Can this Budget bring about land justice? Something has got to give

By Nkanyiso Gumede and Ruth Hall

Budget allocation is one measure of the government’s commitment to deliver on its obligation to enact land reforms – including land redistribution, land restitution and land tenure reform. The government makes resources available to the Minister of Land Reform and Rural Development to fulfill its mandate to implement land reforms. In this section, we look at resources in the form of budget allocation over time, how it has changed, and the budget allocation as a proportion of the total budget.

Land reform as part of the national budget
The land reform department budget has always been below 1% of the total national expenditure with the exception of the two years from 2007-2009, when it peaked at 1.09%. These are also the same years where land delivered through both land restitution and redistribution came close to a million hectares a year.  In general, land reform has received less than 1% of national expenditure, with the exception of the years 2007/8 and 2008/9, during the peak of government funding, and the peak of land delivery. In the current financial year (2025/26), the land reform budget is 0.4 % of the total budget. Read this year’s national budget here.

Land restitution budgets
Land restitution is implemented through the Commission on the Restitution of Land Rights, which enables return of ancestral land to those who were disposed of their property after 19 June 1913 due to past racially discriminatory laws or practices. To achieve this, the government allocates a budget to implement land restitution: to run the Commission, to purchase land, and to pay financial compensation to provide grants to enable settlement and land development.  This time last year, the budget for restitution declined by 11%, and the year before that by 8%. This year, the restitution budget has grown just under 5%, a slight recovery. But is still lower than pre-COVID levels in real terms. The budget is now less than half of what it was at its peak in the year 2007/08.

Graph 1: Land restitution budget over the years

Sources: National Treasury, various (adjusted for inflation to 2024 Rands)

Land reform budgets
The budget for the land redistribution programme has been in decline since 2012/13. The decline has also been reflected in the number of hectares acquired for land redistribution. In 2011/12, 392 000 hectares were redistributed whereas in 2023/24, only 67,376 hectares were redistributed.

Graph 2: Land and tenure reform budget over the years

Sources: National Treasury (NT), various years, up to 12 March 2025 (adjusted for inflation to 2024 Rands)
Note: the National Treasury figures show that prior years’ audited figures are now changed in the latest budget documents. We will ask NT why this is the case. Here we reflect the revised figures from 12 March 2025.

Summary
Both budget lines – restitution and land reform (including land redistribution and tenure reform) – are in long-term decline and have more or less stabilised at levels equivalent to those about 20 years ago. We find this picture far more illuminating than the Treasury’s announcements of “increased allocations”, which do not take into account inflation.

Sources: National Treasury, various years, up to 12 March 2025 (adjusted for inflation to 2024 Rands)

Total delivery of land
Through all aspects of land reform, official statistics from successive Annual Reports show that just under 9.5 million hectares have been transferred through all aspects of land reform combined. This constitutes 11% of commercial farmland. Land transfers can also be seen as going through shifts across different administrations. Here we caution that it’s worth checking our caveats about the statistics.

Table 1: Land reform delivery 1996-2024

Sources: Annual Reports of the Department of Land Affairs / Rural Development and Land Reform / Agriculture, Land Reform and Rural Development, various
Note: The percentage above is based on the Statistics South Africa figure of 86,186,026 hectares of commercial farmland.  

Table 2: Land redistribution hectares distribution over different administrations (1996-2024)

Looking forward: what of the future?
The original target in 1994 was to transfer 30% of land through land reform and was an arbitrary figure. Nonetheless, it is clear that just over one-third of this has been redistributed or restituted in the 28 years from 1996 to 2024. If current trends were to continue – with buying land at market price and with the kinds of incremental budget increases we are seeing – then it would take 48 more years from now to reach the original 1994 target to redistribute 30% of commercial farmland. But hectares do not tell us much about the real social value and impact of redistributing land. Far more important than pushing up hectares is acquisition and redistribution of strategically located land to meet urgent urban and rural land needs and to resolve land claims. With the constrained budget, the government should not be buying up big farms at market price, and should re-orient itself to buying, negotiating for, and where necessary, expropriating properties in the places where land is urgently needed: in and around towns and cities to underpin urban agriculture, small business and other livelihood opportunities; around the communal areas where livestock owners and small farmers need more land; and in those parts of the country where farm workers and dwellers need land of their own, especially those who have been evicted. You can hear more from people affected by this in our public dialogue with Land Reform and Rural Development Minister Mzwanele Nyhontso. Watch it here.

Postscript: PLAAS’s five caveats about land reform statistics
There are several reasons why the data we present here should be critically interrogated. Overall, our research over the years informs the view that the real scale of land reform is substantially less than the official statistics. We give five reasons why land reform statistics do not reflect the full picture.

1. • Inferring hectares: hypothetical amounts of land that were not in fact transferred, but where equity was bought in a business. From our studies, the entire area of a farm would be cited as ‘transferred’ if a worker trust acquired a minority share in it. This means figures can be seriously misleading, inflating the amount of land counted as transferred, which was never transferred.
   • Strategic partnerships: land is transferred, but the claimants or beneficiaries cannot move on to the land, or make land use decisions. Best-case: rental income, often along with displacement of farm jobs to some of the claimants,
   • Land in, land out: when the state sells land and purchases land without reference to whether or not the land is actually allocated to people on secure terms, there’s a risk of double-counting.
   • Land reform reversals: the figures we present do not account for reversals in land reform – such as repossession of redistributed land by the Land Bank and commercial banks – following debt default. We know that repossessions have occurred, but there is no publicly available information about the scale. Further, reversals in the form of distress sales are not evident.
   • Data management: there have also been errors in data management. At times, when we have dug into provincial project lists and reported statistics, we have found anomalies, with cumulative totals suddenly dipping, and decimal place problems throwing out figures. Some of the more significant errors we picked up in the past appear to be resolved now.