General issues

Overview of the regional situation in the context of the Asian economic crisis

P K Mudbhary

Policy Officer
FAO/RAP

Growth and development

• Asia has had impressive, sustained growth throughout the 1970s, 1980s and early 1990s.

• Real GDP increased at rates above population growth, leading to rising per capita income, urbanisation and commercialisation of economic activities and greater integration between economic sectors within countries and with the outside world.

• Notably, most countries in Southeast and East Asia experienced rapid and sustained economic growth with high rates of savings and investment. Rates of inflation were mostly low to moderate, with a few exceptions. Until the mid-1990s, the budget deficit-to-GDP ratio and the current account-to-GDP ratio were either positive or slightly negative, indicating sound macroeconomic fundamentals for sustained growth and development.

Table 1. Macroeconomic indicators in selected Asian economies, 1990–1997

- Not available
PRC: People's Republic of China
SAR: Special Autonomous Region
Source: Asian Development Bank April 1999

Agricultural production

• Agricultural production grew annually by 3–5 percent.

• Crop diversification increased and brought stability to production, boosting food security as well as export earnings.

Sources of growth

• Throughout the 1980s and early 1990s, technology was the cornerstone of agricultural progress particularly for countries with no more land to till.

• Biotechnology has been widely adopted by small fishermen and farming communities in general.

Results and benefit indicators of agricultural development

• Average dietary supplies improved considerably.

• High growth in food and agricultural trade continued.

• Agriculture contributed to export earnings, but its share declined as non-agricultural exports grew faster. Its net foreign exchange contribution, in terms of value added, is expected to be higher than what its reduced share implies, since industrial exports generally proceed from hefty imports of components.

• Annual food imports varied in direct proportion to the changes in average per capita GDP: low in the less developed countries; in advanced economies, the food import burden eased, while the capacity to pay improved.

• The true benefit from agricultural development was the decline in the real price of food.

Poverty incidence

• Poverty indicators responded quite strongly to economic growth.

Table 2. Response of poverty indicators to per capita GNP

Source: FAO. Food Security and Poverty Alleviation in Asia: Lessons and Challenges, 1999

This is consistent with the empirical evidence from a study of 88 countries with episodes of growth. In as many as 77 of them, the income of the poor improved during the period of growth. But when the economy stagnated or contracted, the poor bore the greater burden: the decline in their income was proportionally greater than that of those of the rich.

The Asian crisis

• Began with speculative attacks on the baht, forcing Thailand to float her currency on 2 July 1997.

• The financial crisis became a regional problem as foreign capital fled from the region. Within months, Indonesia, the fourth most populous country in the world, and Korea, the world's eleventh largest economy, were reeling, as was Malaysia. The Philippines also suffered, but avoided the worst.

• Two main interpretations dominate the debate on the origin of the crisis.

1. Poor economic fundamentals and inconsistent policies

   • Large current account deficits - short-term external debt in Indonesia, Korea and Thailand exceeded available international reserves;

   • governments' financial policies, such as those providing implicit guarantees to the banking system, led to reckless lending by the latter; and

   • poor corporate governance led to a large stock of non-performing loans.

2. Self-fulfilling pessimism of international lenders leading to financial panic

   • country debts were largely denominated in foreign currency and of a short maturity.

• IMF and the other multilateral financial institutions put together several billion-dollar rescue packages to help recovery.

• This crisis has forced an unprecedented reappraisal of policies from corporate governance to exchange rate management.

Impact on agriculture and the poor

In principle, depreciation of a national currency should favour agriculture over other activities such as manufacturing, which use more imported components. Higher export prices in terms of domestic currency should spur exports of agricultural exports, and rising relative prices of tradable agricultural products can lead to a transfer of resources from the subsistence sector to the modern, export-oriented sector.

Whether agriculture actually benefits from a readjustment of the national currency depends on a country's ability to adapt to changing conditions. Structural rigidities can prevent production adjustments and put a damper on price transmission and thus eventually reduce supplies. There are also possibly negative effects on the welfare of small farmers and other net buyers of food as domestic prices rise. Shifts of resources to the tradable sector also tend to raise the cost of non-tradable commodities, worsening the conditions of the subsistence sector.

The credit squeeze and high interest rates have narrowed access to operating capital for essential inputs, marketing and distribution of goods, including export and import activities. Structural rigidities in poor areas hampered rapid production response. Such rigidities reflect in part past neglect to invest in, among other things, an efficient distribution and marketing system.

The livestock industry has probably been affected by shrinking demand and rising cost of production of traded inputs such as feed, breeding stock and veterinary medicine. The other parts of the industry using mainly non-traded inputs are thought to have been affected less. The degree of effect also depends on how soon the supply can be attuned to demand conditions. In some cases, lower demand forcing lower prices may result in massive culling of the herd because it is not economical to maintain the stock when there are doubts about future profitability.

In rural areas where most of the poor depend on agriculture for income, the poorest tend to be net buyers of food staples, which represent the largest component in total household expenditure. The sharp rise in food prices in crisis-affected areas after a steep devaluation represents a serious threat to household food security, as well as to the gains in poverty alleviation in the last quarter of this century.

The latest World Bank survey results confirm that the impact of the crisis on poverty incidence has been severe. Estimates for Indonesia indicate an increase of about 10 percent in the incidence of poverty, corresponding to something like 20 million people - the equivalent of a medium-size country. The poverty incidence increased from 11 to 19.9 percent between 1997 and 1998 and the average standard of living fell by 24.4 percent. Significant increases in poverty took place in urban Korea where poverty incidence increased from 8.6 to 19.2 percent and there was a 21.6-percent fall in the average standard of living. Thailand was affected to a lesser extent, with a marginal increase in poverty incidence (from 11.4 to 12.9 percent) and a reduction in the standard of living of 13.6 percent. 1

1 Poverty lines of about US$1/day in Indonesia. US$2/day in Thailand and US$4/day in Korea: survey evidence from selected areas of Indonesia; poverty incidence for Thailand as of 1996.

Outlook for the future

• The worst of the crisis seems to be over. The affected economies are expected to post positive rates of growth in 1999. As indicated in Table 3, with the exception of Indonesia, the countries of East Asia which experienced negative growth in GDP, ranging from 0.5 percent (Philippines) to 13.7 percent (Indonesia), are expected to see their economies grow modestly - by 1 to 2.5 percent.

  Table 3. GDP growth - Asia and the Pacific (%)

  	1997	1998	1999
  			Autum WEO	Spring WEO	Current est.
  Industrial countries	1.7	-1.8	0.7	-0.7	1
  Japan	1.4	-2.8	0.5	-1.4	0–1   3
  Australia	3.6	5.1	2.0	3.1	1/2	2/
  New Zealand	2.1	-0.3	1.7	2.7	21/2
  Asian NIEs	6.0	-2.0	0.7	2.0	3   1/2
  Hong Kong SAR	5.3	-5.1	0.0	-1.3	-1   1/4
  Korea, Republic of	5.5	-5.8	1.0	2.0	4   1/4
  Singapore	8.0	1.5	0.2	0.5	2
  Taiwan, Prov. of China	6.8	4.8	3.9	3.9	4   1/4
  China	8.8	7.8	5.5	6.6	6   1/2
  India	5.5	5.6	4.9	5.2	5   1/2
  ASEAN-4	3.7	-9.7	-0.1	-1.3	1
  Indonesia	4.6	-13.7	0.0	-4.0	-1   1/2
  Malaysia	7.7	-6.7	-1.5	0.9	2   1/2
  Philippines	5.2	-0.5	3.2	2.0	2
  Thailand	-0.4	-8.0	-1.0	1.0	1–2
  Bangladesh	5.7	4.8	3.5	3.4	4–4   1/2
  Vietnam	8.8	3.5	2.2	3.5	3   1/2
  Other Asia	4.9	4.5	6.1	5.9	…

  Source: IMF, June 1999
  Notes: NIE: Newly industrialised economies; SAR: Special Autonomous Region

• Cautious optimism prevails. Recovery remains fragile, and depends on the continuation of the structural reform policies, as well as on an improving environment.

The industrialisation of livestock production in the light of the Asian economic crisis

H Steinfeld

Senior Officer
Livestock Development Planning, FAO, Rome

•

with the assistance of A Kamakawa
Associate Professional Officer, FAO, Rome

The demand-led surge in livestock production in Asia is accompanied by a dramatic transformation of the livestock sector. In terms of organisational evolution, this transformation basically takes three different forms.

First, livestock production tends to concentrate in areas favoured by cheap input supplies, in particular feed, and by good market outlets for livestock products, such as meat, eggs and milk. This, typically, is the case in the vicinity of large cities.

Second, most of the growth in the livestock sector occurs in industrial types and large-scale production. Rapid growth in scales of operation is a common feature globally, but particularly in industrialising Asia. In many countries, these industrial forms enter into direct competition with land-based production, sometimes supplanting them.

Third, vertical integration along the land-livestock-food chain creates economies of scale. Different contractual forms are chosen for this purpose, sometimes leaving space for individual contract farmers, sometimes not.

It appears that the industrialisation of livestock is occurring in different stages for different animal species and products. Poultry production was the first one to go industrial to the extent that in developed countries, backyard and small-scale poultry production has almost disappeared altogether. Pork follows and dairy is next. The structural evolution of the livestock sector is driven by cost reductions in search of optimal scales and patterns of organisation.

These trends are worrying for a number of reasons:

1. Industrial livestock production generates little employment and the benefits, at production level, accrue to few. While cheap animal protein favours indirectly also poor consumers, the poverty and equity effects, as regards livestock production, are on balance largely negative. For agricultural development, the fact that livestock goes industrial takes away the most important growth stimulus that there could be in rural areas. However, in the harvest and post-harvest sectors, many new jobs are created in slaughterhouses, dairy plants, food processing and in retailing and in street food preparation.

2. Because of the land-detached nature of industrial livestock production, units tend to concentrate in certain areas, causing massive and increasing damage to the environment. The nutrient cycling that was previously carried out on-farm is no longer occurring. Nutrients are loaded onto limited space, and very often discharged untreated into open waters.

3. There are a number of diseases that tend to be associated with increasing intensity of production and concentration of animals on limited space. Many of them pose a threat to human health; industrial and intensive forms of animal production may be a breeding ground for emerging diseases (Nippah, BSE, Avian Flu), with unknown consequences. Public health is threatened by other forms of food safety problems, such as manifested by the recent dioxine scandal in Europe, antibiotic resistance and other residue issues.

In the light of the strong trend to larger scales, the question is whether the same would still be the case, and to what extent, if the negative externalities in terms of the environment, equity and public health were appropriately factored in, and what policy conclusions would stem from such considerations.

The Asian economic crisis has brought about some fundamental macroeconomic changes in the countries concerned, with immediate implication for the livestock sector. Demand for animal products contracted in most countries; feed imports became much more expensive as a result of currency devaluation. This, at first sight, would disfavour industrial production, which has limited scope for substituting local feeds for imported ones.

This paper intends to highlight some of the underlying principles of the industrialisation of livestock production and wants to stimulate debate as to how to address livestock research and development in the light of the Asian economic crisis.

Recent trends in animal protein demand and production

Changes in demand: A combination of growing per capita incomes, population growth and urbanisation is translating into a growth of demand for animal product that has reached 66, 71, 140, 27 and 90 percent for ruminant meat, pork, poultry meat, milk and eggs, respectively, in the 1985 to 1995 period. Growth rates will continue to be high, even if checked temporarily by the Asian economic crisis (Delgado et al., 1999)

Table 1. Growth in consumption of animal products in the Asia-Pacific region 1985–1995

Source: FAO-WAICENT, 1997

Adjustments on the production side: The dramatic surge in demand for animal protein has resulted in a massive transformation of the livestock sector. For millennia, livestock has been an integral part of the farming system, indispensable as a source of draught power and for nutrient cycling. Livestock's asset building and insurance function was matched with its socio-cultural importance in many Asian cultures.

With the explosion in demand for animal products, traditional livestock production based on local feed resources simply could not expand at the same pace as demand did. We are thus observing a dramatic transformation of the livestock sector. One of the predominant features is the “industrialisation” of the sector. Large-scale, land-detached, fully commercialised production units, “meat factories” are claiming an ever-increasing share of the livestock sector, currently growing at double-digit figures. Between 1985 and 1995, industrial livestock production increased by about 17 percent (see chart).

Traditionally, livestock production in Asia was based on low-opportunity cost resources, such as crop residues and grazing land with no potential for cropping. The surge in demand is entailing a rapid expansion of the resource base. Local feed resources can no longer sustain demand increases. Imported feeds are drawn in, and feed concentrates offer high value and ease of transport and trade.

Another adjustment on the production side is the trend toward monogastrics. Monogastrics are short-cycle species that offer better conversion of feed concentrates than do ruminants. With monogastrics, animal product supply can more flexibly expand and respond to growing demand. Because they thrive on feed concentrates, they require limited space. Hence, they lend themselves to industrial forms of production. There are notable differences in the initial shares of production of different meat types, for example the strong predominance of pork in East Asia and the reliance on ruminants in South Asia. However, more recently, there is an obvious tendency to increased monogastric production, with a strong expansion of poultry meat throughout the region.

Source: own calculations, based on Sère and Steinfeld, 1996

In line with the structural changes outlined above, the generation, transfer and adaptation of technologies will focus on improving efficiency of feed use and increasing animal productivity. Feed requires land for its production and this continues to be the limiting factor to the sector's expansion even if countries resort to feed imports. Continuing industrial development in the region will also make traditional livestock less competitive because of diminishing returns to labour even though this process will be very slow and gradual.

We are therefore witnessing a dualistic mode of development, with two, often conflicting, components:

• a modern, demand-driven and capital-intensive sector, producing poultry meat, eggs, pork and sometimes milk, which increasingly uses state-of-the-art technology. This sector uses resources, in particular feed, efficiently - with the notable exception of fossil fuel. It is rapidly expanding and meeting urban demand but it is also susceptible to market upheavals such as the ones created by the Asian economic crisis. It generates little employment, poses great environmental risks because it tends to concentrate in areas with good market access, and creates a number of novel challenges to human and veterinary public health. Technologies are designed to maximise the use of scarce resources, notably feed, and involve the development of genotypes, bio-technologies and general improvement of animal husbandry, plus advances in the backward and forward linkages. Technology uptake has been fast, driven by commercial interests.

• a traditional, resource-driven and labour-intensive sector, which provides a multitude of services to subsistence-oriented farms. Though not efficient in terms of introduced inputs, this sector utilises resources of little or no alternative use, and for the same reason, its potential to expand beyond moderate growth rates is constrained by low technology uptake, insufficient market facilities and infrastructure, and small economies of scale. Often, these systems are closed cycles of nutrients, farm labour, energy, and so on. Unless these cycles are broken, technology uptake will remain constrained.

Structural evolution of the livestock sector

The structural evolution of the livestock sector in Asia is taking three different forms:

• geographic concentration of livestock production and processing;

• growing scales of operation; and

• vertical integration.

Geographic concentration: In the light of booming demand and technological changes, producers and processors concentrate in certain areas which are favourable for the production of given products. In livestock production, this relates mainly to input supply, in particular feed, and to marketing of products, meat, milk and eggs. Particularly in developing countries where infrastructure is not well developed, livestock producers tend to concentrate close to urban centres, particularly where human population growth is high, such as in humid and sub-humid areas and in coastal zones. This allows them to reduce transaction costs, which tend to be very high for perishable animal products compared to other commodities.

Unlike crop production where climate, soil and topography still play a major part, the location of poultry production, and increasingly also of pork production, is dictated by proximity to feed sources or processing facilities. Of these two variables, processing tends to be more important.

Growing scales of operation: Scales of livestock production and processing are growing in response to technological development, market requirements and insufficient returns to labour in traditional systems. Where alternative employment opportunities exist, such as in the rapidly industrialised countries of Asia, traditional subsistence-oriented livestock farming is often abandoned, opening up market and expansion opportunities for other farmers or commercial entrepreneurs.

Some examples of the speed of growing scales:

• In Indonesia, the average size of broiler production units grew from 345 animals per unit to almost 14 000, equivalent to an increase of almost 4000 percent (Direktorat Jenderal Peternakan, 1996).

• Also in Indonesia, the number of pig holding households decreased from 1.076 million to 0.636 million between 1975 and 1995 while the total number of pigs increased by more than 100 percent (Riethmueller, 1997).

• In Japan, sizes of operations continue to grow rapidly (see Table 2). At the same time the number of slaughterhouses decreased from 437 in 1985 to 318 in 1997 while that of dairy plants decreased from 1 118 in 1980 to 813 in 1997 (MAFF, 1998).

• In Korea, the average size of pig holdings increased from 1.3 to 32.2 and that of chicken from 17.7 to 585.1 (FAO, 1997).

Table 2. Average size of livestock holdings in Japan

Source: MAFF, 1998

Given identical single-product farms (such as pig-fattening plants) and free entry and free exit, the optimal unit size is expected to correspond to the minimum of the average cost curve where return to size is locally constant (Chavas and Magand, 1988). Alternatively, if farm sizes that are not optimum tend to disappear over time, this leads to the approach where only farms of efficient size are expected to survive in the long run, and thus increase their market share.

Growing scales of operation are a predominant feature of the animal protein revolution in Asia (Delgado et al., 1999). However, within the livestock sector, units typically have a fairly wide size distribution. This suggests that minimum average costs vary across farms. This could result from a situation in which different farms have access to different technologies (different agro-ecological conditions, or different quality of physical and human capital). Also, different adjustment costs may imply that individual farms can grow or shrink at varying rates depending on managerial abilities, market conditions, etc.

Among the different cost items in livestock production, building costs show typically a strong cost digression. For example, building costs per animal have been shown to decrease by half if units of 100 and 1000 animals are compared (Fuchs et al., 1993). Beyond 1000 animals, additional buildings are added so that no further cost reduction can be realised.

The main feature of the industrialisation process is the substitution of capital for labour. Improvements in broiler-farm labour efficiency have been studied by Lance (1994) for the 1950–1990 period in the US state of Georgia. In the early 1950s, when most operations such as feeding, watering and heating were mostly manual, labour requirements were around 53 hours per thousand birds. This decreased to 2.6 hours per thousand birds in 1990 on large farms. This allows one person to operate flocks of 100 000 to 200 000 per batch with only part-time assistance. While labour efficiency increased by 33 percent between 1983 and 1990, this improvement was primarily due to economies of size, as average flock size increased from 50 000 to 112 000 birds, but basically using the same equipment. For dairy production in Japan, Junichi (1992) found technical economies of scale but not for beef fattening.

The requirements and considerations for management change as livestock operations grow in size. Changes in manure production, collection, treatment, storage, transport and use are necessary as size increases. Land area requirements for nutrient use increase as animal numbers increase. As the size of operation increases, economy of scale usually improves the cost-benefit ratio for manure systems. Economic analysis of manure systems in Missouri (Fulhage, 1997) showed that costs of manure management increased by about 50 percent per unit of milk produced as the herd increased from 100 to 500 cows. With expansion of dairy herds, traditional methods of moving manure are often replaced by hydraulic or mechanised means, such as barn cleaners, automatic alley scrapers and piston pumps. However, if animal densities are regulated and under stringent environmental regulations, very large units face diseconomies of scale that stem from the land requirements or extra investments in manure storage and distribution facilities, or in renting additional land for manure application. In Eastern Europe, the threshold is at about 10 000 fattening pigs.

The requirements for animal health management change too, as operations grow in size. The control of major infectious diseases is a prerequisite to intensive and large-scale production in the first place. While animal disease costs probably increase with growing scales, these are usually outweighed by improved animal productivity through the application of animal measures that also target productivity-limiting endemic diseases, such as rhinitis and enzootic pneumonia in pigs. Disease risk management becomes an important cost factor in large units since an infection can easily wipe out the whole stock. This is because maximum performance is usually traded against much reduced resistance in genetically uniform stocks. Disease risks also grow exponentially in areas where animal production units are concentrated, such as evidenced by the outbreak of classical swine fever in the Netherlands.

Vertical integration: With increasing scales and stronger market orientation of production, mechanisms need to be established to ensure a regular flow of standard quality outputs. This is often achieved through contract farming/livestock production where the actual fattening/production process is contracted out by a company which supplies inputs such as feeder stocks, feed and veterinary supplies and buys the final product back at a pre-established price. In particular, poultry production has often developed from simple farm operation to complex vertical operation of related industries and enterprises, including grain production for animal feeds, feed mills, slaughterhouses and processing plants, food chain stores and wholesale enterprises (Chantalakhana, 1996).

Firms operating in several stages of production are said to have vertical economies of scale if the costs of jointly producing several vertically adjacent products are less than the costs of producing the products independently. Vertical economies of scale may arise from technological and organisational complementarity between successive stages (Chandler, 1966) and also from eliminating factor distortions in monopolised or monopsonised markets and transaction costs of contractual or market exchanges (Perry, 1989).

Among the livestock products, poultry production has the highest degree of vertical integration. Most poultry products in commercial systems are produced by growers who have detailed contractual agreements with processors or marketers. Some are grown in units owned directly by processing firms. The reason for this high level of vertical integration seems to lie in the fact that most innovations in genetics, in nutrition and in management were most productive only if applied as a package. Furthermore, these innovations involved substantial economies of scale and implied scales that demanded more than the capital available to the average small farmer. Finally, they resulted in products of more uniform characteristics, differentiated products that could be branded.

Many producers are tied by formal agreements with processors, who often provide specialised medicines, ration ingredients and advice. As producing eggs and broilers become separate specialised enterprises, units grow in size, and the relationships between farmers and processors or providers of inputs become more formal. For example, a feed company might supply chicks and other inputs in addition to feed, or chicken and egg processors might supply farmers with needed inputs and then guarantee to buy the output on some fixed or price-formula basis.

Production contracts and vertical integration in the broiler industry facilitated rapid adoption of new technology, improved quality control, assured market outlets and provided a steady flow of broilers for processing. While this process is near completion for poultry in the developed countries, it is ongoing for pork production. In the US, approximately 40 percent of hog sales to packers were co-ordinated by contracts and integrated operations in 1998, compared with 11 percent in 1993 and only 3 percent in 1980 (Martinez, 1999). In both past developments in the poultry sector and ongoing trends in the pork industry, contracting and vertical integration have been associated with new technology and new (geographic) areas of production.

Contracting and vertical integration in the poultry and pork industries offer incentives for reducing transaction costs associated with asset specificity, measuring and sorting costs of the traded goods, and assuring supplies and market outlets.

An important driving force toward increasing vertical integration is the risk aspect. First, vertical integration along the production chain allows quality control, and in particular disease control, which is the more important a risk factor as scales grow. Second, vertical integration hedges against some or all of the market risk, even if sometimes at a high premium, i.e. contractors operating as quasi-dependent employees.

Changes in vertical integration can reduce farm production costs and costs of marketing services, or marketing costs, which include slaughtering, processing, cutting and merchandising.

Distortions and externalities

Environmental externalities: Environmental problems associated with livestock production in the region mainly stem from the concentration of large numbers of animals in a given area. This concentration results from the cost advantage of being close to consumer centres and input (feed) supplies, from a lack of infrastructure and from a low level of regulations or insufficient institutions to regulate or enforce low animal densities. High animal concentrations result in large amounts of waste, polluting water bodies and soils, and endangered human populations. The same principles apply to processing plants, in particular slaughterhouses, which generate large amounts of waste if not properly equipped with waste recovery and control facilities.

Institutions and policies, until now, have largely failed to address these issues effectively. Concern about the long-term productivity of natural resources, including land, water, air and bio-diversity, is not reflected in market prices unless governments and international organisations define and establish mechanisms to correctly reflect the present and future value of natural resources. It has been estimated that total environmental costs for large-scale pig production range between 10 and 15 percent of total production costs, if operated in densely populated areas (see for example Taiganides, 1992). Depending on the stringency of environmental regulations, and provided sufficient land is available for waste application, very large pig fattening units (with a capacity of 100 000 pigs) incur double the waste application costs (US$4 per pig) than do large units (with a capacity of 10 000 pigs) and more than four times that of medium-sized operations (500 pigs) (Fuchs et al., 1993). In case no land is available, costs of composting manure amount to US$9 per pig.

Costs would be lower for poultry production because of lower moisture content of the manure, reducing storage, treatment and transport costs.

Capital subsidies: Many Asian countries have favoured intensive livestock production through a combination of economy-wide and sectoral policies. Economy-wide policies that have an effect on livestock production include subsidies on energy, urban infrastructure and protection of industries through tariffs. Sectoral policies include subsidisation of feeds and ingredients, genotypes, public services, and other. The combined effect has been that, although public and donor-driven technical assistance has focused on smallholder production, the overall economic framework has favoured commercial and large-scale production, and often penalised smallholder production in many countries of the region.

With regard to infrastructure development, in the past decades there have been massive investments into physical infrastructure including the construction of roads, ports and communication facilities, as well as slaughterhouses, cold storage and retail facilities. All these, especially in the case of livestock providing extremely perishable products, have greatly reduced the transaction costs and have, in numerous places, made possible trade in livestock products in the first place. This equally applies to international and domestic trade. Because of the perishable nature of livestock products, infrastructure development has an extremely stimulating effect on livestock production. However, infrastructure development in most Asian countries has been highly selective and has favoured urban over rural. This has promoted the emergence of intensive and industrial types of livestock production near urban markets, resulting in environmental problems and depriving smallholders' land-based production of important development opportunities.

Political economy in developing countries often favours systems that significantly benefit a few targeted individuals or institutions, as opposed to providing small increments for large numbers of people (Delgado et al., 1999). Inappropriate livestock development patterns, such as high cost and highly capitalised industrial pig, milk and poultry production near urban areas in developing countries, are often the result of deliberate policy choices, whether state farms or private companies are involved. Artificially created economies of scale in production add to technological ones in production of poultry and pork, and to a lesser extent for dairy.

Even in the absence of specific tax breaks or subsidies, distortions in domestic capital markets that provide cheap capital to large enterprises and very limited or expensive loans to small-scale ones, favour the large over the small and the substitution of capital for labour.

Conclusion and recommendations

Is continuing geographic concentration, ever larger scales of operation, and close vertical integration the future for livestock production in Asia as a whole? While there are differences from specie to specie, the overall trend points toward greater product differentiation, greater vertical integration, greater farm specialisation, and greater production and processing concentration. It appears that because of lower technical economies of scale, pig and dairy production will not reach the size or degree of concentration that poultry has reached in developed countries. Dairy, in particular, will continue to be dependent on agro-ecological conditions and land availability, and thus will be less uniform in the type of technology applied and will be faced with diseconomies of scale beyond a certain level (500 to 1000 cows).

While specialisation and integration will continue, environmental impact, logistics, diseases and their interaction will place a ceiling over growth of livestock operations on individual sites. However, that ceiling is far from being reached in most places.

Following similar developments in the developed countries, the traditional family farm is on the brink of becoming obsolete whereas large-scale livestock operations, often vertically integrated with agribusiness, are more and more dominating. The industrialisation process can be characterised by a sectoral and regional concentration process, capitalisation of production, and a decentralisation and specialisation of the management functions (Windhorst, 1989). This process was made possible by an agro-technical revolution.

In Asia, the emergence of large producers employing economies of scale within the industry is favoured by recurrent cycles of excess production or under consumption, as frequently observed in the pork and poultry markets. Such a dip in the rapid expansion of the livestock industry in Asia is likely to push inefficient producers out of the market, and because of their sub-optimal scales, many smallholders.

With regard to the structural evolution of the livestock sector and the Asian crisis, two scenarios emerge:

• Where the economic crisis manifests itself in lasting negative effects on purchasing power and in extended levels of poverty, there is likely to be some reversal of the previous rampant industrialisation. Across the region, this is not a likely scenario, with the possible exception of Indonesia, where there is probably a lasting (mid-term) income depression.

• For the other countries, the economic crisis appears to be a temporary phenomenon leading to an effective restructuring of the economy. In this process, the drive toward efficient scales will be fostered by continued expansion of demand, relatively liberal economy-wide policies across the region and momentum triggered by rapid institutional development and knowledge-building. Here, we are likely to observe a continued, if not accelerated, structural evolution of the livestock sector. In light of the Asian economic crisis there is also currently little inclination to step up environmental regulations, so that large-scale producers will continue to be favoured by opportunities to externalise environmental costs.

In future, there is thus a need to design policies to correct the negative effects of large-scale industrial livestock production. These policies should address the underlying causes of environmental degradation, emerging public health hazards, and growing income disparities. They must be flexible, site-specific and well targeted. Instruments include pricing, regulations and institutional development. The collective purpose of these instruments is to establish feedback mechanisms that ensure that the impact on public goods of industrial livestock production is consistent with overall social objectives.

It is the task of government institutions to establish and enforce a regulatory framework for the protection of susceptible areas and for waste control. Intensive production depends on inputs that contain a high component of natural resources, often not reflected in their market price. These should be priced higher by abolishing subsidies or, in some situations, taxation. Examples include feed concentrates, fossil fuel, inorganic fertiliser, livestock products, land, mechanisation and genetic material. This, in addition to a quantitative effect of reduced consumption, will induce a more efficient use of natural resources, with both economic and environmental gains. It will also favour a more even spatial distribution of livestock production and promote land-based systems. Incentives may also be provided though a levy on waste discharge or tradable manure quota schemes, which have been established in many developed countries (Brandjes et al., 1995).

Regulatory instruments are imposed to control the distribution and concentration of livestock production and introduce technical control systems. Some of these regulations are relatively easy to enforce whereas others, for example the maximum permitted amount of manure per unit area, are more difficult. Compliance with regulations affects the cost of production and may therefore influence regional distribution densities.

References

Brandjes, P, de Wit, J, van Keulen, H & van de Mer, HB. 1995. Environmental Impact of manure Management. Impact domain study for the FAO/WB study on livestock and the Environment. Wageningen, The Netherlands. International Agricultural Centre

Chandler, ADJ. 1996. Strategy and Structure. New York, NY. Doubleday, Anchor Books

Chantalakhana C. 1996. Animal Industries in Thailand and possible impacts of trade liberalisation. In Proceedings of the 8th AAAP Animal Science Congress, 13–19 Oct 1996. Tokyo

Chavas, J-P & Magand, G. 1988. A dynamic analysis of the size distribution of firms: the case of the US dairy industry. In Agribusiness, Vol4, No4, pp315–329. John Wiley & Sons

Delgado, C., Rosegrant, M, Steinfeld, H, Ehui S & Courbois C. 1999. Livestock to 2020 - The Next Food Revolution. IFPRI, FAO, ILRI. Food, Agriculture and Environment Discussion Paper N028

Direktorat Jenderal Peternakan. 1996. Livestock Production Statistics, Jakarta

FAO. 1997. Report on the 1990 World Census of Agriculture. FAO Statistical Development Series 9. Rome

FAO. Various years. Food Consumption Statistics. WAICENT. Rome

Fulhage, CD. 1997. Manure management considerations for expanding dairy herds. In J Dairy Science, 80: 1872–1879

Fuchs, C, Blum-Minkel, K & Schüle, H. 1993. Betriebswirtschaftliche Beurteilung der Schweineproduktion. In Abhängigkeit vom bestandsgröβenbegingten Gülleanfall. In Ber Ldw 71 (1993) 432–444. Paul Parey, Hamburg and Berlin

Junichi, I. 1992. The Determinants of Rates of Return to Scale Expansion Investment in Japanese Livestock Production: A stochastic frontier cost function approach. National Research Centre of Agricultural Economics, MAFF. Tokyo

Lance, GC. 1995. Broiler farm labor efficiency analysis of the 1950 to 1990 period in Georgia. In Poultry Science. Vol74, No2, pp398–401

Marberry, S. 1992. Pork industry's numbers shrink as it heads to maturity. In Feedstuffs. 1992, 64: 15, pl, p19–20.

Ministry of Agriculture, Fisheries and Forestry (Japan). Website consulted Jun 1999

Perry, M. 1989. Vertical integration: determinants and effects. In Schmanelsee, R & Willig, R (eds), Handbook of Industrial Organization. New York, NY. North Holland

Riethmueller, P. 1997. Indonesia Livestock Industries. Report prepared for FAO.

Sere, C & Steinfeld, H. 1996. World Livestock Production Systems. Current Status, Issues and Trends. Animal Production and Health Paper 125. Rome. FAO

Taiganides, EP. 1992. Pig Management and Recycling: the Singapore Experience. Ottawa. International Development Research Centre

Windhorst, HW. 1989. Industrialisation of agricultural production and the role of large-scale farms in US agriculture. In Tijdschrit voor Economische en Sociale Geografie 1989, 80:5, pp270–282. Vechta, Germany

Short-and long-term effects of the Asian economic crisis on the livestock sector

Christopher L Delgado, Mark W Rosegrant
and Claude B Courbois

International Food Policy Research Institute
2033 K St, NW
Washington DC 20006 USA
e-mail: [email protected]
fax: 1-202-467-4439

After two decades of rapid economic growth, five of the most economically dynamic countries of East and Southeast Asia entered a period of serious economic and financial crisis in 1997. These countries are the locus of the Asian economic crisis, although arguably events in a broader area of East Asia, including Japan and China, should be included in a full discussion of the phenomenon. Between mid-1997 and the Spring of 1998, the currencies of Indonesia, Malaysia, the Philippines, Thailand and the Republic of Korea fell by 40–80 percent against the US dollar, encapsulating and perpetuating a financial and economic crisis characterised by massive capital outflows. As a consequence, the gross domestic product for most of these countries fell sharply in 1998. Indonesia's GDP is estimated to have declined by 13.7 percent in 1998, Korea's by 5.5 percent, Thailand's by 8 percent, Malaysia's by 6.2 percent, and the Philippines by 0.5 percent (see Table 1 in annexe).

Such rapid and large swings in domestic demand and in trading conditions could not fail to have major implications for all economic sectors. This has been especially true for the livestock sector, with its well-known dependence on urban demand growth for selling output and on its large imports of intermediate goods such as feed cereals. Indeed, the confluence of the Asian economic crisis with economic stagnation in Japan, the economic crisis in Russia, and structural factors in world meat markets such as the surge of industrial pork production in the United States following a period of very high prices in the first half of 1997, is undoubtedly responsible for the 25-percent decline in world pork and poultry prices in 1998.

This paper has three objectives: first, to isolate the short-term effects of the Asian economic crisis on the livestock sector in the five developing countries most concerned; second, to distinguish the mechanisms of transmission of the economic crisis to the livestock-producing sector by assessing, through a case study of Thailand, the relative importance of domestic demand compression versus rising costs of production and shifting world prices; and third, to summarise results from a recent modelling study of the long-term outlook for livestock in Asia and for world livestock and feed prices under different scenarios, including a severe version of the Asian economic crisis.

Livestock production and trade in the five Asian countries most concerned

Meat - mainly poultry and pork - and milk production in the five countries soared from the early 1980s until the onset of the economic crisis in mid-1997, as did feed grain use (Steinfeld, 1998). The same occurred in the rest of Southeast and East Asia (Delgado et al., 1999). Available FAO production figures suggest that meat and milk production in the five countries did not decline relative to the pre-crisis period, and may even have grown marginally in the case of Indonesia, Korea and the Philippines, but at much lower rates than previously (Table 2).

The livestock trade data in Table 3 show the huge expansion in the trade of animal products between the mid-1980s and the mid-1990s, including growth in net meat exports from Korea, Malaysia and the Philippines. Yet net meat trade (Table 3) remained small at 10 to 15 percent of total production (Table 2), even in the mid-1990s.

Anecdotal evidence from agricultural attaché reports and a few data series, such as frozen fowl exports from Thailand (Bank of Thailand, 1999), suggests that meat exports from the affected countries increased significantly after the crisis set in. This is not surprising, in view of the sharply increased export incentive due to exchange rate depreciation and reduced domestic demand. This trend is expected to accelerate now that Japan is once again reporting growth in its quarterly GDP, as Japan is a primary export destination for countries in the region.

Similar evidence suggests that imports of meat, milk and feed have fallen sharply. In 1997, the five countries concerned accounted for 12 percent of all US agricultural exports (USDA/ERS, 1998). Calculations with the ‘G-cubed’ agriculture model at USDA/ERS suggest that the net effect of a contained crisis for the five countries in 1998 was to lower US livestock exports by about 10 percent and feed-grain exports by about 6 percent overall, with reallocation of exports beyond the 10 percent decline to other destinations (Coyle et al., 1998). The same study confirms that agricultural adjustment in the five countries is occurring primarily through a reduction of imports rather than an expansion of exports. Finally, the forward-looking part of the study predicts recovery beginning in 2000, provided that the Asian economic crisis does not spread further to Japan and China.

How the economic crisis in the countries affects the livestock sector

Certain stylised facts are essential to understanding how the Asian economic crisis affects livestock producers in the countries concerned. More complete analysis is found in USDA/ERS (1998), Rosegrant and Hazell (1999) and ADB (1999). The stylised facts, as illustrated in Table 1, include: (a) declining GDP in 1998 after a long period of robust growth, implying a rapid decline in domestic demand for animal products; (b) very sizeable depreciation of market foreign exchange rates, implying that feed imports are much more expensive in domestic currency, even as livestock exports to non-affected countries become more competitive; (c) high inflation rates, especially for Indonesia; and (d) very significant declines in imports, unevenly compensated by an expansion in the value of exports, and the latter only in the Philippines and Indonesia. Each factor will be discussed in turn.

Demand compression

Demand compression, illustrated by the falling rates of GDP in Table 1, cannot fail to affect the livestock sector, given its relatively high income elasticity of demand (Huang and Bouis, 1996; Rae, 1998). For the countries concerned, for example, Rate estimates expenditure elasticity for animal products of the order of 0.24 to 0.99, implying that a one-percent decrease in income will lead to a 0.24- to 0.99-percent decrease in demand. These estimates are at the conservative end of the scale, since the estimation process also allows for the even stronger impact of urbanisation, typically correlated with income changes, on growth of demand for animal products. In the region, a one-percent return of the urban population to the countryside is equated with a one-percent fall in national demand for animal products. Given that the impact of income compression in the five countries is being amplified by massive transfers of unemployed labour back to the countryside (Rosegrant and Hazell, 1998), domestic demand for animal products is likely to decrease significantly in the countries concerned.

Exchange rates

The precipitous depreciation in exchange rates is a key manifestation of the Asian economic crisis. It is also a key mechanism for transmission of incentives for adjustment to the livestock sector. Between July 1997 and May 1999, market exchange rates depreciated by 30–70 percent (Table 1). The partial reappreciation of market exchange rates from their low point in the Spring of 1998 suggests that the crisis may be working itself out. Yet the current average of 30-percent depreciation since the start of the crisis is still a major change in trade incentives.

A key attribute of the livestock sector in all five countries is that the inputs to livestock production (feed concentrates, medicines, etc.) are far more tradable than the end products of meat and milk. Although there is some variation between the five countries, milk is an importable item, and beef and pork fluctuate between being importable and non-tradable goods, depending on the year and country. Pork exports have occurred, but production increasingly went to meet burgeoning home demand before the crisis began. Imports of meat rose in the past because domestic demand before the crisis sufficiently exceeded supply, so that prices were driven to a point at which it was financially profitable to import these items, even in the face of positive protection (Rae, 1992).

With rapid demand compression following July 1997, animal products have been in many cases nontraded goods whose domestic prices were not high enough to justify further imports at prevailing exchange rates, nor low enough to justify profitable exports. Poultry in Thailand is an exception. There, the development of a highly organised export industry before the crisis has allowed the maintenance of poultry exports, even when faced with higher input costs (Rosegrant and Hazell, 1999). As a rule, prior to the crisis, Thailand exported about 20 percent of its poultry production and imported 80 percent of the feed used in poultry. Feed stuffs, principally US corn and sorghum, are actively imported throughout the region (USDA/ERS, 1998).

Under these circumstances, the impact of a 30-percent market exchange rate depreciation is to raise costs of production by up to one quarter (30 percent times 80 percent), without necessarily raising output prices, or at least not proportionately. Thus in the countries concerned, a depreciation of the market exchange rate tends to reduce the profit margin for livestock production. Even frantic efforts by livestock producers to replace imported inputs with domestic ones are unlikely to bear fruit, since the price of domestic feeds will also be drawn up by the rising import price in domestic currency.

Inflation and the real exchange rate

A third way in which the impact of the Asian crisis is transmitted to livestock producers is through its effects on the general level of domestic inflation. In the relatively open economies concerned, reduced pressure on prices from falling disposable incomes of consumers is offset by higher costs for imports, and in some cases, for exportable goods. Furthermore, demand compression generally hurts the prices of nontradable goods as a whole, since there is no other market for them than the domestic one. This shows that the true incentive effects of major exchange rate changes cannot be assessed in isolation. The overall impact of those changes on domestic prices relative to each other and relative to world import prices depends to a large extent on the tradability of different domestic items.

Economists tend to summarise these concepts in the notion of a ‘real’ exchange rate, which is the ratio of prices for tradable goods such as petrol and non-tradable goods such as labour in the economy. In practice, this rate is often calculated as the market exchange rate adjusted for differential rates of inflation domestically and in the world. Domestic inflation in 1998 in the concerned countries (Table 1) was substantially higher than the OECD country average of 3.7 percent (OECD, 1999), particularly in Indonesia. High rates of domestic inflation relative to world prices cause the real exchange rate to appreciate, even if the nominal exchange rate stays constant. This shifts production incentives in favour of non-tradable goods. Nevertheless, a 30-percent depreciation of the market exchange rates is not offset by a 3- to 7-percent appreciation stemming from the difference between domestic and world price inflation. There is no doubt that the real exchange rate in all the countries concerned has depreciated significantly, shifting production incentives to producers of exports and import substitutes, and away from producers of non-tradable goods.

Import compression and the elasticity of demand for imports

A fourth way that the Asian economic crisis affects the livestock sector is the impact of import compression on the elasticity of demand for imports. The crisis was associated with massive capital outflows, which put pressure on exchange rates in the first place. Livestock producers wishing to import feed need to compete for the remaining supplies of foreign exchange with importers of items such as petrol, medicine, etc. Given the great compression in imports that has occurred to date, it seems likely that the demand schedule for foreign exchange will be fairly inelastic, since remaining imports tend to be strongly desired at any price, which implies that incremental access to foreign exchange will come only at a substantial further depreciation of the market exchange rate. Given that the price elasticity of demand for livestock products is fairly high, derived demand for feed is likely to be somewhat elastic (Rae, 1992, 1998). Under such conditions, it is unlikely that feed imports will rise significantly until the crisis has largely turned around. Table 1 shows the significant compression of overall imports in the region, and further shows that significant export response has only occurred in the Philippines.

A case study of the determinants of changing profit rates during the Asian economic crisis for livestock producers in Thailand

Available data make possible an empirical study of the process of short-term adjustment to the Asian economic crisis in Thailand, which is a significant net exporter of poultry and a significant net importer of feed grains. The time series for Thailand is also slightly longer than elsewhere, since the crisis as manifested by the sharp currency depreciation started there, with a 15-percent depreciation of the baht on 2 July 1997 which spread over the following months to the other countries (USDA/ERS, 1998). Finally, Thailand makes an especially appropriate case study because of the great severity of the crisis there, combined with evidence of recovery starting in the second quarter of 1998.

Some stylised facts pertaining to the impact of the economic crisis on the livestock sector in Thailand are shown in Table 4, which gives quarterly data from 1 January 1997 to 31 March 1999. The impact of the crisis on the market exchange rate shows the sharp depreciation (53 percent) that took place between July 1997 and March 1998, and the 24-percent re-appreciation that followed. The ratio of an index of domestic Thai meat prices to an index of feed prices is shown in the third column, and is clearly collinear to some degree with the currency depreciation, as would be predicted by the greater tradability of importable feeds than of exportable meat and milk. This ratio of domestic meat to feed prices also is a convenient proxy for domestic livestock profit margin because feed is at least 70 percent of the cost of livestock production in Thailand (Steinfeld, 1998). Other production costs mainly relate to non-tradable inputs, such as capital and labour.

A similar ratio of US meat and feed export prices is shown in the last column, since world prices were not static over the period and surely affected Thai prices. The appearance of some correlation between Thai and US price ratios for meat and feed suggests that US relative prices were also affected by events in Asia. Data on Thai consumption of livestock products are not available to the authors for this recent period, but perhaps can be inferred from total national power usage, a statistic as highly sensitive to domestic economic conditions as meat demand. It is worth noting that despite economic hardships, Thai frozen fowl exports have increased substantially during the crisis, largely in harmony with movements in the exchange rate, as predicted. Finally, the ratio of the Bangkok consumer price index to the US wholesale price index provides a measure of by how much Thai inflation exceeds world inflation.

The data underlying Table 4 are on a monthly basis, and can be used in an empirical framework to assess the relative impact of world price swings, rising production costs, and domestic demand compression on livestock producer profit margins in Thailand. The latter are modelled by the dependent variable, PROFIT(t), as the ratio of domestic meat to feed prices. The three determinants are respectively modelled as: WORLD_P(t), the ratio of US export meat to feed prices; RER(t), the real exchange rate which is the market exchange rate (US$/baht) adjusted for differences in inflation between Thailand and US dollar export prices; and DOM_DEM(t), domestic demand as derived from domestic electricity use in MKwh. The real exchange rate (RER(t)) captures both the effect of more expensive imports due to exchange depreciation and domestic inflation, and inflation in other non-traded inputs. More formally,

Profit(t) = A * World_P(t)^b * RER(t)^c * DOM_DEM(t)^d + e(t)

in which A is a constant; b, c and d are elasticity coefficients of the dependent variable with respect to the independent variable in question; t designates the month between January 1994 and March 1999; and e(t) is a random residual.

Results of a log-log regression used to estimate the coefficients are given in the first row of Table 5. These three variables give a gratifyingly close fit, explaining nearly 60 percent of the total variation in the profit margin over the period (adjusted R²=0.59). All coefficients were statistically significant at the 5-percent level, suggesting that they belong in the regression, except the intercept (A), which is not statistically different from zero at a 10-percent confidence level. Note the unexpected property that the sum of the estimated values of b, c and d is one, suggesting a linearly homogenous relationship among the variables. This implies that, if all the explanatory variables increase in constant proportion, the livestock producer profit margin will increase in the same proportion, which seems plausible and supports the choice of specification.

The estimated coefficients may be interpreted as follows: if world prices of meat to feed increase by one percent, Thai price ratios will increase by 0.33 percent. This less than unitary pass-through of price changes supports the view that meat is less tradable than feed in Thailand. Furthermore, real exchange rate depreciation of one percent is associated with a decline in livestock profit margins of 0.40 percent. Finally, a fall in domestic demand, as modelled by a one-percent fall in electricity use, is associated with a 0.17-percent fall in livestock profit margins.

In order to interpret quantitatively how these different factors have affected livestock producer profit margins in Thailand, it is necessary to see how the underlying variables changed over the estimation period. Figure 1 graphs the movements in real exchange rates and PROFIT(t), showing that it is useful to distinguish three distinct periods: a pre-crisis period (before July 1997), the worst part of the crisis (July 1997 to February 1998) and the recovery period (March 1998 to the present).

The monthly rates of change for the regression variables for these three periods are shown in the middle rows of Table 5. They emphasise the differences among the time periods. The rates of change of the dependent variable suggest that livestock profit margins did not change between January 1994 and June 1997. In the eight months beginning July 1997, they fell by 4.3 percent per month. After March 1998 they rose at an average rate of 1.9 percent per month.

The monthly rates of change for the explanatory variables can be applied to the elasticity properties estimated with the above equation to derive the monthly change in profit margin attributable to the variable in question. Results in the bottom rows of Table 5 show that at the height of the crisis, depreciation in the real exchange rate was responsible for a 2.56-percent monthly decline in livestock producer profit margins. Adverse trends in world prices depressed Thai livestock profitability by 0.66 percent per month. Domestic demand compression only decreased profitability by 0.32 percent per month. It is clear that at the height of the crisis, from July 1997 to February 1998, Thai livestock producers were seeing their profitability cut primarily by depreciation of the real exchange rate, which raised imported feed costs, and costs of non-traded inputs such as capital. After March 1998, the major factor depressing profitability was domestic demand compression.

The same procedure can be used to estimate unexplained changes in PROFIT(t), as shown in the right-hand column in Table 5. Prior to July 1997, our simple regression accounted for virtually all of the change in livestock producer profit ratios, and most of it until February 1998, changes in the real exchange rate were clearly driving profit rates. The model is much less accurate after March 1998, since the coefficients explain only 25 percent of the variation in PROFIT(t). It is hypothesised that after March 1998, domestic factors not captured in the model, such as domestic non-tradable feeds, may have become more important.

The long-term impact of the crisis on Asia and the World

The main difference between short- and long-term effects are that the latter occur after considerable iterative adjustments to changed incentives are made by billions of producers and consumers world-wide operating through a very large number of linked product, input and factor markets. The rapid globalisation of markets emphasises the need for a conceptual framework that allows for such interactions to occur at a level of detail that is neither too restrictive nor too vague. Such efforts have typically involved complex general equilibrium models that link the outcome for sectors such as livestock to overall economic recovery and to trade incentives. On the other hand, they do not de-aggregate agriculture to any great degree (Rosegrant and Hazell, 1999), and may not contain many long-term insights specific to livestock.

IFPRI's IMPACT model provides an alternative approach for producing projections to the year 2020. It is a global model of food supply and demand, created and first reported on by Rosegrant, Agcaoili-Sombilla and Perez (1995), and substantially updated in Rosegrant and Ringler (1998), with the livestock component further analysed in Delgado et al., (1999).

IMPACT breaks the world into 37 country groups demanding and supplying 12 crop and six livestock commodities. Market-clearing solutions for world crop and livestock prices are derived annually through an iterative algorithm, and these are fed back into national prices. Interested readers are referred to the references above for details of the model, its advantages and drawbacks, and a full exposition of results. The purpose of the rest of this section is to extract and report on only the Asia-specific livestock and on the world price results from the most recent of those publications, for the insights they give on the likely evolution of the Asian livestock sector over the next twenty years.•

• The rest of this section is drawn from Delgado et al., 1999, and reflects the underlying modelling work done by Mark Rosegrant.

IMPACT baseline projections for growth in Asian livestock consumption and trade to 2020

The baseline projection reflects the authors' best judgement as of June 1998 of the most plausible assumptions to make about income growth, exchange rate policy, and other imponderables which have changed quite a bit under the Asian economic crisis, but which have already benefited from a view of the worst the crisis had to offer. The baseline assumptions about demographic trends, income trends, and supply and demand response parameters are conservative compared to other such models. Yet the results looking to the future show the likelihood of a demand-driven “livestock revolution” starting in Asia and spreading to the rest of the developing world.

Meat consumption in Asia is forecast to grow by about 3 percent per year in most sub-areas of the region, with milk consumption growth being much more variable, depending on location. Southeast Asia is forecast to continue to 2020 as a marginal net exporter of meat, but also as a major world importer of dairy products. Per capita meat and milk consumption in Southeast Asia in 2020 (24 kg of meat and 16 kg of milk) is forecast to remain very low compared to the industrialised countries of the North (83 kg of meat and 189 kg of milk) and lower than the average for the developing world (30 kg of meat and 62 kg of milk).

Sensitivity analysis: Asian scenarios

Two Asia-specific simulations were conducted to assess the sensitivity of the baseline 2020 results to assumptions. Changes in assumptions are summarised in Table 7. The Asia Severe Crisis scenario is an unreasonably pessimistic view that assumes lasting damage from the crisis that would continue to depress growth and exchange rates through 2020. The India High Meat scenario is probably overly strong in its assumptions that Indians will begin eating meat more in conformity with world norms for India's per capita income level, along the lines of current Indian trends in poultry and milk consumption.

Results are shown in Table 8 in terms of percentage differences between the new projections and the 2020 baseline projections in Table 6. The Asia Severe Crisis scenario leads to 8- to 20-percent lower total consumption of meat and milk, and 5- to 20-percent lower use of cereals-based feed, but also to significantly higher net exports of meat, milk and feed (or lower net imports, as the case may be). The India High Meat scenario leads to huge percentage increase in Indian consumption and feed imports, and major increases in livestock product exports from the rest of Asia to India.

The bottom line

The two scenarios tend to pull world markets in opposite directions. Projected market clearing prices for major products are shown in Table 9 for a variety of crop and livestock products, adjusted to constant 1990 US dollars. Basically, the most likely projection is that world crop and livestock product prices will continue their long-term fall in real terms, but at a much slower rate than in the past due to the demand-driven livestock revolution. Forecast 2020 real poultry prices in the 2020 baseline are 11 percent below actual levels in the first half of the 1990s. They will be only 3 percent lower if India High Meat comes to pass, or 17 percent lower if Asia Severe Crisis is a more accurate scenario. Similar orders of magnitude (give or take 5 percentage points) prevail for pork, beef and milk under the three scenarios.

The overall conclusion for the long-run projections in Asia is that the livestock sector will continue to grow rapidly, regardless of the assumptions made about the Asian economic crisis. Given what appears to be the initial stages of a significant recovery as of this writing (June 1999), except possibly in Indonesia, the less pessimistic set of scenarios would seem to be warranted.

Summary and conclusions

Livestock is a major economic sector in Asia, the relative importance of which increased greatly prior to the onset of the Asian economic crisis in July 1997, largely due to rapid increases in urban demand. While meat is frequently exported, the share of exports in total production is low (20 percent in the extreme case of Thai poultry). On the other hand, production has rapidly become industrialised, using large amounts of imported cereals and soybeans as feeds.

The Asian crisis has been most acute in Thailand, South Korea, Indonesia, Malaysia and the Philippines, associated with income compression of the order of 6 to 13 percent, and exchange rate depreciation of the order of 30 percent to-date. It began in mid-1997 and saw rapid economic deterioration through the first quarter of 1998, and some recovery thereafter.

The crisis had a major depressing effect on livestock production incentives in the short run. In Thailand, this was manifested through a 28-percent decline in the ratio of domestic meat prices to feed costs from the third quarter of 1997 through the first quarter of 1998. Since March 1998, a gradual recovery has been underway in livestock production incentives. In Thailand, the ratio of domestic meat to feed prices had fully recovered its decline by the end of the first quarter of 1999.

Analysis shows that the main negative influence on incentives to livestock producers during the height of the crisis was through the depreciation of the real exchange rate, which raised production costs much faster than returns. The five Asian countries most affected by the Asian economic crisis have adjusted in the livestock sector through decreases in imports of meat and milk, and especially of feed grains, rather than through increased exports of meat.

Compression of domestic demand was only a secondary explanation of unfavourable incentives from the onset of the crisis until March 1998, although it has seemingly become more important in a relative sense as producers have adjusted to other feed sources and world prices have sagged. Full recovery for the livestock sector is likely in the next few years only if domestic demand recovers, which is unlikely in the foreseeable future in Indonesia, and may require a little more time in South Korea and Thailand.

Looking at the long term and in a broader regional context, the demand-driven “livestock revolution” in Asia is fuelled by population growth, urbanisation, and income growth. By 2020, these forces will dwarf the effects of the 1997 financial crisis in Asia. There are at least as many reasons for being bullish about livestock demand in Asia as there are reasons for seeing lasting stagnation. Yet these projections are based largely on demand growth in Asia itself, so an important caveat in both the medium and the long term is that the health of the sector will require a resumption of income expansion in urban areas.

References

ADB. 1999. Annual report. Manila

ADB. 1999. Asian development outlook. Manila

ADB. 1999. The financial crisis in Asia. Manila

Bank of Thailand (Economic Research Department). 1999. Data bank. <http: //www.bot.or.th/research/public/DataBank/Databa_e.htm>. Accessed 28 May 1999

Coyle, WT, McKibbin WJ, Wang Z & Lopez M. 1998. The Asian financial crisis: Effects on US agriculture. Market and Trade Economics Division, Economic Research Service Staff Paper No9805. Washington, DC: United States Department of Agriculture

Delgado, C, Rosegrant M, Steinfeld H, Ehui S & Courbois C. 1999. Livestock to 2020: The next food revolution. Food, Agriculture and the Environment Discussion Paper 28. Washington, DC: International Food Policy Research Institute

FAO. 1999. FAO Statistical database. <http:faostat.fao.org/default.htm>. Accessed 28 May 1999

Huang, J & Bouis H. 1996. Structural changes in the demand for food in Asia. Food, Agriculture and the Environment Discussion Paper 11. Washington, DC: International Food Policy Research Institute

OANDA (Olsen and Associates). 1999. The currency site. <http://www.oanda.com/>. Accessed 28 May 1999

OECD. 1999. Latest trends in consumer prices. <http://www.oecd.org/news_and_events/new-numbers/cp/nw99-47a.htm f>. Accessed 10 Jun 1999

Rae, AN. 1992. Interaction between livestock and feeds policies: Evidence from Southeast Asia. In Agricultural Economics 7 (June): pp25–37

Rae, AN. 1998. The effects of expenditure growth and urbanisation on food consumption in East Asia: A note on animal products. In Agricultural Economics 18 (May): pp291–299

Rosegrant, MW & Hazell PBR. 1999. Rural Asia transformed: The quiet revolution. Report prepared for the ADB. Manila

Rosegrant, MW & Ringler C. 1998. Asian economic crisis and the long-term global food situation. Paper prepared for the International Agricultural Trade Research Consortium Symposium on Policy Reform, Market Stability and Food Security, 26–27 June, Alexandria, VA. Washington DC: International Food Policy Research Institute

Rosegrant, MW, Agcaoili-Sombilla M & Perez N.D. 1995. Global food projections to 2020: Implications for investment. Food, Agriculture and the Environment Discussion Paper 5. Washington DC: International Food Policy Research Institute

Steinfeld, H. 1998. Livestock production in the Asia and Pacific region-current status, issues and trends. World Animal Review 90(1): pp14–21

US Bureau of Labor Statistics. 1999. Producer Price Indexes. <http://stats.bls.gov/ppihome.htm>. Accessed May 1999

USDA/ERS (United States Department of Agriculture Economic Research Service). 1998. Events in Asia: lower prospects for US farm and rural economy. Agricultural Outlook 248 (February): pp9–12

USDA NASS (United States Department of Agriculture National Agricultural Statistics Service). Various years. Agricultural prices. Washington DC

World Bank. 1998. World development indicators 1998. Washington DC

Figure 1. The real exchange rate and meat-feed price ratio in Thailand, Jan. 1984-Apr. 1999

	Real exchange rate
Sources: Bank of Thailand (1999) and USDA (various).
	Livestock-feed price ratio

Table 1. Asian developing countries most affected by the Asian economic crisis

Sources: World Bank 1998, OANDA 1999 and ADB 1999a
^a A negative number implies depreciation

Table 2. Livestock production in Asian developing countries most affected by the Asian economic crisis

Source: FAO 1999
Notes: Meat includes meat products from bovine animals, sheep, goats, pigs and poultry. Feed grains as used here include all cereals, soybeans and soybean meal.

Table 3. Trade in livestock and feed in Asian countries most affected by the Asian economic crisis

Source: FAO. 1999

Table 4. Livestock and feed in Thailand, Jan 1997-Mar 1999

Sources: Thailand data from Bank of Thailand (1999). International price ratio of meat and feed is calculated from the USDA NASS (various) price indices for livestock and feed products. US finished goods producer price index is from the US Bureau of Labor Statistics (1999).

Table 5. Decomposition of changes in Thai livestock production profit margin by source, Jan 1994-Mar 1999

Sources: * The elasticity coefficients are estimated using a log-log unrestricted regression fitted to monthly data from the Bank of Thailand (1999), USDA (various), and the US Bureau of Labor Statistics (1999). The intercept (-0.17) was not statistically significant at 10%. All other variables were statistically significant at 5%. Adjusted R² = 0.59. Monthly growth rates are estimated by semi-log regressions.

Table 6. Baseline projections of Asian food consumption and trade for meat and milk, 2020

Sources: Rosegrant and Ringler, 1998, and Delgado et al., 1999.
Notes: Consumption is direct use as food, uncooked weight bone-in. Meat includes beef, pork, mutton, goat and poultry. Milk is milk and milk products in liquid milk equivalents.
The country aggregates are defined as follows:
Southeast Asia: Brunei, Cambodia, East Timor, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand and Vietnam
Other East Asia: Hong Kong, Macao, Mongolia, North Korea and South Korea
Other South Asia: Afghanistan, Bangladesh, Bhutan, Maldives, Nepal, Pakistan and Sri Lanka

Table 7. Changes in assumptions to Asia Severe Crisis and India High Meat Consumption scenarios

Source: Delgado et al. 1999. The Asia Severe Crisis scenario was first described and estimated in Rosegrant and Ringler, 1998.

Table 8. Total percentage changes in projections of aggregate consumption and trade in 2020 due to changes in assumptions relative to the baseline

Sources: Rosegrant and Ringler, 1998, and Delgado et al., 1999.
Notes: ^a The Asia Severe Crisis scenario incorporates lower projected income growth and depreciation in exchange rates for countries in Asia. Depending on the country, income growth projections are 30–45% lower than in the baseline scenario, and exchange rate depreciation results in 5–13% higher domestic prices.
^b" The India High Meat scenario incorporates a shift in tastes toward consumption of animal foods and increases in production response and feed use ratios by Indian livestock producers. Income elasticity for animal products rise to between 1.5 and 2.0, depending on the commodity. The trend growth rate of Indian herd size rises by 0.3 to 0.7% (depending on the commodity), and feed conversion ratios rise above those typical in developing countries.
^c “Mutton” includes sheep and goat meat and edible products.
^d “Milk” includes all dairy products in liquid milk equivalents.
^e Very large percentage changes for meat and feed trade in India only reflect their starting from a very low base, and therefore are reported as plus or minus only. Projected beef consumption in 2020, for example, under the high meat scenario, is still only 31kg per capita, less than China today.
^f “Rest of Asia” includes all other Asian countries, including West Asian and North Asian countries.

Table 9. Real prices of selected feed and livestock products as projected by the IMPACT model

Sources: Rosegrant and Ringler, 1998, and Delgado et al., 1999