Academic journal article Nomadic Peoples

The Economic Contribution of Pastoralism: Case Studies from the Horn of Africa and Southern Africa

Academic journal article Nomadic Peoples

The Economic Contribution of Pastoralism: Case Studies from the Horn of Africa and Southern Africa

Article excerpt

Abstract

This paper reviews the literature on the economic contribution of pastoralism to the national economies of seven countries, two in the Horn of Africa and five in southern Africa. For all the countries covered here, it is difficult to disaggregate at the national level the economic outputs of pastoral livestock from livestock more generally. Research studies have nonetheless identified the distinctive contributions of pastoralism and occasionally quantified these contributions for selected localities and regions. This work demonstrates that the economic role of pastoralism can be quite different even for national economies in the same region. There also exist broad regional differences between the Horn, where pastoralists are heavily involved in producing live animals for export, and southern Africa, where smaller herds are kept primarily for purposes other than commercial sales. At least three methodological shortcomings limit our ability to quantify the contributions made by pastoralists to national economies: aggregated national statistics, the selective recording of data on pastoral production, and analytical confusion about how to impute realistic cash values to the products that herders obtain from their own herds for their own consumption.

Keywords: Pastoralism, Horn of Africa, southern Africa, livestock economics

Introduction

This review takes the proverbial trip from the Cape to Cairo, pastoral style. For pastoral purposes we begin the journey in Somalia and finish in the Republic of South Africa. The object of our journey is to establish the contribution of pastoralism to national economies in the Horn and southern Africa. Along the way we encounter both conceptual forests and conceptual trees.

The trees that might prevent us from seeing the forest are represented by individual African countries. With respect to livestock marketing and pastoral commercial involvement, each of the countries reviewed in this study is remarkably distinct, a fact that is emphasized in the following national case studies.

But there is also a forest to report on, in the form of a single, overall pattern that emerges from the subcontinental breadth of this review. In terms of variations in the level and kind of pastoral commercial involvement, there is a clear regional difference between the Horn and southern Africa.

The economies of Ethiopia and Somalia are poor, technologically underdeveloped, but integrated nonetheless into global capitalism. In terms of livestock exports, these countries produce relatively unprocessed raw commodities--hides, skins and live animals--for regional markets. Weak, non-existent or (from the perspective of livestock producers and traders) parasitic national governments do little to promote their livestock industries. Despite these constraints, pastoral households in Ethiopia and Somalia are increasingly involved in commercial livestock production for sale, with results that in the aggregate are impressive both in terms of the volume and value of international trade and foreign exchange earnings. In return, these pastoral households receive marketed commodities such as grain and clothing that are essential for their survival. Despite their remoteness and their participation in weak national economies, these are pastoral production systems oriented to commercial international exports.

The situation is different in southern Africa, where strong and occasionally benign national governments support and may subsidize their livestock producers. The export product range includes both processed commodities (such as chilled, boneless beef from Botswana) and sophisticated goods for international niche markets (such as Karakul lamb pelts from Namibia for the fashion industry, or branded 'free range' meat from South Africa). The export trade is officially regulated, treaty-bound, technically advanced, and diverse in the geographical spread of its markets.

Despite living in comparatively advanced industrial economies, many livestock-owning households in southern Africa are averse to routine market involvement. This paradox is sometimes explained in terms of pastoralists' 'attitudes', which are variously said to be conservative, traditional, or fixated on the accumulation of livestock wealth for purposes of prestige. The evidence assembled in this review provides an alternative economic explanation for restricted market participation by southern African pastoralists.

Many poorer farmers and herders in southern Africa obtain the bulk of their cash from non-agricultural sources--remittances from relatives, salaried employment, the informal economy, pensions or welfare. These poorer rural residents engage in agriculture not to make money but to save cash by producing for themselves food that they would otherwise need to purchase. This is an attractive strategy because of the different prices at which rural people can expect to buy and sell agricultural produce. Because of their relatively remote location, transport costs and poorly developed marketing systems, they can expect to buy food at inflated retail prices but routinely sell agricultural produce at deflated farm gate prices.

Pastoralists who actively seek commercial marketing opportunities do exist in southern Africa, for example among large Botswana cattle herd owners or among the Himba of northwest Namibia (Behnke 1987; Bollig 2006). But for many smaller producers it makes good economic sense merely to provision their own families by their agricultural labour, to market windfall surpluses in good years or sell livestock capital in times of distress. Beyond this point, any additional family labour or capital is better deployed to secure off-farm or non-pastoral sources of cash that are more lucrative than producing regular agro-pastoral surpluses for sale. This pattern is exacerbated by southern Africa's colonial history. Smaller producers who might otherwise aspire to regularly sell meat, milk or maize for profit must compete head-on against large, mechanized, well-financed farms or ranches with privileged market access and secure land tenure arrangements--the heirs to southern Africa's dualistic settler economy. Many smaller producers do not attempt to sell regularly into markets where distribution networks are controlled and prices are set by firms that achieve economies of scale far beyond the smallholder.

The contribution of pastoralism to national economies is therefore very different in the Horn and in southern Africa. In the Horn, pastoralism generates impressive amounts of foreign exchange and a significant contribution to GDP (gross domestic product) or it could be shown to do so if there was anyone around to enumerate these effects. In Somalia no one is counting because there is no national state and hence no national accounts. In Ethiopia no one is counting because government construes most international livestock trading as smuggling, which it refuses to countenance.

The situation is different in southern Africa. Here national accounts and agricultural statistics are conscientiously kept but do not capture many of the non-traded goods and services provided by pastoralism. For this information we must turn to specialized research studies, some of which are reviewed in this paper.

This paper is organized into seven country case studies--Somalia and Ethiopia on the Horn of Africa, followed by Botswana, Zimbabwe, South Africa, Namibia and Zambia in southern Africa. A concluding section examines the methodological impediments that limit our ability to quantify the contribution of pastoralism to these national economies.

Somalia: Livestock Exports and Food Purchases

The territory that was once the nation state of Somalia is pastoral country: nearly 98 per cent of the agricultural land is pasture (FAOSTAT), and out of a total population of over 9.7 million people, 7.4 million--or about three-quarters of the population--live in semi-arid rangeland areas suitable only for livestock production (ILRI 2002: 24).

Somalia has had no functioning central government since 1991 and the availability of national statistics on rangeland production and the livestock trade is uneven. (1) FAOSTAT gives no livestock population estimates or production figures for the country but does provide statistics for the number and value of live animal exports, which are the most important form in which Somali livestock are exchanged. This data is summarized in Figures 1 and 2, which document sudden and high fluctuations in exports both when Somalia had a unified central government during the 1970s and 1980s, and in the 1990s following the break-up of the state. Three factors explain these fluctuations: periodic severe droughts, veterinary restrictions by importing countries in the Persian Gulf, and insecurity and the disruption of trade routes within Somalia itself, especially since 1991. However, the number of animals exported and their value does not change markedly before and after 1991, when Somalia had and subsequently lost a functional national government.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

It is instructive to examine the Somali livestock export industry at different points in its volatile history. Articles by Reusse (1982) and Swift (1977, 1979) provide a capsule summary of developments up to about 1980. More recently, a series of publications by Peter Little and his colleagues (Little, Teka and Azeze 2001; Little and Mahmoud 2005; Little 2002, 2003, 2005 and 2006) analyses the situation since the late 1980s.

Background: Livestock Exports before 1991

The recorded involvement of Somali pastoralists in international trade went through three relatively distinct phases:

* In the early nineteenth century the northern Somali traded in exotic wild products (ivory, frankincense, ostrich feathers and gum arabic) that were gathered by the nomads but were not livestock products.

* The pattern of trade changed and the volume of trade expanded when the British occupied Aden and the North Somali coast. Hides and skins (bound for the United States and Europe) came to dominate the trade, which also included slaughter animals (which supplied the British garrisons with meat), and clarified butter. Pastoral produce was now being sold, but the wide range of products indicates that the nomads were marketing the by-products and surpluses of a subsistence-oriented system of production.

* Specialized, single-commodity commercial pastoralism only arose in the 1950s with the oil boom in Saudi Arabia and among the Gulf States. In response to rising demands for red meat and slaughter animals for Islamic festivals associated with the Hajj to Mecca, the Somali began to supply these markets with large numbers of small ruminants, most of which left from the northern Somali port of Berbera destined for Jeddah in Saudi Arabia. (The preceding points are based on Swift 1977: 285-87; 1979: 448, 449, 451.)

It would be difficult to overestimate the magnitude of the Gulf trade in live animals and its importance to the national economy of Somalia. According to Reusse:

* In some years in the 1970s Berbera in northern Somalia was the world's number one livestock shipping point.

* Until the Australians shifted to live animal exports in the late 1970s and eroded Somalia's market share, Somalia was the world's major sheep and goat exporter.

* As late as 1976 Somali exports valued at point of shipping constituted one-sixth by value of the world livestock exports (US$340 million).

* About a third of total livestock off-take was for export, and a third of bank deposits in one northern Somali town (Burao) were from nomadic pastoralists. (The above points are based on Reusse 1982.)

* In 1981 live animal exports produced 91 per cent of the country's hard currency (exclusive of remittances from workers living abroad) while the export of all animal products combined accounted for over 80 per cent of Somalia's foreign exchange in five of the seven years between 1975 and 1981 (Holtzman 1982: 9).

1991 to the Present

Since 1991 and the collapse of a centralized government, three cross-border trade routes have evolved. These are:

* the Kenya/southern Somalia border focusing on long-distance, export cattle trade to Kenya, especially the Nairobi market;

* the eastern Ethiopia/central Somalia border--based on local domestic trade in livestock and cereals;

* the eastern Ethiopia/Somaliland border--dealing in small stock export and the importation of food and consumer items (Little 2002:179).

There are several reasons why the livestock trade has been able to survive and expand despite the collapse of the state. Mubarak (1997) estimates that the pastoral sub-sector, which generated more than 80 per cent of annual exports, received only six per cent of public expenditure from 1974 to 1988. Since they had received so little government support, herders suffered less disruption when state services were withdrawn (Little 2005). Moreover, pastoralists were mobile and could move away from conflict, and trekked trade livestock were less dependent upon the deteriorating road network than forms of trade that relied on mechanized transport to move goods to market.

Along the Kenya/southern Somalia border, cattle, consumer electronics and clothes are exported from Somalia, which imports qut (a stimulant), maize, wheat flour, tea and sugar from Kenya. The livestock component of this trade involves 'medium to high quality male and female [cattle], which are used for slaughter in Kenya's major urban centres and for restocking and breeding purposes on commercial ranches in the Rift Valley' (Little 2002: 181). Cattle from the Mogadishu and Lower Shebelli regions are trekked long distances, in some cases more than 450 km. Aklilu (2002) estimates that this trade provides 26 per cent of the beef consumed in Kenya; Little and Mahmoud (2005) estimated that 16 per cent of beef consumed in Nairobi was provided through this channel, or annually about 73,000 head of cattle, 75 per cent of which end up in the terminal market of Nairobi (Little 2002: 184). Garissa market, the main entrepot for this trade route, generated over US$15 million in cattle sales alone in 1998 and was the largest cattle exchange in Kenya outside of the urban markets of Nairobi and Mombasa:

   This amount of annual revenue compares favourably with some of
   Kenya's major coffee and cash crop-producing districts, a
   phenomenon that is rarely acknowledged in Kenya's official economic
   reports. Most policy directives generally undervalue and neglect
   the importance of the pastoral sector to the national economy
   because it is poorly understood and does not contribute much to the
   state's foreign exchange coffers. (Little 2002: 185)

The eastern Ethiopia/central Somalia market route is centred on the western Somali town of Belet Weyne. Maize, sorghum, wheat flour, pasta and imported consumer goods flow into Ethiopia along this route, in return for coffee, goats and sheep, camels and kerosene from Ethiopia. The road linking Belet Weyne to Mogadishu is badly damaged, livestock prices are low, and much of the pre-1991 trade along this route has probably been redirected to other channels. Teka, Azeze and Gebre-Mariam (1999) nonetheless estimated that more than 50,000 head of Ethiopian livestock were informally exported to Kenya, often transiting through Somalia.

Goods traded along the eastern Ethiopia/Somaliland border include wheat flour, pasta, sugar and rice leaving Somaliland in return for sheep, goats, cattle, camels, maize and charcoal from Ethiopia. Berbera in Somaliland and Bossaso in Puntland are at the centre of this trade, which is destined for the Arab Gulf states. Little reports that the trade in sheep and goats out of Berbera at the end of the 1990s was larger than during the pre-1991 years (2002: 194), which were disrupted in the late 1980s by insecurity and a Saudi ban on Somali livestock imports. According to export figures for Somalia (including Somaliland), in 1998 Somalia accounted for 95 per cent of all goat exports and 52 per cent of sheep exports for the eastern African region (Zaal and Poderman 2000:17-18, cited in Little 2002); Somaliland was the regional centre for small ruminant exports. Since 50 to 60 per cent of the small stock leaving Berbera originate in Ethiopia, this trade made an important contribution to the Ethiopian as well as the Somaliland economy.

Little argues that cross-border trade is essential to regional food security, with the export of animals financing the importation of essential foodstuffs such as rice, wheat flour, cooking oil and pasta. When the export of animals declined or was interrupted, food supplies dwindled and became expensive (2002: 198-99). The contribution of cross-border trade to pastoral livelihoods was also diminished by the relatively low proportion of the final sales price that accrued to pastoralists, estimated at between 46 and 65 per cent of final revenue (Little 2006: 181). What was needed were more open, integrated and competitive markets. Instead, in government circles 'trans-border commerce often is still portrayed as smuggling and illegal and consequently remains subject to disruptive border closures and animal confiscations' (Little and Mahmoud 2005). These efforts at policing are generally ineffective, simply pushing the trade into new routes, temporarily suppressing it, or simply adding to the risks and costs incurred by traders to avoid official interference. The decline of pointless policing and rent extraction by officials in Somalia after 1991 may partially explain why the Somali export livestock trade has flourished without the benefit of a functioning unitary state. The liabilities of an actively inept and extractive government are amply illustrated in the next case study.

Ethiopia: Government Suppression of Livestock Trading and Exportation

The importance of pastoralism in Ethiopia depends on the criteria of assessment--land area, human population size or animal numbers. Pastoralists are a minority in Ethiopia. According to the International Livestock Research Institute (ILRI), there are about five million people resident in Ethiopia's semi-arid rangeland areas, or about eight per cent of the total national population (ILRI 2002). However, this pastoral population occupies a disproportionately large area and produces much more than its share of national livestock output. According to FAO (Food and Agriculture Organization) figures, pastures constitute 63 per cent of the agricultural land area in Ethiopia, while the Ethiopian Ministry of Agriculture estimates that pastoralists use 60 per cent of the land area of the country, but own 73 per cent of the nation's goats, 25 per cent of the sheep, 20 per cent of the cattle and the entire camel population (Aklilu 2002: 33). Ethiopia probably has the second largest ruminant population in Africa, after Sudan, though Ethiopian livestock statistics are at best approximations, with official figures unchanged for nearly thirty years and then adjusted upward by a constant rate for a decade (Aklilu 2002: 4).

In the late 1990s, livestock and livestock products provided about 10 per cent of Ethiopia's foreign exchange earnings, with hides and skins constituting about 90 per cent of the total livestock contribution (Table 1). According to data provided by the U.S. Embassy in Ethiopia, in 1998 the leather industry was second only to coffee as a foreign exchange earner, exporting $41 million worth of hides and skins to Europe, Asia and the Middle East (STAT-USA, no date). In contrast to leather exports, live animal and meat exports are modest, averaging between about 5 per cent and 14 per cent of the total value of livestock product exports. Animal and meat exports also fluctuate widely from year to year (Table 2). Some of the fluctuations in trade recorded in Table 2 result from what Aklilu calls 'inconsistent and at times self-defeating government policies' (Aklilu 2002: 41). Ethiopia has suffered two revolutionary regime changes in the last thirty years, with the installation of the Derg in the 1970s, then their ouster and replacement by the current regime in the early 1990s. In both cases, 'Ethiopia effectively exited itself' from export markets because each new regime quickly abolished the marketing and export policies of its predecessor, but then neglected to institute any replacement policies for several years, by which time the market had moved on and new suppliers had laid claim to Ethiopia's previous market share (Aklilu 2006: 187, 194).

Erratic policy has been compounded by weak support for the export sector. The poor provision by government of veterinary services illustrates the problem. Disruption to Ethiopian exports can be caused by veterinary embargos imposed on livestock by importing countries. For example, in 1998 and again in 2000 the Kingdom of Saudi Arabia, the largest consumer of livestock exports from the Horn of Africa, banned imports from Kenya, Somalia and Ethiopia due to outbreaks of Rift Valley fever. Following the first ban, annual exports from the port of Berbera in Somaliland, about half of which originate from Somali Region of Ethiopia, fell from over three million head in 1997 to 1.5 million in 1998, an estimated loss of US$100 million; and export levels were halved again in 2001 with the second ban (Devereux 2006: 57-59). The Ethiopian government has nonetheless been slow to create a unitary veterinary service that can vouch for the health and safety of Ethiopian exports, and it has never managed to create a disease-free zone for livestock quarantine prior to export. Stock routes and holding grounds destroyed in the late 1970s in the Ethiopia-Somalia war have never been repaired (Aklilu 2002: 8).

The lack of investment in livestock production and trade does not reflect the substantial economic contribution of this sector to the national economy:

   Livestock contribute about 40 percent of agricultural GDP or more
   than 20 percent of the total GDP (or even more if other
   intermediate values of livestock are properly assessed) ... A study
   undertaken by OAU-IBAR ... indicate that between 1993/4 and 1998/9
   the Government of Ethiopia allocated only 5 percent of its
   recurrent expenditures on agriculture and less than 0.3 percent on
   livestock or 3 percent of the recurrent agricultural expenses.
   (Aklilu 2002: 6)

The official export figures in Table 2 also do not represent the full extent of Ethiopia's involvement in international livestock trade, much of which is deemed by the authorities to be illegal. Based on interviews with Ethiopian government officials, the BBC reported that the 'smuggling' of live animals, hides and skins from Ethiopia into neighbouring countries cost the country's treasury an estimated $100 million each year' (BBC 2001). The movement of livestock through illegal channels reflects the difficulties and cost of moving them legally. For exporters there is wide variation in local and regional government fees and taxes depending on where animals are sourced and the jurisdictions through which they travel, 'a convenient misunderstanding of the Federal Acts to enrich the [Regional] State's coffer' (Aklilu 2002: 46). Even then, twelve different offices must be visited and fees paid to each in order to legally export animals (Aklilu 2006: 202). As with the export trade, one of the reasons domestic traders avoid official channels is the extractive nature of government's involvement in trading:

   Based on a three year cattle price data (1998-2001) collected by
   the City Bureau of Agriculture, the proportion of transaction costs
   and service fees at the [municipal market in Addis] including
   broker fees is equivalent to an astounding 17 per cent of the
   purchase price of a steer, 29 per cent of a Bullock/oxen and 27 per
   cent of a barren cow. (Aklilu 2002: 38)

Despite these impediments, the marketing of livestock has become fundamental to pastoral welfare and pastoral production strategies (Desta and Coppock 2005). One report on Somali Region of Ethiopia states:

   Over the last three to four decades the Somali pastoralist society
   experienced a significant change in the economy. Basically, it was
   a change from a livestock subsistence society, which lived mainly
   on milk and meat as staple foods complemented occasionally with a
   little grain, to a livestock export oriented market economy
   developing comparatively sophisticated trade links, where the
   marketing of livestock allowed both cash earnings and through
   bartering--the inflow of food- and non-food consumer goods from far
   away. (UNDP 1998)

For northwestern Ethiopia's Somali Region, the main port for livestock exports is Berbera in Somaliland and Bosasso, further east in Puntland (part of former Somalia). The three principal commodities imported through that port--sugar, rice and wheat flour making up more than three-quarters of all imports--reflect the importance of the livestock trade for household food supplies. When trade was interrupted in 1998, local food prices inflated rapidly and shops closed (UNDP 1998). Despite this dependence on markets and trade, Baars (2000) has demonstrated that milk production remains the most important source of revenue from these herds--66 per cent of total returns coming from milk versus 17 per cent from livestock sales. Baas estimated that in 1997 and 1998 the gross income from these herds was equivalent to US$6,382.

This dual dependence on livestock exports and domestic dairy production has proved relatively successful despite government harassment of cross-border traders. Based on the results of a consumption and expenditure survey conducted by the Ministry of Finance and Economic Development, Somali Region of Ethiopia had the lowest poverty levels of any region in Ethiopia (MoFED 2002, cited in Devereux 2004).

Botswana: Livestock as a Route out of Poverty

Ninety-nine per cent of Botswana's agricultural land is rangeland, in which live about 600,000 people or 47 per cent of Botswana's population (FAOSTATS and ILRI 2002). Eighty per cent of the country is covered by the Kalahari Desert; rainfall varies from an average maximum of about 650 mm in the northeast to an average annual minimum of 250 mm in the southwest. Seventy-one per cent of agricultural land is under communal or tribal tenure, 23 per cent is state owned, and six per cent is freehold lease for large-scale commercial ranching (Darkoh 1989).

Cattle were the traditional mainstay of the rural economy and were Botswana's leading export before independence in 1966 (Hubbard 1986). By the mid-1990s livestock accounted for three per cent of GDP and about 28 per cent of the total agricultural gross product (Panin 2000 quoting CSO 1995). In the 1990s the contribution of cattle to the average annual rural household income was estimated at 33 per cent and the contribution of small ruminants at 15 per cent (Panin and Mahabile 1996).

U.K. and European export markets for processed beef were developed before independence by the British Colonial Development Corporation, which invested in the construction of a modern abattoir and in commercial ranches. After independence, the Botswana Meat Commission, a parastatal, was given monopoly responsibility for meat exports. Following the discovery of diamonds in the early 1970s, the importance of beef to national export earnings has been gradually replaced by minerals, but cattle have remained important for the rural economy. There is no doubt that cattle owners, especially large herd operators who can take advantage of export markets, have received generous government support, possibly up to half of production costs, making the livestock sector artificially attractive. According to Cullis and Watson, this support has taken several forms:

* largely free veterinary services supported by general tax revenue;

* direct subsidies for the purchase of bulls, artificial insemination, and borehole drilling;

* indirect subsidies in the form of low-interest loans and tax breaks;

* artificially low rents for state leases on government-allocated commercial ranches;

* the continuation of dual grazing rights, allowing ranchers to move their livestock back onto communal land at will;

* beef producer prices inflated by the Botswana Meat Commission's slaughter policy, which is geared to selling high-priced beef in quota-controlled export markets such as the EU, but ignores alternative low-priced markets for lower-quality meat. (Cullis and Watson 2004: 14, 15)

As a result of the government's supportive policies and the country's harsh climate, beef is essentially Botswana's only agricultural export (Figure 3).

[FIGURE 3 OMITTED]

The position of Botswana's livestock keepers is, therefore, substantially different from that of pastoralists in Ethiopia or Somalia, where governments (where they exist) often extract more in fees and taxes than they return to the livestock sector in services. Also, whereas the bulk of the livestock traded in the Horn of Africa are exported unofficially or 'smuggled', Botswana's beef trade is sanctioned by international treaties (initially the Lome Convention and currently the Cotonou Beef Protocol with the EU) and subject to intense official regulation to guarantee sanitary and health standards. Finally, whereas the Horn of Africa exports high volumes of low-priced live animals to regional markets, Botswana specializes in the export of chilled beef at premium prices to Europe.

For many decades in the twentieth century it appeared that Botswana's capital-intensive and technologically sophisticated approach to livestock exports was more profitable than that of the Horn. This difference is no longer so clear cut. In 1998/99 the Botswana Meat Commission posted its worst year since its formation in 1966 and only its fourth ever loss. Subsequently it has run a financial loss in every year since then except 2001. In 2003 it posted its poorest trading results since its inception (Stevens and Kennan 2005: 12). According to the analysis by Stevens and Kennan, the problem is caused by a combination of escalating export costs, flat prices and competition for beef supplies from Botswana's increasingly prosperous domestic consumers.

Stevens and Kennan conclude that the only long-term way to preserve Botswana's export beef sector is to increase the supply of marketed beef, to meet demand in both domestic and international markets. On the face of it, supply increases look possible. Botswana's cattle off-take rates (at 13 per cent) and average carcass weight (at 175 kg) are lower than in neighbouring South Africa (off-take 17 per cent and carcass weight 220 kg) and slightly lower than in Namibia (off-take 14 per cent and carcass weight 180 kg).

Existing evidence suggests, however, that increasing commercial off-take may not be feasible because the costs of such an increase to other forms of livestock production may be unacceptably high for many livestock owners. Support for this conclusion comes from comparative studies of the economics of goat and cattle production in Botswana. These studies showed that the major component of the total gross revenue for both cattle- and goat-rearing enterprises was milk, a commodity that was predominantly consumed by livestock-owning households rather than commercially traded. Milk output accounted for 70 per cent and 67 per cent respectively of the value of total cattle and goat output, while sales of animals accounted for 23 per cent of the gross value of goat and cattle enterprises (Panin 2000). If increased meat production for sale impinged upon milk production for immediate consumption (by deflecting milk from human to calf growth), then higher rates of animal off-take would probably be neither attractive nor economically prudent for some livestock owners, depending on the costs of replacing home-produced milk with a purchased food alternative.

These studies also explain in part why Botswana's goat population has grown while cattle numbers have fallen since the early 1980s. Goats provide a return on capital investment that is roughly comparable to cattle (28 per cent versus 30 per cent) but at a fraction of the capital cost of investing in cattle keeping.

   The estimated capital invested in cattle per average household was
   five times that for goats. This confirms the widely held view that
   cattle rearing is capital intensive, and may explain why
   considerable numbers of smallholder farmers do not own any cattle
   ... Since capital is the limiting factor for production, it is
   reasonable to suggest that rearing small ruminants is more
   practical for many smallholder farmers in Botswana. (Panin 2000:
   195)

The ownership of goats is less skewed than cattle ownership. In one study area 85 per cent of the households reared goats while only 40 per cent kept cattle (Panin 2000).

The challenges of increasing commercial cattle output are also evident if we examine the comparative performance of large and small cattle herds in communal areas. Summarizing studies available in the mid-1980s, Behnke (1987) concluded that large cattle herds were operated on a commercial basis and small herds were managed in order to meet family subsistence needs. In this case, increased commercial involvement did not mean that large herds could be managed more profitably per head or even that animals could be sold at a higher rate. Because of the intensive way they were used, small herds maintained rough productive parity with large herds on a per animal basis; because of their poverty, small subsistence-oriented producers actually sold under duress a greater percentage of their herd than did large commercial operators. Commercialization by larger herd owners simply meant that purchased inputs displaced domestic labour in the production process, and that specialized single-commodity production replaced the production of a diverse array of goods for home consumption. Subsistence production was advantageous to small herd owners because in-kind animal produce consisted largely of replenishable products. By concentrating on live-animal produce, small herd operators hoped both to maintain a reliable income and maximize long-term herd growth. Only by achieving herd growth was it economic for small owners to adopt recommended modern practices, which required them to first spend money before they could make money (Behnke 1987).

We are now in a position to identify a final benefit derived by rural Tswana from livestock they provide a route out of poverty, in several stages. Figure 4 contrasts the different strategies of environmental exploitation employed by the richest and poorest residents of one communal area in Botswana (Kerapeletswe and Lovett 2001: Table l, p. 7). Grazing is the least important source of income for the poorest and the most important source of environmental income for the richest. The poorest of the rural poor have no livestock and depend for support largely on gathering wild produce and on public and private transfers of cash. The rich have significant livestock wealth, which provides direct benefits and the means to engage in arable production (Kerapeletswe and Lovett 2001). Because they require little start-up capital, goats help poorer households begin the process of stock accumulation (Panin 2000). As small livestock owners shift to cattle, the process of accumulation can be maintained by the labour-intensive extraction of live-animal products from small cattle herds, followed by regular commercial sales if herds can grow sufficiently to sustain regular off-take (Behnke 1987).

Zimbabwe: Livestock as Agricultural Inputs

Eighty-four per cent of Zimbabwe's agricultural land is pasture (FAOSTATS); more than six per cent of the population of Zimbabwe live in areas suitable for rangeland-based production systems based only on livestock (ILRI 2002). More broadly, in the mid-1990s 70 per cent of Zimbabwe's human population lived in communal or smallholder farming areas receiving less than 650 mm of average annual rainfall (Francis, Mudamburi and Chikwanda 1999). More than 85 per cent of these communal area farmers used animal draught power for tillage and transport. Oxen provided more than 75 per cent of this power, though cows and donkeys were becoming more important (Francis, Mudamburi and Chikwanda 1999).

Animal draught power is the key to understanding the economic importance of livestock to communal area residents. Table 3 estimates the relative economic importance of livestock to communal farmers in one semi-arid region. According to this recent study, nearly a quarter of cash and in-kind income is attributable to livestock.

Table 4 from the same study breaks total livestock output into its cash and subsistence components. According to this table, 66 per cent of total livestock output is provided by transport and draught power services, only a small fraction of which is monetarized. If we add the value of manure on to this figure, nearly three-quarters of the output of communal area cattle is in the form of inputs into arable agriculture. Cash income from meat and animal sales constitutes only 4 per cent of the total income from livestock.

Livestock in the communal sector, and cattle in particular, do not primarily provide goods for immediate consumption or sale; they instead support arable production which is (if we combine the income from gardens with dryland cropping) the single most important source of household income (Campbell et al. 2002).

These conclusions are supported by Barrett (1992), who reviewed the literature on communal livestock production up to the early 1990s (Table 5). Barrett's estimates of the relative importance of different kinds of livestock outputs are remarkably similar to those of Campbell et al. (2002), despite the decade that separates the studies. According to Barrett, about 64 per cent of cattle output is in the form of draught power; Campbell et al. (2002) give the comparable figure of 66 per cent. Barrett found that milk constituted 14 per cent of output by value, whereas Campbell et al. estimate 15 per cent. Based on these estimates, Barrett calculated that the total annual gross value of output per animal was Z$210. In the aggregate Barrett calculated that the Zimbabwe communal herd contributed roughly Z$800 million (or about US$250 million) to the national economy at 1991 prices.

A similar attempt to calculate the average value of cattle output was undertaken by Scoones (1989), based on prices in the late 1980s (Table 6). Based on the estimates in Table 6, Scoones calculated that the average per annum output of a beast over a lifetime of eight years was Z$353 for a female and Z$140 for a male, i.e., values that bracket Barrett's estimate of Z$210 for a beast of unspecified sex.

These calculations are important for understanding the relatively insignificant role of cattle sales and slaughter in communal area production systems. According to Barrett, quoting official statistics from the 1960s and 1970s, the combined sales and slaughter off-take rate from communal herds was only six per cent, a figure broadly in line with later estimates by other authorities. The contrast with the commercial ranching sector in Zimbabwe is clear:

   Annual slaughter offtake from the commercial herd has been in the
   range of 15 to 23 per cent, while from the communal herd it has
   been less than three per cent ... In the decade since independence
   the commercial cattle herd has contracted significantly while the
   communal herd has increased ... This has caused a significant
   reduction in the supply of slaughter stock to the meat industry,
   leading to regular meat rationing on the domestic market and public
   pressure to reduce exports. (Barrett 1992)

In light of the preceding discussion, low off-take rates do not necessarily indicate that communal area livestock producers are economically irrational, conservative or inefficient. They simply produce a broad range of goods and services from their herds, but especially traction and transport inputs that are essential for their primary economic activity--arable agriculture. In this agro-pastoral production system, high rates of off-take would be a dysfunctional dissipation of capital and of arable productive capacity, unless there existed a surplus of plough oxen. However, in the late 1990s only about 30-45 per cent of communal area farmers had four or more oxen, which they considered to be adequate animal draught power. Those without a capacity to plough on their own hired animals from other farmers, but were frequently forced to plough late and accept poor crop yields (Francis, Mudamburi and Chikwanda 1999).

Communal livestock management practices that emphasize in-kind live-animal outputs at the expense of terminal outputs (such as sales and slaughter) would appear to do a reasonable job of maximizing agricultural incomes despite the overcrowding and poverty that prevail in the communal areas. In the same district where communal cattle owners were realizing an average annual gross return per beast of Z$140-350, Scoones estimates that commercial beef cattle producers reaped about Z$10 per beast per year (Scoones 1989: 7). Barrett (1992) concluded that communal area livestock husbandry generated on the order of Z$30 per hectare, as compared to Z$20 per hectare for commercial ranches.

In sum, the 'communal' or 'traditional' livestock production systems of Botswana and Zimbabwe can be characterized as poor, intensive and (especially in Zimbabwe) market averse. Total output per beast is high, but only because herds are small, their owners have few alternative employment opportunities, and they can afford to devote relatively lavish amounts of labour to extracting a wide range of products and services from their animals. Herd owners also tend to concentrate on in-kind production of goods for home consumption rather than commodities for sale, and they do this for sound economic reasons. Home-produced products that are consumed directly have a value comparable to their retail price, which is what it would cost householders/herd owners to purchase replacement or substitute goods. Production of commodities for sale, on the other hand, is discouraged because these goods generally fetch a much lower farm gate price. As long as households have non-agricultural sources of cash income (typically through remittances, employment, the informal economy or state social security), it pays them to use their agricultural output to conserve scarce cash resources rather than earn money through sales.

These production systems employ generous amounts of labour but few purchased inputs, produce a broad spectrum of live-animal products, and are oriented to satisfying household consumption needs. They appear superficially to be archaic survivals of traditional subsistence pastoralism, and are often mistakenly interpreted as such. In fact this assessment is accurate only to the extent that these production systems preserve indigenous techniques of animal husbandry. From an economic perspective, these are modern adaptations to a specific niche within industrializing economies: survival on the semi-arid margins of national economies characterized by industrial food supplies and limited or low-wage employment opportunities for unskilled labour.

South Africa: Maintaining a National Labour Force

Eighty-four per cent of South Africa's agricultural land is pasture (FAOSTATS); over 6.3 million people or about 16 per cent of South Africa's total population live in rangeland areas (ILRI 2002). The country is a net importer of livestock products, including large numbers of live animals from neighbouring Namibia (Perry et al. 2005). South Africa also exports some livestock products, particularly into high-value niche markets, and possesses the most technically advanced and sophisticated marketing operation of any livestock-exporting country covered in this survey. The South African Meat Industry Company (SAMIC) was created in 1997 to promote the red meat industry. Its subsequent achievements include winning tariff-free exports to the U.S.A., re-establishing beef exports to Saudi Arabia after a rinderpest scare in 2000, and promoting premium-quality brands of meat products:

   Branding as a marketing tool is becoming increasingly important in
   the South African meat industry and several suppliers have
   successfully developed niche markets for their branded products ...
   Currently a range of brands are audited by SAMIC including
   Woolworth's 'Free Range' meats, Pick 'n Pay's 'Country Reared' beef
   and lamb and the Kalahari Kid Corporations 'Desert Lamb'. (Perry et
   al. 2005: 25)

As Perry et al. note, 40 per cent of the national herd is held by emerging black farmers in a position to profit from selling their produce. But many farmers who reside in South Africa's communal areas have little opportunity to profit from sophisticated marketing arrangements. This conclusion is supported by a recent government survey of agricultural production that compared output in the former black homelands and the former white areas of the country (see Table 7). According to this survey conducted in 2000, there are an estimated 698,000 farming operations keeping livestock in South Africa, 84,000 in the former RSA and 614,000 in the former homelands. Eighty-eight per cent of all farming operations that keep livestock are therefore in the former homelands, despite these farms producing only 18 per cent of the nation's meat, one per cent of its milk, and 14 per cent of the hides and skins. Quite clearly, these are very small livestock operations and they produce primarily for home consumption, 96 per cent of meat and 85 per cent of milk being used in this way. Only animal fibres--wool and mohair--are predominantly sold by communal farmers, who contribute a negligible amount of these commodities to national production (Table 7). To find out more about how these small, communal area farmers operate we must turn from national surveys to research studies of specific areas and communities. C.M. and S.E. Shackleton have been involved in many of these studies and in methodological debates about the valuation of rural in-kind production systems in South Africa (Dovie, Witkowski and Shackleton 2005; Shackleton et al. 2002; Shackleton, Shackleton and Cousins 2001; Shackleton, Shackleton and Cousins 2000; Shackleton et al. 1999). Conclusions to emerge from these studies include:

* Communal area residents hold livestock for a variety of reasons including 'cash from sales, a form of employment, milk for home consumption, for funeral purposes, as a form of investment, inherited the livestock, slaughter for feasts/home consumption, for paying bride-wealth, for sale of hides and skins, have land suitable for cattle farming, to help other, for cow dung and for draught/transport purposes' (Andrew, Shackleton and Ainslio 2003). The uses to which livestock are put shifts from locality to locality depending on circumstances and environmental conditions.

* 'The contribution of livestock has been underestimated in economic and livelihood security terms for several reasons, including a focus on productivity, limited consideration of non-monetised products or services, and a neglect of small stock such as goats or poultry' (Shackleton, Shackleton and Cousins 2000: 2).

* 'In Bushbuckridge, South Africa, farmers obtained a net annual value for livestock goods and services of US$765 per household for cattle owning households; US$79 pr household for goat owning households; and US$25 for non-owning households ... The net return per hectare from this and other studies in communal areas are approximately US$69 per year. By contrast standard valuations of communal livestock systems capture only one quarter of the direct use value, leading to the conclusion that they are unproductive and less efficient than commercial systems' Shackleton, Shackleton and Cousins 2000: 2).

* Adams, Cousins and Manona (2000) estimated the annual contribution of livestock to communal areas at R1,200 per household with a total national value of this sector at R2.88 billion per year.

In sum, livestock held by communal area producers do make a significant contribution to the national economy, but they do not primarily provide commodities for sale or export. The goods and services provided by communal area livestock are consumed at home or exchanged locally, for economic reasons identical to those given in the preceding analyses of subsistence-oriented livestock production in Zimbabwe and Botswana. What the communal areas do contribute to South Africa's wider national economy is people. The homelands were the labour reserves of colonial South Africa, designed to provide holding grounds for a supply of labour for other sectors of the economy. Working-age communal area residents still spend much of their time employed and living away from home, while their families live cheaply in the rural areas. When they are too old to hold down formal jobs, many of these workers also retire to the rural areas to care for grandchildren and engage in agricultural production to supplement their pensions. From the perspective of the national economy, the communal areas are a source not of commodities but of labour. And this labour is affordable because it is fed--typically when it is too young or too old to participate in the formal urban economy--by rural farming and herding. 'In the wider economic context ... [traditional livestock production] goes a long way to support the commercial sector in meeting the protein requirements ofthe population' (Dovie, Shackleton and Witkowski 2006: 269).

Namibia: A Research Deficit

Ninety-eight per cent of Namibia agricultural land is pasture (FAOSTAT), and 54 per cent of the national population resides in semi-arid rangeland areas (ILRI 2002). Ruminant meat production per capita is the highest of any African country covered in this survey--just under 50 kg per person in 2005 (FAOSTAT). On the usual southern African pattern, rural land in Namibia is held under three different kinds of title:

* freehold land occupied by fenced commercial ranches created for European settlers;

* unfenced communal land initially designated for various African ethnic groups;

* state land which is usually desert unsuitable for agriculture and leased to mining concessions.

At Independence in 1990 the distribution of land was heavily skewed, with more than half of the agriculturally usable land occupied by about 4,200 mainly white commercial farmers, and the rest occupied by 120,000 black rural households (Adams and Devitt 1992: 1). Government land redistribution programmes have subsequently somewhat altered this pattern (Werner 2003). The balance of livestock numbers held on commercial ranches and in the communal areas has nonetheless shifted dramatically since independence. By 2000 communal cattle outnumbered commercial cattle by almost 2 to 1, a reversal of the situation in the early 1990s (Institute for Public Policy Research 2002: 4).

The bulk of African-occupied communal land lies north of a veterinary quarantine cordon fence commonly known as the 'red line', which runs continuously from east to west and cuts off the northern quarter of the country. South of the line, livestock are certified as disease free and conform to sanitary standards that meet EU and South African import requirements. Namibia's northern border with Angola is open and livestock north of the quarantine fence are hot certified disease free and cannot be legally exported or moved to southern Namibia without undergoing quarantine. There are significant impediments to marketing northern livestock for export purposes:

   Because of the red line, animals in the northern communal areas go
   through cumbersome processes before they tan be auctioned. The
   livestock are quarantined for 21 days, during which farmers have to
   pay for their fodder, while after slaughter, the frozen meat is
   quarantined for a further 21 days. Farmers say not only do their
   animals lose body mass, but they also pay lots of money to
   transport their livestock to the quarantine facilities ...
   [Commercial] off take from these areas is less than three percent
   compared to an off take of 25 percent south of the fence. (Tjaronda
   2006)

The Namibian economy is diversified and depends upon a mixture of agriculture, mining, fishing and tourism. While the economic contribution of the livestock industry to this mix is modest but significant, its contribution to employment is much larger.

   The contribution of the agricultural sector to GDP is modest and
   has not exceeded 10 per cent since Independence. However, the
   economic impact of the sector is much larger than these figures
   suggest due to forward and backward linkages with the wider
   economy. In order to capture these linkages a multiplier of 1,8 is
   generally accepted. This means that the sector has contributed up
   to 18 per cent to the GDP since 1990. Approximately 70 per cent of
   the Namibian population depends on the agricultural sector in one
   way or another. (Werner 2003: 6)

By almost any standard--per capita ruminant meat production, the extent of rangelands, the growth of communal livestock holdings, contribution to GDP or employment--Namibia's developing pastoral sector is important and should constitute an important case study for this review. Unfortunately, the availability of data does not reflect the importance of Namibia's communal livestock sector.

Prior to Independence in 1990, warfare and insecurity discouraged research in northern Namibia's communal areas. Since 1990 there has been excellent work on the performance of the country's commercial ranching sector (for which there is good longitudinal data; Lange, Barnes and Motinga 1997), but--as best we can determine--no comparable work on indigenous African production systems. New work on livestock in the communal areas has instead focused on environmental impact assessment and on debates about the natural resource management capabilities of rural communities (for an introduction to this literature see the citations in Kuiper and Meadows 2002). Commercial assessments of particular livestock-based industries are also biased towards the commercial ranch sector that has traditionally supplied these industries (Directorate of International Trade 2001).

In short, questions about the economic performance of communal area agricultural systems seem to have been genuinely under-researched in Namibia. Bollig (2006: 44-46, 205-7) provides a brief but excellent account of livestock marketing among the Himba of northwestern Namibia. The picture that emerges is of a pastoral people dependent (like the Somali) upon trade for grain supplies and eagerly seeking marketing opportunities. Bollig's focus, however, is on the contribution of marketing to Himba welfare, rather than the contribution of Himba livestock to the general national welfare. By way of contrast, a short report by NOLIDEP-KFSR/E summarizes the literature up to 1997 on the contribution of livestock to the livelihoods of people in the Okavango Region. Okavango receives more rainfall than Kaokoland where the Himba live, and Okavango residents pursue very diversified household provisioning strategies, which include fishing, hunting, gathering, remittances, informal and formal employment and both arable agriculture and livestock keeping. Livestock sales rates are low, as they would be in comparable economies in the former homelands of South Africa. Commercial sales may be more important for large herd owners, on the Botswana pattern, but data on this point is insufficient.

The preceding synopsis of Himba and Okavango livestock production systems hints at the diversity of these systems in different communal areas of Namibia. This diversity has yet to be adequately documented. It would appear that Euro-American concerns about what African stock keepers do to their environment has set the research agenda on Namibian livestock and rangelands, and there has been much less interest in finding out what African stock keepers are able to extract from that environment.

This narrow environmental focus may turn out to be counterproductive. Water is arguably Namibia's most valuable and scarce natural resource, and natural resource accounting techniques can be used to analyse the economic returns to the consumption of water by different sectors of the Namibian economy. In the commercial sector, these calculations are precise and the results strongly suggest the need to rethink government policy (Lange 1998). Unfortunately, Lange could not achieve equal precision in her analysis of water use in the communal sector, where it was impossible to disaggregate the economic returns from livestock and from cropping, and her figures for the contribution of subsistence agriculture to GDP were based on flawed estimates supplied the national Central Statistics Office (Lange 1998). Debates about Namibia's communal area water policies must take place without reliable information on how much value livestock owners obtain from the water they use, and the environmental costs of flawed policies could be considerable.

Zambia: Extending the Frontier of Commercial Farming

Eighty-five per cent of Zambia's agricultural land is pasture (FAOSTATS), and 14 per cent of the national population lives in semi-arid rangeland areas suitable only for livestock production (ILRI 2002). Per capita annual milk and ruminant meat production is low, as is annual per capital consumption, estimated at 12 kg for milk and for meat between 11 kg (Hicks 1995, citing figures for 1993) to 2.5 kg (Sinyangwe and Clinch 2000), the latter figure being half the average level of African consumption. Since 1964 Zambia has been a net importer of milk and meat (Hicks 1995).

In terms of livestock ownership, the country is roughly divided between mostly southern areas occupied by ethnic groups (Tonga, Lozi, Ngoni and Mambwe) with a tradition of livestock management, and predominantly northern areas without a livestock background because a history of tsetse fly infestation. About 89 per cent ofthe national cattle herd is held in Southern, Eastern, Western and Central Provinces; the remaining 11 per cent is found in Northern, Northwestern, Lusaka and Luapula Provinces (Sinyangwe and Clinch 2000; see also the provincial breakdown of cattle populations in Mwenya, Mwenya and Dibbits 1994: 470).

Cattle in Zambia are subject to an impressive array of diseases: contagious bovine pleuropneumonia (reintroduced in 1997; Mangani n.d.), trypanosomiasis (, brucellosis (Sovjak and Matejickova 2002), bovine herpesvirus-1 (Mweene et al. 2003), Rift Valley fever (Samui et al. 1997), fasciolosis (Phiri et al. 2005) and anthrax (Mwale 2000). When veterinary services broke down in the early 1990s following the withdrawal of government support, cattle numbers declined sharply due to disease outbreaks, especially in the southern provinces where animal densities were high.

Agriculture contributes between 11 per cent and 16 per cent to total GDP in this mineral-based economy. Ninety per cent of domestic milk production for sale and 60 per cent of marketed meat for sale comes from commercial farms situated primarily around cities and along the line of rail on about a quarter of the total available arable land. However, about 83 per cent of Zambian cattle, 97 per cent of the goats and 64 per cent of sheep are held in the traditional sector (Sinyangwe and Clinch 2000). Agricultural officers and administrators have been complaining for at least half a century about low marketed off-take from traditional herds (Larson 1966) and continue to do so:

   Productivity of cattle in the traditional sector is characterised
   by slow growth rates (5-8 years to reach market weight), high calf
   and adult mortality rates (20-30 percent and 9 percent
   respectively) and low reproductive performance. National herd
   growth rate is estimated at 3 percent with an average off take of
   8-9 percent. In comparison, production ratios for the commercial
   sector feature low calf mortality (1-2 percent), high reproductive
   rates (65-70 percent) and an off take between 17-18 percent.
   (Sinyangwe and Clinch 2000: 2)

In a survey in 1995, over two hundred smallholders gave the following reasons for livestock keeping (see Table 8). Both goats and cattle are kept for cash; home consumption of goats is in the form of meat and from cattle in the form of draught power and milk. In Western Province, Baars, de Jong and Zwart (1996) found that cattle sales, home-consumed milk and ploughing were the most important benefits to cattle keeping. In light of the preceding southern Africa case studies in this review, Zambia appears to offer nothing novel on the topic of off-take rates versus household consumption and these issues are not further discussed. What is unique to Zambia is a century-long but continuing process of the geographical expansion of animal draught power for smallholder commercial agriculture.

Oxenization in Northern Zambia

At the beginning of the twentieth century all traditional crop agriculture in Zambia was done by hand with a combination of axes, hoes and fire. Following the example of settler farmers and extension work by Christian missions, the adoption of ox ploughing was complete by the 1930s in some southern areas with a long tradition of cattle keeping. The subsequent spread of animal draught power was slowed by the Depression, the Second World War and a vogue for tractorization in the 1960s and 1970s after independence (Wood and Milimo 1994). The current phase of work on animal draught power began following the collapse of copper prices in the 1970s and the subsequent demise of government-subsidized tractor ploughing schemes, and involved the expansion of oxen ploughing into northern Zambia where previously cattle were uncommon.

This expansion was made possible by rural population growth and corresponding decreases in the length of the fallowing period, which forced an increase in land use intensity in many localities (Loffler 1994). Northern Zambia was sitting on the cusp of a major agricultural transformation--from the hoe to the plough and from shifting cultivation to permanent fields (see Pingali, Bigot and Binswanger 1987 for a discussion of the evolutionary regularities in this process). The process of technical transformation has been retarded by several factors:

* The unavailability of cattle due to low rates of herd growth, massive losses due to disease following the withdrawal of government veterinary services (Kaoma-Sprenkels and Mwenda 2000), and (in a reversal of the usual official laments) the lure of quick profits at the local butchery (Lubumbe 1994).

* Low levels of animal management in communities that had no history of livestock ownership. Unpublished research by Behnke in the Chambeshi flood plain (Northern Province) in the late 1980s revealed high rates of stock mortality due to crocodile attacks, feral cattle that were dangerous to milk and difficult to plough with, and low calving rates due to the total absence of bulls in some village herds, all the males having been castrated for ploughing by owners who assumed that someone else would leave a bull whole for breeding.

* Low farm gate prices for staple crops. Farmers in northern Zambia are adopting plough agriculture to produce surpluses for commercial sale, not primarily for home consumption. The inability of Zambian consumers to pay farmers attractive prices for their produce directly undermines incentives for farmers to expand the cultivated area: 'Increased use of animal traction will depend largely on the positive growth of the national economy, and especially on reductions in inflation and interest rates' (Mwenya 1994: 472).

* The unavailability of suitable equipment and blacksmithing in areas where mechanization is beginning to take off and demand for such services is low (Dibbits 1994).

There have nonetheless been considerable benefits from the adoption of oxen draught power, including increases in farmer income (Kaoma-Sprenkels and Mwenda 2000: 88) and the doubling of cash crop areas under cultivation per farm (Loffler 1994: 358). In a study in a maize growing area of Southern Province, Smith (2008) found that the number of cattle owned by a farming household was the single strongest determinant of maize yields because cattle ownership permitted timely ploughing and planting. In a summary of the results of a national survey of animal traction, Dibbits (1994) reported that small and medium-scale farmers in Zambia prepared about 46 per cent of their cultivated area by hand with hoes and 54 per cent with oxen, with a small and unknown contribution from tractor ploughing. Between 1985 and 1990 the estimated number of trained oxen had grown by 48 per cent to 266,000 head, which ploughed a total area of 468,000 ha. For the preceding decade 1976-86, Wood and Milimo reported nearly a doubling in trained oxen numbers, from 90,000 to 179,000 (1994: 347).

Oxenization would therefore appear to have an immediate appeal to smaller farmers, to increase commercial crop production, and to provide the basis for a permanent transformation of the agricultural systems of the north of Zambia. Whether the animals that power these changes come from herds managed on a pastoral basis is an open question. Though most of the original breeding stock came from adjoining pastoral regions of Zambia, oxen distributed to isolated farms by donor-funded projects may be best described as farm animals. However, this is probably not an accurate description of how cattle are kept in communities where oxenization is occurring spontaneously without outside financial assistance. In these villages animals may be held in collective village herds, individually kraaled only when they are being worked, fed on natural forage and standing crop residues, and moved seasonally short distances up- and down-slope to avoid flooding and obtain fresh vegetation. These are 'range-based' husbandry systems that are situated in areas that can receive 1,000 mm of rain per year. Evaluation of the importance and organization of indigenous oxenization is, to the best of my knowledge, under-researched. Most published reports are on donor projects that are adept at self-promotion but may contribute less to long-term change than spontaneous initiatives undertaken by the farmers themselves.

Conclusions

It is possible in principle to quantify pastoral economic outputs, but published national-level data sources do not routinely distinguish between livestock products coming from pastoral and non-pastoral sources. Disaggregating this national-level data requires the use of local and regional studies to estimate the proportion of total output and the kind of output that comes from the pastoral sector. Except for the communal rangeland areas of Namibia and for Somalia, this review has located field-based estimates of the productivity and economic costs and returns to pastoral herds in all of the countries covered in this paper. Especially for Botswana, Zimbabwe and South Africa, the number and quality of localized studies suggests that it should be possible to derive accurate estimates of the contribution of indigenous African stock keeping to national accounts.

A variety of different kinds of evidence also suggest that these calculations would force a reassessment of the role of pastoralism in national economic affairs. Shackleton et al. argue that standard economic techniques can miss three-quarters of the direct use value of communal livestock systems in South Africa (Shackleton, Shackleton and Cousins 2000). For a maize-producing area in southern Zambia, Smith has shown that cattle ownership is the single most important determinant of returns to cropping, more important even than fertilizer, because of its positive impact on the productivity of both land and labour (2008). In the Somali areas of eastern Ethiopia, milk is the primary source of benefit to pastoralists, providing 66 per cent of the value derived from livestock (Baars 2000). Home consumption of dairy products remains important for these herders despite their involvement in commercial livestock sales, on a scale that is indicated by the magnitude of the losses incurred when Rift Valley fever outbreaks interrupted exports to the Middle East. Between 1998 and 2003 from the port of Berbera in Somaliland (shipment point for eastern Ethiopian animal exports), the authorities lost an estimated US$45 million in reduced taxes and vessel docking fees, and private livestock exporters lost an estimated US$330 million in trade revenue (Soumare et al. 2006). Examining the issue from a different angle, Devereux found the lowest rates of poverty for any region in Ethiopia in the largely pastoral Somali Region (Devereux 2004, citing Ministry of Finance and Economic Development 2002). Finally, based on data collected primarily in the 1980s, pastoral productivity per hectare exceeds ranch productivity in comparable environments by 50 per cent or more in Botswana, Zimbabwe and Borana areas of Ethiopia (Barrett 1992; Cossins 1985; de Ridder and Wagenaar 1984).

In contrast to these results, the undervaluing of pastoralism in national accounts rests on two recurrent methodological flaws--underestimation of the cash value of subsistence production and incomplete recording of the different kinds of pastoral output.

Particularly in southern Africa, indigenous producers reap most of their livestock benefits from untraded outputs which are used as inputs into other productive activities (like manure or traction power in crop farming), are directly consumed by households (milk and dairy products), or accumulated as a store of capital (live animals). These are all benefits that do not have an obvious cash value. Ascribing an appropriate monetary value to these outputs becomes increasingly important when, as in southern Africa, they constitute a high proportion of total output and the incorrect imputation of 'shadow' prices would distort overall estimates of economic performance.

When households consume their own produce, its cash value should be estimated at retail prices and not farm gate prices. Retail prices accurately reflect how much it would cost the household to replace their own production with purchased food, and this is the relevant concern for those who do not, after all, sell the bulk of their produce but eat it or use it. Historic inequalities in land distribution, the small scale at which they operate, and the difficulties they face in producing marketable surpluses all contribute to the low level of involvement by southern African agro-pastoralists in commercial production. But their intimate involvement in the cash economy is, paradoxically, also part of the explanation. In these cash-poor communities, money tends to lose its function as a 'universal measure of value' and is used instead as a special-purpose commodity to obtain 'modern' goods and services--like paying school fees, or buying manufactured clothing and modern medicines. By their farming and herding, households use available, low-value domestic resources to produce food. Their objective is not primarily to make money but rather to conserve cash that is better earned in other ways and, then, best deployed to do other things than buy basic foodstuffs. In addition to undervaluing their output, pricing agricultural production on the pretence that it should be sold obscures the way these households selectively engage and disengage from market involvement, and their reasons for doing so.

Official unawareness of the full spectrum of pastoral output is a second, widespread methodological impediment to accurate estimates of the economic importance of pastoralism. Pastoralism contributes to national economies in at least four material ways: by providing inputs into farming, supplying food and other products for household provisioning, serving as a source of capital and savings for rural households, or generating marketed commodities for trade and export. It is only the latter benefit--marketed commodities--that routinely enters into national economic statistics, and even then only when the trading is done through officially recognized channels. In the sample of countries reviewed here, Ethiopia--where the authorities do not recognize the most important pastoral marketing channels as legitimate--provides an extreme example of official statistics rendered meaningless as a matter of policy. Similar but less dramatic distortions are produced by limiting estimates of pastoral output to livestock sales in larger markets where records are kept.

Aggregated national statistics, the selective recording of data and analytical shortcomings are all part of the problem. Specialized studies of the economic value of pastoralism provide an important corrective to flawed national statistics, and will likely continue to do so for some time.

Acknowledgements

This paper is based on research conducted initially for the IUCN livestock economics and marketing study EARO/76572-000/999 and the World Initiative for Sustainable Pastoralism, whose financial support is gratefully acknowledged. The opinions expressed here are solely those of the author.

References

Adams, M., B. Cousins and S. Manona. 2000. 'Land Tenure and Economic Development in Rural South Africa'. In At the Crossroads: Land and Agrarian Reform in South Africa into the 21st Century, ed. B. Cousins. NLC & PLAAS, University of the Western Cape, Cape Town, pp. 111-28.

Adams, M.E. and P. Devitt. 1992. 'Grappling with Land Reform in Pastoral Namibia'. Pastoral Development Network Paper No. 32a. Overseas Development Institute, London.

Aklilu, Y. 2002. An Audit of the Livestock Marketing Status in Kenya, Ethiopia and Sudan, Volumes I and II. OAU/Interafrican Bureau for Animal Resources.

--. 2006. 'A Review of Policies and Their Impact on Livestock Trade in Ethiopia during Three Regimes (1965-2005)'. In Pastoral Livestock Marketing in Eastern Africa: Research and Policy Challenges, eds J.G. McPeak and P.D. Little. Intermediate Technology Publications, Rugby, pp. 187-202.

Andrew, M., C. Shackleton and A. Ainslio. 2003. Land Use and Rural Livelihoods: Have They Been Enhanced Through Land Reform? Policy Brief No. 5, PLAAS, University of the Western Cape, Cape Town.

Baars, R.M.T. 2000. 'Costs and Returns of Camels, Cattle and Small Ruminants in Pastoral Herds in Eastern Ethiopia', Tropical Animal Health and Production 32:113-26.

Baars, R.M.T., R. de Jong and D. Zwart. 1996. 'Costs and Returns of the Crop-Cattle System in the Western Province of Zambia', Revue d'Elevage et de Medecine V6terinaire des Pays Tropicaux. 49(3): 243-51.

Barrett, J.C. 1992. 'The Economic Role of Cattle in Communal Farming Systems in Zimbabwe', Pastoral Development Network Paper 32b. Overseas Development Institute, London.

BBC. 27 May 2001: 'Ethiopia Concerns over Animal Smuggling'.

Behnke, R.H. 1987. 'Cattle Accumulation and the Commercialization of the Traditional Livestock Industry in Botswana', Agricultural Systems 24: 1-29.

Bollig, M. 2006. Risk Management in a Hazardous Environment: A Comparative Study of Two Pastoral Societies. Springer, New York.

Campbell, B.S. Jeffery, W. Kozanayi, M. Luckert, M. Mutamba and C. Zindi. 2002. Household Livelihoods in Semi-arid Regions: Options and Constraints. Bongor, Indonesia: Center for International Forestry Research.

Cossins, W.J. 1985. 'The Productivity of Pastoral Systems', ILCA Bulletin 21: 10-15.

Cullis, A. and C. Watson. 2004. Winners and Losers: Privatising the Commons in Botswana. IIED, London.

Darkoh, M.B.K. 1989. 'Desertification in Botswana', RALA Report No. 200.

De Ridder, N. and K.T. Wagenaar. 1984. 'A Comparison between the Productivity of Traditional Livestock Systems and Ranching in Eastern Botswana', ILCA Newsletter 3(3): 5-6.

Desta, S. and D.L. Coppock. 2005. 'Linking Pastoralists and Exporters in a Livestock Marketing Chain: Recent Experiences from Ethiopia'. Research Brief 05-06-PARIMA. Global Livestock Collaborative Research Support Program. NDP Emergencies Unit for Ethiopia 'Cessation of Livestock Exports Severely Affects the Pastoralist Economy of Somali Region', Mission of 31 March to 7 April 1998.

Devereux, S. 2004. 'Food Security Issues in Ethiopia: Comparisons and Contrasts between Lowland and Highland Areas'. Seminar organized by UN-OCHA, Addis Ababa.

--. 2006. 'Vulnerable Livelihoods in Somali Region, Ethiopia'. Institute of Development Studies at the University of Sussex, Brighton.

Dibbits, H.J. 1994. 'Surveying Animal Traction Use in Zambia'. In Improving Animal Traction Technology, eds P. Starkey, E. Mwenya and J. Stares. CTA, Wageningen.

Directorate of International Trade. 2001. 'Supply Survey on Hides and Skins, Including Company Profiles'. Ministry of Trade and Industry, Windhoek, Namibia and the International Trade Centre UNCTAD/WTO, Geneva.

Dovie, D.B.K., C.M. Shackleton and E.T.F. Witkowski. 2006. 'Valuation of Communal Area Livestock Benefits, Rural Livelihoods and Related Policy Issues', Land Use Policy 23: 260-71.

Dovie, D.B.K., E.T.E Witkowski and C.M. Shackleton. 2005. 'Monetary Valuation of Livelihoods for Understanding the Composition and Complexity of Rural Households'. Agriculture and Human Values 22: 87-103.

FAOSTAT www.fao.org/faostat/

Francis, J., B. Mudamburi and B. Chikwanda. 1999. In Meeting the Challenges of Animal Traction, eds P. Starkey and P Kaumbutho. ATNESA, Harare and IT Publications, London.

Hicks, C. 1995. 'The Role of Zambian Cattle Populations in Socio-economic Development.' Livestock Research for Rural Development 7(2). http://www.cipav.org.co.lrrd/lrrd7/2/3.htm

Holtzman, J. 1982. The Economics of Improving Animal Health and Livestock Marketing in Somalia. Report for USAID Mogadishu.

Hubbard, M. 1986. Agricultural Exports and Economic Growth: A Study of Botswana's Beef Industry. Routledge and Kegan Paul, London.

ILRI. 2002. Mapping Poverty and Livestock in the Developing World. International Livestock Research Institute, Addis Ababa.

Institute for Public Policy Research. 2002. IJG Business Climate Monitor for May. Windhoek, Namibia.

Kaoma-Sprenkels, C. and N. Mwenda. 2000. 'Community-based Cattle Projects in Zambia: a New Approach to Oxenisation'. In Empowering Farmers with Animal Traction, eds P.G. Kaumbultho, R.A. Pearson and T.E. Simalenga. http://www.atnesa.org

Katakura, K., C. Lubinga, H. Chitambo and Y. Tada. 1997. 'Detection of Trypanosoma Congolense and T. Brucei Subspecies in Cattle in Zambia by Polymerase Chain Reaction from Flood Collection on a Filter Paper', Parasitology Research 83(3): 241-45.

Kerapeletswe, C.K. and J.C. Lovett. 2001. The Role of Common Pool Resources in Economic Welfare of Rural Households. University of York Digital Library of the Commons, York.

Kuiper, S.M. and M.E. Meadows. 2002. 'Sustainability of Livestock Farming in the Communal Lands of Southern Namibia', Land Degradation and Development 13: 1-15.

Lange, G.-M. 1998. 'An Approach to Sustainable Water Management in Southern Africa Using Natural Resource Accounts: the Experience in Namibia', Ecological Economics 26: 299-311.

Lange, G.-M., J.I. Barnes and D.J. Motinga. 1997. 'Cattle Numbers, Biomass, Productivity and Degradation in the Commercial Farming Sector of Namibia, 1915 to 1995'. Directorate of Environmental Affairs, Ministry of the Environment and Tourism, Windhoek, Namibia.

Larson, ED. 1966. 'Cultural Conflicts with the Cattle Business in Zambia, Africa', Journal of Range Management 19(5): 367-370.

Little, E 2005. 'Unofficial Trade When States Are Weak: the Case of Cross-border Commerce in the Horn of Africa'. Research Paper No. 2005/13 EGDI and UNUWIDER.

Little, ED. 2002. 'The Global Dimensions of Cross-border Trade in the Somalia Borderlands'. In Globalisation, Democracy, and Development in Africa: Future Prospects, eds Abdel Ghaffar M Ahmed. Addis Ababa, Ethiopia: Organisation for Social Science Research in Eastern and Southern Africa (OSSREA).

--. 2003. Somalia: Economy without a State. James Currey, Oxford and Indiana University Press, Bloomington, Indiana.

--. 2005. 'Pastoralism in a Stateless Environment: the Case of the Southern Somalia Borderlands', Geography Research Forum 25: 128-47.

--. 2006. 'Working across Borders: Methodological and Policy Challenges of Cross-border Livestock Trade in the Horn of Africa'. In Pastoral Livestock Marketing in Eastern Africa: Research and Policy Challenges, eds J.G. McPeak and P.D. Little. Intermediate Technology Publications, Rugby, U.K., pp. 160-85.

Little, P.D. and H.A. Mahmoud. 2005. 'Cross-border Cattle Trade along the Somalia/Kenya and Ethiopia/Kenya Borderlands'. Research Brief 05-03-PARIMA. Global Livestock Collaborative Research Support Program.

Little, ED., K. Smith, B.A. Cellarius and D.L. Coppock. 2000. 'Avoiding disaster: diversification and risk management among East African Herders', Development and Change 32(3): 401-433.

Little, ED., T. Teka and A. Azeze. 2001. 'Cross-border Livestock Trade and Food Security in the Horn of Africa: an Overview', Research Report of the Broadening Access to Markets and Input, Systems-Collaborative Research Support Program (BASIS-CRSP) and OSSREA.

Loffler, C. 1994. 'Transfer of Animal Traction Technology to Farmers in the North Western Province of Zambia'. In Improving Animal Traction Technology, eds P. Starkey, E. Mwenya and J. Stares. CTA, Wageningen.

Lubumbe, S.L. 1994. 'Transfer of Animal Traction Technology in Luapula Province, Zambia'. In Improving Animal Traction Technology, eds P. Starkey, E. Mwenya and J. Stares. CTA, Wageningen.

Lungu, J.C.N. n.d. 'Mixed crop-livestock production systems of smallholder farmers in sub-humid and semi-humid area of Zambia', www.naweb.iaea.org/nafa/aph/ public/reports

Mangani, M.P.C. n.d. 'Contagious Bovine Pleurophenumonia in Zambia'. Research and Specialist Services, Lusaka, Zambia. http://www.fao.org//docrep/007/y5510e/y5510e0y.htm

MoFED (Ministry of Finance and Economic Development, Federal Government of Ethiopia). 2002. Development and Poverty Profile of Ethiopia for 2002. Addis Ababa.

Mubarak, J.A. 1997. 'The "Hidden Hand" behind the Resilience of the Stateless Economy of Somalia', World Development 25(12): 2027-41.

Mwale, M. 2000. 'Anxieties over Anthrax'. Agriculturalist on-line http://www.independent.co.uk/news/World/Africa/2000-12/wildebeest201200.shtml Mweene, A.S., H. Fukushi, G.S. Pandey, M. Sayakalima, M. Simuunza, M. Malamo, A.

Nambota, K.L. Samui, T. Tsubota, Y. Kakazato, M. Onuma and J. Yasuda. 2003. 'The Prevalence of Bovine Herpesvirus-I in Traditional Cattle in Southern Province, Zambia'. World Organization for Animal Health, Scientific and Technical Review 22(3): 873-77.

Mwenya, E., W.N.M. Mwenya and J.J. Dibbits. 1994. 'Animal Draft Power in Zambia: Constraints to Development and Possibilities for Improvements'. In Improving Animal Traction Technology, eds P. Starkey, E. Mwenya and J. Stares. CTA, Wageningen.

NOLIDEP and KFSR/E. 1997. 'The Contribution of Livestock to Okavango Livelihoods'. Unpublished report. Windhoek, Namibia.

Panin, A. 2000. 'A Comparative Economic Analysis of Smallholder Cattle and Small Ruminant Production Systems in Botswana', Tropical Animal Health and Production 32: 189-96.

Panin, A. and M. Mahabile. 1996. 'Profitability and Household Income Contribution of Small Ruminants to Small-scale Farmers in Botswana', Small Ruminant Research 25: 9-15.

Perry, B., A.N. Pratt, K. Sones and C. Stevens. 2005. 'An Appropriate Level of Risk: Balancing the Need for Safe Livestock Products with Fair Market Access for the Poor'. International Livestock Research Institute. PPLPI Working Paper No. 23.

Phiri, A.M., I.K. Phiri, C.S. Sikasunge and J. Monrad. 2005. 'Prevalence of Fasciolosis in Zambian Cattle Observed at Selected Abattoirs with Emphasis on Age, Sex and Origin', Journal of Veterinary Medicine Series B 52: 414.

Pingali, E, Y. Bigot and H.P. Binswanger. 1987. Agricultural Mechanization and the Evolution of Farming Systems in sub-Saharan Africa. Johns Hopkins Press, Baltimore, Md.

Reusse, E. 1982. 'Somalia's Nomadic Livestock Economy: Its Response to Profitable Export Opportunity', World Animal Review (July-September): 2-11.

Samui, K.L., S. Inoue, A.S. Mweene, A.M. Nambota, J.E. Mlangwa, P. Chilonda, M. Onuma and C. Morita. 1997. 'Distribution of Rift Valley Fever among Cattle in Zambia', Japanese Journal of Medical Science and Biology 50(2): 73-77.

Scoones, I. 1989. 'Economic and Ecological Carrying Capacity: Implications for Livestock Development in the Dryland Communal Areas of Zimbabwe', Pastoral Development Network paper 27b. Overseas Development Institute, London.

Shackleton, C.M., S.E. Shackleton, T.R. Netshiluvhi, F.R. Mathabela and C. Phiri. 1999. 'The Direct Use Value of Goods and Services Attributed to Cattle and Goats in the Sand River Catchment, Bushbuckridge'. Unpublished report No. ENV-P-C 99003. Pretoria: CSIR [cited in Shackleton et al. 2000)]

Shackleton, S., C. Shackleton and B. Cousins. 2000. 'Revaluing the Communal Lands of Southern Africa: New Understandings of Rural Livelihoods', Natural Resource Perspectives No. 62, Overseas Development Institute, London.

Shackleton, S., C. Shackleton and B. 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 18:581-604.

Shackleton, S.E., C.M. Shackleton, T.R. Netshiluvhi, G.S. Geach, A. Balance and D.H.K. Fairbanks. 2002. 'Use Patterns and Value of Savanna Resources in Three Rural Villages in South Africa', Economic Botany 56: 130-46.

Sinyangwe, P.G. and N.J.L. Clinch. 2000. 'Increasing the Efficiency of Livestock Service Delivery--the Experience of the Livestock Sector in Zambia'. In Primary Animal Health Care in the 21st Century: Shaping the Rules, Policies and Institutions. www.eldis.org/fulltext /cape_new/mombasaconference/Sinyangwe&Clinch_IncreasingEfficiency.pdf

Smith, R.E. 2008. 'Cattle's Effect on Land and Labour Productivity: Evidence from Zambia', Journal of International Development. www.interscience.wiley.com Published on line July 2008 in advance of printing.

Soumare, B., E. Thys, D. Berkvens, A. Hashi and G. Van Huylenbroeck. 2006. 'Effects of Livestock Import Bans Imposed by Saudi Arabia on Somalialand for Sanitary Reasons Related to Rift Valley Fever', Outlook on Agriculture 35(1): 19-24.

Sovjak, R. and R. Matejickova. 2002. 'Main Problems of Brucellosis in Zambia'. http://www.tropentag.de/2002/proceedings/node160.html

Statistics South Africa. 2002. Report on a Survey of Large and Small Scale Agriculture.

STAT-USA. n.d. Market Research Report on the Leather Industry, prepared by the U.S. Embassy in Ethiopia and the U.S. Department of Commerce.

Stevens, C. and J. Kennan. 2005. Botswana Beef Exports and Trade Policy. Institute of Development Studies, University of Sussex, Brighton, U.K.

Swift, J. 1977. 'Pastoral Development in Somalia: Herding Cooperatives as a Strategy against Desertification and Famine'. In Desertification: Environmental Degradation in and around Arid Lands., ed. Michael H. Glantz. Westview Press, Boulder, Colorado. Pp. 275-305.

--. 1979. 'The Development of Livestock Trading in a Nomad Pastoral Economy: the Somali Case'. In Pastoral Production and Society. Cambridge University Press, Cambridge, pp. 447-465.

Teka, T. A. Azeze and A. Gebre-Mariam. 1999. 'Cross-border Livestock Trade and Food Security in Southern and Southeastern Borderlands of Ethiopia'. OSSREA Development Research Report Series No. 1, OSSREA/BASIS-CRSP Project on Cross-Border Trade and Food Security in the Horn of Africa, Addis Ababa.

Tjaronda, W. 2006. 'Namibia: Shifting the Red Line'. allAfrica.com, web posting.

UNDP Emergencies Unit for Ethiopia. 'Cessation of Livestock Exports Severely Affects the Pastoralist Economy of Somali Region', Mission of 31 March to 7 April 1998.

Werner, W. 2003. 'Land Reform in Namibia: Motor or Obstacle of Democratic Development'. Paper presented at a Meeting on Land Reform in Southern Africa: Motor or Obstacle of Democratic Development, Friedrich Ebert Foundation, Berlin.

Wood, A. and J. Milimo. 1994. 'Transfer of Animal Traction Technology in Zambia: an Historical Perspective'. In Improving Animal Traction Technology, eds P. Starkey, E. Mwenya and J. Stares. CTA, Wageningen, The Netherlands.

Note

(1.) The Somali Republic has an internationally recognized government with no effective control over much of the country. Somalialand, in what was once the British Protectorate of Somalialand, declared its independence from the central government in 1991, but is not recognized internationally. Puntland has also declared its independence from the central government but likewise is not recognized internationally.

Roy H. Behnke was trained in Islamic civilization and social anthropology at the Universities of Chicago and California, and has worked on problems of extensive livestock production and rangeland management for thirty years. He has lived and worked for extended periods in Libya, Iraq, Sudan, Somalia, Zambia, Namibia, Botswana, Kazakhstan and Turkmenistan. His major research has been on analyses of pastoral market involvement and the shadow pricing of subsistence livestock production, bio-economic theories of indigenous pastoral tenure systems, and the ecology of grazing systems in areas of low and erratic rainfall. His current research focuses on the decollectivization of pastoral production in ex-Soviet Central Asia and impacts of enclosure on the Tibetan plateau in China. He is a Research Associate with the Macaulay Institute, Aberdeen, U.K. and belongs to Odessa Centre Ltd. U.K., which carries out research and consultancy on rangelands and extensive livestock production.

Email: roy_behnke@msn.com

Table 1: Share of Livestock and Livestock Products in Foreign Exchange
Earnings

          % contribution    % contribution   % contribution
           of animal and     of hides and      of meat to
          animal products      skins to      animal exports
          to total export   animal exports

1995-6         12.38              96               4
1996-7         10.47              91               6
1997-8          9.36              90               8
1998-9          8.43              87               11
1999-00         8.43              86               10

          % contribution
          of live animals
             to animal
              exports

1995-6           0
1996-7           3
1997-8           3
1998-9           2
1999-00          4

Source: Aklilu 2002: 40.

Table 2: Ethiopian Live Animal and Meat Exports 1993-2004

       Cattle--     Goats--     Sheep--       Beef       Goat
       head live   head live   head live   and veal--   meat--
                                             metric     metric
                                              tons       tons
1993      5,600           0       8,800          0           0
1994      2,250       5,293      17,975         15          69
1995         25           0      22,800         73         243
1996          0           0       4,000         73         243
1997        800           0       4,000         15       1,490
1998      1,218       1,896      15,515         47       2,302
1999        549       1,300      30,704          1       1,818
2000        326         160      39,960          0       1,149
2001         44           0      15,000          0         222
2002        544           0       1,140          8         879
2003      2,217       3,080      11,706          0       2,094
2004     56,658       4,080      53,348        177       2,094
2005    267,596      13,636      38,104        --          --

         Mutton
       and lamb--
         metric
          tons
1993         33
1994        124
1995        259
1996        259
1997        312
1998        155
1999         87
2000         13
2001         20
2002        184
2003      1,501
2004          8
2005        --

Source: FAOSTATS

Table 3: Household Income by Source: Masvingo Province, Zimbabwe

Income source               % contribution to average household
                            income in cash and kind--total 100%

Wages and home industries                    12
Remittances                                  21
Gardens                                       8
Woodlands                                    15
Dryland crops                                21
Livestock                                    23

Source: Campbell et al. 2002

Table 4: Average Cash and Subsistence Gross Income from Livestock
per Household per Year ($Z)

               Subsistence    Cash    Total    % contribution
                 income      income   income      to total
                                                 livestock
                                                   income

Manure             382          0       382           7
Milk               810         48       858          15
Meat               457         14       471           8
Transport         1062         36      1098          19
Draft             2588         86      2674          47
Hides               16          0        16           0
Animal sales         0        228       228           4
Overall           5318        412      5727         100

Source: Campbell et al. 2002

Table 5: Estimate of the Total Annual Economic Output from a Herd of
100 Communal Cattle

Draught            Number of    Unit of       Annual       Total value
                    animals     output     output 1991Z$

Draught                40      Days work       1,200           13,404
Milk                   30      Lines           3,375            2,869
Manure                100      Cartload          1.5            1,800
Local offtake           3      Kg meat           480            1,200
Commercial sales        2      Head              2.5            1,000
Herd growth             2      Head                2              800
Total                                                       Z$ 21,073

Draught               % of
                   total value

Draught                63.6
Milk                   13.6
Manure                  8.5
Local offtake           5.7
Commercial sales        4.7
Herd growth             3.8
Total                 100

Source: Barrett 1992

Table 6: The Economics of Livestock Production in the Communal Areas

Activity          Z$ per working beast per annum

Ploughing               112
Manure                   10
Milk                    290 per cow
Calf production         105 per cow
Work                     68
Sale for beef           350 (at age 8 years)

Source: Scoones 1989

Table 7: The Production and Use of Livestock Products and Maize
Meal in South Africa

Products                     Total           Total       % homeland
                           production      production    production
                        former homelands   former RSA   kept for home
                             x 1000          x 1000      consumption

Maize meal kg                 53965           233274          98
Meat kg                       30691           139454          96
Milk litres                   25665          2955134          85
Butter kg                       107              101         100
Other dairy
  products kg                    20              116         100
Wool kg                        2645            57990           1
Mohair kg                        53            16182           3
Hides and skins
  number                         86              520          64
Other animal products            32               16          99

Products                  % homeland
                        production in
                        total national
                          production

Maize meal kg                 19
Meat kg                       18
Milk litres                    1
Butter kg                     51
Other dairy
  products kg                 15
Wool kg                        0
Mohair kg                      0
Hides and skins
  number                      14
Other animal products         66

Source: Statistics South Africa 2002, calculated from Tables
8.3.2 and 8.3.3.

Note: 'Former RSA' refers to the whites-only area of the old
apartheid state and therefore excludes black homelands.

Table 8: Reasons for Livestock Rearing, Percent of Respondents
Surveyed

                        Cattle   Goats

Cash income              30.5     37.5
Consumption               2.8     35.1
Draft power              33.3      0.2
Manure                    5.5      8.5
Milk                     12.9      2.0
Bride wealth payments    10.8      1.3
Social occasions          3.5     14.8
Social status             0.8      0.7

Source: Lungu n.d.: 138

Figure 4: Environmental Incomes from One Communal Area in Botswana

Environmental income--poorest 20%

livestock
 grazing           8%
building
 material         17%
crafts            18%
firewood          21%
wild fruits
 and vegetables   36%

Environmental income--richest 20%

wild fruits
 and vegetables   9%
livestock
 grazing          76%
building
 material          6%
crafts             2%
firewood           7%

Source: Kerapeletswe and Lovett 2001

Note: Table made from pie chart.
Author Advanced search

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.