve. My first two points concern the tensions that exist within the programme between national outcomes and local process. The other five points relate to certain non-programmatic constraints on land reform – i.e. constraints that are rooted not in the failures or weaknesses of political will or particular conceptions or practices around land reform, but in structural issues that will continue to confront and constrain alternative programmes and a different set of implementers as well.

My first point is about the disjuncture that exists between the national political debate around land reform and its targets on the one hand, and local imperatives around actual land reform projects on the ground on the other. Today the national debate is focused on disposing/transferring as much land as possible – it takes place at the level of overall targets, broad transformation goals and politically informed deadlines. The local – the project level of implementation - requires attention to process in beneficiary identification, participation, institutional development, planning, and the provision of services, and probably more rather than less official time (meaning larger operating budgets) per project to work better. In many cases there are issues and conflicts to be resolved which cannot be rushed to an externally imposed, deadline-driven resolution.

The second, related point is that speeding up the pace and scale at which land is transferred overall will not resolve the underlying constraints on sustainable implementation – may even exacerbate these. The national obsession with overall targets and total numbers works against improved quality, stronger local institutions, more appropriate development plans. And here it is important to note that in particular, if we are serious about bringing more women, especially more poor women, into the land reform programme, we have to pay more attention and give more time to process, to facilitation, to investigating the menu of different, gendered options and outcomes and risks. In pushing the national debate, therefore, we need to understand that there are tensions between national and project-level requirements and priorities, and difficult political trade-offs. (This is not a challenge for land reform alone.)

The other points I want to raise relate to what I describe as the non-programmatic limits to land reform – all huge issues, which I can only touch upon here, which also need to be thought of as interacting with, rather than as operating in isolation from, each other.

The first of these involves demography, including population growth and urbanization. In the course of the 20th century the population of South Africa grew nine-fold, from a little over 5 million in 1904 to c 45 million today. Inevitably this translates into increased competition for the more favourable land and more pressure on resources, and requires state regulation and planning around settlement options and land use. Furthermore, today over half the population (58%) is classified as urban. Urban constituencies are politically dominant, and even though, as we know, many people living in urban areas straddle rural- and urban-based identities and livelihoods in complex ways during the course of their lives, nevertheless, the aspirations and expectations of growing numbers of South Africans, the youth in particular, are urban rather than rural.

The second point concerns ecological constraints. We know, or should know, that South Africa is a water-poor country and that access to water is set to become an increasingly serious regional issue in the coming decades. The optimal management of water also poses major policy, planning and regulatory challenges for the state. Linked to this, there is also a limited amount of arable land in the country – only 12% of the country is classified as arable, most of it on the already densely settled eastern seaboard. Much of the country is not suitable for large-scale resettlement or intensive land use, e.g. the Northern Cape, which has already been mentioned. (The issue here is not whether one can or cannot "do land reform" in arid areas. The issue is what sort of land reform, for what densities of population, what sorts of return, and at what cost to the state and the environment.)

The third point concerns the difficulties facing agriculture as an economic sector in today’s world. As others have pointed out, neither the local nor the global policy and economic environment is favourable for those wanting to make a living out of farming, whether at the small or large-scale end of the spectrum. (Hence the need to look more carefully at the requirements for a land reform programme not tied so tightly to agriculture.) Neither nostalgia nor romantic identity politics are sufficient basis for a strong agrarian policy.

The fourth point, which is of particular relevance for redistribution projects and proposals that are premised on the state acquiring land pro-actively (including my own), concerns the reluctance of many (not all) poor people to relocate from their social networks, from the familiar, and move long distances in order to access land. Unless we want to propose massive state engineering programmes of population relocation, which would be ominously reminiscent of apartheid-era planning, social factors of this nature also set limits to a radical redistribution programme, especially to the opportunities offered by redistributive land reform to the heartland districts of the former bantustan territories, notably the Transkei. This is not to say that there would be total resistance in these areas to migration in search of new opportunities – migration is after all woven into their history over the past 150 years. However, one recent national study of migration has documented that those who are most vulnerable in terms of poverty are the least likely to migrate, with poor women-headed households, one of our target groups, particularly strongly represented in this category (Kok et al, 2003).

My final point concerns the as yet inadequately articulated and poorly understood impact of the HIV/AIDS pandemic not only on poor people’s priorities and needs in relation to land reform (and other state programmes) but also on state capacity to design and deliver such a programme effectively. There are many difficult issues and unknowns here, but numerous studies have shown that the capacity of AIDS-affected households and individuals to use their land productively and defend their rights in land is strained and/or eroded, while tensions over land rights, tenure systems and land use between men and women and across the generations may be exacerbated. Elevated morbidity and mortality rates can be expected to undermine community institutions, such as land management structures, and shift local and national priorities around land reform objectives and outcomes.

Were there time I would have liked to finish my presentation with a case study that illustrates the complexities and ambiguities of land reform at project level. It seems to me that if we really want to develop land reform policy, we need to move between both the macro and the micro levels of analyses, and build up a much more extensive set of local case studies than are currently being drawn upon. I am currently writing up the history of Cremin, near Ladysmith in north-western KwaZulu Natal, which is interesting because it was the first land claim to be settled in KwaZulu Natal, in 1997, and is, I think, one of the more successful restoration restitution projects to date. It is a former ‘black spot’ which was expropriated and removed in 1977/78. The claim was settled without acrimony and with some memorable moments of genuine goodwill. Community leadership has been reasonably cohesive and effective and continues to display a strong commitment to farming in principle. The practice is, however, proving more difficult, even though the land, while not prime, is not bad, and the farm is relatively well located in terms of roads, access to town, a railway line and markets. After 6 years, only a handful of claimants (17 out of 83) have returned. The young people I have spoken to are reluctant to move away from the township to which their parents were relocated 20 years ago, in which many have

been born. It is not sentimental ties that keep them there, so much as a reluctance to leave the better amenities and services of the township and a desire to find urban work. Furthermore, on the farm the infrastructure that was promised in terms of the restitution settlement grant has not been fully installed and is not operational. The local council is seen by claimant representatives as overly politicised and has not taken on the service development responsibilities delegated to it by the Department of Land Affairs, while the Department of Agriculture has never visited the project.

Cremin is not a disaster, but it is struggling to become the productive agricultural community that its leadership and the state since 1994 have both aspired to recreate. For those of us who are in the business or politics of developing policy blueprints for land reform, it is important that we understand better what the actual constraints and challenges and opportunities facing Cremin and all the other land reform projects already out there are.

I end with my current, working conclusions taken from the paper I referred to earlier:

Land reform … is overloaded with the claims of history and the twinned but incongruent imperatives of redress for the past and development for the future that that has bequeathed us. It is also hobbled by the constraints of the present, including not only the relative marginality of the rural areas politically and economically, but also the indifferent – uncooperative – natural environment in which it is to work its remedy. Popular expectations have been shaped by a ‘master narrative’ of quintessentially rural dispossession and restoration that, while not, broadly, untrue, is no longer directly relevant to today’s developmental challenges. It focuses too narrowly on the so-called ‘white’ countryside, underplays the importance of urban land reform and the former reserves, and under-estimates the contemporary challenges to agriculture.

It is not that land issues and land reform are not important for the millions who do look to the land to provide or supplement a living. …. It is that successful programmes of restitution, redistribution and enhanced tenure security will, at best, provide only some of the preconditions for emancipation from oppression and poverty ….


Aliber, Michael and Reuben Mokoena (2003). ‘The Land Question in contemporary South Africa’ in John Daniel, Adam Habib and Roger Southall (eds) State of the Nation. South Africa 2003 – 2004. HSRC publishers, Cape Town.

Kok, Pieter Michael O’Donovan, Oumar Bouare and Johan van Zyl (2003). Post-apartheid patterns of internal migration in South Africa. HSRC publishers, Cape Town.

Statistics South Africa (1996). ‘Agricultural surveys 1994, 1995 and 1996.’ Report no. 11-01-01 (1996), Statistics South Africa, Pretoria.

Walker, Cherryl (forthcoming). ‘The Limits to Land Reform: Reviewing the ‘Land Question’’ in Jo Beall and Shireen Hassim (eds) State and Society in South Africa: Ten Years of Democracy and Development.

Walker, Cherryl (2003). ‘Piety in the Sky? Gender Policy and Land Reform in South Africa’ in Journal of Agrarian Change, Volume 3, Numbers 1 and 2, January and April 2003.