Previous Page Table of Contents Next Page

Policy issues in livestock production in arid regions and the management of extensive grazing lands

by T.T. Treacher


Arid and semi-arid regions may be defined as areas where rainfall, relative to the level of evapotranspiration, is inadequate to sustain reliable crop production (eg Meigs, 1953) These areas are covered by grasslands, shrublands, savanna, semi-arid woodlands or desert. Kassas (1975) estimated that 43% of the world's surface is arid and Harrington (1981) suggested that in 1973 more than 40% of the world's population of sheep, 30% of goats and 25% of cattle were found in the arid zone.

Exploitation Systems

It is important to emphasise that the exploitation of extensive arid areas is concerned with land use and not the production of crops. Social organisation, ownership and access have profound effects on grazing management. There are four main systems of exploitation defined by the movements of the flocks and the extent to which the herders are sedentary.

Nomadic systems: These systems are found in desert and desert fringes, where rainfall is extremely erratic and the flocks and their herders move to wherever forage is available, with no set seasonal patterns. True nomadic systems are becoming rare.

Transhumant systems: Characterized by regular, seasonal movements of the flocks between grazing areas, often at different altitudes. In west Asia and north Africa flocks often move into the cultivated areas to utilise stubbles and by-products. The many changes that have occurred in transhumant systems recently are discussed below.

Semi-sedentary systems: These systems are found mainly in Sub-Saharan Africa, where herders live in non-transportable dwellings but abandon them, if necessary, to move to other areas at times of feed shortage.

Sedentary systems: Extensive grazing lands, where sedentary systems are practised, occur mainly in the developed countries of the Americas, Australia and southern Africa, where properties have boundary fences and are often divided into fenced paddocks. In other areas an important, but declining, area of extensive grazing land close to villages is exploited by the flocks of village farmers. Now, with the availability of trucks to transport water, sedentary systems are extending rapidly into large areas of extensive grazing land that formerly were only grazed seasonally because drinking water was not available for large periods of the year.

Private ownership of extensive grazing land is only found in South America, southern Africa and parts of the USA. In Australia, arid rangelands are owned by the State governments and in the USA parts of the range are owned by Government agencies. The State, or Government agency, leases the land to graziers and, ultimately, they have the power to control the worst abuses of the land. Elsewhere, extensive grazing is generally an open-access or common-property resource, which may or may not have well defined regulations in relation to stock numbers and the duration and timing etc of their grazing. Regulation often existed on tribal lands or within tribal groups and, although it was often maintained under colonial rule, it has generally disappeared with independence.

It is widely assumed that degradation of extensive grazing areas is linked to over grazing and over exploitation of communally grazed areas, but it is clear that degradation has occurred and is occurring in areas where extensive grazing is owned or leased. There have been problems on leased land in Australia (Harrington et al., 1990) and on freehold range in USA (Stoddart et al., 1970). On privately owned ranches in Patagonia and Argentina stocking rates are now 25–30% lower than 50 years ago because overstocking caused degradation of the natural vegetation and serious erosion (Mueller, personal communication). Long-term sustainability of extensive grazing land can, therefore, be a problem under any system of land tenure.

The remainder of the paper will discuss the factors in ruminant production, predominantly sheep and goats, in west Asia and north Africa (WANA), which lead to the widespread degradation of the extensive grazing land in these areas.


The recent and detailed analysis by Boutonnet (1989) of the Algerian ruminant livestock industry, which is dominated by small ruminants, has been taken as a starting point in this discussion. The number of ewes in Algeria has risen from 3.9 million in 1966 to 6.1 million in 1976 reaching 9.5 million in 1986. Seven million of these ewes, together with cattle, goats, horses and camels equivalent to 3 million ewes, are kept in the steppe. Between 1971 and 1985, it is estimated that the carrying capacity was reduced by a half from 0.18 to 0.09 ewe equivalents per hectare. The 11 million hectare of steppe, therefore, can only provide about 10% of the feed requirements of the animals kept on it. Consequently, three quarters of the requirements of the sheep in the steppe are brought into the area and this necessitates the transport of 4.7 million tonnes of feed annually, comprising of 500,000 tonnes of barley, 400,000 tonnes of bran and 3,800,000 tonnes of straw. The remaining 15% of feed requirements is obtained by spring grazing in the desert and in the areas of cultivation after harvest.

The demand for red meat is high, mainly as a result of the rapidly increasing population (12.0 million to 23.0 million between 1966 and 1987), and a slight increase in consumption per head (eg 3.4 to 4.1 kg of sheep meat/head/year between 1970 and 1987), partly as a result of urbanization. A key factor, however, in the expansion of sheep numbers has been the distortion of the market, created by regulations enforcing the compulsory sale of cereals to the Government, at fixed prices, while the market for red meat has remained free. Thus, the wholesale price of sheep carcase, adjusted for the cost of living index, increased from 38 DA/kg in 1970 to 102 DA/kg in 1987. In the same period the adjusted price for barley rose from 1.29 to 1.55 DA/kg. The ratio between the prices of sheep meat and barley widened from 30:1 to 66:1.

A further factor in this complex picture is the wide spread speculation in sheep. With a national flock of approximately 9 million ewes, Boutonnet estimates that nine million sales of ewes and ewe lambs occur annually, with large fluctuations in price. He states that the art of the flock owner in Algeria is less in the maximization of meat production and more in the management of his capital of live animals.

Before independence, Algerian sheep production was characterised by periodic large fluctuations in sheep numbers with the affects of climate on feed availability. In bad years part of the flock was sold and the worst effects of drought on the steppe prevented. For 30 years since independence, the sheep population has steadily increased, as the state has always made sufficient imported barley available for the sheep numbers to be maintained in any particular year. Boutonnet quotes a peasant “in the past, in bad years, the shepherd sold part of his flock and kept only the sheep he could feed; now he sells his wife's jewellery to buy barley”.

The increase in sheep population has been accompanied by a decrease in productivity. Carcase production per ewe and per year has declined as a result of a reduction in the number of sheep sold per ewe, in spite of an increase in individual carcase weight.

 Sheep sold/eweCarcase wt
Carcase wt/ewe

Boutonnet's description of the changes in Algeria and their causes has been presented in considerable detail, because it is derived from a careful analysis of a large amount of data and gives many insights into a complex situation. It emphasises the importance of the ratio between the prices of cereals and of sheep meat in stimulating flock expansion. In these circumstances, sheep enterprises, which in the past played a role as a reserve against uncertainty, have become a way of making speculative gains. Boutonnet says “there is no need, in these circumstances, to invoke obscure reasons of tradition or socio-cultural prestige to understand the general propensity for an increase in the number of animals, since the economics of sheep production permit it, or for the lack of interest in improving flock productivity, in spite of the high price of meat”.

A further point made in Boutonnet's analysis is that, although the rural population of Algeria has declined as a percentage of the total population, the actual rural population has not decreased because of the dramatic increase in total population (the Algerian statistics quoted by Boutonnet differ radically from the FAO data in Table 1.) This strong and continuing pressure of rural people is in itself a contributory factor to the increase in sheep numbers. Where the ratio of meat to barley prices is favourable and there is access to grazing land, crop residues or even vegetable matter from urban waste, starting a small flock is a way of creating capital from a small initial investment. The large families mean that there is no shortage of children to shepherd the flock.

Boutonnet's report identified examples of this amongst farmers surveyed in the area of Sidi-Bel-Abbes. Small farmers with less than 20 ha of land, generally had no flock and sold surplus feed and by-products. If, however, they had a flock, the stocking rate was more than 10 ewes per hectare, far higher than could be supported by their land, and the flock was fed mainly on purchased feed. These flocks were generally financed from outside the farm, usually by some agreement with urban investors or landless shepherds.

Tables 1 and 2 present FAO data for four countries in north Africa and four in west Asia and show that similar trends to those in Algeria are occurring in the other countries of the region. There are large increases in the human population in all these countries, except Jordan. The rural population, as a percentage of the total population, has declined in all these countries but the changes in the actual rural population vary considerably: no decline in Turkey, very slight in Morocco and small in all the other countries, except Jordan and Algeria (see the previous paragraph). Sheep numbers have increased in all these countries, except Iraq and Turkey, in the latter, between 1975 and 1987 the numbers increased and then declined again. The FAO statistics appear to be too inaccurate to make any judgement as to whether the decrease in off-take of slaughter sheep and the increase in slaughter weight found in Algeria is a general trend. The area of barley cultivation and the production of barley increased in all countries except Libya and Jordan, both small producers.

Table 1. Total and Rural Populations in 8 Countries of north Africa and west Asia (FAO Production Yearbooks)

CountryTotal Population
Rural Pop. as % of totalRural Population

1 = mean of 1974, 75 & 76
2 = mean of 1986, 78 & 88

Table 2. Area and Production of Barley and Sheep Population in 8 Countries of north Africa and west Asia

CountryBarley Area
(1000 ha)
Barley production
(1000 tons)
Sheep population

1 = Mean of 1974, 75 and 76.
2 = Mean of 1986, 87 and 88


In the last 15 years, major degradation has occurred in the Syrian steppe, where 75% of the sheep population is kept. This has happened in spite of the adoption of a range of measures intended to prevent continued over-utilization.

In the mid-1970s it was possible to describe (eg Nygaard et al., 1982) the traditional seasonal movements of flocks from the steppe to the agricultural land in June to graze stubbles and other byproducts. The flocks then returned to the steppe from November until the following June. Movements still occur but they are now less regular and, when they return to the steppe, the flocks stay in a small area around the family's house and are fed purchased feed.

A number of factors have contributed to the degradation. Masri (1979) suggests that a key factor in starting the process of degradation was the introduction of tractors in the 1950's, which allowed easy cultivation of the steppe, with the areas with the best rainfall and soils being cultivated first. Later, in the 1960's, drilling of new wells by the Government allowed grazing in areas which were formerly protected from grazing for long periods by lack of water. Then the purchase of water tankers permitted richer flock owners, with capital, to graze the remoter areas for longer periods. Before, these remote areas were a reserve, not because of tribal tradition or agreement, but because of their isolation and lack of regular watering facilities. Pick-ups enabled large areas to be covered looking for good grazing and the flocks could be moved rapidly by lorry. Thus the pressure on the remaining steppe increased. This was further intensified by the progressive abandoning of the tribal control of range land and the permitting of areas to be transferred to individual owners, which is still continuing.

From about 1960, the Syrian Government introduced a series of measures, which experts at the time (see Draz, 1974) thought would prevent further decline of the steppe. About 50 sheep and range cooperatives were established between 1968 and 1978 to provide credit to nomadic flock owners for the purchase of feed and livestock, with the aim of reducing fluctuations in stock numbers. The survey by Nygaard et al. in 1982 makes almost no reference to the improvement of range, which was initially a major objective of these co-operatives. To help decrease the grazing pressure in the steppe, about 50 fattening co-operatives were also set up to take light lambs from steppe flocks and fatten them on cereals. Membership of these co-operatives gave preferential access to feeds at controlled prices from the Government purchasing organisation.

The results of these initiatives have not been, as was intended, a stabilisation of the fluctuations in sheep numbers, and in the cycle of low prices for sheep meat in dry years and high ones in good years. Instead it has resulted in a steady expansion in sheep numbers since about 1970. This has led to a rising demand for barley, although the ratios between the prices of sheep meat and barley have been much less favourable than those in Algeria. For example, in 1988 the ratio was 15:1 with the free market price of barley and 18:1 with the controlled market price, and in 1989, when drought increased sales of sheep and depressed the price, the rations fell to 9:1 and 10:1 respectively. The demand for barley has caused a steep rise in the area planted to barley (Fig 1), which further intensifies the grazing pressure in the decreased area of steppe. The palatable bushes are over-grazed and disappear, and the unpalatable ones are often removed for fuel, especially in settled areas, which now extend far into the steppe. Even unpalatable bushes are important for the survival of the degraded steppe, as they help stabilize the soil and offer some protection from grazing for a few annual plants until they can flower and set seed. Further rapid degradation and complete loss of bush cover results from speculative barley growing financed by urban investors and which now extends into areas with less than 200mm of rainfall.


This description of the situation and changes in the west Asia and north Africa region indicates the enormous scale of over-grazing and environmental degradation in the region. The only possibility of achieving long-term sustainability in animal production systems is for governments to tackle the social and economic situation of the flock owners and farmers in these areas, so that it becomes economic for them to adopt more sustainable systems. It is almost impossible to believe that this can be done by any government, whether democratic or not, before further major damage to the environment occurs. This is mainly because the pressure of rapid population increase is the major determining factor and this cannot be altered in the foreseeable future.

Other policy issues, which could have some effect in reducing problems of over-grazing are discussed below.

Limiting Rural population

Rural depopulation and the availability of alternative employment will undoubtably have some effect in reducing the agricultural population towards a level that the land can sustain. In general, however, in these areas, rural populations are not declining quickly and some are still increasing. If stocking rates are to be reduced to levels that are sustainable in the long term, governments will have to introduce strong legislation and strictly enforce it. Again, it seems unlikely that this will happen: in the short term such legislation would cause enormous hardship to one of the poorest sections of society and its introduction would require either the provision of alternative employment or compensation for loss of a traditional way of making a living.

Pricing Policy for Inputs

It is clear that low prices for purchased feed are a major cause of flock expansion, because flock owners make rational economic responses. Government policy should be directed to eliminating subsidies and market distortions. To be effective, it must extend to bread flour, as subsidised bread becomes an animal feed when the price ratios are right. Fuel pricing policy must also be considered, as low diesel prices reduce the costs of cultivation and of the transportation of feed, animals and water.

Investment Policy

It is clear that sheep production is a good investment in many of these arid, developing countries, where there may be very limited ways for private investment. Speculation, however, in barley growing in the Syrian steppe, financed from the towns, is a serious cause of degradation. Availability of alternative investments in countries with poor financial services, might help to divert speculative investment from these forms of environmentally hazardous production. In the Islamic countries, where the gaining of interest on loans is forbidden by the sharia, there are particular difficulties.


Although change in socio-economic factors is undoubtedly necessary to limit degradation, research has a clear contribution to make in defining problems and providing solutions, where and when socio-economic conditions suggest that technical change may have an effect.

Some of the topics that require research are set out below.

Socio-economic Studies

Research, similar to Boutonnet's study of Algeria, is needed to define the socio-economic situation in each country. This needs to make full analyses of feed supply from all sources, including any unofficial market and that grown illegally. It is also important to quantify trends in off-takes of slaughter animals to establish whether systems are becoming less efficient in most countries.

Research is needed into the patterns of rights of access to rangeland: even if traditional systems have disappeared or been abandoned, it is too simplistic to assume that access is entirely free. The current situation, after a long period of ploughing in the steppes, should be described. It is clear that planting a barley crop establishes rights to the crop in its harvest year, but does it establish rights to the land subsequently? ICARDA is actively working on this in Syria.

Farmer Participatory Research (FPR).

Farmer participatory research is needed to define the farmers' view of their problems and discuss with them possible technical solutions. I see FPR as the only way to tackle the problem of low off-take of slaughter animals, which may be an increasing problem. Very close contact with flock owners is needed to identify specific causes and seek solutions.

Improvement of Extensive Grazing Land.

The possibilities for improving productivity and sustainability of rangeland need investigating. Because productivity is limited by the amount of rainfall, technologies requiring costly inputs will never be economically viable in these arid areas. Options do exist and ICARDA has programmes on these for the WANA region.

In order of increasing input requirements, they are:

Improved grazing management. In areas with annual plants, management directed to improving seed set will slowly increase productivity. Currently, these plant communities are grazed heavily in late winter, limiting early growth and making very little contribution to the animals' nutrient requirements. Lower grazing pressures and the deferring of intense grazing would slowly allow an increase in herbage production. For areas where the imbalance between animal numbers and pasture resources is not too great, guidelines for grazing management have to be established for flocks that are continuously shepherded. Traditionally high grazing pressures (animals/unit area/unit time) are used to achieve very high utilisation of the herbage available at a particular time. Continuous stocking systems are very difficult, if not impossible, to introduce for shepherded flocks.

The benefits of deferring the start of heavy grazing in early spring needs demonstrating, especially when shown that flocks walking long distances to graze very sparse pasture actually require more supplementation than continuously penned animals. Thompson (personal communication) showed that a flock taken to graze for 6 hours daily on poor pasture required 2.2kg/head of supplement to maintain weight, while a continuously penned flock only required 1.7kg/head.

Application of fertilizer. In many soils in the region low soil phosphorous (P) limits growth and small applications of P in the presence of legumes can rapidly increase herbage production, if the grazing pressure is controlled. For example, legume seed numbers increased from 3.1 to 18.6 thousand per m2 over 4 years with annual applications of 25kg of P2 O5 (Osman et al, 1990). Stocking rate was 2.5 times that of the unfertilized land.

Introduction of Legumes. Cocks (personal communication) suggests that, in Mediterranean areas, very small-seeded legumes of genera, such as Astragalus and Trigonella, may warrant research for areas with rainfall of about 250 mm. Species from these genera possess reproductive strategies adapted to these dry areas: very small seeds, that pass largely undigested through animals, large numbers of seeds per pod and high levels of hard seededness.

Introduction of Edible Shrubs and Trees. Shrubs, such as Salsola vermiculata and Artemisia, were an important component of natural steppe, not only as feed sources, but also to stabilise soils. Much research has been done on the establishment of plantations of bushes, particularly Atriplex species, both native and introduced, but there is a lack of long-term data on carrying capacity and animal performance. Although wide spread growing outside government stations is not common at present, they probably offer the only possibility for regeneration of degraded areas of steppe, particularly as low cost establishment from seed shows considerable promise.


Boutonnet J, P., 1989. La speculation ovine en Algerie. INRA. Serie Notes et Documents No. 90.

Draz O. 1974. Management and Rehabilitation of the Natural Grazing Lands. Paper presented at the Regional Sheep & Forage Production Workshop held in Beirut, Lebanon, February 11–14, 1974.

Harrington G. 1981. Grazing Arid and Semi-Arid Pastures. In Grazing Animals ed F H W Morley. Elsevier. Amsterdam. p 181- 202.

Harrington G., Wilson A. and Young M. 1990 Management of Australia's Rangelands. CSIRO Division of Wildlife and Ecology, East Melbourne. 354 pp. Kassas, M., 1975. U.N.E.P. Overviews in the priority Subject Area:Land, Water and Desertification (UNEP/PROG/2,Nairobi), 14pp.

Meigs, P. 1953. World distribution of arid and semi-arid homoclimates. Reviews of Research on Arid Zone Hydrology. UNESCO, Paris, pp. 203–210.

Masri A. 1979. Rangelands Development in Jordan in Relation to Controlled Grazing Systems and Rehabilitation. Jor/79/010.

Nygaard D., Martin A. and Bahhady F. 1982. Range and Sheep Co-operatives and Fattening Co-operatives in Syria Supplemental Feed Purchases and Credit Requirements. ICARDA, Project Report No.4 141 pp.

Osman A.E. (1990) Response of Mediterranean grasslands to phosphate and stocking rate: biomass production and botanical composition. Journal of Agricultural Science (Cambridge) in press.

Stoddart L.A., Smith, A.D.and Box, T.W. 1975. Range Management. 3rd Edition. McGraw-Hill, New York. 532 pp.

Previous Page Top of Page Next Page