Getting Customary Land Rights Wrong: The DA’s Private Member’s Bill
This statement is authored by Ruth Hall and Andries du Toit, Institute for Poverty, Land and Agrarian Studies (PLAAS)
The DA’s notice in respect of the introduction of a Private Member’s Bill on Communal Land Rights is right about one thing: poor people’s tenure rights in the former Bantustans have been languishing in a policy vacuum for too long. The Department of Rural Development and Land Reform has failed to provide any leadership on this matter. This vacuum has been particularly serious since the Constitutional Court struck down key provisions of the Communal Land Rights Act in May last year. A debate on the future of the former Bantustans is sorely needed.
Unfortunately, the DA is wrong about almost everything else. The proposals they put forth are misguided, outmoded, and based on a mistaken analysis of the problem. They are likely to make matters worse, not better. In addition, they are entirely unrealistic and probably almost impossible to implement.
Communal rights and Freehold Tenure
The DA’s proposal is that communal tenure should be replaced by freehold tenure. According to their spokesperson, Lindiwe Mazibuko, the DA believes that giving people unencumbered title to their land is essential to secure their tenure and to ensure increased productivity, rural job creation, and food security.
This is based on ignorance of how tenure works in areas governed by customary practices. It also ignores the lessons of history. Ambitious titling programs such as the one proposed by the DA have been tried in other countries in sub-Saharan Africa, and they have failed to achieve their aims. They have proved to be expensive to implement; they have not resulted in the hoped-for boosting commercial agriculture; and they have in many cases exacerbated poverty and inequality.
- The first thing the DA gets wrong is that it does not understand the value and importance of communal tenure. The history of land tenure systems in Africa and elsewhere indicates that communal and customary tenure can play a positive role in reducing poverty and vulnerability. Access to communal lands is an important safety net, and allows many people to survive who otherwise would be forced to migrate to the cities, there to become part of the urban unemployed. Rather than to try to demolish customary tenure systems, government should try to strengthen them and make them more transparent and democratic.
- Secondly, communal tenure is not necessarily insecure tenure. Properly managed, they can be instruments for the flexible allocation of land rights and access to those in need. All over the developing world, productive smallholder and subsistence farming takes place on land that is not under freehold title.
- Thirdly, the DA misunderstands the reasons for poverty in the rural areas. The unresolved nature of tenure systems is not a major obstacle in the way to rural development, growth and employment. The reasons lie elsewhere. The most important obstacles are the lack of appropriate agricultural policies, the lack of properly designed and implemented extension and support for smallholder farming, and the lack of infrastructure and investment. None of those are dependent on freehold title.
- Fourthly, the DA’s proposals are likely to have more negative than positive consequences. Freehold title on its own cannot create the conditions for rural commerce or enterprise. The notion that freehold title will give poor people access to capital and finance that would not otherwise be available is an illusion. It has not worked in poor urban slums, and it will not work in poor rural areas. Banks are unlikely to provide accept land in poor and overcrowded communal areas as collateral, no matter what the form of title is. And many of the poorest and most vulnerable are likely to end up being excluded.
- Finally, the course of action they propose is impossible to implement. The Communal Land Rights Act (11 of 2004) proposed registration of the land rights of ‘traditional communities’, which would have involved rights enquiries, defining physical boundaries, defining community membership and drawing up community rules – a complex process that was expected to take a matter of a few years per community. It was estimated that approximately 20,000 such ‘traditional communities’ exist within South Africa. Even with very substantial funds and institutional capacity, it would be expected the registering land to communities would take a matter of several decades. What the DA proposes is even more ambitious. Registering the rights of individuals within communities, and transferring these in private ownership, would be an enormous undertaking. Who should get title, and who should not? Who will adjudicate disputes and how? International experiences of titling programmes being imposed in situations where customary practices govern land tenure suggest that they are usually undermined by (a) the problem of records being chronically out of date, (b) conflicts emerging because of competing and overlapping claims to land within and between communities, and (c) the privileging of men and elites, and subordination of secondary rights holders, usually women. In Kenya, for instance, the infamous Swynnerton Plan, a titling drive very similar to that being proposed by the DA, implemented from the 1950s onwards, turned out to be highly costly. It not only failed to stimulate rural economic development, but has also been widely recognised as a long-term cause of land-based conflicts in that country.
It is disappointing that the DA is seeking to recycle once again failed development notions from more than fifty years ago. The notion that one can modernize rural areas by imposing freehold tenure from above is outmoded and discredited.
Instead, two alternative routes should be explored. First, legislation should be developed to provide statutory recognition of existing occupancy as constituting a real property right, enforceable in law and equivalent to private ownership (even where these are informal rights that are not registered).
Second, the option should be made available to communities within these areas to register their land and to formalise their systems of land allocation and administration, with state support. Registration should not, though, be a precondition for legal recognition of rights. And registration should be available for a wide array of rights, not just for freehold tenure.
This two-fold approach is based on the Mozambican model, and is widely accepted as best practice on the continent.
Other aspects of land management
Two further provisions of the DA proposals do merit comment. Firstly, they propose that legislation must be drafted to ensure that all land in the former homelands is surveyed so that it is known exactly how much land exists and who is living on it. Currently, this land is categorised as “unsurveyed, unregistered state land” and “trust land” and is not properly registered in any database. This is a good suggestion, but it does not require new legislation. It can be done under current law.
Secondly, the DA proposes that the rates collection system that applies in municipal areas should be applied in these areas as well, so that the entire country is covered by a simple and uniform rates collection system. In addition, these areas to be demarcated as municipalities and therefore subject to the provisions of the Municipal Systems Act. This is a strange proposal. The communal areas of South Africa already fall within the provisions of the Municipal Systems Act and are part of demarcated municipalities which already cover the entire country. The proposal to bring these areas within this municipal system is simply redundant. The proposal that the poor and marginalized people who live on this land should be made to pay property rates is unrealistic, and implementing it is likely to lead to civil unrest.
Back to the drawing board
South Africa needs better tenure solutions. But the property models that work in wealthy suburbs can’t be enshrined as the only valid approach. The DA’s bill is poorly thought through, and should be rejected in its entirety.