4.6 The livestock production

Contents - Previous - Next

The projections of production and consumption of livestock products were presented and discussed in Chapter 3, together with the implications for the growth of demand for cereals and oilseed proteins for feed. This section provides some further discussion of the likely evolution of the main parameters and other issues underlying the projections of production. In the first place, the trend for the share of the pig and poultry sectors in total meat production to increase is likely to continue (Table 4.15). In particular, the growth of the poultry sector would continue at high rates, though lower than in the past, while that of the pig sector may be in the future much less rapid than in the past, because of the expected slowdown in East Asia, the region which accounts for 90 percent of total pigmeat production. Possible developments in China account for much of this slowdown, as discussed in Chapter 3.

Concerning the main production parameters, the increases in livestock numbers and in the offtake rates (the percentage of animals slaughtered each year) have been the dominant sources of growth in meat production in the past in the developing countries. This will continue to be so in the future. However, it must be noted that the historical data on animal numbers are often not sufficiently reliable to document developments in the sector. For example, some recent research indicates that the existing data may underestimate significantly actual livestock populations, particularly of small ruminants (Wins and Bourn, 1994, see below).

In many countries, there is only limited potential to increase livestock numbers in extensive systems. Overall, there has been a trend towards more intensive production systems in the very diverse production systems prevailing in the developing countries. This trend will continue and in the future much of the increased output of pigmeat, poultrymeat, eggs and, to a lesser extent, dairy would come from the further expansion of intensive and semi-intensive production systems with the use of supplementary-feeds. This would make it possible for increases in yields per animal (carcass weight, milk, eggs) to be a more important source of growth than in the past.

Livestock production systems differ in their ability to respond to changing market conditions, reflecting primarily differences in the biological characteristics of the production process. With industrial-type systems being increasingly employed in the poultry sector of developing countries, production can respond relatively quickly to changing market conditions because of fast reproduction cycles and proximity of operations to urban markets. A wide range of commodities in elastic supply can be used to provide the required feed. Thus, compared with ruminants, poultry has a flexible feed resource base, and feed conversion efficiencies are high. However, the production systems of developing countries are usually highly dependent on imported technology and inputs. Eggs, pork and, to a lesser extent, dairy production systems tend also to be relatively responsive to changing market conditions. Due to the technological characteristics of intensive production, poultry, pig and most dairy systems cannot be gradually transformed from traditional to intensive production. Thus, at the level of the individual production unit, the growth process tends to be discontinuous rather than evolutionary.

In contrast, the production of ruminant meat and, to a smaller degree, dairy tends to be much less responsive to changes in demand, because of the long reproduction cycles, low feed conversion efficiencies and low degree of specialization. Thus, movements away from traditional systems toward more intensive methods tend to be slower and of an evolutionary character. Mixed farming systems are also unlikely to be highly responsive to growing demand for ruminant livestock products, simply because of the other functions that livestock have to fulfil at the farm. The possibility of expanding production through a gradual transformation of traditional farming systems is usually insufficient to respond effectively to growing consumer demand. Consequently, modern production systems, similar to those in the developed countries, have emerged in almost all developing countries alongside traditional systems. As the latter systems prove increasingly unable to meet the rising demand, in the future a larger share of total supply is likely to come from more intensive systems.

Some countries will probably find it difficult to move along the path of significant intensification and resulting productivity gains in the foreseeable future. For example, the low feed resource base and the import requirements for intensification of production would limit progress in this direction in most Sahelian countries. The difficulties in the transition from extensive to more intensive livestock production enhance the risk of environmental degradation. One of the main threats comes from overgrazing. It is believed that in many countries, especially in semi-arid ones, livestock numbers already exceed the carrying capacity of unimproved grazing land. There are major institutional and economic constraints on the path to achieving a sustainable balance between livestock numbers and forage and other feed resources. These problems will be difficult to overcome in the short to medium term and are likely to grow rather than diminish in scope and gravity.

The above-mentioned recent study (Wins and Bourn, 1994) provides interesting new insights in the evolutionary processes taking place in African livestock systems. The study is based on extensive surveys of livestock populations in Mali, Niger, Nigeria, Chad and the Sudan, and covers both pastoral and village livestock systems as well as arid, semi-arid, sub-humid and humid agroecological zones. The general conclusion is that in most situations the intensity of livestock production activities is closely correlated with the intensity of human activities (as measured by habitation density and share of land cultivated) and not, or only weakly, related to the distribution of natural grazing resources. This holds both for village and pastoral livestock activities. These findings suggest that a trend exists for livestock systems to become less dependent on the availability of extensive rangelands and for livestock production to be more closely related to the more secure feed resources associated with proximity to human settlements. They tally with the knowledge that feed requirements for ruminant and non-ruminant animals have traditionally been provided by a mix of grazing, crop by-products and, to a lesser extent, cultivated fodder crops. Ruminants depend the most on these feed resources. Reduced communal grazing resources due to increasing population, arable land expansion and degradation of pasture are making livestock increasingly dependent on crop residues and marginal feed resources.

Though pasture and forage sources remain the most important animal feedstuffs in developing countries, their supply has increased only slowly and has been inadequate to meet demand for livestock products. Concentrate feeds, mainly feed grains, have been increasingly used to supplement other fodder. Grain output has been growing much faster than pastures and fodder and its use as feed has increased considerably in the past 30 years (see Chapter 3). Higher proportions of intensive dairying, poultry and pigs will increase the use of cereals as feed. Its share in total livestock feed is expected to increase further as natural grazing resources become scarcer and the institutional changes required to control overgrazing and eventually reverse its consequences will take many years. Pasture improvement and the growing of forage have yet to be widely accepted by pastoralists or settled farmers, except in parts of North Africa, the Near East and China. Expected crop production in some countries will not provide sufficient by-products to meet protein and, in some cases, the metabolizable energy needs. Some nutritional deficits can be met by the use of feed additives, notably urea and molasses, but most of the shortfall in roughages will have to be met by the use of concentrates.

Concern is often expressed whether the developing countries will be able to increase cereal supplies by as much as needed to support the growth of their livestock sector. For example, Nordblom and Shomo (1993) consider that foreign exchange shortages in the Near East/North Africa region will make it difficult for the feed deficits to be met by cereal imports. Livestock production in the Near East/North Africa region is indeed fairly intensive in the use of concentrate feedstuffs and other regions are projected to follow on this path, though they will continue to use much smaller quantities of cereals per unit of livestock output than the Near East/North Africa region. The cereals projections underlying the projection of the livestock sector were presented in Chapter 3. They assume that for the Near East/North Africa region the cereal intensity of livestock production (amount of cereals used per unit of product) will not continue to grow as in the past following gains in the efficiency of livestock production.

Table 4.15 Meat production by species in the developing countries (including China)


Meat production

Livestock numbers

million tonnes

annual growth(%)


annual growth (%)







1988/90 2010





1988/90 2010

(93) Developing countries
cattle and buffalo 12.1 18.6 19.3 20.5 32.3 2.2 2.7 798 1005 1369 1.3 1.5
sheep and goat 3.0 4.9 5.0 5.6 9.5 2.8 3.1 869 1129 1578 1.5 1.6
pig 9.7 28.3 28.6 33.5 64.0 6.1 4.0 291 486 860 2.2 2.8
poultry 3.7 12.9 13.3 16.7 36.9 7.0 5.1 2504 6469 12318 5.4 3.1
total meat 28.5 64.7 66.1 76.3 142.7 4.6 3.8          
Sub-Saharan Africa
cattle and buffalo 1.7 2.3 2.2 2.3 4.2 1.6 2.9 129 159 200 1.2 1.1
sheep and goat 0.7 0.9 0.8 0.9 1.8 1.5 3.3 203 259 344 1.4 1.4
pig 0.2 0.3 0.4 0.5 0.8 3.3 4.3 6 It 16 3.8 1.8
poultry 0.3 0.9 0.8 0.9 2.2 5.0 4.6 339 630 1097 3.5 2.7
total meat 2.9 4.4 4.2 4.6 9.0 2.2 3.5          
Near East/North Africa
cattle and buffalo 0.8 1.4 1.3 1.4 2.4 3.1 2.6 37 37 52 -0.1 1.7
sheep and goat 1.0 1.4 1.5 1.5 2.7 2.2 3.0 203 240 326 1.2 1.5
poultry 0.4 1.6 1.8 1.9 4.6 8.1 5.1 206 677 1125 6.9 2.5
total meat 2.2 4.4 4.6 4.9 9.7 4.1 3.8          
East Asia (incl. China)
cattle and buffalo 0.9 2.3 3.2 3.6 6.4 4.7 5.0 118 153 332 1.5 3.8
sheep and goat 0.3 1.1 1.1 1.3 2.0 7.4 3.0 157 220 371 1.5 2.5
pig 7.3 24.6 24.8 29.5 57.2 7.0 4.1 216 388 727 2.4 3.0
poultry 1.6 5.1 5.2 7.1 17.3 6.3 1.0 1178 3335 7415 6.1 3.9
total meat 10.1 33.1 34.3 41.6 82.9 6.7 4.5          
South Asia
cattle and buffalo 1.8 2.6 3.1 3.4 4.5 2.2 2.6 293 335 419 0.8 1.1
sheep and goat 0.5 1.1 1.2 1.4 2.3 4.0 3.5 148 247 337 2.9 1.5
pig 0.2 0.4 0.4 0.4 0.6 2.9 2.6 7 11 14 2.8 1.1
poultry 0.2 0.5 0.5 0.6 2.0 6.5 6.4 235 571 857 5.0 2.0
total meat 2.7 4.6 5.3 5.9 9.4 3.0 3.4          
Latin America/Caribbean
cattle and buffalo 6.9 10.0 9.4 9.7 14.8 1.9 1.9 218 319 364 1.9 0.6
sheep and goat 0.5 0.4 0.4 0.4 0.7 -0.6 2.7 152 153 187 0.2 1.0
pig 1.9 3.0 3.0 3.1 5.3 2.3 2.8 63 75 103 0.8 1.5
poultry 1.2 4.8 4.9 6.1 10.8 7.8 3.9 546 1256 1822 4.5 1.8
total meat 10.5 18.2 17.7 19.4 31.6 3.0 2.7          

*Revised data as known in May 1994, but not used in this study.

Only in very few developing countries have meat and dairy industries developed to the stage where they can provide safe and regular supplies to their rapidly expanding urban populations. Over 90 percent of the livestock in developing countries is owned by rural smallholders with inadequate links to urban markets. The failure of existing structures and organizational patterns to cope with present and future demand for livestock products is evident in a number of aspects. First, demographic expansion has increased technical and infrastructural difficulties in meeting the effective demand, which are sometimes reflected in high price differentials between rural and urban areas. Second, environmental pollution, mainly waste from industrialized livestock production units and processing (in particular, slaughterhouses) is becoming ever more serious because of insufficient structures and the absence of adequate regulations or the lack of their enforcement. Lastly, insufficient food safety standards, because of technical and institutional constraints, are a continuous and growing human health hazard.

Animal genetic diversity

While only a few species of livestock are used to produce livestock products such as meat, milk, skins, fibre and draught power, those species have each been developed to produce specific products in a wide array of production environments, giving rise to a large number of unique breeds each with their own gene pool. It is the range of genetic diversity formed by this array of breeds which provides the key to future increases in efficiency and sustainability of livestock production.

Indigenous breeds have been developed to produce within their own specific environments and often possess attributes which are not immediately apparent, such as, for example, the ability to withstand local stress conditions which may not occur each year. In many cases, improved breeds are introduced which, under different conditions, have greater output and the resulting crossbreeds may well be better than the local pure breeds. Subsequent backcrosses to the improved breeds may, however, result in lower overall productivity due to lower rates of reproduction and lower survival chances, greater disease susceptibility, and the inability to cope with a high share of roughages in their feeds.

Crossbreeding can be a very valuable strategy as it exploits the benefits of hybrid vigour. However, sustainable systems of crossing are difficult to achieve in certain species, particularly in those with low rates of reproduction such as horses, cattle, buffalo and some sheep and goats. Practical bottlenecks are the more complicated logistics and reliable provision of sufficient crossbred replacement stock.

Nevertheless, new technologies such as embryo cloning combined with other modern techniques such as artificial insemination, in vitro fertilization and semen or embryo sexing, offer some potential for the better utilization of heterosis. For example, the continued production of first cross (F1) females for milk production, exploits, to the maximum extent possible, the advantages from both indigenous and exotic genes. Such use of F1 animals is commonplace in species with higher reproduction rates (pigs, poultry, some sheep).

The ability to provide now and in the future animals best suited to the various specific environments, depends on the maintenance of a wide spectrum of diversity within each domestic species. The diverse strains can then be used as required to cope with the inevitable change in production environments which occurs with development. The maintenance of genetic diversity is crucial for the full exploitation of all genetic means to provide in the most efficient way animal products and the most efficient means of sustaining domestic animal diversity is through well designed breeding operations.

Animal health

Meat, milk and egg production are still severely limited by pests and diseases. Some estimates show that at least 5 percent of cattle, 10 percent of sheep and goats and 15 percent of pigs die annually due to diseases. Apart from the direct loss of animals, indirect losses are incurred as a consequence of poor reproduction efficiency, retarded rates of growth and low levels of production. Growth in meat production between now and 2010 is expected to be derived for the greater part from increases in animal numbers with the remainder coming from improved productivity. Together with improved management, veterinary measures to control major epizootics and various disease vectors (such as ticks and tsetse fly) and preventive medicine will play an important part in increasing yields from both single animals and herds.

Major infectious animal diseases are those that are of significant economic importance, or have public health implications (such as rabies or brucellosis), or have recently been introduced and threaten to disrupt the sector (such as African swine fever (ASF) in Latin America in the 1980s). Disease eradication in developing countries has been fraught with difficulties but the successes are notable. ASF has been eradicated from Cuba, Brazil and the Dominican Republic, and babesiosis from large areas of Argentina and Mexico. Contagious bovine pleuropneumonia has been eliminated from the Central African Republic. Foot and mouth disease has been eradicated from all Central American countries and also from Chile, though it is still present in all other countries in Latin America.

There is a large group of chronic diseases that have more insidious effects than the major infectious diseases. Their importance is often overlooked and seriously underestimated. Though less obvious, they often have a serious economic impact through their effects on production or reproductive performance. Examples are helminth infestations, enzootic pneumonia of pigs, mastitis in dairy cattle, and chronic respiratory diseases in poultry. While managerial procedures and prophylactic animal health measures are facilitated when stock is raised under intensive production methods, the higher stocking rates and heightened stress can increase the risk of catching diseases. The development of more intensive production systems causes also a change in the disease spectrum. While in cattle rinderpest and pleuropneumonia decline, other diseases such as brucellosis, leptospirosis and mastitis usually become more important. For poultry, the conversion from extensive village poultry practice to intensive commercial systems shifts importance from, for example, Newcastle disease towards chronic respiratory disease, Marek's disease and Gumboro disease.

In Africa, existing and planned production facilities should make the continent largely self-sufficient in key vaccines which will be a base for substantial improvements in livestock health. The pan-African rinderpest campaign will require around 100 million doses annually, and protecting the cattle population at risk from bovine pleuropneumonia requires some 60 million doses per year. The major disease constraint in Africa however remains trypanosomiasis, which is virtually the only disease which, without preventive measures, entirely precludes the introduction of cattle, though some cattle strains have developed a degree of resistance. Many sub-humid tsetse-infested areas with good agricultural potential could be utilized in a more sustainable manner with mixed farming systems using draught animals. Recent research has provided new control techniques which require low technology inputs and, unlike older methods, do not rely on large applications of insecticides. In many parts of sub-Saharan Africa, however, attempts to improve livestock breeds and dairy production will continue to be hampered unless efforts are made to contain trypanosomiasis. As in other developing regions, the likely expansion of intensive pig and poultry production in Africa will depend on regular measures to control diseases such as avian encephalitis and infectious bronchitis.

Unlike Africa, Latin America is free from diseases such as bovine pleuropneumonia, rinderpest and peste de petite ruminants. Foot and mouth disease however continues to be a problem. After the successful eradication of the New World screw worm from the United States and Mexico as well as North Africa, efforts are concentrated on screw worm control in the Caribbean and Central America.

Asia is also largely free of major infectious animal disease problems in cattle and buffalo. A programme for rinderpest control has been implemented in the Near East and another one is planned for South Asia, both having substantive vaccine requirements. Control of foot and mouth disease and brucellosis in the Near East is of particular concern. Developing preparedness to deal with emergency animal diseases has become an important activity in the region. In many Asian countries, notably the Philippines, Thailand, Indonesia and Bangladesh, the poultry sector is very important and control of Newcastle disease crucial, particularly at the village level.

Where intensive pig industries have been established, classical swine fever is one of the main threats facing production and trade in pigs and pig products. Diagnosis needs sophisticated laboratory testing and trained personnel, while facilities are necessary for control and eventual eradication. African swine fever would have devastating effects on the large pig population in Asia and careful monitoring is needed to prevent it from entering the region.

Further development of good quality low-cost vaccines against most bacterial and viral diseases is expected in the near future. The failure of some international eradication campaigns (e.g. rinderpest) was however not due to the lack of a suitable vaccine, but to poor veterinary infrastructure. One of the preconditions for improved veterinary health will be further investment in expansion of diagnostic facilities and in training of veterinary personnel.


1. There are a number of indications that data on cultivated area in China under report land actually in use (e.g. see USDA, 1991, 1993b). Some sources indicate that under-reporting amounts to about 30 percent, i.e. instead of a cultivated area of about 96 million ha (official data), actual cultivated area would be 125 million ha. Data on average cropping intensity (151 percent) seem to be fairly accurate as are data on production. If correct, this would imply a harvested area of 189 million ha instead of 145 million ha and yields and inputs per harvested hectare would be lower than official data show. Until such uncertainties about the actual data on area and yield by crop are resolved, it is difficult to make meaningful projections for such variables.

2. See, for example, Delgado and Pinstrup-Andersen (1993) for a critique of the land expansion projections to 2000 in the 1987 edition of this study (Alexandratos, 1988). Yet the actual data, halfway into the projection period, show that expansion of harvested area for the major crops in the developing countries (see below) had been taking place largely on the projected path.

3. These crops are: millet, sorghum, maize, spring wheat, winter wheat, barley, bunded rice, upland rice, sweet potato, cassava, white potato, phaseolus bean, groundout, soybean, cowpea, chickpea, oil palm, sugarcane, banana, olive and cotton.

4. For example, for maize in the moderately cool tropics, the maximum constraint-free yield is 10.9 tonnes/ha at the high technology level and 2.7 tonnes/ha at the low technology level while in the warm tropics these yields are 7.1 and 1.8 tonnes/ha respectively (see FAO, 1978-81, Vol. 3: 124-31). Thus, land which could produce less than 20 percent of these yields (2.2 tonnes/ha of maize under high technology, 0.54 tonnes/ha under the low one in the moderately cool tropics) was classified as not suitable for maize though it was included in the agricultural land if it met the minimum yield criterion (20 percent of MCFY) for one or more of the other crops. 5. For a more extensive explanation of the methodology, see FAO (1978-81, 1982). The estimate of total potential (2573 million ha) presented in Table 4.1 (and in Appendix 3 tables) is well above the 2142 million ha that had been estimated for the 1987 edition of this study (Alexandratos, 1988) for a number of reasons. Although also the 1987 estimates were based on soil data of the FAO-UNESCO Soil Map of the World (SMW, FAO, 1971-81), measurements now were made with the aid of a Geographical Information System while the 1987 estimate was based on a manual grid count. This improved reading of data led to a number of changes. The number of crops for which suitability was tested was increased from 11 in 1987 to 21 in the present study. Likewise, suitability was evaluated on the basis of the yield results under any one of the three technology levels, while the 1987 estimate was based on the yield results under the low technology level only. Furthermore, a number of rules applied in the procedure have been changed to reflect refinements in the methodology developed since 1987. Finally, a number of ex-post adjustments to the mechanically derived results were made in 1987 to take into account man-made changes to land resources (for example, terracing of slopes deemed unsuitable for agricultural production in the mechanical procedure). Such ad hoc expert judgement-based changes were not made in the present study.

6. The data underlying the maps are somewhat different from those in the tables because for map-drawing each map unit was allocated in toto to the agroecological cell and land class which occupied 50 percent or more of its area. For this reason, the term "dominant land class" is used in the legends of the maps.

7. Human settlement areas were estimated as follows: the only country with systematic data is China, for which there are data on both population density and nonagricultural land use per person (residence and infrastructure areas) for about 2000 counties. Based on these data a function was estimated linking non-agricultural land use per person to population density (the higher this density, the lower area per person used for non agricultural purposes). This function was subsequently used for all countries and all agroecological zones in each country. Estimates of population densities by agroecological zones were derived from the data used in the study. "Land Resources for Populations of the Future" (FAO/UNFPA, 1980). All these estimates are tentative and probably subject to large margins of error. The most recent world data-set for land use has data on built-up areas for 17 developed countries and none for the developing countries (World Resources Institute, 1994).

8. Environment Canada (1989). The editor is grateful to Dr J. Dumanski of Agriculture Canada for bringing this document to his attention.

9. The land classified as not suitable for rainfed crop production (class NS) was tested for its suitability for forest tree species using the same methodology employed for the evaluation of suitability for rainfed crop production, except that the crop growth requirements were substituted by those for trees.

10. Classes I, II, III, IV of the Categories for Conservation Management of the International Union for the Conservation of Nature (IUCN) (IUCN, 1990).

11. Maps and inventories of National Parks, Conservation Forest and Wildlife Reserves were made available by the World Conservation Monitoring Centre, Cambridge, United Kingdom.

12. The data presented earlier for the land with rainfed crop production potential by LGP classifications can be used to draw inferences about water supplies for rainfed agriculture, as discussed in Chapter 2.

13. The term "pest" here is used to describe organisms that cause crop damage, such as animal pests including insects, mites, nematodes, and pathogens such as fungi and bacteria, and weeds. The term "pesticides" refers to all chemical means to control pests. "Active ingredients" refers to the biologically active part of pesticides.

14. Some cattle, sheep and goat strains in Western Africa have developed tolerance to trypanosomiasis through natural selection. Limited information (from the International Livestock Centre for Africa) on the productivity of trypanotolerant cattle suggests that under light tsetse pressure their loss of performance is negligible and comparable with breeds outside trypanosomiasis risk areas. However, under medium and high tsetse pressure the productivity index falls around one-quarter and one-half respectively. No evidence suggests that the trypanotolerant sheep and goats have a lower level of productivity than other sheep and goats in Africa.

Contents - Previous - Next