Land Use and Rural Livelihoods in South Africa: Emerging Evidence from the Eastern Cape

Abstract

This article examines whether rural households, both land reform beneficiary and aspirant, with access to land, use the land they have or not. Micro-data on crop and livestock production from land reform beneficiary and aspirant households in the Eastern Cape (South Africa) are analyzed to establish if households that have access to land actually use the land they have. Analysis of crop and livestock production data suggests that although land may not be used to full capacity in many of the households, the majority of them are using the land they have to produce different crops and livestock products for home consumption and sale. This is contrary to media reports that there is no production taking place on the overwhelming majority of land reform projects and in communal areas of former homelands. It is also apparent from the data presented below that the land reform beneficiary households on an average produce more crops and livestock than the non-beneficiary households.

Keywords

land reform household land use beneficiary non-beneficiary

Introduction

South Africa has been implementing land reforms since 1994, through three main programmes: land redistribution, land restitution and security of tenure reforms.^{By the end of the 2010/11 financial year, a total of 6.8 million hectares of land had been either transferred through redistribution and tenure reform, or restored through restitution.¹ Apart from widespread dissatisfaction with the pace of land reform, there has been general discontent with what is happening to the land that has been transferred. Broadly, the displeasure about the transferred land hinges on two logically related issues: one is whether the transferred land is being used and the other is whether this land contributes meaningfully to the livelihoods of the beneficiaries. Several studies have looked at these two issues. While few studies report that land reform has made a positive contribution (Chitonge and Ntsebeza 2012; Keswell and Carter 2011),² most of the studies conclude that the contribution to the well-being of the majority of the beneficiaries is insignificant for both redistribution (see Anseeuw and Mathebula 2008; Bradstock 2006; HSRC 2003) and restitution projects (see Aliber et al. 2011, Lahiff et al. 2012). Most of the studies conclude that land reform has been associated with failure and declining production. These findings resonate with media stories as well as statements by some senior government officials who have repeatedly asserted that over 90 per cent of land transferred is not productive (Nkwinti 2010; Xingwana 2009). This article contributes to this debate by analyzing crop and livestock production data from land reform beneficiary and aspirant (non-beneficiary) households to determine whether land is being used and whether using land contributes to the livelihoods of these households.}

By analyzing crop and livestock production data for beneficiary households vis-à-vis the non-beneficiary households it is possible to determine whether land is being used and if the level of production in these households makes a significant contribution to well-being. This article focuses on crop and livestock production at the household level and raises two critical methodology issues. One is whether land reform projects should be assessed using project or household level data; and the other is whether production on land reform projects should be benchmarked to the production levels before the farm was transferred. Undoubtedly, the approach adopted in assessing the performance of the transferred land affects the conclusion reached.

A close look at the data suggests that while there are few (10 per cent) beneficiary households which are not involved in any agriculture production, the majority of the households are using their land and are producing different crops and livestock products for home consumption and sale. The other main point that arises from the data is that beneficiaries, like many rural households, are not homogenous; they are highly differentiated in terms of command over factors of production (including investments resources, land and labour), as well as their social and political connections and influence. Because of the differentiated nature of the beneficiaries, even in the same project, project-level assessment (which most of the studies so far have adopted) is more likely to miss the nuanced outcome of land reform among the different classes of beneficiaries. Data presented below also indicates that a beneficiary household, an average, produces more crops and livestock than an average non-beneficiary household, suggesting that land, in general and land reform in particular, do make a difference.

This article begins by presenting the context of South Africa’s land reform, focusing on the debates regarding the role of land in rural livelihoods. Following this, the data on crop and livestock production will first be described and then presented in detail. The last section will summarize the key points emerging from the data.

The Context of Land and Agrarian Reform in South Africa

The broader context of land reform in South Africa is the well-known situation of highly skewed distribution and ownership of land, arising largely from the widespread and extended land dispossession during the colonial and apartheid eras. Existing evidence suggests that inequality in access and ownership of land has not significantly changed after 18 years of implementing land reform. According to the official agricultural statistics, 82 per cent (86 million hectares) of the total land mass is owned by 40,000 commercial farmers, mostly white, while only 11.8 per cent (14.5 million hectares) is shared by the 2.4 million smallholder and subsistence farmers, mostly in former homeland areas (DAFF 2010).³ Given this scenario, very few people (locally and internationally) contest the idea that land redistribution is necessary to deal with past injustices and address the highly skewed ownership of land along racial lines. However, there are many people who are critical of the idea that land reform can actually make a significant contribution to improving the livelihoods of beneficiaries. For example, critics of land reform have argued that ‘[l]and redistribution is unlikely to play a significant role in the strategy to reduce poverty in South Africa’ (Jenkins and Thomas 2000: 23). Although those critical of land reform offer various reasons for embracing this position, there are three broad explanations that are often cited.

The first is the idea that the South African economy has undergone structural transformation which has reduced the role of land and agriculture related activities in the overall national output. Broadly, this view is linked to the two-sector model proposed by Arthur Lewis (1954) and adapted by Ranis and Fei (1961), which has influenced development and agricultural policy since the 1960s (Johnston 1993). The key argument of the two-sector model is that, as a country undergoes the process of economic transformation, the share of agriculture in national output and labour force, relative to industry and other sectors, falls quite rapidly and its contribution to the national economy diminishes (see Hazel et al., 2007: 4).⁴ This, together with the view that income elasticity of demand for agricultural produce is always less than one and usually declining, has led some analysts to refer to agriculture as a sector in ‘secular decline’ (see Johnston and Mellor 1961). The influence of this view is not just a thing of the past; it is very much alive today, as evident in the view that economies grow when people ‘move from farms to firms’ (World Bank 2012). This view has, of course, been contested, with some analysts arguing that this is a misunderstanding of the role of agriculture in economic development (Johnston 1993).

In South Africa, the idea of agriculture being a sector in secular decline has been invoked in various ways to support the argument that there is little prospect for viable livelihoods in subsistence farming. It has been argued that, unlike other countries in the continent, where agriculture accounts for a large share in Gross Domestic Product (GDP) and the largest share of the labour force, South Africa has displayed all the signs of a transformed economy.⁵ Evidence cited is to show that this includes the low share of the agricultural sector in GDP (currently at 2.4 per cent), low ratio of labour force in agriculture (currently 9 per cent), increasing farm size, a general shift to high-value commodity production in agriculture (such as horticulture, which has grown more rapidly than any sub-sector (DAFF 2012)) and the steadily rising productivity of agricultural labour (Liebenberg et al. 2010).

In a fundamental way, this pessimistic view about the role of land in promoting sustainable livelihoods in rural areas is not so much directed at agriculture in general, but more towards smallholder and subsistence farming (see Collier 2008), leaving many convinced that the role of land reform in creating sustainable livelihoods has been, and will continue to be, minor. While many people see land reform making little or no contribution beyond equity concerns, there are some analysts who see it playing an important role beyond remedying past injustices: ‘[i]n addition to redressing past injustices, land reform in South Africa has the potential to improve rural income and employment, even though it has not yet lived up to that potential’ (van den Brink et al. 2009: 201).

The second set of explanations for the view that land reform is not going to contribute significantly towards improving the well-being of the beneficiaries is the idea that most rural households have diversified their livelihood strategies such that land-based strategies contribute only a tiny share to rural livelihoods. The central argument of this view is that as more and more rural households diversify their livelihoods, land-based activities (mainly agriculture) contribute an increasingly declining proportion of household income and livelihood, a phenomenon which has been referred to as ‘de-agrarianisation’ (Bryceson 1999) and sometimes ‘depeasantisation’ (Bryceson 2002, 2009). Accordingly to Rigg (2006: 181),

not only are non-farm activities becoming central to rural livelihoods but also…an increasing number of rural households have no commitment to farming whatsoever. It is not, therefore, just a question of weighing and balance, but of a more profound transition from one way of making a living to another.

In South Africa, this view is linked to the argument about the impact of colonial capitalist development and apartheid on African households. For some analysts, colonial and apartheid capitalist development in agriculture, mining and industry during the first half of the twentieth century resulted in the complete proletarianization of the peasantry, leading to the ‘virtual destruction of the peasantry’ (Wolpe 1972: 25). Expressions such as the ‘rise and fall of the peasantry’ (Bundy 1972), as well as the notion of ‘linear proletarianisation’ (Wolpe 1972), have all being employed to suggest that capitalist development successfully transformed the peasantry into a ‘reserve army of labour’ (Legassick and Wolpe 1976). Proponents of this view argue that this reserve army has over time become disinterested in farming; its interest has irreversibly shifted to wage labour. This view has been recently articulated, more particularly, by the Centre for Development and Enterprise (CDE 2005, 2008), which has consistently argued that there is little prospect for viable livelihoods in subsistence agriculture and that ‘only 9 per cent of black people who are currently not farmers have clear farming aspirations’ (ibid 2005: 14).

The view that the peasantry was completely wiped out by colonial and apartheid capitalism has been contested, with some analysts pointing out that the notion of total destruction of the peasantry in South Africa is empirically inaccurate and theoretically misconstrued (Neocosmos 1993). There have been studies which have shown that although most residents in former homeland areas had limited access to land with limited agricultural activities, many families continued to engage in small-scale agriculture producing both for home consumption and for local markets. Studies by May (1987) in former homeland of KwaZulu-Natal and by Makhura et al. (1998) in former KaNgwane homeland (in currently Mpumalanga Province), show that households that had migrant labourers continued to use the little land they had to produce crops and livestock and this has continued in post-apartheid South Africa (see Stats SA 1999). Thus, the linear proleterianization or depeasantization theses seem to underplay the resilience of the peasantry in South Africa. In many parts of Africa, including countries that had a large settler population, the peasantry has been a resilient layer of society, to the extent that terms such as ‘re-emergence of the peasantry’ (van der Ploeg 2010) can be misleading.

The third explanation for the insignificant role of land reforms in improving rural livelihoods is that South Africa has a broad-based social assistance system in the form of social grants, particularly for pensioners and children.⁶ Proponents of this view argue that unlike in most African countries, where the majority of the rural poor have subsistence farming as their major means of survival, the grant system in South Africa constitutes a significant source of income for poor households, such that the recipients often survive without having to engage in subsistence agriculture. Indeed, there have been a number of studies that have reported that most rural households, including many land reform beneficiaries, are relying increasingly on social grants (Mencarnini 2000; Anseeuw and Mathebula 2008) and that social grants account for the largest share of household income and livelihood among most land reform beneficiaries (Bradstock 2006). A study commissioned by the Development Bank of Southern Africa (DBSA) reported that social grants accounted for 50 per cent of the total income of smallholder and subsistence households, followed by wages at 22.9 per cent and remittances at 18.6 per cent, with income from agriculture accounting for only 3.7 per cent (Vink and van Rooyen 2009: 14).⁷ Many analysts take these figures as evidence that subsistence agriculture makes very little contribution to rural livelihoods and that the prospect for viable livelihoods in subsistence farming is low.

In this context, it has been suggested that increased access to social grants provides poor households with alternative income which, in turn, reduces the pressure to embark on subsistence agriculture (NPC 2011: 198), although evidence of the disincentive effect of social grants has not been carefully and empirically examined (Baiphethi and Jacobs 2009). Evidence on whether social grant recipient households in rural areas are able to survive entirely on grants, especially in cases where there is only one grant recipient, is scarce. Certainly, producing own food would improve the households’ well-being more than relying on social grants only.

It is important to note that in all the three explanations above there is a strong belief that what will improve the well-being of most people in former homelands are jobs in urban areas or commercial agriculture and not subsistence or smallholder farming. It has been alleged that ‘far fewer black South Africans want to farm than is commonly supposed; most blacks regard jobs and housing in urban areas as more important priorities’ (CDE 2005: 14). However, the urban formal job solution seems to be problematic under the current situation, given the country’s stubborn lump of unemployment resulting from the failure of the economy to create enough jobs and there are no clear indications of an economic turn-around that would result in rapid job creation anytime soon.⁸

Because of the reasons summarized above, it has been observed by Kepe et al. (2008: 144) that,

…few people in South Africa appear to believe that land reform will contribute much to economic growth and development, or poverty reduction. There is widespread scepticism about the potential of agriculture to reduce poverty for large numbers of people.

The pessimism around land reform is linked to the widespread view that the land transferred through land reform has not been productive for various reasons, including the fragmented way in which the reforms have been implemented, resulting in the failure to adequately and sustainably support land reform beneficiaries (Hall 2011).

Data and Methods

The data presented below, unless stated otherwise, are drawn from a household survey conducted in the Chris Hani District Municipality (CHDM) in 2010.⁹ Household level data from land reform beneficiaries as well as non-beneficiaries were collected through a multi-topic household survey questionnaire. Additional data were collected through participatory research techniques.¹⁰ Land reform beneficiary households were selected from all eight local municipalities that make up CHDM, with assistance from CALUSA, a local NGO that works with land reform beneficiaries and aspirants. For the non-beneficiaries, only households that were waiting for land from the land reform programme (aspirant land reform beneficiary) and only those from the same community as the beneficiaries were selected. Since most beneficiary households had similar livelihood strategies as the non-beneficiaries prior to becoming beneficiaries, the non-beneficiary group provides a counter-factual group for assessing whether land reform is making a difference in the livelihoods of the beneficiaries. If there is no difference between the two groups, it is most likely that land reform is making no significant contribution. However, the difference between the two groups may not be entirely attributed to the impact of land reform, since there are many unobserved factors that may be responsible for the difference. Nonetheless, selecting non-beneficiaries from the same communities, with similar livelihood strategies, as the beneficiaries, helps to reduce the bias introduced due to unobserved initial conditions. Evaluation literature acknowledges that in cases where the groups being compared are not randomized, ensuring that they come from the same residential environment can reduce the selection bias and increase the matching of non-observable characteristics significantly (Heckman et al. 1997). Nevertheless, selecting the beneficiary and non-beneficiary households from the same community does not guarantee a match between households in the two groups prior to land transfer and as such the difference observed between the two groups should be cautiously interpreted. One of the common features for households in this study is that most of them (both beneficiary and non-beneficiary) come from the former homelands or Bantustans where the majority of those engaged in farming, due to shortage of land, are still producing primarily for home consumption. In total, 99 beneficiary and 156 non-beneficiary households were included in the study.

As evident from Table 1, both beneficiaries and non-beneficiaries can be disaggregated into different sub-groups depending on the type of landholding and means through which the household accesses land. For the beneficiaries, the two main sub-groups are: households with individual family farm and households with land belonging to a group. In the case of the non-beneficiary households, though there is often an assumption in the literature that they all have small plots in former reserves (communal areas), a close analysis of these households suggests a more complex configuration: ten households bought land outside of land reform (hence to be referred to as ‘Bought’), some even before land reform started in 1994; 29 households inherited land (referred to as ‘Inherited’), mainly from relatives within the communal areas; 49 households were allocated land by the traditional leaders (referred to as ‘Chiefs’); 68 households reported having no plot of land of their own (referred to as ‘No Land’). May & Roberts (2000: 8) and the Rural Survey (Stats SA 1999) report similar sub-groups, differentiated by the way they accessed land. It is important to note that while households in the non-beneficiary group are still waiting for land under land redistribution programme, most of them (58 per cent) have their own small plots, mainly in communal areas. Furthermore, households in the ‘No land’ sub-group have access to common grazing land, especially in communal areas, making it possible for them to own livestock.

Table 1

Surveyed Household Disaggregated by Landholding

	Households (n)	%	Average Land Size (Ha)	Household size
Beneficiary Group
Individual	25	25	207.5	8.5
Group	74	75	56.1	8.3
Total	99		102.1	8.3
Non-Beneficiary Group
Bought	10	6.4	205	4.8
Inherited	29	18.6	69.7	8.2
Chiefs	49	31.6	1.6	6.9
No Land	68	43.4	0	6.3
Total	156	100		6.5
Sample	255			7.4

Source: Collected and compiled by the Author.

Note: The average hectarege for the beneficiary group in group projects may not be accurate, because this was estimated using the number of members in the group which may not correspond to the actual number of households, since some members belong to the same household. This is because the LRAD programme targeted individuals and not households. For those households which inherited land, the majority (24 households or 82 per cent) inherited land within the communal area, while the remaining (five households or 18 per cent) inherited land outside the reserves. For households that bought land outside of land reform, six have an average plot size of less than 20 ha, while the other four have plots of between 280 and 1200 ha, suggesting that they bought land inside and outside of communal areas, respectively.

In this article, both aggregated and disaggregated data for the land groups are presented.

The household is here adopted as a unit of analysis for the reason that the contribution that land makes is more likely to be aptly captured at the household level. Assessing whether land is used and whether land use makes any significant contribution to livelihoods using project level data, may lead to a distorted picture of the role land plays in rural livelihoods in general, especially for households in group projects. This is because in many group projects individual members are allowed to keep livestock (and in some projects, even to grow crops on individual plots), which are not accounted for at the project level. In such situations, the project level data may not be able to account for all production and land-use.

The other crucial methodological issue to note is that when assessing land reform projects, the common approach adopted is to compare the current production and income levels to that before the farm was transferred. For example, Anseeuw & Mathebula (2008: 36), in their study of land reform projects in Limpopo province, compared gross income on the farms before and after transfer and concluded that ‘in total land reform has led to an 89.5 per cent decrease in gross income of the restituted or redistributed farms’. Many land reform projects have been assessed in this way. Generally, land reform beneficiaries ‘no matter how poor or how numerous—are required to step into the shoes of former white owners and [are expected to] continue to manage the farm as a unitary, commercially oriented enterprise’ (Lahiff 2009: 184). Large, well-supported farms, owned and run by one farmer, after being transferred to a group of, sometimes, 60 beneficiaries, with little or no support, have been expected to produce at the same level as before land was transferred. As it has been observed, it is virtually impossible to compare white farmers and land reform beneficiaries from former homelands (van der Brink et al. 2009: 203). The approach of assessing the production of land reform projects against the production levels of previous farm owners is highly questionable given the difference in capacities and production constraints under which land is used (Moyo et al. 2009: 52).

Agricultural Production

The two main land-based productive activities considered here are crop and livestock production, which are the major agricultural activities among households in the study area. Patterns of crop and livestock production are analyzed here as a rough indication of land use, for the simple reason that production of crops and/or livestock, at any scale, is impossible without the use of land. Establishing whether land is used or not is important not just for land reform beneficiaries, but all rural households with access to land, given the strong belief that fields in former homelands are often abandoned (Ardington 1988; Perret 2002) or under-utilized (NPC 2011: 197–98).

It is also important to note that the production data reported here are based on cross-sectional data and, as such, it is not possible to determine the trend in land use and production levels, which require panel data. So far, most of the assessments, especially of land reform projects, have relied on cross-sectional data. The reported production figures for both crop and livestock are gross values; net production values could not be estimated due to the unavailability of input costs data.¹¹

The agriculture sector output in South Africa is broadly divided into three sub-sectors comprising of field crops, livestock and horticulture. Nationally, in 2008, livestock had the largest share by output value, accounting for 44 per cent, followed by field crops at 33 per cent and horticulture at 23 per cent (Liebenberg and Pardey 2010).

In the Eastern Cape, where this study was conducted, livestock production has a much larger share of the provincial agriculture output, mainly because of the semi-arid conditions prevailing in the province. A national survey found that the Eastern Cape accounted for close to half (43 per cent) of all the livestock farm units in the country, with 55 per cent of dairy cattle, 46 per cent of beef cattle and 48 per cent of goat farm units (Stats SA 2002: 32–34). However, as the data presented below show, rain-fed crop production for most of the households in the study is also common, though at a much smaller scale.

Crop Production

As Table 2 shows, fifteen different crops were reported to be grown by households in the area. Not surprisingly, maize was the most common crop, followed by potatoes, pumpkin and cabbage. Figures from the Rural Survey also show that maize is the most common crop grown by households in former homeland areas, with 93 per cent of households involved in crop production indicating that they grew maize. For the Eastern Cape Province, 95 per cent of respondents indicated that they grew maize in the previous season, followed by potatoes at 27 per cent and onions and tomatoes both at 4 per cent (Stats SA 1999). The popularity of maize among smallholder growers, especially in former homeland and land reform groups, may be an indication that the majority of these households engage in farming primarily to produce food for home consumption. This is evident from the quantities produced. The dominance of maize, across all land groups, is understandable, given that maize is used as a staple food for the majority of households in the area (see also McAllister 1999, Shackleton et al. 2001). Although there are reports of many rural households relying on maize meal purchased from super markets (D’Haese and Huylenbroek 2005), most rural households engage in growing maize for own consumption. As such, increased maize productivity and production among rural households can improve their food security (Baiphethi and Jacobs 2009).

Table 2

Percentage of Households Growing a Particular Crop

Crop	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	Non-ben (%)	Total (%)
Maize	56	60	66	49	61.4	60
Potato	42	60	48	55	60.2	51
Pumpkin	43	50	28	45	46.6	45
Cabbage	40	30	31	51	44.3	42
Onion	33	70	34	45	48.9	41
Spinach	39	60	21	41	39.8	40
Beans	32	30	28	41	39.8	36
Other Vegetable	26	30	14	22	22.7	25
Fodder	32	50	17	22	11.4	22
Tomato	17	20	3	12	26.1	21
Peas	3	0	3	18	11.4	7
Lettuce	2	10	7	10	11.4	5
Squash	4	10	3	4	4.5	4
Citrus	4	0	0	2	1.1	3
Sorghum	3	0	0	2	0.1	2
Sample size (n)	99	10	29	49	88	187

Source: Collected and compiled by the Author.

Note: Households without own plot of land (‘No Land’ group) are excluded from the analysis on crop production. ‘Non-ben’ refers to non-beneficiary.

Even though maize is the most common crop grown among households in the study area, only 60 per cent of households with access to own plot of land reported to have grown maize in the season preceding the interviews. This is a much lower rate when compared with other areas, such as KwaZulu-Natal (KZN) where a much higher rate (95 per cent) of households has been reported (Kirsten et al. 1998). Kirsten et al. (1998) reported that 89 per cent of households in KZN grew maize, with 85 per cent indicating that they grew for home consumption. The low ratio of households reporting growing maize in the study areas can be attributed to the semi-arid conditions in the study area. However, recent studies are suggesting that households in former homeland areas are preferring livestock to crop production (Aliber 2009), especially in cases where households are unable to obtain farm inputs. It is, however, not clear if there is an overall drop in the land used for crop production vis-à-vis livestock production among smallholders.

While land reform beneficiaries have a lower ratio of households growing a particular crop, except for fodder and vegetables, the average quantity produced by beneficiaries, on average, is much higher than the average for non-beneficiaries, as Table 3 illustrates.

Table 3

Household Maize Quantity Produced by Land Group

Kg	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	Non-ben (%)	Total (%)
50–100	16.4	16.7	21.1	20.8	20.41	18.3
100–200	10.9	0.0	10.5	33.3	20.41	15.4
200–500	34.5	50.0	36.8	16.7	32.65	33.7
500–1000	16.4	0.0	5.3	20.8	14.29	15.4
1000–2000	3.6	33.3	21.1	4.2	16.33	9.6
>2000	18.2	0.0	5.3	4.2	6.12	12.5
Average Kg	1808	791	657	437.5	575	1180
Std. Deviation	3832	863	686	558	663.1	2783

Source: Collected and compiled by the Author.

The types of crops grown and quantity reported by households in the study suggest that most of the crops are oriented towards subsistence production. The absence of commercial cash crops such as wheat, peas, rye, citrus fruits, soya beans and barley, which are grown on commercial farms with irrigation facilities in the same areas, may also be indicative of the subsistence orientation of most of households in this study. Quantities of each crop produced also suggest that production is on a much smaller scale for the majority of the households.

However, it is important to note that the non-beneficiary group is over-represented in the lower quantity categories and under-represented in the higher quantity categories, particularly for the category above two tons. Similarly, the average number of kilos produced by beneficiaries is more than three times the average for non-beneficiaries. This is a probable indication that some of the beneficiary households (especially those with individual family farms) are producing maize primarily for the market.

A similar picture emerges when potato production is considered. Although more than 70 per cent of households produce less than 20 sacks of potatoes annually, there are some households that are producing more than 100 sacks of potatoes, as Table 4 shows.

Table 4

Household Potato Quantity Produced by Land Group

10 kg Sacks	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	Non-ben (%)	Total (%)
1–5	14.3	33.3	50.0	22.2	38.3	23.6
5–20	54.8	16.7	35.7	59.3	46.8	50.6
20–50	21.4	33.3	7.1	14.8	14.9	18.0
50–100	9.5	16.7	7.1	0.0	4.3	6.7
>100	0.0	0.0	0.0	3.7	2.1	1.1
Average (sacks)	23.8	24.8	15.3	20.2	18.2	20.6
Std. Deviation	20.8	24.4	25.6	33.4	28.4	25.3

Source: Collected and compiled by the Author.

Notably, land reform beneficiaries, though they do not have a single household producing more than 100 sacks, they have the largest share of households producing between 50 and 100 potato sacks, with a higher average than non-beneficiaries.

In the case of onion production, non-beneficiaries have a higher ratio of households in the high quantity categories, as well as a higher average than beneficiaries as Table 5 shows. The explanation for this is that onion and other crops that require close attention are mainly grown in the plots next to the homestead in the communal areas. Since many land reform beneficiaries do not stay on the acquired farms (Kirsten and Machethe 2005), they tend to focus on growing crops that require less attention, such as maize, fodder and potato.¹² The majority of the beneficiaries that grow crops like onion are those with individual family farms who have moved from communal areas to the acquired farms.

Table 5

Household Onion Quantities Produced by Land Group

10 kg Sacks	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	Non-ben (%)	Total (%)
1–5	39.4	57.1	50.0	31.8	41.0	40.3
5–10	15.2	0.0	10.0	4.5	5.1	9.7
10–20	15.2	0.0	20.0	13.6	12.8	15.3
20–50	21.2	28.6	10.0	27.3	23.1	25.0
50– 100	6.1	14.3	10.0	13.6	12.8	9.7
>100	0.0	0.0	0.0	4.5	2.6	1.4
Average	16.2	24.7	15.1	26.5	23.4	20.1
Std. Deviation	21.3	36.6	21.7	28.9	27.7	25.2

Source: Collected and compiled by the Author.

A slightly different scenario surfaces from the analysis of cabbage production. The average number of kilos for the beneficiary group is about 35 per cent higher than the average for non-beneficiaries, as Table 6 shows.

Table 6

Cabbage Production Categories by Land Group

Kgs	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	Non-ben (%)	Total (%)
1–100	50	33.3	33.3	44.0	45.9	48.1
100–500	30	33.3	66.7	36.0	43.2	36.4
500–1000	12.5	0.0	0.0	16.0	10.8	11.7
1000–2000	5	33.3	0.0	4.0	5.4	5.2
>2000	2.5	0	0	0	0	1.3
Average	384	541	120	269	246	316
Std. Deviation	830	592	74	347	329	634

Source: Collected and compiled by the Author.

Again, while there is a concentration of households producing in the lower quantity category for both land groups, there are comparably more beneficiary households producing more than two tons of cabbage. One of the reasons why many households may be growing smaller quantities of cabbage is access to markets. Many respondents complained about the difficulty of finding the market for some of the crops, such as cabbage, which are not easily stored; it is only households with own transport that engage in production at a large scale.

In terms of the contribution of land to rural livelihoods, it is important to look not just at levels of production, but also at what happens to the produce. Table 7 gives an idea of how maize is utilized among household in the study. Interestingly, there is a high percentage of households reporting that they sell part of the maize produced, across the groups.

Table 7

Household Maize Utilization Profile by Land Group

	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	Non-ben (%)	Total (%)
Sell Part	45.5	60.0	36.8	20.8	31.3	38.8
Sell & Buy	5.5	20.0	21.1	20.8	20.8	12.6
Keep All	38.2	0.0	15.8	37.5	25.0	32.0
No Sell but Buy	10.9	20.0	26.3	20.8	22.9	16.5
Months Consumed Maize from Own Production
1 to 3	26.5	60.0	47.1	39.1	44.4	34.0
3 to 6	34.7	20.0	17.6	43.5	31.1	32.0
6 to 9	14.3	0.0	0.0	0.0	0.0	7.2
9 to 12	24.5	20.0	35.3	17.4	24.4	23.7
Average Months	6.4	5	5.9	5.2	5.6	6

Source: Collected and compiled by the Author.

On average, about 40 per cent of households that harvested maize sold part of what they harvested, with this number being significantly higher for beneficiary compared to non-beneficiary households. These figures are certainly much higher than those reported by smallholder farmers in Zambia (25 per cent), Mozambique (16 per cent), Malawi (15 per cent), but close to figures for Kenya (39 per cent) (Anderson 2011; Jayne et al. 2010: 13). It has been observed that in some poor households, maize is sold for various reasons, including unforeseen cash needs (e.g., medical emergencies), school fees and other expenses. Studies have shown that households sell ‘some of their maize, despite the fact that total harvest is insufficient for a year’s household needs’ (Kirsten and van Zyl 1996). Thus, selling of maize does not suggest that the household has more than it needs; selling can be a way of smoothing household cash flow. ‘Distress selling’ has been reported among poorer smallholder farmers in a number of countries in Africa, especially during the harvest time (Jayne et al. 2010). In this study, as shown in Table 7, more than a third of households keep all the maize they produce for home consumption. It has been suggested that the moderately poor households often keep all that they produce and only buy when they run out of what they produce (Shackleton et al. 2000). Generally, it has been observed (Jayne et al. 2010: 1389) that:

only 5–15% of the rural population buys and sells the main staple commodity in the same year. They comprise both the relatively large farms that sell grain and buy back small quantities of processed meal, as well as relatively poor households that make distress sales of grain after harvest…

The second part of Table 7 shows the number of months that households rely on maize from own production. Because of the differences in how households utilize maize grain, it is not easy to infer food security levels from the number of months a household consumes maize from own production. Generally, however, it has been acknowledged that households that rely on food produced in the household throughout the year are more food secure than those that rely on buying (Valdes et al. 2009). For households in this study, the average number of months a household consumes food from own production is six months. Again, there is differentiation across land groups in terms of the reported number of months a household consumes maize from own production: the majority of households, or 75 per cent, consumes maize from own production for less than six months, while roughly 25 per cent consumes food from own production for between nine to 12 months. Livelihood contribution from consuming food produced in the households can be quite substantial.

However, as noted above, two groups of households are highly differentiated in several ways. Other than the difference according to the way a household accessed land, there is also differentiation along gender lines. Table 8 below presents the average production figures for households according to the gender of household head.

Table 8

Selected Crop Production by Gender of Household Head

	HH Head	Benef	Bought	Inherited	Chief	Non-ben	Total
Maize (kg)	Male	1489	590	513	294	463	948
	Female	3600	1800	670	933	850	2072
Potato (sacks)	Male	22.4	20	5	18.2	14	16.5
	Female	35.7	48	14.5	14.3	18.9	26.7
Onion (sacks)	Male	16.6	21	24.7	38.3	30.6	24.6
	Female	47.8	120	1	55	64.2	56
Cabbage (kg)	Male	643.5	51	566	1403	1063	861.9
	Female	938.3	45	34	446.7	237	615

Source: Collected and compiled by the Author.

Interestingly, on an average, female-headed households produce more than the average male-headed households for all crops except cabbage. This is completely the opposite of figures on livestock (see Table 16 below). One possible explanation for this is that female-headed households are more concerned about food than male-headed households, where the household head may be more interested in income generating crops. This may explain the higher average for cabbage among male-headed households, since cabbage grown in large quantities is often meant for sale due to storage problems.

It is important to note that the figures reported above are not based on written records; the figures are based on recall by the most knowledgeable member of the household and they may not reflect the exact quantities produced in a year. Generally, there is a higher likelihood of under-reporting the total output figures, especially for crops produced for home consumption. Under-reporting of crop production among smallholder and subsistence growers often arises from a respondent’s failure to recall and report small quantities that might have been consumed, given away, or spoilt during harvest or in post-harvest stages (see Shackleton et al. 2001). Related to the problem of recall is the fact that most of the crops grown for home consumption are harvested throughout the growing season in smaller quantities and are often excluded from the reported total output (McAllister 1999). This has been noted as one of the problems that lead to underestimation of crop output among smallholder growers (see Kirsten & van Zyl 1996).

Livestock Production

As noted earlier, livestock production is a common agricultural activity and one of the major sources of livelihood for smallholder households in the study area. For smallholder farmers in former homeland areas, livestock serves different purposes, ranging from the supply of milk, draught power, meat and manure, to accumulation of assets and prestige and are used in traditional ceremonies and anniversaries. In the Eastern Cape, livestock ownership is common, with 64 per cent of households in former homelands reporting that they own cattle, 55 per cent goats, 52 per cent pigs, 34 per cent sheep (Stats SA 1999). This is similar to the livestock ownership proportions found in the study area, with the most common livestock being cattle, sheep and goats, while a few households own pigs, donkeys, horses and chickens (see Table 9 below).

Table 9

Percentage of Household Owing Livestock

	Beneficiary (%)	Bought (%)	Inherited (%)	Chiefs (%)	No Land (%)	Non-ben (%)	Total (%)
Cattle	87.9	60.0	82.8	65.3	38.2	56.4	68.6
Sheep	60.6	50.0	51.7	40.8	17.6	33.3	43.9
Goats	37.4	40.0	41.4	32.7	20.6	29.5	32.5
Pigs	15.2	20.0	13.8	16.3	8.8	12.8	13.7
Horse	28.3	20.0	24.1	16.3	8.8	14.7	20.0
Donkey	6.1	0.0	6.9	6.1	1.5	2.4	4.7
Chickens	20.1	30	31.0	18.3	7.4	16.7	15.7
(n)	99	10	29	49	68	156	255

Source: Collected and compiled by the Author.

Note: Households without own plot are included in the livestock analysis since they own livestock on communal grazing land.

Often, studies on livestock focus on cattle, ignoring smaller livestock, such as sheep, chickens, pigs and goats, which have a significant contribution to the livelihoods of poorer households (Shackleton et al. 2001). For households in this study, the majority own, at least, one mature livestock unit.

Beneficiary households have the highest proportion of households owning livestock in all the livestock categories. The average ownership percentage for the non-beneficiary group is heavily affected by the ‘No land’ group, which has relatively low livestock ownership ratios. There is a surprisingly higher proportion of households, across the groups, owning horses and donkeys. On average, horses seem to be more common than pigs and donkeys, across the land groups. One of the reasons given for this is that horses are used as a means of transport as well as a symbol of status in the areas; owning a horse is associated with middle or upper class status in the area.

Other than having a higher ownership ratio, beneficiary households, on average, have a higher average number of livestock per household in all categories, except for pigs and horses, as Table 10 shows.

The average number of cattle, sheep and goats owned by beneficiary households is more than twice that for non-beneficiaries. Similar levels of cattle ownership have been reported in a study conducted in Moapane in Rustenburg (Schwalbach et al. 2001: 200).

Table 10

Average Number of Livestock Owned by Land Group

	Beneficiary	Bought	Inherited	Chiefs	No Land	Non-ben	Total
Cattle	34.5	24.5	16.6	13.3	11.7	14.5	24.4
Std. Dev	42.6	24.4	23.3	10.7	10.2	16.1	33.6
Sheep	117.9	65.4	59.8	57	55.9	58.9	89.9
Std. Dev	268.6	43	81.1	51.7	47.4	58.4	201.2
Goat	29.7	34.3	19.9	13.6	12.6	16.6	22.4
Std. Dev	46	32.5	13.7	13.3	8.6	15.1	33
Pig	3.2	3	1.2	6	4.5	4.3	3.8
Std. Dev	3.1	1.4	0.5	5.8	4.7	4.7	4.1
Horse	3	3	1.4	4.5	1.8	2.7	2.8
Std. Dev	2	0	0.8	2.9	1.9	2.3	2.1

Source: Collected and compiled by the Author.

Note: Number of livestock excludes livestock owned jointly for households in group projects.

The fact that beneficiary households own more livestock than an average non-beneficiary can be accounted for by several factors, including the fact that beneficiaries on average have more land than non-beneficiary households, especially households with individual family farms (see Table 1 above). As expected, households without access to own plot of land have the lowest number of livestock in every category. This suggests a strong correlation between plot size and number of livestock owned, which has been confirmed from regression analysis, especially for cattle, sheep and goats.¹³

Households without own plot of land reported that their livestock number often does not increase, because of overgrazing on communal land, which results in a higher incidence of livestock morbidity and mortality than for households with own plots of land.

As with crop production, the higher standard deviation for the beneficiary group indicates that ownership is more widely dispersed among households in this group than among non-beneficiaries. This is confirmed when the households are grouped into categories according to quantities produced, as Table 11 shows. In the case of cattle, while about 43 per cent of the beneficiary households own between one and 10 cattle, some households (10 per cent) own more than 100 cattle. Like crop production, the beneficiary group has a higher proportion of households in the categories that own more than 50 cattle than non-beneficiaries, suggesting that beneficiary households own larger herds of cattle on average. However, the scenario is different with sheep ownership, which shows that the ratios for the two groups are close in most of the categories. This could be an indication that sheep can be reared more easily on communal grazing land, on which many non-beneficiary households rely.

Table 11

Household Livestock Ownership by Category (%)

Category	Beneficiary	Bought	Inherited	Chiefs	No Land	Non-ben	Total
Number				Cattle
1–5	23.3	33.3	29.2	22.6	38.5	29.9	26.6
5–10	19.8	0.0	33.3	32.3	19.2	26.4	23.1
10–20	16.3	16.7	16.7	29.0	26.9	24.1	20.2
20–50	16.3	33.3	12.5	16.1	15.4	16.1	16.2
50–100	14.0	16.7	4.2	0.0	0.0	2.3	8.1
>100	10.5	0.0	4.2	0.0	0.0	1.1	5.8
Number				Sheep
1–5	3.5	20.0	7.7	0.0	7.7	5.9	4.6
5–10	5.3	0.0	23.1	5.0	7.7	9.8	7.4
10–20	8.8	0.0	15.4	30.0	7.7	17.6	13.0
20–50	40.4	20.0	23.1	25.0	38.5	27.5	34.3
50–100	19.3	40.0	7.7	20.0	15.4	17.6	18.5
>100	22.8	20.0	23.1	20.0	23.1	21.6	22.2
Number				Goats
1–5	10.5	25	16.7	25	18.8	20.8	16.3
5–10	39.5	0.0	8.3	43.8	25.0	25.0	31.4
10–20	13.2	25.0	33.3	12.5	31.3	25.0	19.8
20–50	23.7	25.0	41.7	18.8	25.0	27.1	25.6
50–100	7.9	25.0	0.0	0.0	0.0	2.1	4.7
>100	5.3	0.0	0.0	0.0	0.0	0.0	2.3
Number				Pigs
1–5	85.7	100.0	100.0	62.5	85.7	81.0	82.9
5–10	7.1	0.0	0.0	12.5	0.0	4.8	5.7
>10	7.1	0.0	0.0	25.0	14.3	14.3	11.4
Number				Horses
1–5	82.1	100.0	100.0	75.0	85.7	87.5	84.6
5–10	17.9	0.0	0.0	12.5	14.3	8.3	13.5
>10	0.0	0.0	0.0	12.5	0.0	4.2	1.9

Source: Collected and compiled by the Author.

Since there is no baseline data on livestock ownership for these households, it is not possible to establish if the number of livestock in each household is increasing or not. However, the study collected data on the number of livestock that a household had in 2001 and the current number, including livestock sold over the year. As Table 12 shows, the average livestock growth between 2001 and 2010 was highest among beneficiary households, for both sheep and cattle. Most respondents attributed this growth among beneficiary households to the fact that their stock is not mixing with other people’s stock, thereby reducing the occurrence of diseases and also that livestock on private fenced grazing camps is less prone to stealing.

Table 12

Percentage of Livestock Growth by Land Group (2001–2010)

	Cattle	Sheep	Goat	Pig
Beneficiary	16.6	51.4	4.7	0.2
Bought	10.2	13.9	9	0.6
Inherited	4.2	7.8	6.2	0.13
Chiefs	2.4	15.2	–1.8	0.7
No Land	2	9.2	2.2	0.3
Non-Beneficiary	3	10.9	1.8	0.3

Source: Collected and compiled by the Author.

Households without own plot and those with land from traditional authorities (chiefs), reported the lowest growth in the number of livestock, suggesting that having own plot of land contributes significantly to the observed difference in livestock growth.

Livelihood effects through livestock is complex, though it can be mediated through various ways, such as the influence that livestock has on crop production through ploughing, transport and assets diversification (Cousins 1996). Empirical literature on livestock production among smallholder farmers has established a strong relationship between livestock ownership and other farm activities, such as crop production (Barrett 1992). This relationship can be explained by many factors, including the ploughing power provided by large livestock, the manure produced from livestock and also the income generated, which can be used to invest in crop production (Scoones 1992).

Earlier studies on livestock production in communal areas suggested a very limited use for livestock, with little economic value. For instance, in a study of Nkandla in KZN, it was reported that the ‘wide incidence of cattle ownership and the importance attached to stockholding should not be seen as a reflection of their economic value to the community— something which often did not extend beyond their value as draught animals’ (Ardington 1986). Delali et al. (2003) also makes a similar observation that, in the past, livestock in communal areas had often been given the symbolic value of social status. But some studies have challenged this view, affirming the multiple value and benefit of livestock (Scoones 1992; Adams et al. 2000). Some analysts point out that subsistence livestock farming provides more complex benefits than it has often been assumed in conventional economic models on livestock valuation (Stroebel et al. 2008). Although lobola payment, social status and draught power are still important benefits, it has been observed that ‘lack of accounting for all uses in a multipurpose livelihood system potentially results in significant underestimation of livelihoods…’ (Delali et al. 2003). Schwallbach et al. (2001) argue that, while in the past, livestock owners in communal areas may have been reluctant to sell their livestock, for various reasons, many small-scale livestock farmers are now reporting higher livestock sales than before.

Table 13 shows the proportion of households reporting selling livestock, as well as the average number of livestock sold in a year. Except for cattle, the proportion of households slaughtering smaller livestock for home consumption is quite high across land groups and selling of livestock is common among all land groups, contrary to views that livestock among smallholders is often given symbolic value.

Table 13

Livestock Sales and Consumption by Land Group

	Ben	Bought	Inherited	Chiefs	No Land	Non-ben	Total
Percentage of Households Selling Livestock
Cattle	63.2	33.3	37.5	50	30.8	36.4	49.7
Sheep	60	40	33.3	40	58.3	42.3	51.8
Goat	40.5	50	41.7	31.3	42.9	39.1	39.8
Pig	20	50	0	37.45	50	35	28.6
Average number of Livestock Sold per year
Cattle	3.8	2.6	1.4	1.3	0.4	1	2.1
Sheep	8.2	2.4	4.4	1.7	0.9	1.9	4.3
Goat	1.9	3.5	1.5	1.1	0.3	1	1.3
Pig	0.1	0.2	0	0.08	0.08	0.07	0.08
Percentage of Households Slaughtering Livestock for Home Consumption
Cattle	14.9	50	25	18.8	7.7	19.3	17.1
Sheep	68.3	20	66.7	60	83.3	63.5	66
Goat	35.1	50	41.7	18.8	28.6	30.4	32.5
Pig	26.7	50	25	12.5	16.7	15	20
Average Number of Livestock Slaughtered for Home Consumption
Cattle	0.3	0.5	0.3	0.1	0.05	0.2	0.2
Sheep	1.9	1	1.7	1.1	0.8	1.1	1.4
Goat	0.1	0.2	0.2	0.1	0.1	0.1	0.1
Pig	0.04	0.1	0.02	0.02	0.01	0.01	0.02

Source: Collected and compiled by the Author.

In terms of livestock sales, the beneficiary group has a higher ratio of households reporting animal sale and also the highest number of livestock sales across all categories. This could be a sign that most beneficiary households are involved in livestock production for sale. However, the average of live herd sale (off-take), even among beneficiaries, reported at 11.1 per cent of existing livestock, is still low, but much higher than the off-take for small-scale farmers in other parts of South Africa (see Stroebel et al. 2008).¹⁴

As noted earlier, the contribution of livestock to livelihood is not just in terms of income from sales or meat consumption. Livestock has other uses which are frequently reported as some of the major reasons for keeping livestock among small-scale livestock farmers (Adams et al. 2000; Barrett 1992; Delali et al. 2003). While the value attached to each of these benefits from livestock in different households may differ, what is common among most households is that livestock farming brings several benefits which cannot be captured by estimating the monetary value of livestock and related products. Thus, in assessing the contribution of land, one has to take into account the different benefits derived from owning livestock.

In the area where the study was conducted, as evident from Table 9 above, sheep farming is very common and one of its by-products is wool. Table 14 shows that almost two-thirds of households that own sheep produce some wool for sale. As expected, beneficiary households, on average, reported a higher proportion of households producing wool, as well as a higher average quantity of wool produced compared to non-beneficiary households.¹⁵

Table 14

Households Wool Production by Land Group

	Beneficiary	Bought	Inherited	Chiefs	No Land	Non-ben	Total
Wool (%)	71.7	40	86.7	45	41.7	55.8	64.3
Average(Bales)	4.2	0.75	2.9	4.2	2.9	3.1	3.7
Number of Bales
0.5–3	62.8	100.0	76.9	66.7	60.0	72.4	66.7
3.1 to 5	25.6	0.0	7.7	11.1	40.0	13.8	20.8
5.1 to 10	7.0	0.0	7.7	11.1	0.0	6.9	6.9
>10	4.7	0.0	7.7	11.1	0.0	6.9	5.6

Source: Collected and compiled by the Author.

Note: An average bale of wool was estimated at 140–150 kg by the Cape Mohair and Wool Ltd in Port Elizabeth.

As is the case with other products, households without access to own plot of land have the lowest proportion of members producing wool, especially in the higher quantity categories.

When we look at other livestock products, it is also apparent that beneficiary households have a higher proportion of households owing chicken, producing fodder, eggs and milk, with a higher mean quantity of each of these products, particularly fodder, eggs and milk than non-beneficiary households, as Table 15 shows.

Table 15

Other Livestock Products

	Beneficiary	Bought	Inherited	Chiefs	No Land	Non-ben	Total
Chicken (%)	20.2	30.0	31.0	18.3	7.4	16.7	15.7
No. of Chickens	13.0	6.8	17.8	25.7	1.8	12.6	12.7
Fodder (%)	32.3	20.0	3.4	12.2	0.0	5.8	16.0
No. of Bales	5.8	1.0	0.4	1.3	0.0	0.6	2.6
Eggs (%)	16.2	30.0	24.1	14.3	8.8	14.7	15.3
No. of Trays	8.5	4.9	5.4	3.1	0.8	2.6	4.9
Milk (%)	59.8	66.7	45	37.5	23	37.5	48.5
No. of litres	1448	491	1031	1432	1650	1182	1339

Source: Collected and compiled by the Author.

Unlike crop production, there is a marked difference between female- and male-headed households in terms of livestock ownership, as Table 16 shows. Male-headed households on average own more livestock than female-headed households in all livestock categories (except chicken) and across land groups.

Table 16

Livestock Ownership by Gender of Household Head

	Cattle		Sheep		Goat		Pig		Chicken
Male	Female	Male	Female	Male	Female	Male	Female	Male	Female
Beneficiary	32.9	20	85	26.5	12.2	8.7	0.5	0.1	21.4	30.9
Bought	16.3	0	35.8	5	15.2	0	0.7	0	10.8	0.0
Inherited	14.1	12.8	34.2	22.3	8.5	7.5	0.04	0.8	10.9	26.3
Chiefs	10.5	4.7	23.7	22.5	4.3	4.7	1.1	0.7	13.2	7.4
No land	5.8	3	17.9	3	3.4	2.5	0.7	0.2	7.7	2.1
Non-ben	10.1	4.8	25	11.5	5.9	3.8	0.5	0.2	10.4	15.3
Total	20.2	9	51.5	15.6	8.7	5.1	0.6	0.3	15.2	14.7

Source: Collected and compiled by the Author.

The reasons for the difference in livestock ownership between male- and female-headed households are many, including cultural practices regarding livestock ownership (see Guy 1990),¹⁶ but also the difference in command over assets, such as land and labour, as well as income. Overall, the majority of beneficiary households (90 per cent) produce either crop or livestock, or both. Only 10 per cent of beneficiary households reported having no agricultural production of any kind. For the non-beneficiary households, more than a third reported no agricultural production in the year preceding interviews, as Table 17 shows.

For households without own plot of land, two-thirds of households did not engage in agricultural production of any kind. It is also apparent from Table 17 that majority of the households involved in agriculture combine crop and livestock production. The contribution of these activities to livelihoods, food security and well-being, even for small quantities producers, can be substantial, especially for the poorest households with limited alternative sources of income.

Table 17

Household Agriculture Production Profile by Land Group (%)

	Beneficiary	Bought	Inherited	Chiefs	No Land	Non-ben	Total
Livestock & Crop	71.7	60	58.6	40.8	0	39.9	46.2
Crop only	6	0	13.8	6.1	0.0	4.9	5.2
Livestock Only	12.1	40	20.7	28.6	33.1	30.6	26.9
No Agri Income	10.1	0	6.9	24.5	66.9	24.6	21.6

Source: Collected and compiled by the Author.

The difference between the beneficiary and non-beneficiary group in terms of crop and livestock production is evident when the average quantities produced are considered. Table 18 below presents the means test results for the two groups for selected crops and livestock. The results indicate that the difference in means is statistically significant except for pigs, wool, and onion. The difference in means for onion, the only crop for which non-beneficiaries have a higher average than beneficiaries, is not statistically significant.

Table 18

Means Test for Selected Crop and Livestock Production

Maize (Kg)	(n)	Means	p-value	Cattle	(n)	Means	p-value
Beneficiary	55	1808	0.0221*	Beneficiary	87	34.5	0.0029*
Non-ben	57	575		Non-ben	88	14.5
Potatoes (Pockets)			0.0126*	Sheep
Beneficiary	42	23.8		Beneficiary	60	117.9	0.0000*
Non-ben	53	18.2		Non-ben	52	58.9
Onion (Pockets)			0.2610***	Goats
Beneficiary	33	16.2		Beneficiary	37	29.7	0.0363*
Non-ben	43	23.4		Non-ben	46	16.6
Cabbage (Kg)			0.4952***	Pigs
Beneficiary	40	384		Beneficiary	15	3.2	0.9176***
Non-ben	39	246		Non-ben	20	4.3
Fodder (Bales)			0.001*	Wool(bales)
Beneficiary		1.565		Beneficiary	43	4.2	0.4441***
Non-ben		0.250		Non-ben	29	3.1

Source: Collected and compiled by the Author.

Note: A non-parametric means test (the Mann-Whitney test) which does not assume normal distribution and equal variance within samples was used. [*]=significant at 5%; [**]=significant at 10 %; [***]=not significant.

Conclusion

This article has considered the question of whether land in general and land received through land reform, is used and whether this land is contributing positively to rural livelihoods. Using crop and livestock production as an indicator of land use, it is clear that the majority of rural households that have access to land, including land reform beneficiaries, are actually using their land to produce different crops and rear different types of livestock. As it would be expected, different households have different capacities to use land and this is reflected, largely, in the difference in the quantities of crop and livestock produced. For land reform beneficiaries in particular, although the form and level of land use may not be the same as before land was transferred (and there is no reason why these should be the same), the majority of households are using land, mainly for crop and livestock production, according to their different capacities. Similarly, using crop and livestock quantities produced in the household as a reflection of the contribution of land to livelihood, it is evident that having access to land does make a difference in these households. Even if the exact contribution to household income is not estimated in this article, the quantities of crop and livestock produced, whether used at home or sold, suggest that the contribution is substantial for most households. This is contrary to findings in most studies using project level data, which report little or no contribution.

From the data presented above, it is also apparent that households in the study are highly differentiated as reflected by the difference in the quantities produced, even for households in the same project or land group. Differences among land reform beneficiaries and aspirants point to the need for policy to take into account the different capacities, constraints and aspirations when designing and implementing programmes. The strong emphasis and bias towards commercial production has lead to unrealistic expectations from most land users, especially land reform beneficiaries. Many land users in rural areas have no aspiration and capacity to produce at commercial scale; most of them would like to produce enough to feed themselves and their families. The undue emphasis placed on commercial production has often resulted in ignoring those who are unable or have no intensions to produce at that level and also in downplaying crop and livestock production among smallholder and subsistence farmers.

In terms of the beneficiary and non-beneficiary households, it is apparent from the data that land reform beneficiaries perform much better, on average, for both crop and livestock production compared to non-beneficiaries. While this difference between beneficiaries and non-beneficiaries may not be entirely attributed to land reform, it is likely that land reform has contributed meaningfully to the observed difference between the two groups. Despite the fact that it is not possible to establish whether the current levels of production for beneficiary groups signifies a positive trend (due to lack of data), the possible contribution from using land, even in households that produce only for home consumption cannot be overlooked. Analysis of crop and livestock production data in this study suggests that the contribution of these activities to rural livelihoods in general is much higher than what is often reported in media and many land-use studies that assess land reform projects at project level.

Looking at figures on crop and livestock production, and taking into account the fact that many of these households achieve the reported levels of production with limited or no support (Chitonge and Ntsebeza 2012; Hall 2011), it is apparent that there is potential for these households to increase production from the current level if appropriate support were provided. Although the current production capacities and outputs are relatively low, there are signs that productivity can be increased.

Footnotes

Acknowledgements

The author wishes to acknowledge the financial support for this study from the NRF Chair on Land Reform at the University of Cape Town and is grateful for the comments and suggestions by the three anonymous reviewers.

Notes

References

Adams

Martin

Benjamin

Cousins

Sheona

Manona

(2000). Land tenure and economic development in rural South Africa. In Cousins

Ben

(ed.), At the crossroads: Land and agrarian reforms in South Africa into the 21^st century (pp. 111–28). Cape Town: PLAAS.

Aliber

Michael

(2009). Exploring Statistics South Africa’s National Household Surveys as sources of information about household-level food security, Agrekon, 48(4), 384–409.

Aliber

Michael

Themba

Maluleke

Tshililo

Manenzhe

Gaynor

Paradza

Benjamin

Cousins

(2011). Livelihoods after land reform: Trajectories of change in Limpopo Province, South Africa. Bellville: Institute for Poverty, Land and Agrarian Studies (PLAAS), University of the Western Cape.

Anderson

Agnes

(2011). Maize remittances, smallholder livelihoods and maize consumption in Malawi, Journal of Modern African Studies, 49(1), 1–25.

Anseeuw

Ward

Ntombifuthi

Mathebula

(2008). Evaluating land reform’s contribution to South Africa’s pro-poor growth pattern. Paper presented at the TIPS Annual Forum (n/d). Available at http://www.tips.org.za/publication/evaluating-land-reforms-contribution-south-africas-pro-poor-growth-pattern (accessed 18 September2012).

Ardington

Elisabeth

(1984). Nkandla revisited: A longitudinal study of the strategies adopted to alleviate poverty in a rural community, Centre for Social and Development Studies (CSDS), Rural-Urban Studies Working Paper No. 16, University of Natal.

Ardington

Elisabeth

(1984). Poverty and development in a rural community in KwaZulu, Centre for Social and Development Studies (CSDS), Rural-Urban Studies Working Paper No. 9, University of Natal.

Baiphethi

Nompati

Peter

Jacobs

(2009). The contribution of subsistence farming to food security in South Africa. Pretoria: Centre for Poverty, Employment and Growth, Human Science Research Council (HSRC).

Barrett

John

(1992). The economic role of cattle in communal farming systems in Zimbabwe, Overseas Development Institute (ODI) Pastoral Development Network, Working Paper No. 32b, London.

10.

Binswanger-Mkhize

Hans

Glen

Thomas

Rogier van den

Brink

(2009). Introduction and summary. In Binswanger-Mkhize

Hans

Bourguignon

Camille

van den Brink

Rogier

(eds), Agricultural land redistribution: Towards a greater consensus (pp. 3–42). Washington, DC: World Bank.

11.

Boyle

Brendan

(2010). Massive Farm Failure in SA, Timeslive, 2 March. Available at www.timeslive.co.za/local/article334448.ece (accessed 10 April 2010).

12.

Bradstock

Alastair

(2006). Land reform and livelihoods in South Africa’s Northern Cape Province, Land Use Policy, 23, 247–59.

13.

Bryceson

Deborah F.

(2002). The scramble in Africa: Reorienting rural livelihoods, World Development, 30(5), 725–39.

14.

Bryceson

Deborah F.

(2009). Sub-Saharan Africa’s vanishing peasantry and the spectre of global food crisis. Available at http://monthlyreview.org/2009/07/01/sub-saharan-africas-vanishing-peasantries-and-the-specter-of-a-global-food-crisis (accessed 12 September 2012).

15.

Bryceson

Deborah F.

(1999). Sub-Saharan Africa betwix and between: Rural livelihood practices and policies, De-agrarianisation and Rural Employment (DARE), Afrika-Studiecentrum (ASC), Working Paper No. 43. Available at www.asc.leidenuniv.nl/general/dare.htm (accessed 13 May 2009).

16.

Bundy

Colin

(1972). The rise and fall of the South African peasantry. Woodbridge: James Currey.

17.

Centre for Development and Enterprise (2005). Land reform in South Africa: A 21^st century perspective, CDE Research Report No. 14, Johannesburg.

18.

Centre for Development and Enterprise (2008). Land reform in South Africa: getting back on tract, CDE Research Report No. 16. Available at www.cde.org.za/article.php?a_id=284 (accessed 17 June 2010).

19.

Chitonge

Horman

Lungisile

Ntsebeza

(forthcoming 2012). Land reform and rural livelihood in South Africa: Does land matter?, Review of Agrarian Studies, 2(2). Available at http://ras.org.in/index.php?id=HomePage (accessed 7 January 2013).

20.

Chitonge

Horman

(2010). The state of KwaZulu-Natal labour market. In Nzimande

Nompumelelo

(ed.), State of the population of KwaZulu-Natal: Demographic profile and development indicators (pp. 54–72). Pietermaritzburg: KZN Provincial Department of Social Development.

21.

Collier

Paul

(2008). The politics of hunger: How illusion and greed fan the food crisis, Foreign Affairs, 87(6), 67–79.

22.

Cousins

Benjamin

(1996). Livestock production and common property struggles in South Africa’s agrarian reform, Journal of Peasant Studies, 23, 166–208.

23.

D’Haese

Marijke

Guido

van Huylenbroeck

(2005). The rise of supermarkets and changing expenditure patterns of poor rural households: Case study in Transkei area, South Africa, Food Policy, 30, 97–113.

24.

Deininger

Klaus

Julian

May

(2000). Can there be growth with equity? An initial assessment of land reform in South Africa, World Bank Policy Research Working Paper No. 2451. Available at www.worldbank.org (accessed on 19 Februrary 2009).

25.

Delali

B.K.

Dovie

C.M. Shackleton

Witkowski

E.T.F.

(2006). Valuation of communal area livestock benefits, rural livelihoods and related policy issues, Land Use Policy, 23, 260–71.

26.

DAFF [Department of Agriculture Food and Fisheries] (2010). Abstract of agricultural statistics. Pretoria: DAFF.

27.

DAFF [Department of Agriculture Food and Fisheries] (2012). Agriculture food and fisheries: Integrated growth and development plan 2012. Pretoria: DAFF.

28.

Guy

Jeff

(1990). Gender operation in Southern Africa’s pre-capitalist society. In Walker

Cheryl,

(ed.), Women and gender in Southern Africa until 1945 (pp. 33–47). Cape Town: David Phillips.

29.

Hall

Ruth

(2011). Agrarian change and agroecology. Paper presented at the conference entitled Agroecology and Agrarian Change: Towards a Democratic Participatory Agroecology, 24–26 May, Cape Town.

30.

Hazel

Peter

Colin

Pulton

Steve

Wiggins

Dorward

(2007). The future of small farms for poverty reduction and growth, International Food Policy Research Institute (IFPI) 2020 Discussion Paper No. 42, Washington, DC.

31.

Heckman

James

Ichimura

Todd

(1997). Matching as an econometric evaluation estimator: Evidence from evaluating a job training programmes, Review of Economic Studies, 64, 605–54.

32.

HSRC (2003). Land redistribution for agricultural development (LRAD): Case studies in three provinces. Pretoria: Human Science Research Council.

33.

Jayne

Thomas

David

Mather

Elliot

Mghenyi

(2010). Principal challenges confronting smallholder agriculture in sub-Saharan Africa, World Development, 38(10), 1384–98.

34.

Jenkins

Carolyn

Lynne

Thomas

(2000). The changing nature of inequality in South Africa, World Institute of Development Economics Research (WIDER), Working Paper No. 203, United Nations University, Helsinki.

35.

Johnston

Bruce

John

Mellor

(1961). The role of agriculture in economic development, American Economic Review, 51(4), 566–93.

36.

Johnston

Gale

(1993). Role of agriculture in economic development revisited, Agricultural Economics, 8, 421–34.

37.

Kepe

Themebla

Ruth

Hall

Benjamin

Cousins

(2008). Land. In Shepherd

Nick

Robins

Simpson

(eds), New South African key words (pp. 143–55). Athens: Ohio University Press.

38.

Keswell

Malcolm

Michael

Carter

(2011). Poverty and land ownership in South Africa: Evidence from natural experiment. Available at http://agecon.ucdavis.edu/people/faculty/michael-carter/docs/KeswellCarterJuly2011.pdf (accessed 20 February 2011).

39.

Kirsten

Johan

Charles

Machethe

(2005). Appraisal of land reform projects in the North Western Province. Report submitted to the Department of Agriculture. Available at http://mpra.ub.uni-muenchen.de/31614/1/Appraisal_of_land_reform_projects_in_the_North_West_Province_of_South_Africa.pdf (accessed 12 May 2012).

40.

Kirsten

Johan

van Zyl

(1996). The cost and benefits of providing agricultural support services to rural households in the developing areas of South Africa, Development Southern Africa, 13, 415–28.

41.

Kirsten

Johann

Robert

Townsend

Chris

Gibson

(1998). Determining the contribution of agricultural production to household nutritional status in KwaZulu-Natal, South Africa, Development Southern Africa, 15(4), 573–87.

42.

Lahiff

Edward

(2009). Land redistribution in South Africa. In Binswanger-Mkhize

Hans

Bourguignon

Camille

van den Brink

Rogier

(eds), Agricultural land redistribution: Towards a greater consensus (pp. 169–200). Washington, DC: World Bank.

43.

Lahiff

Edward

Nerhene

Davis

Shililo

Manenzhe

(2012). Joint ventures in agriculture: Lessons from land reform projects in South Africa. London: International Fund for Agriculture Development (IFAD).

44.

Legassick

Martin

Harold

Wolpe

(1976). Bantustan and capital accumulation in South Africa, Review of African Political Economy, 3(7), 87–107.

45.

Lewis

Arthur

(1954). Economic development with unlimited supplies of labour, Manchester School, 22, 139–91.

46.

Liebenberg

Frikkle

Philip

Pardey

(2010). South Africa agricultural production and productivity patterns. In Alston

Julian

Babcock

Bruce

Pardey

Philip

(eds), The Shifting Patterns of Agricultural Production and Productivity Worldwide (pp. 383–408). Iowa: Ames.

47.

Lipton

Michael

Frank

Ellis

Merle

Lipton

(1996). Introduction. In Lipton

Michael

de Klerk

Mike

Lipton

Merle

(eds), Land, labour and rural livelihoods in South Africa, Vol. 1 (pp. v–xxii). Durban: University of Natal Press.

48.

Makhura

Moraka

Frank

Goode

Gerhard

Cotzee

(1998). A cluster analysis of commercialisation of farmers in developing rural areas of South Africa, Development Southern Africa, 15(3), 429–48.

49.

May

Julian

(1987). Differentiation and inequality in the Bantustans: Evidence from KwaZulu-Natal, Centre for Applied Social Science Working Paper No. 7, University of Natal, Durban.

50.

May

Julian

Benjamin

Roberts

(2000). Monitoring and evaluating the quality of life of land reform beneficiaries: 1998/1999. Summary report prepared for the Department of Land Affairs, Final report.

51.

McAllister

Pat

(1999). Agriculture and cooperative labour in Shixini, Transkei, South Africa, ASC Working Paper No. 33. Available at https://openaccess.leidenuniv.nl/handle/1887/420 (accessed 13 May 2011).

52.

Mencarnini

Letizia

(2000). A note on landholding and fertility in rural South Africa, Genus, 56(3/4), 109–19.

53.

Mokgope

Kgoptotso

(2000). Community and diversity: The complexity of interests in land reform: A case study of Gallawatera Farm in the Eastern Cape. In Cousins

Benjamin

(ed.), At the crossroads: Land and agrarian reform in South Africa into the 21^st century (pp. 107–10). Cape Town: PLAS.

54.

Moyo

Sam

Walter

Chambati

Tendai

Murisa

Dumisani

Siziba

Charity

Dangwa

Kingston

Mujeyi

Ndabezinhle

Nyoni

(2009). Fast track land reform baseline survey in Zimbabwe: Trends and tendencies, 2005/06. Harare: African Institute for Agrarian Studies.

55.

NPC [National Planning Commission] (2011). National Development Plan: Vision 2030. Available at http://www.npconline.co.za/medialib/downloads/home/NPC%20National%20Development%20Plan%20Vision%202030%20-lo-res.pdf (accessed 17 October 2012).

56.

Neocosmos

Michael

(1993). The agrarian question in Southern Africa and ‘accumulation from below’. Research Report No. 93, Uppsala: Scandinavian Institute of African Studies.

57.

Nkwinti

Gugile

(2010). Speech by the Minister of Rural Development and Land Reform, budget vote for 2010/11. Available at http://ww2.ruraldevelopment.gov.za/DLA-internet//content/document_library/Ministry/Ministry_Minister_Speeches/Speech2.pdf (accessed 24 March 2010).

58.

Perret

Sylvain

(2002). Livelihood strategies in the Transkei (Eastern Cape Province): How does wool production fit in? Department of Agricultural Economics, Extension and Rural Development, University of Pretoria, Working Paper No. 2002-20. Available at http://web.up.ac.za/sitefiles/File/48/2052/2002-20.pdf, (accessed 23 May 2012).

59.

Ranis

Gustav

John

Fei

(1961). A theory of economic development, American Economic Review, 51(4), 533–65.

60.

Reardon

Thomas

Peter

Timmer

Christopher

Barret

Julio

Berdegue

(2003). The rise of supermarkets in Africa, Asia and Latin America, American Journal of Agricultural Economics, 85(5), 1140–46.

61.

Reardon

Thomas

Valerie

Kelly

Eric

Crawford

Thomas

Jayne

Kimseyinga

Savadogo

Daniel

Clay

(1996). Determinants of farm productivity in Africa: A synthesis of four case studies, Michigan State University (MSU) Development Paper No. 22, Department of Agricultural Economics.

62.

Rigg

Jonathan

(2006). Land, farming, livelihoods, and poverty: Rethinking the links in the rural South, World Development, 34(1), 180–202.

63.

Schwalbach

L.M.

Groenewald

I. B.

Marfo

C. B.

(2001). A survey of small-scale cattle farming systems in the North West Province of South Africa, South African Journal of Animal Science, 31(3), 200–04.

64.

Scoones

Ian

(1992). The economic value of livestock in communal areas of southern Zimbabwe, Agricultural Systems, 39, 339–59.

65.

Sen

Amartya

(1966). Peasants and dualism with and without dualism, Journal of Political Economy, 74(5), 425–50.

66.

Shackleton

Sheona

Charlie

Shackleton

Benjamin

Cousins

(2000). Revaluing the communal lands of Southern Africa: New understandings of rural livelihoods, Natural Resource Perspectives, 62, 1–4.

67.

Shackleton

Charlie

Sheona

Shackleton

Benjamin

Cousins

(2001). The role of land-based strategies in rural livelihoods: The contribution of arable production, animal husbandry and natural resource harvesting in communal areas in South Africa, Development Southern Africa, 18(5), 581–602.

68.

Stats SA [Statistics South Africa] (1999). Rural survey (P0360). Pretoria: Stats SA.

69.

Stats SA [Statistics South Africa] (2002). Report on the Survey of Large and Small Scale Agriculture. Pretoria: Stats SA.

70.

Stats SA [Statistics South Africa] (2011). Quarterly labour force survey: Quarter 3, Statistical Release P0211. Available at www.statssa.gov.za/publications/P0211/P02113rdQuarter2011.pdf (accessed 29 November 2011).

71.

Stroebel

Swanepoel

F.J.

Nthakheni

N.D.

Nesamvuni

A.E.

Taylor

(2008). Benefits obtained from cattle by smallholder farmers: A case study of Limpopo Province, South Africa, Australian Journal of Experimental Agriculture, 48(7), 825–28.

72.

UNDP [United Nations Development Programme] (2012). Africa human development report 2012: Towards a food secure future. New York, NY: UNDP.

73.

Valdes

Alberto

William

Foster

Gustavo

Anrique

Carlo

Azzarri

Katia

Covarrubias

Benjamin

Davis

Stefania

Digiuseppe

Tim

Essam

Tom

Hertz

Ana

de la O

Esteba

Quinones

Kostas

Stamoulis

Paul

Winters

Alberto

Zezza

(2009). A profile of the rural poor, ESA Working Paper No. 09–04, Food and Agriculture Organisation (FAO), Agricultural Development Economics Division. Available at www.fao.org/es/esa (accessed 24 July 2012).

74.

van den Brink

Roger

Thomas

Glen

Hans

Binswanger-Mkhize

(2009). Land reform in South Africa: Additional data and comments. In Binswanger-Mkhize

Hans,

Bourguignon

Carmille

van den Brink

Rogier

(eds), Agricultural land redistribution: Towards a greater consensus (pp. 201–14). Washington, DC: World Bank.

75.

van der Ploeg

Jan Douwe

(2010). The peasantries of the 21^st century: The commoditisation debate revisited, Journal of Peasant Studies, 37(1), 10–30.

76.

Vink

Nick

van Rooyen

Johan

(2009). The economic performance of agriculture in South Africa since 1994: Implications for food security, Development Planning Division Working Paper No. 17, Development Bank of Southern Africa.

77.

Wolpe

Harold

(1972). Capitalism and cheap labour power in South Africa: From segregation to apartheid, Economy and Society, 1(4), 425–56.

78.

World Bank (2012). World development report 2012: Jobs. Washington DC: World Bank.

79.

Xingwana

Lulu

(2009). Speech at Phaphamangma Africa Resuscitation Project, 11 March 2009, Ostrich Oasis Farm, http://www.ruraldevelopment.gov.za, (accessed 30 March 2010).