Thinking about South African Land Reform.
How might land holding be made more equitable without doing economic harm?
This article will answer this question first by considering relevant African experience. Next, it will turn to the record of land re-distribution in South Africa. Why have South African efforts to undertake land reform failed to satisfy public expectations? And why has land-redistribution become such an urgent policy issue? Finally, what practical steps can the government take that would address effectively demands for land reform without seriously risking economic and political stability?
Two African national experiences of land reform are especially relevant for South Africans, those of Kenya and Zimbabwe. As in the case of South Africa, both countries are ex-British colonies in which a significant portion of land was alienated from the indigenous population and transferred to a white settler population, mainly between 1890 and 1910. As in South Africa, people of local descent were restricted or completely excluded from the possibility of acquiring individual land ownership, and confined to areas designated as ethnic reserves in which land was allocated to households through communal tenure arrangements administered through local chiefs.
Let us consider the case of Kenya first. Here the British settler population was relatively small, at 60,000 a tiny fraction of the country’s eight million inhabitants at the time of independence. The alienated land the settlers occupied they held on 99 year leases. The settled land was the most fertile, relatively well-rained “highland” area, free from malaria and tsetse fly, though only a small proportion of Kenya’s arable land. Much of the land held by the settlers was underused or farmed inefficiently.
Kenya’s independence struggle was closely tied to land rights. It mobilised Kikuyu labour tenants living on white-owned farms in a Land Freedom Army (the Mau-Mau). Political power would eventually pass from the British colonial authorities – ex-patriate officials, not local settlers – to the Kikuyu elite. The Kikuyu were the group most affected by colonial land alienation, and also the group most likely to have experienced western education and urbanisation.
Independence was preceded and accompanied by a land settlement funded by the British exchequer, in which land was bought on a “willing seller, willing buyer” basis. The settlement had three types of beneficiaries: small scale “Yeoman” farmers, who would cultivate mainly for the market and employ a limited amount of help, peasant family-worked plot holders, and big farmers, often Kikuyu politicians leading the ruling party, the Kenya African National Union. This last group took over the larger colonial estates, and in effect the land settlement preserved intact the major part of the colonial agrarian domain. Much of the land allocated to yeoman and peasant farmers was of more marginal quality.
For a time, the land settlement supplied the basis for political stability because in the short term it satisfied land hunger within the ethnic group most affected by land alienation, the Kikuyu. But it privileged ethnicity as an organising principle in Kenyan politics as the land allocations were made on the basis of ethnic group membership. Indeed it used as a starting point colonially conceived ethnic group boundaries.
Simultaneously, the settlement spread the availability of inheritable and sellable private property rights – land holders under the various settlement schemes held individual titles. The state through a range of different agencies would continue to allocate to other ethnic groups remaining public land – all land not held under individual title was crown or state land. Political power and political competition in Kenya increasingly became linked to ethnically assigned land allocation. The original and subsequent settlements entrenched the social inequality that was a feature of the colonial political economy, and especially disadvantaged pastoral groups such as the Masai. Large-scale landed estates have grown proportionately.
Kenyan land reform was a success in three respects. It deflected early social demands for land reform, though the growth of the urban industrial economy in the 1960’s and early 1970’s was also important in providing livelihoods for rural landless people. Secondly, the settlement preserved and maintained Kenya’s export-oriented capitalist sector. Thirdly, to the extent to which the land reform created a small farm sector it arguably resulted in more agricultural efficiency. In the Kenyan experience, small farmers seem to use land more effectively than large estate owners. But today land based political conflict is extensive.
So the Kenyan experience does suggest that major land reform can be undertaken in a way that avoids economic disruption. But in the longer term the way it was done may have perpetuated inequalities. And the original settlement scheme was externally funded.
What about the Zimbabwean experience? In colonial Zimbabwe there was a roughly 50/50 split between settler farms and ethnic reserves. The reserves, the Tribal Trust areas were less fertile and more likely to be arid than white farms. As in Kenya, the independence struggle mobilised rural people, on farms to an extent but particularly in the Tribal Trust areas. At the time of Independence the rural white settler population was quite small – not more than 10,000 households.
White Rhodesians were mainly urban. In the Lancaster House settlement negotiated in 1979-1980, the British envisaged a land settlement similar to Kenya’s – at that time still considered by policy-makers to be an arch-typically successful decolonisation. However planners and Zimbabwean policy makers would pay much more attention to small holder agriculture and to the possibilities of upgrading agriculture in the ethnic reserves through the extension of agricultural services, granary facilities, improving market access and making loans and tenure rights available.
The British committed themselves to funding Zimbabwean land reform and indeed would provide around £40 million until the mid-1990’s (roughly, the equivalent of an annual budget at that time for a small university in the United Kingdom). The British suspended aid for land reform after it became clear that a proportion of the funds they supplied was being used to buy farms for political leaders. In fact, though, the British provision of funding, even while the commitment was maintained, was less generous than in the case of Kenya. Land purchases were initially on a “willing buyer, willing seller” basis.
Land reform in Zimbabwe in the 1980s and 1990s would be at a much slower pace than in Kenya, partly because of shortages of external financing but also because of the resilience of a comparatively efficient settler sector. Land purchases that would provide livelihoods for about 100,000 households by the 1990s were quite successful, though. The major growth in output amongst black Zimbabwean farmers, though, was in the former Tribal Trust areas which began to produce a large proportion of the maize used in domestic consumption.
Redistribution was slow, too slow to even begin to match demand, even after the government began a policy of compulsory acquisition in 1995. Illegal land occupations began in the late 1990s, partly as a consequence of increasing economic hardship in the cities. A hastily implemented structural adjustment programme removed tariffs and subsidies that had protected the local industrial sector that had developed during the sanctions era. After 1991, urban unemployment rose sharply.
What were the effects of the state-sponsored land seizures that began in 2000 and which by 2003 had expropriated the majority of the historic settler sector?
The land invasions especially affected farms producing for the domestic market. Export-oriented agriculture received a measure of political protection though it also was damaged substantially. Maybe more than a million people moved onto the seized farms, between 160,000 and 300,000 households, depending on which calculations are believed. But much of the land they occupied has remained unused. Many of the new occupants of these farms had little or no farming experience. The government’s agricultural extension services were quite unprepared and under-resourced for the huge expansion of need for their support. Zimbabwe’s economy lost export receipts and in effect a food exporting country became food dependent – and remains so today. Wider economic consequences included a currency collapse, accelerating urban unemployment and an exodus of a million plus migrants to South Africa in search of jobs. At least two hundred thousand farm workers lost their livelihoods. The seizures helped to reinforce the rural support for the ruling party and were the key to its continued retention of political power. They were undeniably popular.
Over the longer term there is some evidence of more benign effects. The beneficiaries of land seizures are probably today better off. There are probably today slightly more agriculturally-based rural livelihoods in Zimbabwe than was the case in 2000. A proportion of the farmers are producing export crops quite efficiently, on medium sized tobacco farms for example, in which they employ some labour. Many of the bigger farms, though, that have remained intact are now owned by politicians and are inefficient. And research suggests that most rural households, even in areas in which surplus market oriented production has resumed remain heavily dependent on migrant remittances from South Africa. Indeed without the possibility of massive and illegal migration to South Africa, many more people would be severely impoverished.
The Zimbabwean land seizures certainly resulted in a major re-configuration of land ownership. They helped to reduce public demands for social justice and enabled the ruling party to stay in power. The cost has been the destruction of a relatively efficient labour-employing commercial sector and the wider damage to business confidence. Today the Zimbabwean government is offering compensation to the some of the expropriated white farmers as well as proposing the possibility of their resumption of commercial-scale farming under a 99 year leasehold system.
For many South Africans, it seems, Zimbabwe remains an attractive model. The facts that the state has survived and that the political elite remains resilient and that the economy has been partly reconstructed with the help of Chinese investment in mining enhance its appeal to part of the ANC’s following as well as to its populist opponents. It is a model that would cause even more disruption if it was imitated in South Africa, though.
In South Africa there was a much higher share of land alienation than in Kenya or Zimbabwe. Peasant-or household based agriculture mostly lost its surplus generating capacity at some point between 1930 and 1960 depending on different geographical locations. South Africans were and are much more dependent on urban-generated incomes than was the case in Zimbabwe or Kenya before land reform. Accordingly, South Africa’s political struggle for democracy, its equivalent of anti-colonial mobilisation, was urban based, not really rural until its later stages. The rural people most likely to be engaged in liberation politics were school children and unemployed school leavers. In 1994 when the ANC conducted opinion polling to establish its policy priorities it discovered that demand for land reform was quite low – there was much keener interest in housing, education and health care, for example.
In 1994, South Africa’s commercial sector was relatively efficient, if environmentally costly. It was weaned off state supports during the 1980s when the government began to liberalise its management of the rural economy. The sector had become in certain areas very capital intensive and could be criticised for over-using marginal land. South Africa then and today was a major food exporter, especially to other African countries and was and is domestically food self –sufficient. Agriculture supports a large local food processing industry as well as other industries. And land reform would have needed to be internally financed for unlike Kenya or Zimbabwe South Africa had no prospects of obtaining foreign aid for funding redistribution. In the light of these considerations, the South African government’s relative neglect of land reform in the 1990s was not altogether irrational. Land redistribution would be implemented very slowly and would have to compete with other more pressing demands – housing, water reticulation, education, health and social welfare - for budgetary allocations. The government has remained committed to a constitution that protects property rights and which stipulates that except under particular circumstances government land purchases would have to be based on a fair assessment of market value.
Since 1994, there have been three types of land reform. First there has been a land restitution procedure, through which people who believe they themselves or their families lost land through forced expropriation after 1910 can claim restitution. In the case of a favourable finding by the Land Court, the government either buys back the expropriated land or provides equivalent compensation which might be in the form of a cash payment. Then, secondly, the government has sponsored its own settlement schemes, buying up commercial farmland and allocating it since 2013 on a leasehold basis to aspirant farmers who qualify for joining such schemes in various ways. Thirdly the government has tried to introduce reforms of customary tenure in the homelands so that households can have secure access to land of a kind that can enable them to raise loans to invest upgraded farming. To win support for such plans from traditional leaders, often patriarchally predisposed, conceptions of rights have been codified that discriminate against women.
We don’t know precisely how much land that was white-owned in 1994 has since been transferred to black South Africans and retained by them. Estimates begin at a minimum of 9 per cent of privately owned commercial land though certain academic assessments calculate a much higher proportion. Unfortunately, the management of land registry records has deteriorated, so the numbers of farms that may have been sold back to white South Africans is unknown. There is no record that can tell us how much land had been purchased by black South Africans independently of any government supported schemes. Official figures supplied by the Minister of Rural Development in 2017 indicated that the government had since 1994 bought 11,000 million hectares – about 12 per cent of total farmland. Beneficiaries of restitution or redistribution totalled around 500,000 in 2013. By that year the government had spent R12 billion on buying nearly 5,000 farms for redistribution to nearly 250,000 households and a further R16 billion on land claims, a sum that included cash compensations but which created roughly comparable numbers of new landholders to those who benefitted from redistribution. With respect to the latter group, in 2013 the government stopped transferring land to beneficiaries, leasing it instead, and requiring business plans and commercial partnerships from the new leaseholders, arguably slowing down implementation and making the scheme more restrictive. In February 2018 the Department of Rural Development and Land Reform released figures that suggested that Africans owned a mere 4 per cent of individually owned farms – which constitute 37 per cent of total farmland but that figure does not indicate the extent of state sponsored land transfers which have been characteristically undertaken through leases to groups.
What do we know about the impact of South African land reform up to now? We know that much of the land transferred has been of poor quality, located on the borders of or close to the historic homeland boundaries. Land purchases next to game reserves have been one quite frequently used option, with the farms handed over to the historically dispossessed community to manage as an extension to the neighbouring wildlife reserve. We know from research conducted at the University of the Western Cape that new settlement schemes have been quite successful in relieving poverty among beneficiaries but they have yet to create a large group of assertive market-oriented “yeoman” farmers of the kind that was beginning to emerge in Zimbabwe in the 1990s. A Dutch- sponsored study in the Waterberg area indicates that redistribution schemes are more likely to generate agriculturally-based livelihoods than restitution schemes, a reflection partly of the “business plans” government has required from redistribution beneficiaries under more recent schemes. South African beneficiaries of land reform have not had the same quality of supportive extension services available in Kenya in the 1960s and Zimbabwe in the 1980s and 1990s. In South Africa local announcements of projected land settlements prompt rises in land prices, putting further strain on land reform budgets. Meanwhile the commercial sector remains efficient in terms of business criteria, and increasingly important as a generator of export income. But agricultural employment on the mainly white-owned commercial farms has been decreasing steadily, down by a million since 1994 to around 600,000 today. Reductions in the workforce have been prompted partly by wage rises in compliance with minimum wage requirements as well as evictions by farmers anxious that their workers may claim occupancy rights. Changing patterns of land usage also explain declining farm employment.
What explains the present upsurge in demand for land reform in South Africa?
It is stoked by the social inequality that has intensified since 1994. Some of demand is urban – in effect a demand for housing. It is fed by widespread perceptions of social injustice which for many people remain heavily racialised. When land occupations or land invasions have happened, the authorities have been unable or unwilling to check them effectively. But there is rural demand as well, not least because of losses of commercial farm jobs. Since 1994, the ANC has become more firmly rooted in rural politics. It has also become provincialized, because of its own organisational adaption to the requirements of a federal political dispensation. Because the provinces were partly delineated along old ethnic boundaries, the ANC has become partially ethnicised. Following this development land-based or territorial notions of political identity have become more important for its supporters. The demand for land reform without compensation is a moral or an emotive demand. It reflects widespread perceptions that present society is unfair, and that the historical beneficiaries of racial justice remain privileged. Put another way, it is about punishment and retribution, directed at white South Africans in general, not just the 35,000 or so white farmers. It is hardly likely to be assuaged with a few token confiscations of land held by criminal, absentee or untraceable owners, as certain ANC leaders evidently hope.
The Zimbabwean experience shows land confiscations would be a high risk strategy, though. And as noted above, Zimbabwean officials today are contemplating payment of compensation. Even if it was done in a tokenistic fashion so as to make a symbolic point, confiscation without compensation could damage confidence across the economy, not just in agriculture, among investors in the urban economy who can also be affected by land claims and land occupations. Another obvious lesson from the Zimbabwean and Kenyan experiences is that for effective land reform more is needed than land transfers, whether paid for or not. New settlers need a wide range of services that probably in the longer term could cost much more than the commercial price of the land.
So, what should the South African government be doing with respect to land?
There is no question that the state has to be seen to be acting much more assertively than appears to be the case at present.
A sensible first step would be to confirm what has been achieved so far. Re-organising the land registry should be a priority.
More extensive land reform does not require constitutional amendments, really. This was acknowledged at the ANC’s Land Summit this year on May 19-20 which resolved in favour of expropriation without compensation under the fairly restrictive terms permitted by Section 25 of the Constitution. Since then, though, the government has decided in favour of a constitutional amendment that will spell out more precisely under what circumstances land can be confiscated without compensation. Speaking to journalists about these plans in August President Ramaphosa implied that the amendment will simply clarify existing powers but obviously his administration will be under political pressure to broaden them. It may well be that senior ANC politicians hope that even a minor change to the constitution will deflect political pressure from the left but an effective land reform programme will probably require more that whatever land can be obtained through the application of even a broadened mandate under section 25.
The government will probably still need to argue the case for the merits of paying compensation, in language that is accessible, though it should also act more forcibly to determine what it should pay. Even if they were technically legal, widespread expropriations without compensation would destroy any prospect of investor confidence, internally and externally. With or without compensation, extensive land reform can only be expensive and it will require a major administrative effort, well beyond the scope of the present Department of Rural Development and Land Reform. There will need to be a major expansion and refinement of agricultural extension services so that they can meet new needs. A land reform strategy that is calculated to meet land hunger-generated demands will require breaking up very large farms into small units that can be managed by households.
“Deconcentration” of a significant part of the commercial sector if it is to be done in a way that protects food security and maintains export receipts will require massive investment In rural infrastructure and services as well as very extensive training for hundreds of thousands if not millions of new farmers. It cannot be undertaken overnight. If government is to take land reform seriously it will require a much larger share of the budget. Today total expenditure to date, that is in 24 years, on land reform is roughly equal to what in one year the government spends on housing. Obviously other areas of government expenditure will have to be reduced; there is very limited scope for raising additional revenue.
More effective and better use of state land in peri-urban areas might help to meet fresh demands for housing, though housing policy itself requires a major shake-up. More RDP-type settlements won’t meet the demand for inner city housing. In Gauteng alone, estimates of the housing backlog have reached 600,000 dwellings. Addressing the needs of people involved in urban land occupations such as the Marikana and eNkanini settlements in Durban’s Cato Manor requires imaginative and careful town planning rather than the present regime of armed evictions. By international standards, South African cities are extensive, not densely settled; creating additional homes need not involve expropriations. Urban land transfers may be productive in other ways too. Recent research on peri-urban land occupations indicates significant numbers of occupiers making livelihoods or at least producing foodstuffs through small scale farming.
Rural land demand can be checked by better protection for farm labour. At present slow processing of land restitution claims in the Land Court has led to a climate of uncertainty in which farmers worried about claims on the basis of occupation rights and as labour tenants are evicting workers and confiscating their livestock. At the present rate of progress, land restitution will not be complete for at least another decade.
What agricultural and livelihood gains might result from communal land tenure reform needs more research. Former homelands accommodate four million or so farmers, apparently, though how many of these actually succeed in making secure livelihoods from agriculture is uncertain. We know that land in the former homelands is underutilised. Many rural households are women headed – in certain areas most households – and they are often discriminated against in customary land provision. In any case lack of formal and legally recognised forms of title or use rights is a barrier to raising loans for investment in reclaiming neglected ex-homeland land. More investment in small loans facilities for small farmers is a key need. The Venda Building Society may have been acting irregularly in taking in short term deposits from municipalities but the loans it was making were to people whom most banks view as ineligible for borrowing, but who need credit desperately. In areas of communal tenure there is also the problem of traditional leaders or tribal authorities selling or leasing land to developers from outside the communities that share customary usage rights. The Ingonyama Trust Board in KwaZulu-Natal earns R90 million a year from corporate (that is, big business) leaseholders.
Within the existing framework of laws and regulations, then, much more could be achieved if the government wanted to really demonstrate its commitment to land reform. But it will require hard choices that will not always be popular and such choices will need skilful leadership to justify and defend.
Tom Lodge is a Professor of Peace and Conflict Studies in the Department of Politics and Public Administration at the University of Limerick in Ireland. Before this, Lodge was a Professor of Political Studies at the University of Witwatersrand. He has also held positions at the University of York, the Social Science Research Council in New York, and the Electoral Institute of South Africa.
This article first appeared in Focus 83, the journal of the Helen Suzman Foundation.
John Harbeson, “Land reform and Politics in Kenya, 1954-1970”, Journal of Modern African Studies, 9, 2, 1971, 231-271.
Karuti Kanyinga, “The legacy of the white highlands: land rights, ethnicity and the post-2007 election violence in Kenya”, Journal of Contemporary African Studies, 27, 3, 2009, pp. 325-344.
Charles Laurie, The Land Reform Deception: Political Opportunism in Zimbabwe’s Land Seizure Era, Oxford University Press, Oxford, 2016.
Ian Scoones et al, Zimbabwe’s Land Reform: Myths and Realities, Jacana, Cape Town, 2010.
Special Issue on Fast Track Land Reform In Zimbabwe, Journal of Peasant Studies, 38, 5, 2011,
On South Africa:
Ruth Hall and Thembeke Kepe, “Elite capture and state neglect: new evidence from South Africa’s land reform”, Review of African Political Economy, 44, 2017, 151, pp. 122-130.
R Jacobs, “An urban proletariat with peasant characteristics”, Journal of Peasant Studies, 2017.
Avhafunani Netshile, Simon Oosting, Edzisani Raidami, Majela Mashiloane and Imke de Boer, “Land Reform in South Africa: Beneficiary participation and impact upon land use in the Waterberg district”, NJAS – Wageningen Journal of Life Sciences, 83, 2017, 57-66.