Menu Close

South Africa’s land debate is clouded by misrepresentation and lack of data

Harvest season on a wine farm in Stellenbosch, South Africa. The country is struggling with land redistribution. Shutterstock

South Africa’s parliament has passed a resolution to amend the constitution and allow expropriation without compensation. The decision has generated a storm of gigantic proportions as political parties, citizens, white farmers and commentators anticipate either the moment of salvation (‘real land reform at last!’) or disaster (‘the collapse of the market economy!’).

Sadly, few contributions to the public debate are informed by the available evidence. And poorly informed commentators often misrepresent the issues. Compounding this is a serious problem – the absence of reliable national data on many aspects of the land issue.

Land policy, at the centre of the storm, is flailing around in the dark. South Africa’s land policy is based on three main pillars: restitution, redistribution and tenure reform.

Restitution involves people claiming back land taken away from them after June 1913, or compensation for their loss. Land redistribution involves acquiring and transferring land from white farmers to black farmers, for a variety of purposes, including farming and settlement. Tenure reform aims to secure the land rights of those whose rights are insecure as a result of past discrimination.

Land reform has been slow, with government reporting that, so far, around 9% of commercial farmland has been transferred through restitution and redistribution. Tenure reform has been remarkably ineffective, with many poor people as insecure as they ever were.

But, in reality, there is only the haziest of understandings of how well or how badly land reform is doing, and why.

The woeful record keeping of national and local government departments is partly to blame. But they are not the sole culprits. The last census of commercial farming conducted in 2007 underestimates the true numbers of farm owners as it only reports on farms that are registered for VAT – currently those with a minimum turnover of R1 million. And StatsSA agricultural reports don’t distinguish farms by size or value of output. Also, official survey data on smallholder farming are also thin and even less useful.

In general, the lack of accurate information on land reform and the rural economy allows much of the public debate to be misinformed, and is a serious constraint on policy making.

Data deficit

Nobody knows precisely how much agricultural land has been privately purchased by black farmers and how much has been acquired via land reform.

Consider two national land audits released in recent months, one by the agricultural lobby group AgriSA and the other by government. Both are based on analysis of information derived from title deeds in the national registry.

The AgriSA land audit of 2017 argues that the initial target of transferring 30% of agricultural land via land reform is close to being met. It concludes that the market is much more effective at transferring land than the state.

But the market is not redistributing land to black people to the extent AgriSA claims. Its methodology and most of its conclusions are fundamentally flawed. For example, much of the 4.3 million hectares of land it says were acquired through private purchases by previously disadvantaged individuals includes transfers of land as a result of land reform. In these cases, government has provided funds and served as an intermediary in the transaction. So they were not private transactions.

Government’s latest land audit is also not particularly useful. It provides some evidence of continuing patterns of racial inequality in land ownership. But it can’t identify the racial, gender and national identity of the 320 000 companies, trust and community based organisations that own 61% of all privately owned land.

Neither of these audits is able to identify zones of need and opportunity for land reform. Information of this kind, crucial for well planned redistribution, simply doesn’t exist.

There is almost zero information on how many people have actually benefited from land reform, patterns of land use after transfer, and levels of production and income.

A few reports on these issues have been published, but they aren’t a substitute for systematic data collection. Similarly, case studies by academics can’t serve the wider purposes of guiding planning for land reform at scale.

In relation to deeds registry data, there are vast discrepancies between official records for black land owners, both rural and urban, and realities on the ground. In our 2017 book, ‘Untitled. Securing land tenure in urban and rural South Africa’, we estimate that close to 60% of all South Africans hold land or housing outside the formal system, and the deeds registry can tell us little or nothing about these realities.

Debunking myths

Official data, although inadequate, does allow common misrepresentations of land reform to be refuted. For example, one widely held view is that the great majority of land restitution claimants have chosen cash compensation rather than restoration of their land. This is nonsense.

Around 87% of land claims lodged by the cutoff date in 1998 were to urban properties, and in most cases claimants were offered (and accepted) a standard cash settlement, because restoration was clearly impracticable. But the great majority of rural claims, involving a great many more people since most are group claims, have opted for restoration.

Another misconception is that land reform can involve the redistribution of state owned land. The reality is that most state land in rural areas comprises densely settled communal land which obviously isn’t available for redistribution. The recent government land audit confirms this, and shows that state land comprises only 18% of the total.

In urban areas, however, state owned land can be used for low cost housing if it is in appropriate locations close to economic opportunities.

The single most misleading ‘fact’ endlessly repeated in the media is the assertion by former minister Gugile Nkwinti that 90% of land reform projects have failed. This has no foundation in any research evidence – a fact that he himself later admitted.

Empirical evidence suggests that around 50% of the projects have improved the livelihoods of beneficiaries to a degree.

This is not to say that these land reform projects have been highly productive. The real potential of rural land reform and agricultural development, as well as urban land reform, to reduce poverty and inequality has not been realised in South Africa to date.


How to acquire and transfer land, the focus of much current debate, is the least difficult aspect of land reform. It simply requires increasing the tiny budget and paying just and equitable compensation in line with the constitution.

Larger challenges involve targeting beneficiaries, identifying well located land, ensuring that water is reallocated along with land, effective district based planning, and enabling small scale economic activity in both rural and urban spaces.

And new legislation that secures the rights of people in the social tenures found in communal areas, farm dweller communities and informal settlements is also an urgent requirement.

Much of the current commentary on land policy is ill informed and fails to identify these challenges. More important, the lack of robust official data on land and agriculture is a key problem that must surely be high on the agenda of the new president.

Want to write?

Write an article and join a growing community of more than 186,800 academics and researchers from 4,994 institutions.

Register now