Previous PageTable Of ContentsNext Page

Intensification or extensification: which has the lowest environmental burden?

M. Upton

The author's address is: The Old Schoolhouse, Stratfield Save (near Reading), RG7 2EJ, UK; fax: 0044 1734 756467; e-mail: AESUPTON@RDG.ACUK.

INTENSIFICATION OU «EXTENSIFICATION»: LAQUELLE PÈSE LE MOINS LOURDEMENT SUR L'ENVIRONNEMENT?

Si l'«extensification» a un coût moins lourd au niveau de l'environnement, une intensification de la production animale est néanmoins plus probable dès lors que la demande augmente. Il s'agit donc de ramener à un niveau acceptable et durable les effets négatifs sur l'environnement.
 Dans le domaine de la production animale, l'objectif de la recherche-développement, vulgarisation comprise, devrait être d'accroître la productivité par tête de bétail, tout en préservant les ressources limitées et en minimisant la pollution, par exemple à travers l'amélioration de la conversion des aliments pour réduire les déchets.
 La croissance de l'offre de produits animaux est allée de pair avec celle de la demande, malgré les changements importants intervenus au niveau de la structure de production. La question qui se pose est toutefois de savoir si cette évolution accélérée de l'industrie animale mondiale peut s'inscrire dans le temps sans porter atteinte à l'environnment, compte tenu, notamment, de l'augmentation des besoins en céréales fourragères et en légumineuses pour l'alimentation des animaux.

INTENSIFICACION O «EXTENSIFICACION»: ¿CUAL REPRESENTA LA MENOR CARGA PARA EL MEDIO AMBIENTE?

La «extensificación» tiene menores costos ecológicos, pero es probable que se produzca intensificación de la producción animal siempre que lo exija el crecimiento de la ganadería. Por consiguiente, el objetivo debe consistir en limitar los efectos negativos para el medio ambiente a un nivel aceptable y sostenible. La investigación sobre la ganadería y su fomento, sin olvidar la extensión, debe orientarse a incrementar la productividad por animal, conservando al mismo tiempo los escasos recursos disponibles y reduciendo al mínimo la contaminación, por ejemplo mediante la mejora de la conversión de los piensos, de manera que se reduzca la cantidad de desechos. El crecimiento de la oferta de productos pecuarios ha mantenido el ritmo de aumento de la demanda, aunque se han registrado cambios importantes en la estructura de la producción. Sin embargo, se plantea la cuestión de si esta industria pecuaria mundial, que cambia y crece con rapidez, se puede mantener sin efectos adversos para el medio ambiente, particularmente teniendo en cuenta la mayor dependencia de los cereales y legumbres utilizados como pienso para la nutrición animal.

THE BACKGROUND: DEMAND GROWTH FOR LIVESTOCK PRODUCTS

The FAO global study World Agriculture: Towards 2010 (FAO, 1995) predicted that "structural change in the food consumption of the developing countries towards more livestock products will continue with significant increases in per caput consumption of meat in all regions except South Asia and sub-Saharan Africa. However, their per caput consumption of such products will still be below those of the high-income countries in 2010."
These predictions are based on the knowledge that meat, milk and eggs are favoured by consumers over other more basic food items, even though the cost per unit of energy is much higher. Hence, as per caput incomes rise, so too does meat consumption as a proportion of the normal diet and total household expenditure. (In economic terms the income elasticity of demand, i.e. the percentage change in the quantity demanded for a 1 percent change in income, is high, often greater than unity.) South Asia is exceptional in that meat consumption is limited for cultural reasons, although milk and egg consumption is expected to grow. In sub-Saharan Africa the stagnation in the level of consumption is related to the lack of growth in per caput income.
Growth in the demand for animal protein is influenced not only by rising incomes but also by the increase in human population. When both factors are taken into account, the predicted growth rate in demand for livestock products D is given by the simple relationship: D = p + ng

where p = rate of population growth, n = income elasticity of demand for the product and g = rate of increase in per caput incomes. Given that in most of the developing world all these variables are positive for livestock products, the total demand is predicted to grow rapidly.
In fact, total demand worldwide is estimated to have risen by 4.8 percent annually between 1970 and 1990, faster than the human population growth rate (FAO, 1995).
Overall, growth in the supply of livestock products has kept pace with increases in demand, although there have been major changes in the structure of production. Other livestock contributions, particularly draught power and manure, are of importance to mixed farmers in many parts of the world. However, the structural changes mentioned above are associated with a shift in emphasis to the production of meat, milk and eggs. The question arises as to whether this rapidly changing and growing world livestock industry can be sustained without adverse environmental impacts, particularly in view of the increased reliance on feedgrains and legumes for animal nutrition.
In 1990, roughly 38 percent of the world's grain was fed to livestock (Durning and Brough, 1991), while the figure for the United States was 70 percent.
International trade in feedgrains and legumes is relatively greater than the trade in livestock products. Declining world prices of major feedgrains from the mid-1970s to the early 1990s may have contributed to the rapid expansion of pig and poultry production in developing countries. The sustainability of some of these systems is now open to question in view of recent increases in the real price of grain.

 

W5256t08.JPG (16719 bytes)

High mountain pastures in Kyrgyzstan, Central Asia, provide essential summer grazing and help relieve pressure on pastures in the foothills close to settlement areas
Les pâturages dans les hautes montagnes au Kirghizistan, Asie centrale, fournissent l'essentiel des prairies pendant l'été et soulagent la pression sur les pâturages des collines proches des zones habitées
Las praderas de alta montaña en Kirguistán, Asia Central, proporcionan el pasto esencial de verano y disminuyen la presión en las praderas de las faldas de las montañas cercanas a las zonas de los asentamientos humanos
Photo/Foto: R. Mearns

 

W5256t09.JPG (17657 bytes)

Intensive pig production in China
Production intensive de porcs en Chine
Producción porcina intensiva en China
Photo/Foto: FAO

 

W5256t10.JPG (21166 bytes)

Battery hens in Malaysia
Poulets de batterie en Malaisie
Batería de gallinas en Malasia
Photo/Foto: R.D. Branckaert

HORIZONTAL VERSUS VERTICAL EXPANSION OF PRODUCTION

For the purposes of this discussion, production intensity is defined in terms of the number of livestock units and quantities of associated inputs per unit of land area. On this basis, industrial pig or poultry production qualifies as an intensive system whereas pastoral herding on natural rangelands is viewed as being extensive. Intensification then implies an increase in input use, with a corresponding increase in output produced per hectare.
This measure of intensification excludes increases in the productivity per livestock unit, resulting from improved breeding, feeding and management, although the supply of livestock products may also be increased in this way. Despite the fact that the majority of the world's livestock, of all species, also inhabit the developing countries, most meat, milk and eggs are produced in the developed countries of Europe and America, where more than four times as much beef and six times as much milk are produced per animal. Clearly these differences in productivity are important but, for the purpose of this paper, the alternative measure of intensity in terms of livestock units per hectare is preferred.
The intensity of livestock production may be influenced by the characteristics of the physical environment and also by product and input price levels which, in turn, are affected by location in relation to the market or point of sale. Wint and Bourn (1994) show that in most situations the intensity of production is closely correlated with human population density and the share of land cultivated rather than with the distribution of natural grazing resources.
Using the theoretical framework developed by von Thünen (1826), it is argued that the price received by the producer declines while costs of manufactured farm inputs rise with increasing distance from the market, since transport costs are greater (Figure 1). The overall effect is that the gross profit margins that can be earned fall with increasing distance from the market.
Thus, the distance-decay function of Figure 1 applies to product price, gross margin per livestock unit (LSU) and intensity of production. The most intensive forms of production are found near the market (e.g. peri-urban livestock keeping) at the "intensive margin", while the least extensive systems are found at the "extensive margin" distance, OX, from the market in Figure 1. Beyond OX there are no commercial profits to be made from meat production.
If the demand for meat increases without any corresponding change in supply, then the price will increase, as shown by the broken line in Figure 2. There is a corresponding increase in gross margins, leading to intensification of production. Furthermore, meat production becomes profitable and is extended on to land between X and Z km from the market. The main conclusion to be drawn from this analysis is that increased demand provides incentives for intensification by all the existing suppliers, regardless of their current levels of intensity, together with an extension of the supply area. Intensification and extensification are not really alternatives, since both occur concurrently.
In practice, further extension of the supply area may not be possible. Land that is submarginal for one market may already supply other markets. The real question is whether there remains any unused land on to which production can be extended. While there may be some scope for extending the area dedicated to livestock production in Africa and Latin America, this may be at the expense of forestry and wildlife reserves. In Asia there is very little scope for any extension of cultivation or livestock production (FAO, 1995).
Pigs and poultry are usually the most intensive industrial system, since for economic reasons they are located nearest to markets (Figure 3).
Dairying is often the second most intensive system, as milk is a perishable product with relatively high transport costs. However, it also depends on supplies of bulky fodder which are more readily available in rural areas. The gross margin per LSU is lower than that for pigs and poultry in the immediate vicinity of the market but, since it declines less rapidly with distance, a boundary is reached where dairying becomes more profitable (Figure 3).
In more remote areas, meat-producing ruminants yield a higher gross margin per LSU (milk may be important for household subsistence but less so for the market). For this system, few purchased inputs are used while the product is relatively cheaply transported on the hoof (Figure 3). As a result, a series of "rings" of different production systems of declining intensity may be observed around an urban market centre (Figure 4).
It is now possible to incorporate the notion that the scope for increasing the productivity and intensity of production may differ between different systems. Indeed, this is found to be the case. Whereas extensive ruminant grazing systems have shown very little improvement in productivity over recent decades, that of the more intensive industrial livestock systems has increased significantly. Furthermore, the intensity of stocking, while fixed within narrow limits by the availability of grazing under extensive systems, can be increased almost without limit for pigs and poultry.

 

W5256t11.gif (1979 bytes)

FIGURE 1
The distance-decay function for prices, gross margins and intensity
La fonction distance-détérioration pour les prix, les marges brutes et l'intensité
Función de disminución de la distancia en relación con los precios, los beneficios brutos y la intensidad

 

W5256t12.gif (1663 bytes)

FIGURE 2
The effect of increased prices on productivity
L'effet de la croissance des prix sur la productivité
Efecto del aumento de los precios en la productividad

 

W5256t13.gif (2425 bytes)

FIGURE 3
The choice of livestock species and system
Le choix des espèces animales et des systèmes d'élevage
Elección de especies de ganado y del sistema

 

W5256t14.JPG (19477 bytes)

Intensive sheep production in Cyprus
Production intensive de moutons à Chypre
Producción intensiva de ganado ovino en Chipre
Photo/Foto: FAO

ENVIRONMENTAL COSTS OF INTENSIVE PRODUCTION SYSTEMS

The intensive systems with which this article is concerned are those involving the "industrial production" of the monogastric livestock, pigs and poultry, as well as ruminant livestock for dairying or feedlot fattening. Links with the agricultural land base are broken and feedstuffs produced elsewhere are transported to the production site which is often peri-urban.
The environmental costs of such systems are often claimed to include the large inputs of grain and legumes and the energy used in producing them. Conversion rates of grain into meat are low - between 14.5 and 38.5 percent by weight in the United States (Durning and Brough, 1991) - so more people could be fed per hectare if the crops were consumed directly rather than being converted into livestock products.
The question of whether high income earners should consume less meat in order to reduce the demand for feedgrains and allow for better nutrition of hungry people and/or the conservation of energy reserves for future generations is an ethical issue open to subjective judgement. It may be avoided if the international, interpersonal and intergenerational distribution of income is taken as given. Then we may argue that, provided consumers are willing to pay prices that cover the full costs of the purchased feeds as well as the energy used in their production, the allocation of resources does not need to be modified.
Until recently, however, industrial systems in many countries have greatly benefited from policy distortions, such as subsidies on feedgrains, which reduce the price consumers pay to less than the true cost of production. In such cases, excessive input use, intensive livestock production and consumption will occur, with a consequent loss of welfare to society (Figure 5). The effects are similar when prices are distorted through price support, trade restrictions or maintenance of an overvalued currency. Policy distortions, which promote overproduction from intensive livestock systems, have been applied in the European Union, the former centrally planned economies and in many countries of Africa and the Near East. It is highly desirable that such distortions be eliminated.
The other environmental costs associated with intensive livestock production are the disposal of manure and other waste products that may cause pollution (Ap Dewi et al., 1994); the increased disease risks to animals and possibly humans, for example, bovine spongiform encephalopathy (BSE), salmonella and leptospirosis; and the reduction in biodiversity as a result of concentration on a small number of specialized breeds. Some of these are experienced as direct "internal" costs to livestock producers, for instance the costs of manure disposal and disease control.
However, other environmental costs are "externalities" which fall not on the producers but on other sectors of the public. These include the pollution of soils, air and water supplies, resulting from the inadequate disposal or control of waste products; greenhouse gas emissions; unsightly and smelly waste dumps; acid rain; and possible human health hazards from zoonotic diseases or from pollution. The external costs of the loss of biodiversity fall on future generations in terms of the loss of "option value" associated with the reduction of the gene pool.
The basic analysis of the effects of externalities is similar to that for subsidies illustrated in Figure 5. While, in that case, the gap between marginal private cost and marginal social cost was due to the subsidy, it now reflects the external costs (Figure 6). The analysis here shows that, since the external costs do not impinge on the individual producer (who is motivated solely by private gain), an excessive use of inputs and output of intensive livestock products will result. Again, there is a net loss in social welfare, the losers being members of the public who suffer from the externalities. Hence, controls are needed to reduce or avoid this loss in social welfare. It should be noted, however, that at the socially acceptable level of production the external impacts are not eliminated completely; there remains an external cost equivalent to AE (in the figure) per unit of output.
There are many policy instruments that may be used to correct the pattern of resource allocation in the presence of external costs, including:

In theory, all these methods could be equally effective. With reference to Figure 6, public control or marketable permits could be applied to limit livestock production, and the associated environmental costs to Qs. The same effect would be achieved by charging a tax equal to the cost, AE, per unit of output, or paying a subsidy for abatement equal to this "damage cost".
However, they have rather different effects on the distribution of income. A pollution tax, for instance, would both reduce the price obtained by livestock producers (from Pp to Pe in Figure 6) and raise that paid by consumers (from Pp to Ps in Figure 6), the overall result being a transfer of income from both producers and consumers to the government. A subsidy would transfer income in the opposite direction.
A serious problem in applying any of these policy instruments is that of measuring the external environmental impact and evaluating the cost. Given these difficulties, the best that may be achieved is to establish "safe minimum standards", for animal waste generation for instance. If quotas are allocated to producers and these quotas are allowed to be marketed, a value will be established through the market. Safe minimum standards may be based on scientific measurements, but they are also likely to be influenced by the relative political power of producers, consumers and those affected by the externalities.
If controls are to be imposed on pollution from solid, liquid and gaseous wastes, the establishment of safe minimum standards and marketable quotas may be appropriate. The abatement of disease risks to humans and loss of biodiversity requires public sector investment.
An alternative approach to reducing external environmental costs is to inform and educate consumers. International workshops may contribute to this process. The effect may be to reduce the demand for intensively produced meat, milk and eggs or to increase the pressures for the control and abatement of environmental damage. In response, some producers may turn to extensive methods of production.

 

W5256t15.gif (4237 bytes)

FIGURE 4
Concentric rings of decreasing intensity around a central market
Cercles concentriques d'intensité dégressive autour d'un marché central
Anillos concéntricos de intensidad decreciente alrededor de un mercado central
 

W5256t16.JPG (15456 bytes)

Intensive cattle production in Kenya
Production intensive de bétail au Kenya
Producción intensiva de ganado vacuno en Kenya
Photo/Foto: FAO
 

W5256t17.gif (4854 bytes)

FIGURE 5
The effect of a subsidy on production and welfare
L'effet d'une subvention sur la production et le bien-être
Efecto de una subvención sobre la producción y el bienestar
 

W5256t18.gif (5016 bytes)

FIGURE 6
External costs and the need for control
Coûts externes et nécessité d'un contrôle
Costos externos y necesidad de control

ENVIRONMENTAL COSTS OF EXTENSIVE PRODUCTION SYSTEMS

Extensive land-based livestock production is generally more environmentally friendly than intensive systems. This may explain why producers, motivated by the need for environmental conservation, tend to adopt less intensive systems such as organic farming, and free-range pig or poultry keeping. Extensive systems also include rangeland-based pastoral ruminant-livestock keeping.
Most of the external environmental costs associated with intensive industrial production are reduced or avoided: there are fewer problems in acquiring or producing balanced feed rations, disposing of waste products and controlling disease, and many consumers believe that there are fewer human health risks from food produced extensively. Moreover, a better maintenance of biodiversity is assured if environmentally adapted breeds are retained.
One environmental cost, however, which can be greater under extensive livestock systems is the emission of greenhouse gases, particularly methane. Produced as a by-product of food digestion, mainly by ruminants, methane emissions are highest when feed quality and the level of production are low (Steinfeld, de Haan and Blackburn, 1996). These conditions prevail in extensive rangeland production of the arid and humid tropics and subtropics. Pigs and poultry cannot digest fibrous grasses and other bulky forage so their emissions are relatively low. However, methane emissions from manure stored under anaerobic conditions are associated with high levels of productivity and intensity.
External environmental costs may arise from attempts to intensify rangeland-based production at the expense of virgin forest and wildlife reserves. Although, by increasing the stocking rate, intensification might cause rangeland degradation, the extent of such damage in practice has been greatly exaggerated. Climatic variation has a much greater impact on the condition of the rangelands, while recovery after drought is quite rapid (Behnke, Scoones and Kerven, 1993).
The external costs of degradation may be felt by other livestock owners who suffer from declining feed resources. Overexploitation of natural grazing resources may result from "open access" and the so-called "tragedy of the commons", owing to the fact that no individual has an incentive to conserve these resources. There is a widespread view that under traditional, communal forms of tenure, rangeland resources were effectively conserved. Restoring communal rights and responsibilities to the pastoralists may prevent any further degradation. However, if pastoralists manage the resources at their disposal efficiently, there would be little scope for further intensification, unless grazing resources are supplemented with crop by-products from agropastoral systems.
The spread of extensive livestock production is only one of many causes of deforestation. Only in Latin America, where the annual rate of deforestation is less than 1 percent, does conversion to pasture represent a significant cause. In Asia and Africa, the spread of shifting cultivation and other forms of crop production are much more important (Amelung and Diehl, 1992). In Brazil, and probably elsewhere in Latin America, deforestation to allow for the spread of agriculture has been encouraged through subsidies, price exemptions and land title regulations which provide strong financial incentives. Without these incentives it is unlikely that extensive livestock husbandry would provide sufficient returns to justify the costs of forest clearing.
It is concluded, therefore, that the environmental costs of extensive livestock production systems are much lower. Unfortunately, there is little scope for expanding the area under such systems other than at the expense of virgin forest and bush, or by conversion of the crop cultivation areas essential to feed production for intensive systems.
Intensification involves a greater reliance on crop products in agropastoral or mixed farming systems. As argued elsewhere (Steinfeld, de Haan and Blackburn, 1996), such systems allow for the exploitation of crop-livestock interactions and the establishment of a nutrient balance between plants and animals.

IMPLICATIONS OF INTENSIVE OR EXTENSIVE SYSTEMS FOR DEVELOPED AND DEVELOPING COUNTRIES

It would be wrong to assume that intensity of livestock production is associated with the national level of economic development. A broad spectrum of intensities is found in both developed and developing countries. However, there are marked differences in other respects. Generally the ratio of livestock to humans is higher in the developed world, where productivity per animal and meat consumption are much higher. Extreme examples are the United States where the average annual consumption of all meats per person is 115.2 kg, while that for India is 2.2 kg (FAO, 1994). The pressures of livestock keeping on the environment are therefore greater in the developed countries. The problems of waste disposal are most severe in countries such as the Netherlands, where much of the livestock feed is imported.
Moreover, in developed countries there is hardly any growth in the demand for livestock products, with little or no increase in human populations, while public concerns are growing over the health effects of consuming livestock products. Tastes are changing towards low cholesterol diets, vegetarianism, and organically produced food. In response, some producers have shifted to less intensive methods such as organic farming and free-range livestock keeping. However, price supports for animal products and concentrate feeding continue to encourage the adoption of intensive livestock production systems, while organic farmers and free-range livestock keepers need to charge higher prices and have difficulties in competing with the majority of "intensive producers".
At the same time there is increasing public awareness and concern over the external costs of pollution. For example, European Union directives limit the number of animals kept per hectare to avoid excessive manure deposits. There is a ban on the direct discharge of waste into surface waters, and tradable manure quotas have been introduced in the Netherlands (FAO, 1996).
In developing countries, expansion of peri-urban intensive production has been rapid, often supported by direct or indirect (overvalued exchange rate) subsidies. This expansion has not always been sustainable, economically or environmentally. Economic costs have been a continuing drain on scarce resources. Existing waste disposal and pollution abatement facilities may be inadequate for the increased demands placed on them. Also public pressure for the control of polluting waste materials is likely to be weak, since environmental concerns have a lower priority in poor societies, unless public health and other problems become acute.
The most promising option for sustainable growth in livestock production is integrated mixed farming, which exploits crop-livestock interactions. These include the feeding of livestock on crop by-products, the use of animal manure to fertilize the crops, the use of animal draught power and the supplementation of human diets and household income.

CONCLUSIONS

 

Bibliography

Amelung, T. & Diehl, M. 1992. Deforestation of tropical rain forests: economic causes and impact on development. Kieler Studien 241. Tübingen, Germany, Mohr (Paul Siebeck).
Ap Dewi, I., Axford, R.F.E., Fayez, I., Marai, M. & Omed, H.M., eds. 1994. Pollution in livestock production systems. Wallingford, UK, CAB International.
Behnke, R, Scoones, I. & Kerven, C., eds. 1993. Range ecology at disequilibrium: new models of natural variability and pastoral adaptation in African savannas. London, Overseas Development Institute.
Durning, A.B. & Brough, H.B. 1991. Taking stock: animal farming and the environment. Worldwatch Paper No. 103, Washington, DC, Worldwatch Institute.
FAO. 1994. FAOSTAT (computerized database). Rome.
FAO. 1995. World Agriculture: Towards 2010. Edited by N. Alexandratos. Chichester, UK, FAO & John Wiley.
FAO. 1996. Environmental impact of animal manure management. Edited by P.J. Brandjes, J. de Wit, H.G. van der Meer & H. van Keulen. FAO Consultancy Report for Livestock and Environment Study, Rome.
Steinfeld, H., de Haan, C. & Blackburn, H. 1996. Livestock-environment interactions. Issues and options. EU (European Commission Directorate-General for Development, Development Policy, Sustainable Development and Natural Resources).
von Thünen, J.H. 1826. Der isolierte Staat in Beziehung auf Landwirtschaft und Nationalökonomie. Rostock, Germany, Verlag von Gustav Fischer.
Wint, W. & Bourn, D. 1994. Anthropogenic and environmental correlates of livestock distribution in sub-Saharan Africa. Oxford, UK, Oxford Environmental Research Group.

Previous PageTable Of ContentsNext Page