Using the land the poor have

William Beinart and Peter Delius says it is striking how little land in former homelands is being used for crops


No 5, October 2018

Smallholders and land reform: A realistic perspective


The current heated debate on land reform is fuelled by bold pronouncements about how white-owned land should be secured for African people. But it is marked by an absence of detailed policy discussion about what should be done with the land that is redistributed. There are some suggestions that African smallholders – families that occupy a residential plot with garden, a field of about one or two hectares and have access to communal grazing – should be the main beneficiaries. In this scenario, they would play a central part in agricultural production and boosting national food security on transferred land. These ideas are partly underpinned by an assumption that the black rural population contains a substantial dormant peasantry waiting to be revived.

Smallholders occupy a considerable area of agricultural land in South Africa, perhaps 23-25 per cent (over 20 million ha), and we agree that policy should facilitate their livelihoods by expanding agricultural production. Where possible, they should be assisted to intensify production and market their output. But at present they produce relatively little and evidence suggests that they are, as a whole, withdrawing from farming.

Before transferring large amounts of new land, and imagining that smallholders will make a major contribution to national agriculture, we should analyse historical trajectories as well as the current levels of production. Policies based on profoundly inaccurate data and assumptions are doomed to fail and may do grave damage to both the agricultural sector and the national economy.

Moreover, as a policy priority, it is vital to identify where the major demand for land lies. Continuing urbanisation, and the incidence of shack settlements, suggests that this is for securely held sites in the urban and peri-urban areas, including sites with gardens.

The absence of a South African peasantry

In the late nineteenth century and early decades of the twentieth century, South African smallholders intensified their production with new crops such as maize, some of them producing for new markets. This historical success together with the more recent experiences of land reform in societies such as the ‘Asian Tigers’ has provided a model for some South African policy makers. But South Africa differs from the latter examples in that conquest, land dispossession, taxation and other pressures set in motion the tragic destruction of black smallholder production in the twentieth century.

Instead, it was the pervasive and increasingly coercive system of migrant labour that left the most indelible mark on most African rural communities and fundamentally restructured their economic and social systems. These pressures ensured that the development of black farming was stunted and the vast majority of black families on the land were unable to achieve even basic subsistence. Their position deteriorated from the 1950s as the intergenerational transfer of agricultural skills and resources withered. Wages from migrant workers and grants from government became central to the livelihoods of the great majority of rural families.

During the apartheid era (1948-1994) there was a major roll out of primary and secondary education in the Bantustans; this has expanded further since democracy in 1994. As a result, the labour of youths and children has largely been withdrawn from crop production and especially for herding livestock. Even more importantly young people have developed different expectations and wider horizons. Risky and poorly paid smallholder farming features very low in their imagined futures. Herds have stagnated and growing amounts of land in the former homelands have been left fallow as a lack of capital, labour and prospects sapped the desire of many to risk their hopes, time and resources on farming.

Is land shortage the real problem?

Government statistics, as well as a land audit by AgriSA, suggest that the pace of land transfers from white to black landholders has increased significantly in recent years. The former homelands, in which land is very largely held by African people, represented about 14-15 per cent of the land (17-18 million ha) in 1994. It is not correct to say that this is the worst land in South Africa. Much of it is in the higher rainfall areas in the eastern half of the country.

Government reported in 2017 that it had redistributed 9 per cent of the agricultural land, about 8 million ha out of about 90 million ha, in both the restitution and the redistribution programmes. Additional cash compensation was paid under the restitution programme. Some sources suggested that another 4 million ha is owned by the state, although not all of it is available for land reform.1

Since 2011, the government has generally not transferred ownership of the land purchased for redistribution to the beneficiaries. Instead, it leases land to them. In addition, an unknown amount has been privately purchased – perhaps 2-3 per cent (2-3 million ha) or possibly more. Together this amounts to around 26-27 per cent (25 million ha) in black hands with additional redistributed land owned by the state. Whites do not own the rest as a good deal is urban or state land (much of it reserved for environmental purposes such as national and provincial parks, forests, and watersheds).

In 2017 AgriSA employed researchers to examine overall landholdings. Although this study has been criticised, they arrive at a similar amount: 27 per cent of land by area is now held by ‘previously disadvantaged individuals’. They calculated that this represented 29 per cent of land by value. Because most of this land lies in the wetter, eastern half of South Africa it has relatively high potential both for cultivation and livestock. According to the AgriSA audit, over 70 per cent of land in KwaZulu-Natal is now occupied by black people. There is a good deal of newly acquired land available for smallholders.

Fallow lands and limited smallholder production

If you travel through the former homelands, it is striking how little land is being used for arable agriculture. This applies not only to drier areas   in the North-West Province, and former Lebowa and Gazankulu, but to the well-watered Transkei. Some people have ploughed gardens next to their homes but most of this potentially rich land is used for free-range livestock. Detailed research and the limited available government statistics confirm this picture in most areas occupied by smallholders.

This is a longer-term trend. Commenting on his fieldwork in Sekhukhuneland (Lebowa) in the late 1980s, Delius noted that ‘every year, good or bad, substantial areas of land lay fallow’.2 Ntsebeza argued in 1995 that most rural people were no longer interested in farming because they viewed it as ‘hard and risky’ and would prefer to be ‘fully proletarianised’, earning wages.3 Detailed case studies from both Ciskeian and Transkeian districts suggested that the proportion of household income from agriculture was very limited. In a Lusikisiki village, farmers produced less than 25 per cent of their maize requirements in the early twenty-first century..4

African smallholders previously cultivated both gardens next to their homesteads and bigger fields that were usually further away. Since about 1990, studies have found that fields were being abandoned and most people were concentrating on the smaller garden plots next to homesteads.5 Manona recorded in 2005 that 12 per cent of fields were being used in villages that he surveyed in Flagstaff district.6 Mtero found in Matatiele district that about 40 per cent of households had arable fields but very few used them. Some were being invaded by wattle. Overall, around 60 per cent of households cultivated gardens. By 2008 an estimated ‘71 per cent of those with access to land did not use it for production’.7 In KwaZulu-Natal, a ‘high proportion of fields’ also remained uncultivated.8

In a recent study of 120 households in two villages in the former Ciskei in 2016, about 10 per cent of households cultivated fields and 50 per centcultivated gardens with an average size of about 400 sq. metres.9

Only two households produced on a small commercial scale. The villages as a whole produced about 15 per cent of their maize needs and 25 per cent of vegetable needs. In 2017 in the highly fertile Zoutpansberg area the vast majority of fields were unused despite excellent rains.10

The existing evidence suggests over 80 per cent of former fields are no longer used, amounting to hundreds of thousands of hectares – perhaps close to a million. This is a huge amount of land, a third of the amount that commercial farmers use for maize. If farmed with relatively effective inputs, it could potentially produce a great deal more maize, and other crops, leaving sufficient land for grazing as well. This would transform the economy of the 25 million ha currently occupied by African landholders.

There is no agricultural census or systematic government survey of smallholder production. The Department of Agriculture, Forestry and Fisheries (DAFF) produces separate figures for maize production in the ‘non commercial agricultural sector’.11 These are not a census or count but may be a reasonable estimate based on satellite images and local surveys. In 2017 smallholder (non-commercial farmer) production was estimated at 731,000 tons of maize on 366,650 ha (2 tons/ha) compared with 17 million tons on 2,628,600 ha (6.4 tons/ha) for commercial famers. The estimate for 2018 (August 2018) is 594,000 tons

on 314 835 ha (1.9 tons/ha), compared with 13,207,000 tons on 2.3 million ha (5.7 tons/ha) in the commercial sector. According to these figures, smallholder farmers planted about 500,000 ha in the five years from 2000-4, but in 2014-18 this declined to under 400,000 ha. Smallholders are producing less than 5 per cent of the maize in South Africa; they are probably producing a higher percentage of the vegetables, but even less of many other crops and fruits.

Smallholders and livestock production

Livestock are still seen as valuable investments. With reference to the former homeland areas, peak livestock ownership in African hands was in the 1930s. Numbers have dropped gradually since then. As population has increased so the number of households without livestock has increased greatly. Perhaps half of households in rural villages own livestock of some kind, with the highest proportion keeping chickens. It is likely that 5 per cent of rural households have 10 or more cattle. Some studies show an increasing concentration of livestock ownership in communal areas.

Nevertheless, the number of cattle in black ownership is increasing overall because transferred land is largely used for livestock. While the costs of keeping livestock (including veterinary medicines) are not insignificant, they are lower than those for cultivation. In communal areas, grazing is free and fodder is not generally used; commonage is still the majority of land in African occupied rural areas. Prices for slaughter animals are good (more than R1,000 for goats and R6-7,000 for cattle); the commercial and cultural value of livestock is generally perceived as higher than arable land and the risks lower.

Alcock argues that Msinga in KwaZulu-Natal is getting hotter and drier with less grass and more bush.12 (Bush encroachment is more general and related both to changing land use and, Bond and Midgley argue, to increased carbon dioxide in the atmosphere.13) As a result, Alcock believes that the old maize and cattle African smallholder economy, which coexisted alongside migrant labour is being displaced by a goat and chicken economy that requires less investment – reflecting social and environmental realities. Goats are hardier, and better adapted to drier climates. They require less labour in that they ‘self-herd’ and generally return to their kraals in the evening.

This analysis is interesting as a pointer to one possible trajectory for agriculture. But the scenario of even lower investment is of concern. The question also arises: what do unherded goats preclude? They may offer significant value, and they can of course be farmed alongside cattle and chickens. But they undermine the capacity to sustain other forms of production, notably gardens and arable production in unfenced fields. In fact, it is unlikely that this picture is accurate on a national scale: there are still probably more cattle in African hands than goats.14

Kenyon emphasizes the importance of the livestock economy in the African areas of the Eastern Cape including transferred land.15 Land gained by subsidised transfer has largely been put under livestock. He estimates a take-off rate of 14 per cent per year when slaughter for the ‘ritual economy’ is included. Like Alcock he emphasizes the hidden value of livestock in the smallholder economy – slaughtered cattle, little of it marketed, could have a value of R2 billion a year. But African owned livestock do not meet the demand for meat even in the rural areas: substantial numbers of goats and cattle are purchased from commercial farmers for slaughter. The same applies to chickens, even though many smallholders keep free range birds at their homesteads for eggs, meat and ceremonial purposes. In addition to supermarket sales, egg- laying chickens from commercial poultry enterprises are sold into informal markets after their laying life is ended.16 Overall, there is still considerable scope for production of all types of livestock in order to meet local demand for meat.

How many smallholders are there and what do they do?

There may be 3-4 million households in the former homelands and transferred areas, of which roughly 2 million might be considered as smallholders in some sense, at least with gardens and chickens.17 About 10 per cent or 200,000 make some attempt to sustain production beyond gardens, with fields of about 1-2 ha, and also keep some livestock.18 There is also evidence of investment into agriculture by those with capital or salaries, either into livestock in their home villages, or into the purchase of private land.

Beinart, Mnwana and Wotshela analysed private land purchase in the Isidenge valley in Stutterheim.19 African purchasers held 23 farms, with an average size of 42 ha, out of 33 farms in the area surveyed. Black landowners purchased 17 farms since 1994: at that date they held about 15 per cent of the land in this area, a figure which had increased by 2018 to about 55 per cent. These private owners, however, are potentially commercial farmers rather than smallholders and a couple have invested heavily into mechanised vegetable production – the largest was planting 60 ha with 30 ha under irrigation.

A strong strand in the literature insists that smallholder production tends to be underestimated and that communities also glean a significant proportion of their livelihoods from the commonages: grazing resources, medicines, wild fruit, wood, clay and thatch.20 Yet it seems clear that the great bulk of food in the rural areas is purchased. There are huge opportunities for African growers to market locally in the rural areas and small towns. Even in the rural areas, households produce only about 20 per cent of their food on average, and there is scope for a wide range of other produce as well as meat, timber and charcoal from invasive alien trees such as black wattle (Acacia mearnsii.)

Impediments to expanding production

It is clear that land shortage is not a key impediment to expanding production. Other impediments are far more important. Recent interviews indicate that it is possible to get access to fields of about 2 ha in some of the communal rural areas.21 However, bigger fields are often located away from homesteads and are difficult to control or fence. Often this land is used for grazing after the harvest season (usually April to June) discouraging effective investment that requires control of land throughout the year. Access to sufficient water is a critical issue and investment into water is problematic without such control and ideally private ownership. Moreover, it is difficult to expand landholdings beyond this area in communal systems.

Agricultural technology is a bottleneck. Few smallholders have access to a team of oxen for ploughing, and tractors are expensive to buy, maintain or even hire. It is not possible to meet the costs of investing in a tractor on the proceeds of 2 ha of maize and vegetables. The use of fields sometimes depends on a tractor owner who has been able to purchase the machine using other income and who then ploughs for a number of landholders.

Land acquired through land reform is usually held within Communal Property Associations (CPAs). These have some of the same benefits as communal areas, providing access to land and grazing for a large number of families. But they also lead to the same disadvantages for individual families in that they are managed collectively. Collective investment in a farm is difficult because families generally wish to benefit financially from their new access to land, rather than to contribute to the costs of infrastructure such as fencing and irrigation. In CPAs, as well as on communal land, especially in Kwa-Zulu Natal, chiefs have tended to take control22

Interviewees in communal areas and on CPAs speak of the difficulty of protecting crops against livestock and theft. Livestock is no longer carefully herded because young boys are at school and owners have to look after the animals themselves or hire herders. Protecting crops against livestock is a major headache and there is little fencing except sometimes around gardens. In some areas birds, as well as wild animals, such as baboons, monkeys and bush pigs can be a problem.

Trends away from small field farming

In their detailed discussion of two villages in the former Ciskei, Hebinck and Lent note that smallholders are concerned about the risks involved in crop production.23 Most people perceive the costs of seed, fertiliser, ploughing and equipment to be prohibitive. Manona concludes that the costs of engaging arable production are too high, given the risks, for the majority of smallholders and this is a significant reason for withdrawal from arable fields.24 This illustrates what appears to be a more general trend.

In earlier decades, rural families were generally large enough to meet agricultural labour demands when men migrated, as women did much of the cultivation. In the last fifty years, however, African rural families have declined in size, many women also migrate to find work and homestead heads have gradually lost control of family labour. Women are hard-pressed to meet the demands of childcare and household chores; these pressures have been exacerbated by poverty and the effects of HIV/AIDS. Patriarchy has not entirely died, but withdrawal of women’s labour was both a symptom of pressure on poorer households and a choice by women with more agency. Child labour, which was also very significant for rural production, has moved to the margins because of the demands of education. While communal consumption at weekend events is still ubiquitous, communal work parties, amalima in isiXhosa-speaking areas or letsema in Sepedi, are infrequent.

People see smallholder farming as hard work for limited returns. Many associate it with poverty and a backward way of living which is no longer compatible with their ideas of a modern and viable way of life. Family labour is less available but the transition to wage labour has not generally been made. Low wage levels are accepted only by the destitute or by the rightless, such as illegal immigrants. In some areas, poorer men are employed as herders and this seems to be more general in Lesotho where there is no grants system.25 In South Africa most potential employees have sufficient alternative sources of income to avoid such low-wage employment.

Judging by the rapid improvement in housing in many villages, money is still invested in the rural areas – not into arable production, but into homes, education, consumer goods and, to some degree, livestock. It appears that there is enough cash coming into the rural villages to enable the majority of households to purchase grain and proteins. However, there are still reports of widespread under-nutrition and stunting in children in the rural areas.26

Pockets of dynamism

Although the overall picture of smallholder farming suggests a decline, there are exceptions and it is important to report these ‘pockets of dynamism’ because they assist in analysing where investment and assistance may be most productive. Some   of the main sites are gravity fed irrigation schemes, mostly built and maintained with subsidies in homeland era. Cousins details a successful irrigation project at Tugela Ferry which supplies a thriving local market as well as vegetables for urban markets transported by bakkies.27 70 per cent of the commercial farmers at the scheme are women with an average age of 58.

Manyelo et al describe an irrigation scheme at Dzindi near Thohoyandou, Limpopo, consisting of 106 plots on 140 ha.28 Growers depend on selling to local traders. They specialise in locally popular foods such as green maize, Chinese cabbage, cabbage, nightshade and pumpkins. Fresh maize is cooked for street sales and fetches three times the price of maize for grain; pumpkins are grown for leaves, flowers and young fruit, which are cooked together for relish. On average, growers generate about 30 per cent of their household income and 15 (out of a total of 97) are larger scale market-oriented producers. Overall it is clear that vegetable and green maize production is more profitable than grain production on irrigated smallholdings.

The National Development Plan ambitiously estimates that 500,000 ha could be added to the existing 1,300,000 ha of irrigated land by 2030. This is very unlikely to be achieved. However, there is roughly 110,000 ha of land in irrigation schemes in the former homelands, perhaps half of which are not operating effectively.29 Efforts have been made to regenerate some of the schemes, with limited success. Some progress has been made in identifying the longer-term problems on these sites such as management difficulties and incapacity or unwillingness by landholders to pay irrigation levies. New projects will be even more expensive than resurrecting existing schemes. Finding routes forward is a priority and there is certainly scope for increased water capture along the eastern seaboard of the country, where the bulk of African landholdings are located, whether through tanks, at the household level, or dams. 

Joint ventures or partnerships with the private sector have been an important innovation. These are now very diverse – with different variants in sugar, forestry, dairying and wool. The sugar outgrower schemes, started by the industry, and supported by the KwaZulu homeland government, in the 1970s, has probably been the most successful over the long term. The numbers involved have, however, reduced from over 50,000 at its peak in about 2000 (producing 14 per cent of sugar nationally) to 22,000 in 2012-3 as a result of reductions in support and credit.30 Forestry currently seems most promising. Only 4 per cent of forestry land is currently held by 30-40,000 smallholders whose trees largely go into charcoal and wood chip markets. It is estimated that outgrower production could expand by 140,000 ha.31 Richards Bay has a thriving woodchip facility for export to Japan and China. In Umzimkulu a Mondi-sponsored scheme works on 1,320 ha in a communal area creating 100 jobs.

A different system is sponsored by the National Wool Growers Association in communal and transferred areas where extension officers provide training and advice on improving wool output amongst smaller-scale sheep-owners. They claim on their website that this has resulted in ‘communal farmers increasing income from wool through the formal auction system 88-fold to more than R233 million/annum since 1997’.32 Wool is sold through formal channels and the value of wool from African smallholders and communal areas has gradually moved closer to the average value realized by commercial growers.

Land restitution is resulting in a variety of leasebacks where successful claimants gain ownership of the land but lease it back to established or new commercial operators. The Department of Rural Development and Land Reform is now keen on such arrangements especially where restituted land includes valuable resources such as citrus plantations. In yet another model, the local traditional council in Willowvale, together with BEE entrepreneurs and a specialist company, is planting Macadamia nut trees on a large scale on existing customary land.

In all of these cases, organised corporate agencies, private companies or groups of commercial farmers – sometimes funded partly by the state - provide start- up capital, inputs, knowledge, extension and links to formal marketing chains which otherwise are very difficult for smallholders to command. In Limpopo, the Spar Rural Hub is providing access directly to some supermarkets. In the medium term they diffuse up-to-date knowledge to landholders who may not have easy access to expertise and the internet.

These partnerships have also attracted criticism because they can involve contracts that are unfavourable, and priorities shaped by producing for external markets and corporations rather than local household food security. However, they do not necessarily preclude smallholders engaging in both market-oriented and household food production. When CPAs are involved tensions have arisen because they wish to redistribute a greater percentage of profits while strategic partners wanted to invest in future production. Money can also be diverted by committees who control the CPA.

In sugar, forestry and wool for example, there may now be as many as 80,000 smaller black farmers who connect with major companies both for production and marketing and this is probably the fastest growing area of smallholder production.

Implications for policy and land reform

A central implication of our approach is that there is great scope for innovative thinking and projects on the land already occupied by black smallholders in the former homeland areas or on recently transferred land. This may now constitute around 25 per cent of agricultural land much of it in areas with over 500mm rainfall. It is essential that the current landholders receive upgraded, secure tenure in these areas. Unlocking the productive potential of these areas should be a policy priority.

Equally, the state should proceed cautiously in respect to calls to transfer large commercial farms however they are acquired. Production on these units expanded during the ten years 2006-2015 by 2.5 per cent a year, faster than overall economic growth. Food security and exports rest primarily on the success of this sector and they are not mutually exclusive. South Africa is the second largest citrus exporter globally but the 33 per cent of produce sold domestically provides a cheap and healthy fruit as well as juice. Commercial farms do not need subsidy but they do need certainty to maintain investment.

While the focus of this article has been on the potential for land-based production, we are not suggesting that agricultural innovation should be the only or even the primary route to rural development and improved livelihoods. Especially in the northern provinces, mining – largely controlled by external enterprises – is transforming some rural areas although often in highly conflictual relationships with their neighbours. More effective models need to be developed to make these relationships more symbiotic and to provide greater benefits to the landholders rather than intermediaries.

Perhaps the single most important element of state transfers should be educational investment to ensure that rural youths are in a better position to innovate in the rural areas and to compete on a national job market. Tourism also provides great potential but at present tends to follow existing lines of infrastructure. The former homelands include many scenically attractive places and some national or provincial parks. But at present the political will and resources are not present to develop the wider and deeper heritage attractions of many areas of the country. Part of the problem is the absence   of demarcated routes, trained guides, relevant literature and websites. Many of the small towns and villages lack attractive places to stay, even for adventurous self-guided tourists.

Nevertheless, there remains a strong argument to place a high priority on agriculture in rural development. The land currently held by smallholders should be made more secure. Policy should focus on existing landholdings in communal areas as well as land that has already been transferred. It is in this area that South African agriculture is underperforming. Comparisons do not provide solutions but food for thought. Malawi, about the same size as the former homelands and with the same population, usually produces 3-4 million tons of maize a year worth roughly   R9 billion or perhaps six time that produced by South African smallholders. This has been achieved partly because of a less diversified economy, with lower levels of labour migration over the last few decades, and far fewer livelihood options for rural families. However, there has also been an innovative input subsidy programme that has helped to trigger production.

There is of course a major argument for diversification towards a range of higher value vegetable crops. Enhanced agricultural production in the areas of South Africa already occupied by smallholders would generate considerable, widely distributed income and provide the basis for rural businesses in spheres such as transport, trade, solar energy generation and services. The outgrower and irrigation schemes provide routes forward but there is scope for a wider range of pilot schemes, neglected since the democratic transition.

The challenges of land reform

Restitution was a valuable response to the injustices of the apartheid era. But it was intended as a limited and short-term process. Despite some of the rhetoric that has been used, restitution has not been sensitive to issues of production. It is a matter of urgency that the huge number of outstanding claims should be resolved so that the focus of land reform can move to redistribution programmes that prioritise production. Cash payments, rather than expropriation of effectively farmed land, are often a more suitable route for resolving restitution claims.

Successful redistribution that facilitates production will require far more support than has been provided so far. This should be a gradual process. An average of roughly 0.4 per cent of agricultural land has been transferred annually since 1994. If production is to be maintained and increased, it would be a huge achievement to increase this to 0.5 per cent a year for the next ten years, totalling over 4 million ha of land. This figure should include changes in ownership as a result of restitution and other redistribution programmes. The state does not at present have the resources to provide adequate extension services and agricultural education for those who already hold land and this will require major investments in training and staff. Equally,   it is vital that any new dispersed agricultural settlements are well-serviced with schools, clinics and housing. Again, this will require long term planning and considerable additional expense. It is vital that private sector resources are mobilised as well. The issue is not so much the area of land involved but the provision of capital, inputs and skills to bring it into production and to enhance rural livelihoods rather than spread rural poverty.

Large numbers of people continue to move to the cities and the government should follow the people and prioritise provision of residential sites and housing in secure forms of private tenure. In peri- urban areas and in smaller towns, some of which still have a good deal of municipally owned land, there is scope for larger plots that allow for urban gardens.

A fast track process to land redistribution is likely to be a disaster. Land reform and agricultural policy should be discussed together with the wider aims of food security, cheap food, expanding production, exports, environmental sustainability, deracialisation of corporate enterprises, and the creation of more employment. This requires supporting the existing commercial, highly capitalised farming economy which can also contribute to the growth of more successful smallholder agriculture.

William Beinart is emeritus professor, Oxford University.

Peter Delius is emeritus professor, Department of History, University of the Witwatersrand.

VIEWPOINTS is a series of reports on issues critical to South Africa’s development.

CDE has drawn on the expertise of South Africans in a range of organisations to encourage informed debate on these topics.


1 Wandile Sihlobo and Tinashe Kapuya, ‘Land Policies try to solve imaginary issues at expense of real problems’, Business Day, 6 June 2017.

2 P. Delius, A Lion Amongst the Cattle: Reconstruction and Resistance in the Northern Transvaal (Oxford, James Currey, 1996), p. 222; for a detailed survey in this area at the time, see Rupert Baber, ‘The structure of livelihoods in South Africa’s Bantustans: evidence from two settlements in Northern Province’, D.Phil. University of Oxford (1998).

3 L. Ntsebeza, ‘Land Demand , Availability and Use: Preliminary Reflections on the Possibilities and Limits of the Land Reform Program’, Comparative Studies of South Asia, Africa and the Middle East, 15, 1 (1995), pp.72–5.

4 F. Hajdu, Local Worlds: Rural Livelihood Strategies in Eastern Cape, South Africa (Linkoping, Linkoping University, 2006), pp. 162–3.

5 M. Andrew, ‘A Geographical Study of Agricultural Change since the 1930s in Shixini Location, Guyana (Willowvale) District, Transkei’, MA dissertation, Rhodes University, 1992; M. Andrew and R. Fox, ‘“Undercultivation” and Intensification in the Transkei: A Case Study of Historical Changes in the Use of Arable Land in Nompa, Shixini’, Development Southern Africa, 21, 4 (2004), pp. 687–706.

6 S.S. Manona, ‘Smallholder Agriculture as Local Economic Development (Led) Strategy In Rural South Africa: Exploring Prospects in Pondoland, Eastern Cape’, MPhil thesis, Land and Agrarian Studies, University of the Western Cape, 2005, p. 85.

7 Z. Mfono et al., ‘Directional Trends in Selected Aspects of Rural Poverty in the Eastern Cape Province: Changes during the First Ten Years of South African Democracy’, Development Southern Africa, 25, 1 (2008), pp. 37–60.

8 L. Fenwick and M. Lyne, ‘The Relative Importance of Liquidity and Other Constraints Inhibiting the Growth of Small-Scale Farming in KwaZulu-Natal’, Development Southern Africa, 16, 1 (1999).

9 H J Brooks, ‘The Role of Field and Garden Cultivation for Food Security under a Changing Climate: the Case of Fairbairn and Ntloko Villages, Eastern Cape’, MSc, Rhodes University, 2017.

10 Delius, personal observation.

11 Website: DAFF Crop Estimates.

12 Rauri Alcock, paper to PLAAS, University of the Western Cape, conference on Rethinking Agriculture, Constraints and Opportunities, 11-13th October 2016 13 William J. Bond and Guy F. Midgley, ‘Carbon dioxide and the uneasy interactions of trees and savannah grasses’. Philosophical Transactions of the Royal Society B (2012) 367, 601–612.

14 William Beinart and Karen Brown African Local Knowledge and Livestock Health: Diseases and Treatments in South Africa (Wits University Press, 2013).

15 Mike Kenyon, paper to PLAAS, University of the Western Cape, conference on Rethinking Agriculture, Constraints and Opportunities, 11-13th October 2016.

16 Thomas Cousins and Michelle Pentecost, ‘The Second Shift: Informal Economies and Avian Influenza in South Africa’ Current Anthropology July 2018, website.

17 Estimates of smallholders vary from 1.4 million in the quarterly labour force survey to 2.6 million in the general household surveys because of different criteria. See also Michael Aliber, ‘Exploring Statistics ;South Africa’s national household surveys as sources of information about household-level food security’, Agrekon, 48, 4 (2009), 384-409.

18 Ben Cousins has calculated the figure of 200,000 and offered it as an estimate in various presentations, including at the CDE Roundtable, The Growth Agenda: Agriculture, land reform and jobs – can South Africa make this work? June 2018.

19 William Beinart, Sonwabile Mnwana and Luvuyo Wotshela, ‘Land Reform, Rural Inequality and Agrarian Change: The Case of Isidenge, Stutterheim, Eastern Cape’ paper for SWOP conference on Inequality, University of the Witwatersrand, September 2018.

20 Summarised in W. Beinart, ‘Transkeian Smallholders and Agrarian Reform’, Journal of Contemporary African Studies, 11, 2 (1992), pp. 178–99; Thembela Kepe, ‘Beyond the numbers: Understanding the value of vegetation to rural livelihoods in Africa’, Geoforum. 39, 2 (2008), 958-968.

21 William Beinart, Peter Delius and Michelle Hay, Rights to Land: a Guide to Tenure Upgrading and Restitution in South Africa (Jacana, Johannesburg, 2017); Matthew de la Hey and William Beinart, ‘Why Have South African Smallholders Largely Abandoned Arable Production in Fields? A Case Study’, Journal of Southern African Studies, 43, 4 (2017), 753-770; Paul Hebinck, Nosiseko Mtati and Charlie Shackleton, ‘More than just fields: Reframing deagrarianisation in landscapes and livelihoods’, Journal of Rural Studies, 61 (2018), 323-334

22 Tessa Cousins and Donna Hornby, ‘Leaping the Fissures: Bridging the Gap between Paper and Real Practice in Setting Up Common Property Institutions in Land Reform in South Africa’ in Donna Hornby, Rosalie Kingwill, Lauren Royston and Ben Cousins (eds.) Untitled: Securing Land Tenure in Urban and Rural South Africa (University of Kwa-Zulu Natal Press, Pietermaritzburg, 2017), 320-360.

23 Paul Hebinck and Peter C. Lent (eds.), Livelihoods and Landscape: The People of Guquka and Koloni and Their Resources (Brill, Leiden, 2007).

24 Manona, ‘Smallholder Agriculture’.

25 Margot Leger, ‘Cattle and Rocks: A Social History of Shepherds in the Mountain Kingdom’, MSc African Studies dissertation, University of Oxford, 2015.

26 The South African Demographic and Health Survey, 2017 reported that 27% of children under five suffer from stunting. As a result their growth is delayed, and it is likely that this may also affect their cognitive capacity.

27 B. Cousins, ‘Smallholder Irrigation Schemes, Agrarian Reform and “Accumulation from Above and from Below” in South Africa’, Journal of Agrarian Change, 13, 1 (2013), pp. 116–39.

28 Kgabo Manyelo, Wim van Averbeke and Paul Hebinck, ‘Smallholder irrigators and fresh produce street traders in Thohoyandou, Limpopo Province, South Africa’ in P. Hebinck, J-D. van der Ploeg and S. Schneider (eds.), Rural Development and the Construction of New Markets (Routledge, London, 2015), 131-148. 29 Michael Aliber, paper to PLAAS, University of the Western Cape, conference on Rethinking Agriculture, Constraints and Opportunities, 11-13th October 2016.

30 Alex Dubb,‘The Rise and Decline of Small-Scale Sugarcane Production in South Africa: a Historical Perspective’, Journal of Agrarian Change, 16, 4 (2016), 518-542.

31 Jeanette Clarke, ‘Job creation in agriculture, forestry and fisheries in South Africa: An analysis of employment trends, opportunities and constraints in forestry and wood products industries’, PLAAS Working Paper 52 (2018).

32 National Wool Growers Association Website, September 2018.