Livestock feeds and mixed farming systems in West Asia and North Africa

T L Nordblom, A V Goodchild and F Shomo¹ - Pasture, Forage and Livestock Program, International Center for Agricultural Research in the Dry Areas P O Box 5466, Aleppo, Syria

Introduction
Food in WANA
Oil wealth and population
Land policy
Seed policy
Livestock policy
Macromixes: projections to 2020
Micromixes: projections to 2020
Acknowledgements
References

¹ Agricultural economist, ruminant nutritionist and economic research associate: opinions and ideas expressed in this paper are those of the authors and do not necessarily reflect the policy of ICARDA

Introduction

This paper looks forward to the year 2020 in the West Asia/North Africa (WANA) region. The region extends from Morocco in the west to Pakistan and Afghanistan in the east, and from Turkey in the north to Ethiopia and Yemen in the south. The prospects for crop (and crop residue) production, rangeland capacities and livestock and feed deficits are reviewed. Several case studies of flock level diet calendars for small ruminants and whole farm economic views of mixed crop-livestock systems are presented. The ways in which microlevel balances help in understanding how livestock management options affect cropping decisions while macrobalances at country level determine the economic contexts of livestock production are demonstrated.

WANA is the centre of origin of wheat, barley, lentils and chickpeas, and of sheep and goats. It is also the birthplace of Judaism, Christianity and Islam. WANA held the non-European parts of the Greek, Roman, Ottoman and Persian Empires. The region comprises the Arab countries and several of today's largest non-Arab Islamic countries. This "social ecology" of WANA must be respected and remembered in any discussion of "ecoregions", particularly when it comes to matters of management and policy in agriculture. WANA farming systems thus differ from areas in Australia, California, Chile and South Africa with similar crops and climates. "Farming does not take place in a cultural vacuum" (Fadda, 1992).

WANA is characterized by high population growth, low and erratic rainfall, limited areas of arable land, some of the world's biggest and harshest deserts and limited water resources for irrigation. Climates vary from Mediterranean to monsoonal and from temperate to tropical (Tutwiler and Bailey, 1991).

There is great diversity in the countries of the region in terms of trends in human and ruminant livestock diets over the 1970-1990 period and the differential prospects for domestic food and feed production in relation to projected human populations for the years 2000, 2010 and 2020. Food and feed production prospects are contrasted in 18 of the 25 countries considered by ICARDA as part of its WANA region: Afghanistan, Algeria, Egypt, Ethiopia, Iran, Iraq, Jordan, Lebanon, Libya, Morocco, Oman, Pakistan, Saudi Arabia, Sudan, Syria, Tunisia, Turkey and Yemen. Other sources (TAC/CGIAR, 1992) consider Ethiopia and Sudan to be part of sub-Saharan Africa and Pakistan as part of Asia so these countries are considered separately here.

Food in WANA

The Mediterranean diet, rich in cereal starch and olive oil, is famous for its healthy qualities (Spiller, 1991). Human diets, with the exception of a few countries, clearly improved over the 20-year period 1969-1971 to 1988-1990 (Figure 1) in terms of greater daily per caput protein and energy intakes (FAO, 1993b),

Protein derives chiefly from plant sources (mainly wheat) in ratios of at least 3:1 with animal sources including milk products, eggs, meat and fish (Figure 1). Sudan, where pastoral cattle keeping is important, and the oil states of Saudi Arabia and Libya are exceptions with ratios of less than 2:1. Diets in WANA contrast strikingly with those in the US, UK, France and Japan where most protein is from animal sources. Plant sources, such as wheat - as bread, couscous or burghul - potato and sugar contribute the predominant shares of energy in WANA diets (Figure 1). The contrast with Western countries is again remarkable, due to the high levels of animal products in their diets.

National diet figures mask differences within countries. There is no doubt the wealthiest segment of each population includes greater proportions of animal products in its diet whereas the urban poor has the least. The prospects for large deficits in domestic feed production in WANA in the future may mean even lower quality diets for the poor.

Oil wealth and population

Discussions about world food prospects find it convenient to generalize about geographic groups of countries but the plight of the poorer countries in a region is masked by averaging their lot with that of richer neighbours. This criticism can be levelled at three recent analyses (Alexandratos, 1993; Mitchell and Ingco, 1993; Rosegrant and Agcaoili, 1994) which seem over optimistic in concluding that WANA will not face a food crisis (Nordblom and Shomo, 1994). WANA countries need to be sorted into groups based on their oil wealth and populations (Table 1) as a first step in ensuring that the region's food and agricultural problems will not be ignored.

The first two identifiable groups are oil and remittance sensitive because they are either importers or minor exporters of oil (Janssen, 1993). In addition all are exporters of labour from whom remittances are received. The two groups are separated on the "medium variant" projections of the population trajectories (Table 1, Figure 2) (UN, 1993). Countries of the first group, with fast population growth, are expected to show increasing annual increments through 2020 whereas those of the second group with transitional population growth are expected to have slightly decreasing annual increments over the same period. The third (large populations) and fourth (small populations) groups are (or should be) major oil exporters. A final group "South and East WANA" is included because of the social and economic linkages with WANA-proper and for ICARDA's own interest in seeing these countries in the same perspective.

Figure 1 - Protein (g/person/day) - Sources of protein and energy in human diets in WANA countries (arrows show changes from 1970 to 1990: in this and subsequent figures country codes are Af, Afghanistan, Al, Algeria; Eg, Egypt; Et, Ethiopia; In, Iran; Iq, Iraq; Jo, Jordan, Ln, Lebanon; Ly, Libya; Mo, Morocco; Om, Oman; Pa, Pakistan, SA, Saudi Arabia; Su, Sudan; Sy, Syria; Tn. Tunisia; Tk, Turkey; Ye, Yemen) (Source: FAO, 1993b)

Figure 1 - Energy (Kcal/person/day) - Sources of protein and energy in human diets in WANA countries (arrows show changes from 1970 to 1990: in this and subsequent figures country codes are Af, Afghanistan, Al, Algeria; Eg, Egypt; Et, Ethiopia; In, Iran; Iq, Iraq; Jo, Jordan, Ln, Lebanon; Ly, Libya; Mo, Morocco; Om, Oman; Pa, Pakistan, SA, Saudi Arabia; Su, Sudan; Sy, Syria; Tn. Tunisia; Tk, Turkey; Ye, Yemen) (Source: FAO, 1993b)

Table 1 Per caput GNP and human populations and projections for selected WANA countries

Country	Per caput GNP (US $)	Human population (millions)
		1960	1970	1980	1990	2000	2010	2020
Fast population growth: oil and remittance sensitive
Yemen	515	5.2	6.3	8.2	11.7	16.4	22.6	30.1
Afghanistan	675	10.8	13.6	16.1	16.6	26.5	33.5	41.6
Syria	1000	4.6	6.3	8.7	12.4	17.5	24.2	31.6
Jordan	1120	1.7	2.0	2.9	4.0	5.6	7.6	9.7
Average/total (1990)	763				44.7
Transitional population growth: oil and remittance sensitive
Egypt	620	25.9	33.0	40.9	52.4	64.8	77.7	88.5
Lebanon	812	1.9	2.5	2.7	2.7	3.3	3.8	4.2
Morocco	1030	11.6	15.0	19.4	25.1	31.6	38.7	44.6
Tunisia	1510	4.2	5.0	6.4	8.1	9.8	11.3	12.7
Turkey	1820	277.5	35.3	44.4	55.9	68.2	79.0	88.0
Average/total (1990)	1210				144.2
Oil exporters with large populations
Algeria	2020	10.8	14.0	18.7	25.0	32.9	41.5	48.0
Iran	2320	21.6	28.0	39.3	58.0	77.9	104.0	132.0
Iraq	3657	6.8	9.0	13.0	18.0	24.8	32.9	41.8
Average/total (1990)	2484				101.0
Oil exporters with small populations
Libya	460	1.3	2.0	3.0	4.5	6.5	8.7	11.4
Oman	5650	0.5	0.7	1.0	1.5	2.2	3.0	4.1
Saudi Arabia	7060	4.1	5.7	9.4	14.9	20.7	27.8	36.1
Average/total (1990)	5538				20.9
South and East WANA
Ethiopia	120	24.2	30.6	38.7	49.8	67.0	89.0	116.0
Sudan	370	11.2	13.9	18.7	25.2	33.2	43.0	54.5
Pakistan	400	50.0	65.7	85.3	118.0	154.8	197.7	240.9
Average/total (1990)	324				193.0

Source: GNP data from USDA, 1993b for 1990 or nearest year except Afghanistan. Iraq, Lebanon, Libya Sudan and Yemen for which source is EIU, 1992 and for which data are GDP; population data are from UN, 1993

These groupings are somewhat arbitrary but serve to demonstrate the disparities in WANA and between WANA and other areas. Average GNPs per caput of the five WANA groups in 1990 were US $ 763, US $ 1210, US $ 2484, US$ 5538 and US $ 324. GNPs in developed countries in 1990 were US $ 16000 in the UK, US $ 19400 in France, US $ 21900 in the USA and US $ 25800 in Japan (World Bank, 1993). The western stereotype of the "rich Arab" is most inappropriate for the great population masses of WANA.

In addition to large differences in GNP, oil wealth, significance of remittances and the size and expected trajectories of their populations, the WANA countries show great differences in agricultural potential due to unequal endowments of soil and water resources.

Figure 2 Relative population trajectories in WANA countries (Source: UN, 1993)

Land policy

Contested land ownership and tenure in WANA has a 5000-year history (Nordblom, 1992). Successive civilizations have rejected and borrowed parts of previous institutions in forming new ones for dealing with land resources. After the Islamic conquests only urban property was privately owned whereas farm and pastoral areas were community property. Towards the end of the Ottoman Empire (19th to early 20th centuries), however, communal or tribal types of land tenure in the main agricultural areas were replaced by forms of private ownership.

The Ottoman Land Code of 1858 introduced order to the mass of earlier legislation and was the legal basis of tenure in Syria and Iraq until the 1930s. It defined five categories of land (Warriner, 1962):

· Mulk, held in absolute freehold and governed by sacred, not civil statute law, conferred the right of absolute ownership and usufruct of the land, both rights belonging to the individual;

· Miri is in the absolute ownership of the state but usufruct rights belong to the individual in a form of heritable "ownership" through which the state leases land to him;

· Waqf is land dedicated to religious and benevolent purposes such as mosques, schools and hospitals;

· Matruka land is reserved for public use such as roads, rivers, buildings, market places and village threshing floors; and

· Mawat or Mubah is unreclaimed desert or empty land in the property of the state in the sense of absolute ownership but bringing such land under cultivation over a fixed period confers Miri rights.

Land reforms were implemented by most Arab states and Iran in the 1950s and 1960s. The stated aim was to redress gross inequalities in the distribution of land and rural incomes from the Ottoman era and the colonial/mandate period. Reform broke up large holdings and distributed them to former peasants and share croppers but tribal control of rangelands, virtually "states-within-states", was also revoked in many countries. The unintended result of this was to take rangelands out of traditional common property management and move them to open access and subsequent uncontrolled use and heavy degradation.

Farmland taken from large landholders was often distributed among peasants in small, marginally economic units (Beaumont, 1989; Hopfinger, 1991). Land reforms in Syria have been followed by introduction of schools, electricity, clean water and all weather roads, resulting in enormous improvement in peasants' lives (Rabo, 1986).

Small and highly fragmented farms in Jordan and the need for soil and water conservation in rainfed areas that require modern equipment and proper management have led the agricultural sector away from the laissez-faire philosophy to land use characteristic of the Ottoman and post Ottoman periods. This philosophy resulted in land speculation, absentee landlords, share cropping on short leases without protection for tenants and ploughing of submarginal land with tractors. Pilot projects in Jordan have been based on the idea of government guided agriculture through legislation and compulsory farm associations or farm cooperatives for supply of credit and inputs and, with land pooling, for proper soil and water management and conservation where needed (Aresvik, 1976). It has been said:

"Some land reform programmes have given rise to the establishment of new rural institutions, of which the most important has been the government managed cooperative. These were intended to provide credit, the means for production and help with marketing. In many countries such co-operatives have not lived up to their expectations. Credit has been insufficient and machinery supply and maintenance a problem Even more important, though, is the fact that farmers do not manage their lands individually. As a result, a feeling of alienation has occurred, with the co-operative officials viewed in much the same light as the landowners they replaced. In other cases state farms have been established in which members are paid either a fixed wage or in relation to the number of hours worked. With such organizations the workers do not feel a close tie with the land and so innovation and commitment have been lacking." (Beaumont, 1989)

Some policy issues needing consideration in respect of rehabilitation and sustainable productivity of rangelands are the appropriate form of institution and the length of tenure. Possibilities include government cooperative, tribal cooperative and private management and long term lease of state land or communal or private ownership with zoning or other restrictions on land use.

Determined efforts have been made to revive the communal 'hema' system of rangeland management in WANA (Draz, 1969). This ancient indigenous system provides reserve supplies of feed while protecting rangeland soils and vegetation (Masri, 1991; Qureshi, 1991). Unfortunately, "owing to the specific socio-economic and political changes of the past two decades in the Near East the idea of reviving the 'hema' system has not met with success" (Janzen, 1991).

The idea of communal land ownership and management seems foreign and unworkable to "westerners" living in societies ingrained with English Common Law. To them, individual responsibility, freehold ownership or contractual leases are the normal way of life. As experiences in Australia, the USA and Canada attest, there certainly can be problems with long term land degradation under their legal systems of land tenure and covenants.

The delineation of these problems in Australia and the unfinished debates on how best to solve them, have been well elaborated (Chisholm and Dumsday, 1987; Harrington et al, 1990) The possibility of communal ownership or leasing of farm or pasture land is never discussed in Australian debates but many of the points and proposals made, particularly for leased public lands, are worth considering in the context of institution building and revival in WANA. There are, however, enormous differences in human cultures and institutions. There are also profound differences in farm size and grazing management with typical Australian farms and flocks being perhaps 100 times larger than farms in WANA. Virtually all sheep grazing management in Australia is done inside boundary fences. Fences are unknown and shepherded grazing is the norm in WANA where farm wages are less than 10 per cent of those in Australia.

Several countries have attempted central planning and control of cropping patterns in their main farming areas, even going so far as to define the proportions of land in each district, and each farm in the district, which must be in which crops. Egypt and Algeria have recently abandoned this policy on grounds of economic efficiency. Among sources of efficiency are the new flexibility that farmers have to match production more closely to price signals and to resources, including their skills, family labour, soils and marketing and employment opportunities.

Numerous studies on optimizing farm plans to maximize profits of crop/livestock enterprises under various prices and constraints have been conducted for specific sites in WANA. These include Turkey (Satana, 1974), Pakistan (Qureshi, 1975), Sudan (Abdelmagid, 1986; 1992; Shomo, 1992), Lebanon (Kizirian, 1970) and Syria (Sting, 1987; Maerz, 1990; Nordblom et al, 1992; Nasser, 1994). Further such studies by NARS will be needed to understand farmers' options under new technologies and new price sets.

Solutions to land use management problems in this region must come from within and be adapted to the conditions in each state.

Seed policy

Most formal plant breeding and conservation and characterization of genetic resources in WANA is done by government institutions with some help from the IARCs. The formal seed sector crosses, selects, tests, multiplies and distributes seed for farmers. Seed is expected to meet certain quality standards for purity and germination. It is through the plant breeding and selection process that high yielding varieties are introduced and germplasm shared among countries.

These services yield public goods, especially of self pollinating species such as wheat, rice, barley, lentils and chickpeas which can be multiplied many generations while retaining their genetic character and advantages. This makes such species unattractive for private breeding companies. Large state farms and public seed corporations in many countries that were established to multiply and distribute improved seed have, however, proved ineffective and failed to meet the diverse crop and varietal requirements of farmers (Jaffee and Srivastava, 1994).

The formal sector releases and multiplies named varieties of high genetic and physical purity. In the case of pasture species such as annual Medicago, however, a pure stand of a single variety is often not as desirable as a mix of species due to the ecological advantages of each species under different weather conditions. The formal seed sector could release pure lines to be blended later and sown as pasture mixes. This practice Unfortunately further divides the small proportion of effort that NARS allocate to pasture and forage seed and multiplies the cost per species as well as reducing the volume of each species offered.

ICARDA and NARS are working to bypass this obstacle outside the formal sector in Algeria, Lebanon, Morocco and Syria. Local pasture species are multiplied and seed harvested by cheap hand sweepers (Christiansen, 1993). Seed pods of the pasture species are sown with a cereal crop. Some of the pasture plants germinate and form a pasture in the next year (Mitri, 1994). Pods can also be sown to rehabilitate non-arable pastures.

An open question for policy discussion is the desired balance in support for the formal and informal seed sectors in view of the fact that the latter is often larger and touches the greatest numbers of poor farmers. The question cannot easily be asked of the mass of farmers comprising the informal sector but NARS and IARCs would do well to seek answers through diagnostic studies in the neglected environments where farmers save their own seed.

Livestock policy

Public investments have tended to be heaviest in veterinary services because major public good can be derived through preventive measures. Veterinary epidemiology, research and extension have been identified as services yielding public goods whose provision is economically justified in most countries (Umali et al, 1994). Veterinary surveillance and quarantine can provide first line defence against introduction of infectious disease. Due, however, to the permeability of many national borders in the region surveillance and quarantine face great practical difficulties.

Governments can also effectively provide public goods through veterinary drug quality control and in food hygiene inspection especially in dairies and abattoirs to avoid moral hazards in these issues. Governments may also sometimes be justified in organizing vaccination campaigns, vector control and diagnostic support. Private goods with consumption externalities include clinical interventions such as diagnosis and treatment and preventive medicine including vaccination, vaccine production and vector control (Umali et al, 1994).

Emphasis and investment in veterinary services in many WANA countries appears greater than in some industrialized countries where livestock contribute a major share to the value of agricultural production. The extreme case is Egypt (Table 2) with one veterinary practitioner for each 431 Veterinary Livestock Units (VLUs). Only Japan in the developed world approaches this intensity. Iraq and Syria also have heavy veterinary cover. Jordan, Lebanon, Tunisia and Oman have about 4500 VLUs per veterinarian compared to 5000 in France. Morocco and Saudi Arabia in WANA proper have the highest ratios. Afghanistan, which can here be classed with the countries of South and East WANA has a ratio of over 20000 to 1 (Table 2).

Several countries forbid slaughter of young or productive female sheep. The practical result is that government abattoirs with inspection facilities mainly process male sheep whereas female sheep are slaughtered in the informal sector (Heylen, 1993). Carcasses without testicles intact cannot be hung on display by many butchers in WANA. The public health hazards of uninspected slaughter are the main problem where market forces quietly override a traditional regulation. Of equal importance, however, is that health inspection facilities are usually available only in the larger cities.

Table 2 Livestock economy and livestock health services in selected WANA and non-WANA developed countries

Country	Livestock share (%) of 1988 value		Veterinary Livestock Unitsa) in 1989 per		Ratio government: private veterinarians
	Agriculture	GDP	Veterinarian	Trained auxiliary	1984	1989
Fast population growth: oil and remittance sensitive
Afghanistan	41.6	-	24828	13659	-	9.4
Syria	33.9	8.1	1238	2804	-	0.3
Jordan	53.4	3.1	4786	18824	1.9	45.3
Transitional population growth: oil and remittance sensitive
Egypt	26.5	6.2	431	642	31.7	8.6
Lebanon	38.9	-	4806	8278	1.3	0.6
Morocco	35.1	-	17082	4534	3.7	2.9
Tunisia	29.5	4.1	4499	5138	25.3	6.8
Turkey	22.7	5.2	6482	-	-	1.3
Oil exporters with large populations
Algeria	41.5	1.7	7813	6105	309.0	8.0
Iraq	34.7	-	1126	1839	-	18.2
Oil exporters with small populations
Libya	49.7	1.5	5256	2766	-	42.2
Oman	-	-	4840	3267	-	9.0
Saudi Arabia	54.1	1.1	16170	10054	-	-
South and East WANA
Ethiopia	40.1	22.9	117754	24363	-	-
Sudan	58.3	21.2	34390	36852	-	-
Pakistan	41.7	14.9	20215	9241	-	-
Non-WANA industrialized countries
Australia	6.4	3.4	8173	14691	0.2	0.2
France	51.6	1.9	5008	15468	0.1	0.1
Japan	54.1	0.3	569	-	-	0.6
United Kingdom	63.8	1.2	2489	7673	0.1	0.1
USA	46.7	1.3	2912	5660	0.1	0.1

Note:
^a) Veterinary Livestock Unit calculated as 1 cow/camel, 2 horses/donkeys, 10 sheep/goats, 100 fowl (these differ from Tropical Livestock Units)

Source: Umali et al, 1994 drawing on various other sources

A further regulation in the main rainfed districts of northern Syria and northern Iraq requires all sheep flocks to move out of the area from mid February until after harvest of the crop. Flocks are allowed to graze weeds and young cereal crops in winter but must be removed later due to the impossibility of preventing night grazing of crops by flock owners. Such regulations are in direct conflict with efforts at better integration of livestock and crop production using sown pastures grown in rotation with cereals.

Veterinary and auxiliary staff numbers do not always translate into services. Compared to their counterparts in developed countries veterinary personnel in WANA often lack equipment, are underpaid and do not have access to transport or communications. Migratory flocks and herds add to the challenge faced by the veterinary services.

Health and productivity are closely linked to nutrition and management. Research and extension on nutrition and management have received lower priority than they deserve in most WANA countries and usually trail behind veterinary services and efforts at genetic improvement. With "public good" justification most WANA countries have embarked on expensive and ineffective breed improvement programmes. Livestock nutrition and management often remain "traditional" and are considered something that farmers already know everything about, thus justifying research neglect.

Vitamin or trace mineral deficiencies in feeds offer the prospect of high economic response through better livestock performance. Sheep are known to be deficient in sodium and vitamin E across wide areas of Jordan, Syria and Turkey, suffer from copper deficiency in the Euphrates valley in Syria, are deficient in selenium in Turkey and are low in zinc in many WANA countries (White, 1994). Supplementary feeding of trace elements and minerals, which is common in the developed countries, remains virtually unknown in WANA.

Use of near infra-red spectroscopy techniques on small samples of straw linked to voluntary intake of the same straw by sheep provides the possibility of inexpensive early screening for quality in plant breeding programmes (El-Haramein and Goodchild, 1993). More efficient integration of livestock nutrition and better cereal/legume crop rotations offer the prospect of more profitable and stable rainfed farming systems (Nordblom et al, 1994).

The open policy question to be faced by NARS and IARCs is the optimal balance of emphasis on veterinary services, animal genetic improvement or better nutrition and management. The last are closely linked to or part of crop and resource management because animal nutrition encompasses crop residues and forage crops, grazing of native pastures and feed grains and feeding of concentrates.

Crop residues and rangeland production are marketable only via livestock. Poor productivity by WANA livestock is linked to poor management on the farm and indigenous livestock often perform well below their genetic potential. In these conditions veterinary services and breed improvement cannot overcome the effects of poor nutrition and management. There is still little quantitative data but considerable work is needed simply to approach the policy question of balances between veterinary, breeding and management work on livestock and the broader balance between crop and livestock research.

Some WANA countries open or close their borders to imports and exports of live animals with sometimes disturbing consequences for their own and their neighbours' enterprises. Some also try to prevent (or require) internal movement of livestock and other commodities among their own provinces.

Egypt and Syria have feed subsidies, credit for sheep fattening cooperatives and systems of feed delivery and storage in the dry rangelands. They have also renovated Roman cisterns for stock water. These measures have contributed to overstocking of rangelands in both countries.

Livestock market information and well differentiated price signals to producers are often poor or totally lacking (Rodríguez et al, 1993). Information is typically limited to verbal contacts and personal observation in local livestock markets where animals are sold at a negotiated price either singly or in lots with condition and weight being judged by feel and sight. Public market transactions lower costs for buyers and sellers and limit the scope for misinformation. Good road and telephone links further serve these purposes as do quality standards or grades for live animals or carcasses. These, however, are usually missing. Government imposed regulations are often intended and used to keep urban retail meat prices down (Mahmood and Rodríguez, 1993). It is not always possible to enforce these regulations and they may, in any case, have negative consequences on both producers and consumers.

Macromixes: projections to 2020

Grain production and consumption

Reviews of grain production and consumption from 1970 to 1990 and projections to 2020 reveal large differences among countries in the range of plausible outcomes (Nordblom and Shomo, 1994). Population and production data from standard sources (FAO, 1993a: USDA, 1994) for two 5-year periods centred on 1970 and 1990 for 18 countries were plotted as 20-year trend lines (Figure 3). Consumption patterns (for both livestock feed and direct human consumption) were also calculated for these periods. Grain use in Jordan, Libya and Saudi Arabia rose very rapidly from 1970 to 1990. Much of the increase was for livestock production as reflected in changes in human diet composition (c.f. Figure 1). Livestock production is chiefly for domestic use in Saudi Arabia and Libya. Syrian and Jordanian merchants and farmers' cooperatives import live animals, grow and fatten them with domestic and imported feed and re-export them to the Gulf states. The poorest countries have not been able to balance population growth with increased production or imports and human diets have declined in quantity. These countries have all suffered civil wars and major economic dislocations. Production in Tunisia, Egypt and Syria has kept pace with population growth while in Morocco it has not. Income growth and concessionary pricing has enabled the import of grains for food but also as poultry and livestock feeds. Demand for grain was greatly in excess of production in Jordan, Libya and Saudi Arabia.

Turkey is the greatest agricultural country of the region and its per caput production is nearly double that of its nearest contenders. Turkey is unique for having promoted a large, sustainable and economically sound agricultural sector and is further developing its substantial renewable water sources for irrigation. Turkey is, and will probably remain, WANA's largest (if not only? net exporter of cereal grains and pulses.

Figure 3 Cereal grain plus pulse consumption and production from 1970 to 1990 and projected production of all graims to 2020 (production trend projected at annual rates of increase of 3 per cent (outstanding), 2 per cent (excellent), 1 per cent (good) and 0 per cent)

Algeria has a substantial agricultural sector producing below its capacity. It imports most of its total grain consumption and funds this by oil exports. Like the other "oil-exporters with large populations" Algeria has given highest priority to industrial development and neglected, in relative terms, rural development and agriculture (EIU, 1992). This group of countries is considered to have, in addition to oil, good physical potential for sustained agricultural growth (Janssen, 1993).

Jordan, Libya and Oman produce the smallest shares of their own consumption and import most of what they use. Changes in levels of domestic production would have negligible impact on human diets. This fact may ultimately limit popular support for major uneconomic investments in increasing grain production instead of pursuing other opportunities such as intensive production of higher value crops.

Saudi Arabia has developed substantial grain production capacity based on irrigation with nonrenewable water supplies and heavy fertilizer use. This is a unique achievement in the region although Libya is proceeding with its "Great Man Made River" project (EIU, 1992). Costs of production in Saudi Arabia are, however, several times higher than world prices even without accounting for the scarcity price of water or costs of subsidies on investments in irrigation (Nurul Islam, pers. comm.)

Opportunities for expanding rainfed or irrigated areas in WANA are few (Alexandratos, 1988). It is in this context that sustainable increases in rainfed agricultural productivity of grain, pasture, forage and livestock are the key focus of ICARDA.

The role of pulses in human diets is greater than their small quantities suggest (Figure 3) due to their high protein and energy contents and their use in diets of the poor as substitutes for animal products (Oram and Belaid, 1990). The faba bean, for example, is known as the "poor man's meat" and lentils and chickpeas are known to have similar nutritive qualities.

Per caput consumption is not projected beyond 1990. An imaginary flat line of constant 1990 per caput consumption (Figure 3) shows the gap with production under the various assumptions. Increases in per caput consumption over 1990 levels are badly needed in the poorest countries. Consumption per caput is unfortunately not a function of want but of effective demand and is directly related to income level. Per caput consumption can be expected to rise to 400 kg or more per year in countries with moderate incomes. Any gap between production and consumption in these countries will be made up by grain imports which in many cases will consist of feed grains.

Current levels of production of all grains can be maintained in most countries although, as in Saudi Arabia, it will not always be economic. Continued efforts will, however, be needed to keep inputs in the form of improved seeds, fertilizers and fuels flowing and for maintaining infrastructure. Attempts to increase production by varying proportions (Figure 3) would require greater effort and technical, economic and political help. High anticipated costs for sustaining very rapid production increases must be balanced against the low costs of imported grain.

Optimization of future national grain production trajectories involves economic analyses of comparative advantages and the taking into account of other sectors, domestic resources and world prices. There is also a large element of political choice capable of ignoring or embracing such arguments.

These analyses are beyond the scope of this paper and no comment is made on the current capabilities for enlightened political choice in the countries of the region. In principle, however, each country should seek out its comparative advantages and may wish to improve them through investments in human capital and infrastructure.

The range of possibilities for economically viable rates of increase in grain production could run to more than five per cent in countries where agriculture has been neglected in favour of industrial growth and where there is still considerable potential for intensification, as in Morocco, Algeria and Iraq. In Iraq, for example, cereal (mainly wheat and barley) areas increased by 100 per cent from 1989 to 1990 and production increased by 130 per cent. Production was down by 21 per cent in 1991, however, due to policy changes, changed incentives for farmers and the weather (USDA, 1993b).

Irrigated cereal production by oil exporting countries with small populations could be reduced by 10 per cent or more annually if they choose to restrict allocation of non-renewable water supplies to high value vegetable crops and domestic uses. They could then buy the needed food and feed grains on the world market and maintain strategic food stocks as appropriate.

Other countries in the region do not have such options and the range for manoeuvre is more restricted. The poorest countries could increase production at more than three per cent annually under peaceful conditions following civil war. Such "peace dividends" are not automatic, however, and depend on wise policy and investments.

The effects of failure to achieve higher production in the face of population increase cannot be quantified in terms of human misery. High population growth rates drag down per caput projections of grain production in most countries. The more "downward-bending" the production projections the higher the projected rates of population growth (Figure 3).

Livestock and feed

Livestock include ruminants for meat and milk, camels and equines for draught and poultry (chickens and turkeys) for eggs and meat (Appendix Table A). Poultry products are gaining importance and account for a major share of animal products in human diets in many countries (USDA, 1993a).

Aggregation of animal populations for each country was done by taking the numbers of each kind and multiplying by a factor for conversion to a Tropical Livestock Unit (TLU) equivalent to a bovine of 250 kg live weight at maintenance. The aggregate TLUs were then plotted against total livestock feed (Figure 4). Feeds include crop residues of all sorts, grazing from native pastures, feed grains and other concentrates (Appendix Table B).

Figure 4 Numbers of Tropical Livestock Units and availability of all feeds in selected WANA countries (arrows are 20-year trends of 5-year averages centred on 1970 and 1990)

Crop residues are not usually reported in national or international statistical series but quantitative estimates can be derived with some confidence by using relevant crop data and multiplying by a factor specific to each (Appendix Table B).

Native pasture or rangeland grazing is the most important feed source for several countries (FAO, 1993a). Offtake quotients are not reported in any regular statistical series but was estimated for 10 countries by multiplying areas by specific assumptions (Appendix Table B).

Feed grains and other concentrates include domestic and imported barley, maize and cottonseed (whole, meal and cake). They are the smallest feed category for most countries in WANA but exceptions are Saudi Arabia, Jordan and Libya (Figure 3) where spectacular increases in feed imports have occurred over the past 20 years.

The mega-trend chart (Figure 4) illustrates the close match between aggregate TLUs and total feed availability in WANA. The envelope of annual dry matter consumption, which assumes feeding levels between three per cent and six per cent of live weight per day, covers the middle range of the region's feed and livestock balances. The three feed categories of crop residues and forage crops, natural pasture grazing, and feed grains and other concentrates, nominally add up to 100 per cent of feed use. Contrasts in feed resource endowments and general shifts towards greater proportions of feed grains and crop residues and away from natural grazing are apparent most spectacularly in Jordan, Lebanon,

Libya and Saudi Arabia (Figure 5). All countries except Afghanistan and Yemen show natural pastures as diminishing proportions of livestock diets. This trend does not take account of rangeland degradation for which there are no quantitative data but is the outcome of increased livestock numbers, increased cropping activities and more use of feed grains.

Figure 5 Livestock diet composition by feed category for selected WANA countries (arrows are 20-year trends of 5-year averages centred on 1970 and 1990)

The general trend towards greater use of feed grains and concentrates is expected to continue. Much of this will be to supplement diets based primarily on crop residues and natural grazing but increasing amounts will be profitably used in expanded feeding operations for dairy, poultry and meat production (Khaldi, 1984; Sarma, 1986).

Domestic feed production in most WANA countries will probably increase with developments in crop and sown pasture production. Quantities of both domestic crop residues and feed grains should increase and there should be better management of cereal crops in rotations including forage and food legume crops (Osman et al, 1990). In spite of this, domestic feed production is unlikely to keep up with demand in most countries.

Feed deficits

If it is assumed that crop production increases are allied to increases in crop residues, forage crops and feed grain production in equivalent proportions it is possible to make projections of the numbers of TLUs that can be supported by domestic feed sources in the future. The analysis further assumes, somewhat optimistically, that natural grazing sources will continue to provide current levels of feed. Subtracting feed imports from each country's total feed resources gives a median estimate of 0.2 TLU per caput for 1990 (Nordblom and Shomo, 1993).

Owing to abundant domestic crop residue and rangeland feed resources some countries show surpluses well above 0.2 TLU per caput (Figure 6). The TLU figures represent, however, the standing populations of livestock and not the annual offtake of livestock products. A deficit in TLUs may be satisfied either by imports of livestock products or imports of feeds at the- rate of 2738 kg dry matter per TLU but such deficits may simply remain unsatisfied. Libya is projected to have a feed deficit of under one million tonnes by 2020 if the country's crop production increases at an annual rate of three per cent and a deficit of about 2.5 million tonnes at one per cent whereas Syria's and Algeria's larger populations, with feed deficits already apparent in 1990, are projected to have very large deficits by 2020 (Figure 6a). Scenarios of one per cent rates of increase in crop production would leave some countries with large deficits in domestic feeds given their projected populations (Figure 6b, Figure 6c).

Figure 6 Projected surpluses and deficits in Tropical Livestock Units and livestock feed assuming constant demand of 0.2 TLU per caput (lines begin with a node at 1990 and branches spread to nodes for 2000, 2010 and 2020 for 1 per cent, 2 per cent and 3 per cent rates of increased crop production)

It is most likely that oil rich countries such as Libya, Saudi Arabia and the Gulf states will buy more live stock products and feeds as their populations grow. Poorer countries with the majority of WANA's population will have access to ever smaller amounts of animal protein per caput beginning from an already low base (c.f. Figure 1). Pulses, which substitute for livestock products at lower incomes, will probably gain in importance in absolute terms. Even substantial increases of three per cent per annum will not reduce the decline in per caput TLUs capable of being supported by domestic feed production in most countries of the region.

Macro conclusions

"Oil exporters with small populations" must be considered separately from other countries of the region based not only on their oil wealth but also on poor prospects for economically sustainable large scale agriculture. If, as is done elsewhere (Alexandratos, 1993; Mitchell and Ingco, 1993; Rosegrant and Agcaoili, 1994), these countries are included in overall analyses the prospects for food purchase of "food-deficit WANA" are overstated.

"Oil exporters with large populations" are another matter in terms of agricultural potential and the financial resources to develop them. They form a subgroup which should be distinguished from the rest of the region. Each country in this group is projected to have annual feed deficits of some 10 million tonnes by 2020. The combined (and rapidly growing) population of 101 million means that these countries deserve special attention.

The "oil and remittance sensitive countries" are at various levels of development but rely on rainfed agriculture and pastures for much of their production. With the exception of Turkey, and possibly Tunisia, countries in this group will all face food and feed deficits by 2020.

Egypt and Pakistan each have large populations, agriculture that is heavily dependent on rivers (although there is some rainfed farming) and extensive dry rangelands. Both countries deserve separate attention and Egypt in particular has the current largest feed deficit in WANA. Turkey should also be considered separately and not aggregated with the rest of WANA because its good soil and water resources otherwise mask the conditions of the poorer countries. Turkey and Tunisia are the only countries in WANA proper projected to support more than 0.2 TLU per caput in 2020. Other analyses (Alexandratos, 1988; 1993; Mitchell and Ingco, 1993; Rosegrant and Agcaoili, 1994) do not treat Turkey separately and therefore overstate the agricultural capacity of the "food-deficit states".

The problems in WANA agriculture cannot be solved by averaging the poor with the rich on paper but by raising the capacities of the poor. A perceptive person recently remarked: "The WANA region has a glorious past, a worrying present and a frightening future". Perhaps the "fright" comes with the realization that food self sufficiency is not attainable for most WANA countries in view of their enormous population increases. Food demand has outgrown domestic production in most countries and will probably continue to do so.

Greater integration with world markets will become more essential as will investments in human capital, research and technology development. Each WANA country will benefit by developing its agriculture in a manner consistent with its particular comparative advantages. Being a large importer of food and feed grains is not a catastrophe provided a country develops stable means of financing imports through service industries, manufacturing and advantageous nonagricultural and agricultural exports. Where these means are not actively developed serious food crises can be expected in the future.

In respect of comparative advantages in meeting livestock/feed deficits by importing feed, an alternative is to import live animals and frozen or processed livestock products. Policies favouring feed and livestock product imports at prices below domestic production costs could benefit consumer diets but damage domestic livestock and feed producers. In some cases this could have favourable environmental impact, through lower price incentives and by discouraging risky cereal cultivation and overgrazing in the rangelands. This, and the obvious prospects of a rising import market for low cost feed grain and livestock products, should attract the interest of the major feed and livestock exporting countries.

Public goods will be needed over the next 20 years to improve human welfare and sustain natural resources. Public goods resulting from international research but for which full funding from national or private sources cannot be justified or expected include:

· human capital and institutional strengthening in rangeland tenure and the making of decisions at the farm level (pastoralists and farmers with insecure tenure cannot be expected to consider sustainable resource use but major economic gains appear feasible by devolving property rights and decision making from state bodies to individual farmers or well organized small groups);

· evaluation, conservation and improvement of plant (particularly wheat, barley, lentils, chickpeas, faba bean, forage crops and pasture species whose centres of origin and wild relatives are in WANA) and animal genetic resources (for example, the hardy fat tailed sheep found from Tunisia through central Asia and prized for milk and meat production) by international institutions working in partnership with NARS;

· helping the informal seed sector to provide improved services by support for strategic research on methods followed by well informed extension; and

· giving increased attention to improving livestock management and nutrition on the farm and encouraging economic balance with the current emphasis on national veterinary and animal breeding investments in WANA.

It is no coincidence that ICARDA's mandate is focused on the strategic research areas designed to yield public goods information and the building of human capital for the sustained benefit of mankind. Partnerships with NARS are as much a key as are concerns for sustaining natural resource use and the welfare of resource users and consumers. Partnerships with other CGIAR centres and non-CG centres of excellence will be essential to bring resources and critical masses of research skill to bear on many of these issues.

Micromixes: projections to 2020

Feeding/grazing calendars in mixed systems

System specific feeding and grazing calendars for sedentary farmers in northeast Syria, agropastoralists in northwest Syria and sedentary (semi-intensive sheep/cereal) farmers and pastoral (goat) herders in the Bekaa Valley of Lebanon are presented here (Goodchild et al, 1995). Northeast Syria is a semiarid rainfed crop zone with about 400 mm mean annual precipitation but where there is also some irrigated cotton (Nordblom et al, 1993a). The northwest Syria study focuses on flocks with winter bases in rangeland-barley areas with about 230 mm mean annual precipitation that move in summer and autumn to graze crop residues in higher rainfall and irrigated areas (Wachholtz et al, 1993). Pastoral goat flocks in Lebanon typically graze foothills and rangelands in spring, summer and autumn whereas sheep of the sedentary farmers stay on or near their home farms the whole year (Hamadeh et al, 1994).

Feed calendars were developed (Figure 7) based on a single round survey of more than 30 farmers in each system (Nordblom et al, 1993b). Characteristics common to all systems are the marked seasonal diversity in feed sources, hand feeding of cereal straw and concentrates in the winter and stubble grazing in summer. There are differences, however, in the importance of range grazing and in the starting and ending dates and importance of this feed source. Access to various crop residues, feed grains and agro-industrial by-products such as sugar beet pulp, wheat bran and cottonseed cake is also reflected in the differences in diets among the four systems.

Figure 7 Composition of small ruminant diets as reported by farmers in single round surveys in a. northeast Syria, northwest Syria, e. a semi-intensive system in Lebanon, and d. a pastoral system in Lebanon (one symbol, 4 per cent of diet; upper case symbol, hand feeding; lower case symbol, grazing; C, cereal straw; L, legume straw; E, energy feeds; P. protein supplements; s, cereal stubble; r, higher nitrogen crop residues; u, unharvested crops; g, pasture and range grazing)

This type of system description allows evaluation of farmers' current practices before designing trials aimed at improving mixed farming systems. It is also useful in economic analyses comparing current practices with proposed improvements.

Whole-farm models of mixed farming systems

Production practices, costs and constraints were established in the single round survey (Nordblom et al, 1993a; Nasser, 1994). A long term rotation and grazing trial had been established with the Syrian Ministry of Agriculture near Kamishly in northeast Syria in 1986. Lentils, vetch, fallow, wheat and medic pastures were managed in 2-year rotations with wheat, both phases of each rotation being present every year. Crop yield and pasture offtake estimates for eight years were included in a whole-farm model (Shomo et al, 1995). Market price survey data of inputs and outputs of the mixed system from 1986-1994 were also used, these being deflated to 1992 Syrian Lira (SL 42 = US $ 1) using the Laspeyres general wholesale price indices for 1986-1993 and a projected index for 1994.

Medic pasture competes with other crops for land and labour and grazing offtake in spring and summer compete with other feed sources on the farm, including rangeland and roadside grazing, cereal straw and stubble residues and purchased feeds such as barley grain and cottonseed cake. Medic pasture may be used by breeding flocks of ewes and their lambs or by lambs bought for fattening. Optimal grazing management of medic pasture is affected by:

· quantity and quality of pasture production over time;
· feed demand of sheep over time and as affected by pregnancy and lactation;
· quality, quantity, time of availability and costs of all other home grown or bought feed resources; and
· prices of sheep products, other farm outputs, other farm inputs and wages.

All these are explicitly accounted for in the whole-farm linear programming model (Shomo et al, 1995). Pasture offtake estimates were derived from recorded grazing days multiplied by estimates of dry matter intake per sheep per day and scaled to appropriate values for all four seasons.

The number of ewes is a "strategic" model variable and is optimized over the 8-year run in relation to prices and availability of all feed sources and prices of lambs and milk. Sheep diets are simultaneously optimized for minimum cost while meeting minimum dietary levels of crude protein and metabolizable energy and maximum levels of dry matter intake among the four seasons in each year. The amounts of each feed in each season are "tactical" variables in the model as are the amounts of family and hired labour used and the quantities of crops sold. The model can be used to solve with profit maximizing or non-linear objectives, the latter being for risk aversion where farmers wish for the highest possible average income while avoiding low outcomes in any particular year. The profit maximizing results for each rotation are shown separately for an assumed farm size of 16 ha, typical family labour availability and no access to range grazing.

The optimum number of ewes- and their share of lambs, rams and replacement yearlings is part of each solution of the model. With a model run for each of the crop rotations separately the sequence of annual whole-farm incomes after all costs over the 8-year period can be shown (Figure 8). The results in terms of optimal ewe numbers and average income can also be shown (Figure 9). The lentil-wheat rotation outperforms others in terms of highest net whole-farm income partly because of the quantities of lentil straw and leaves produced and which can be stored to reduce winter feed purchases.

Figure 8 Annual whole-farm income for a 16 ha farm for crop rotations with wheat (Kamishly area, northeast Syria)

The model results are best viewed as expected means and variances of net whole-farm income over time (Figure 9). Standard deviations and coefficients of variation (CV) are used as measures of income variation to allow easy comparison among rotations. Lentil-wheat, for example, offers not only the highest mean but the lowest CV in whole-farm incomes and there is little doubt why this rotation is the most common in the area. Wheat-wheat and wheat-medic, in contrast, provide the lowest means and highest CVs. Fallows and vetch are not widely used in the study area and medic remains virtually unknown to farmers except as a local weed.

Figure 9 Mean and standard deviation of whole-farm income for a 16 ha farm for crop rotations with wheat (Kamishly area, northeast Syria)

Continuous wheat-wheat has been required by government regulations in recent years in order to maximize production of this strategic crop. The resulting low yields and incomes have now been recognized and the regulations rescinded.

Similar results were obtained with a model based on price and growing conditions for a slightly lower rainfall zone (330 mm) in northwest Syria over the period 1985-1993 (Nordblom et al, 1995). Crop and pasture offtake estimates are from a large crop rotation and grazing trial at ICARDA's main station, 35 km south of Aleppo. Farm and market surveys in the area provided the economic data.

Farms with 16 ha and 64 ha of arable land are considered as typical and large farms in this area. Barley yielded better than wheat after medic because it matures earlier. Barley prices are always lower than wheat but medic-barley is an additional rotation option in northwest Syria. The lentil-wheat rotation was again best in this model as it was in northeast Syria. With the model set to find an optimal mix of rotations barley-medic entered the farm plan at low but significant levels with the major proportion of the farm going to lentil-wheat. This is explained by a low correlation between lentil yields and medic pasture offtake such that the combined potential for economic crop and feed production is better than either rotation used alone.

All model runs were made with the "standard" set of prices and yields observed over the 8-year study period and the effects on the medic/barley area of changes in prices of sheep were considered for risk averse and profit maximizing (risk neutral) farmers.

Price levels were set at 80, 90, 100, 110 and 120 per cent of the standard and a sensitivity analysis was done representing the outcome of 40 runs plotting only the "highest-utility" results among 20 runs with sheep and 20 without (Figure 10). The no sheep option removes the labour needed for herding. At the 100 per cent cost/price condition the risk averse solutions for both 16 ha and 64 ha farms call for greater areas of barley/medic pasture than the risk neutral solutions. In addition, and for both risk levels, barley/medic takes a larger share of crop land of the 64 ha than of the 16 ha farm. This can be explained by the relative shortage of family labour on the larger farm linked with the low labour demand of medic compared to lentils.

Reduced or increased sheep prices relative to the standard 100 per cent (Figure 10) call for corresponding decreases or increases in the optimal barley/medic pasture area on both farm sizes if the farmer is risk averse. With sheep prices at 80 per cent of standard, ceteris paribus a no sheep-no medic strategy would be best for risk averse farmers on both large and small farms. With sheep prices at 120 per cent of standard nearly a quarter of the area of small farms and a third of large farms would be devoted to the barley/medic rotation. The effects of different sheep prices are less pronounced for risk neutral farmers.

Figure 10 Optimal (per cent of total) of barley/medic rotation on "typical " 16 ha and large 64 ha farms at variable sheep prices in northwest Syria

The effects are dramatic if lentil prices are allowed to oscillate around the standard level, (Figure 11). As lentil/wheat in the rotation is the strongest competitor for land with barley/medic, higher lentil prices lead to an increase in the share of lentils in the farmer's portfolio of rotations. Lower lentil prices, on the other hand, open the way for greater allocations to barley/medic.

Figure 11 Optimal area (per cent of total) of barley/medic rotation on "typical " 16 ha and large 64 ha farms at variable renal prices in northwest Syria

Effects of access to rangeland on interest in medic pasture

The adaptability of sheep to various diets and the ability of shepherds to direct them to the cheapest available feed source in a given season and year have so far been reflected in the model on the assumption that sheep must stay on the farm. Many farmers, however, have access to grazing off the farm, including rangelands and roadsides. The use of such areas is typically heavy and persistent by multiowner flocks often under open access conditions but sometimes under common access by a closed group to a village or cooperative grazing area. Exclusive use of a range grazing area by a single farmer is very rare in Syria (Nordblom, 1993).

The grazing offtake potential from a given area of rangeland and yields of rainfed crops vary from year to year depending on the weather. Rangeland offtake in the study area, derived from a long term grazing trial, is estimated at 50 per cent of biomass production (Osman et al, 1991; Osman, 1992; Osman and Bahhady, 1993). It can also be assumed (Nordblom and Thomson, 1987) that rangeland can be grazed in winter and/or spring. Taking all winter grazing would preclude spring grazing, and vice versa. Linear combinations of the options are available each year (Nordblom et al, 1995). Costs associated with the rangeland in addition to herding, are assumed to include an annual maintenance application of triple superphosphate fertilizer similar to that under which the trial observations were made.

The assumption limiting range grazing to the winter and spring months (January-May) is justified by results of surveys of flock owners in northern Syria in which rangeland use was reported to be neligible at other times mainly due to the seasonal availability of other feeds and especially cereal stubbles and straws (Nordblom, 1987; Wachholtz et al, 1993).

Model runs showed consistent results for both farm sizes and both levels of risk in that exclusive access to 10 ha of rangeland led to lower strategically optimal areas of barley/medic and larger flock sizes (Figure 12). Another consistent result was that risk averse farmers of either farm size wood prefer smaller flocks and larger barley/medic areas than those chosen by risk neutral farmers regardless of the area of rangeland up to 50 ha. With 30 ha of rangeland the risk neutral farmer with 16 ha of farmland would drop barley/medic from the optimal portfolio while maintaining a substantial flock size.

Figure 12 Optimal area (per cent of total) of barley/medic rotation and sheep flock size on "typical" 16 ha and large 64 ha farms with exclusive access to additional areas of range grazing in northwest Syria

Due to substitution of cheap range grazing for relatively expensive medic pasture (opportunity costs of not growing higher value wheat crops) greater access to rangeland means lowered interest in medic pasture. In the major farming districts, however, many farmers have little, and very rarely exclusive, access to substantial areas of rangeland.

The implications of access to rangeland on choice of farming activities have not yet been fully explored but these results provide new insights. The income generating capacity added to a rainfed farm by well defined access to rangeland raises the question of economic incentives for maintaining and increasing range production through better management. This is likely to be a key to supporting institutional reforms for rangeland property rights.

The analysis reveals a positive but limited economic niche for medic pastures in northwest Syria. The size of this niche is sensitive to a number of variables such as sheep and lentil prices, farm size and access to rangeland. Prices are subject to change as the economy develops in relation to world markets. Current tariff protection of domestic sheep and lentil prices, for example, provides plus and minus incentives for medic pasture.

Micro conclusions

The clear lesson to be drawn from the micro-analyses of the diet calendars and whole-farm models in the light of the macroanalysis which demonstrates vast differences in national feed resources among countries is that different farming systems with different optimal mixes of crop and livestock activities will be found across the region. From the farmers' viewpoint optimal mixes of activities will shift with changing prices of inputs and outputs as well as with changes in technology or policy that alter production possibilities. Both technology and policy are susceptible to improvement for more sustainable and productive mixed farming systems in WANA. This provides some relief to the region's rather gloomy prospects for rapidly declining self sufficiency in livestock production in view of the expected large increases in human populations to 2020 and beyond.

Acknowledgements

This paper benefited from inputs by Euan Thomson, Gus Gintzburger, Faik Bahhady, Abelardo Rodríguez, Hazel Harris, Tony van Gastel, Scott Christiansen, Liz Bailey and Rick Tutwiler at ICARDA. Maurice Saade (IDRC-Cairo), Nurul Islam, Peter Oram and Steve Vosti (IFPRI-Washington DC) are thanked for critical comments on earlier drafts. The seed policy section benefited from discussions with A van Gastel and S Christiansen (ICARDA). E F Thomson, F Bahhady and A Rodríguez contributed to the livestock policy section. The help and collaboration of S Hamadeh and M Abi Said (American University of Beirut), A Toum and A Raad (Lebanese Ministry of Agriculture), A Jumaa (Lebanese University), E Showyri and R Shaaban (ARI, Lebanon), K Dendal and S Nasser (Syrian Ministry of Agriculture and Agrarian Reform), R Wachholtz (Hohenheim University, Stuttgart), A Osman, A Nassar, H Sawmy Edo and A Semman (PFLP, ICARDA) and G Arab (Aleppo) is acknowledged for the diet calendar work. Important contributions to the whole farm modelling were made by David Pannell (University of Western Australia) as well as the other co-authors of the cited sources.

References

Abdelmagid SA. 1986. An economic analysis of dairy/forage enterprises in the Rahad Agricultural Project, Sudan (M. Sc. Dissertation). Faculty of Economics and Rural Development, University of Gezira: Wad Medani, Sudan.

Abdelmagid SA. 1992. The economic impact of faba bean introduction in smallholdings: a case study of the Gezira Scheme, Sudan. In: Doppler W (ed) Farming systems and resource economics in the tropics, Volume 13. Wissenschaftsverlag Vauk: Kiel, Germany.

Abu Akkada AR. 1984. Evaluation of present status and potential development of animal feed resources in Arab countries. Volume 10. Arab Republic of Egypt. Arab Center for the Studies of Arid Zones and Dry Lands: Damascus, Syria [Arabic].

ACSAD/AOAD. 1985. Evaluation of present status and potential development of animal feed resources in Arab countries. Arab Center for the Studies of Arid Zones and Dry Lands/Arab Organization for Agricultural Development: Khartoum, Sudan.

Alexandratos N. 1988. World agriculture: toward 2000. Belhaven Press: London, UK.

Alexandratos N. 1993. Agriculture: towards 2010. Food and Agriculture Organization: Rome, Italy.

Aresvik O. 1976. The agricultural development of Jordan. Praeger: New York, USA.

Beaumont P. 1989. Environmental management and development in drylands. Routledge: London, UK.

Chisholm A and Dumsday R (eds). 1987. Land degradation, problems and policies. Cambridge University Press: London, UK.

Christiansen S. 1993. A medic sweeper and a thresher help farmers produce pasture seed. In: World agriculture 1994: the international review of agribusiness in developing markets. Sterling Publications Ltd: London, UK. 111-114.

Draz O. 1969. The "hema" system of range reserves in the Arabian Peninsula: its possibilities in range improvement and conservation projects in the Near East. Food and Agriculture Organization: Rome, Italy.

EIU. 1992. Country profile 1992-93. Annual survey of political and economic background (various countries). The Economist Intelligence Unit, Business International Limited: London, UK.

El-Haramein FJ and Goodchild AV. 1993.. Using near infra red spectroscopy to predict the voluntary intake and composition of barley straw Pasture, Forage and Livestock Program Annual Report 1992 (ICARDA-070)). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Fadda NR 1992. Foreword. In: Ryan J and Matar A (eds) Fertilizer use efficiency under rain-fed agriculture in West Asia and North Africa (ICARDA-067). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

FAO. 1993a. Agrostat tapes. Food and Agriculture Organization: Rome, Italy.

FAO. 1993b. Production yearbook. Volume 47. Food and Agriculture Organization: Rome, Italy.

Goodchild AV, Nordblom TL, Shomo F. Hamadeh SK and Wachholtz R. 1995. Using feed calendars to interface feeds with needs in Syria and Lebanon. Paper presented at the 3rd international livestock fanning systems symposium, 1-2 September 1994. Aberdeen, Scotland.

Hamadeh SK, Shomo F. Goodchild AV and Nordblom TL. 1994. A rapid survey of small ruminant production in the Bekaa Valley of Lebanon: August 1993 (Pasture, Forage and Livestock Program Annual Report, 1993). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Harrington GN, Wilson AD and Young MD (eds). 1990. Management of Australia's rangelands. Commonwealth Scientific and Industrial Research Organization: Melbourne, Australia.

Hassan NI, Abou-Raya AK, Abou-Akkada AR and Wardeh MF. 1982. Evaluation of present status and potential development of animal feed resources in Arab countries. Volume 17. Yemen Arab Republic. Arab Center for the Studies of Arid and Drylands: Damascus, Syria [Arabic].

Heylen K. 1993. Sheep marketing in Aleppo (M. Sc. Dissertation). Institute of the Tropics, University of Leipzig: Leipzig, Germany [in German].

Hopfinger H. 1991. Oeffentliche und private Landwirtschaft in Syrien (Erlanger Geographische Arbeiten Sonderband 19). Universitätsbuchdrückerei Junge & Sohn: Erlangen, Germany.

Jaffee S and Srivastava J. 1994. The roles of public and private sectors in enhancing the performance of seed systems. World Bank Res. Obs. 9: 97-117.

Jahnke HE. 1982. Livestock production systems and livestock development in tropical Africa. Kieler Wissenschaftsverlag Vauk: Kiel, Germany.

Janssen WG 1993. Economic and agricultural development in West Asia and North Africa: the need for agricultural research. Food Policy 18: 507-522.

Janzen J. 1991. The revival of traditional pastoral systems in the Near East: a survival strategy for pasture lands and mobile livestock keeping? Keynote address presented at the FAO/CARDNE workshop on pastoral communities in the Near East: traditional systems in evolution, 1-5 December 1991, Amman, Jordan.

Kategile JA, Said AN and Sondstol F (eds) 1981. Utilixation of low qualify roughages in Africa (Agricultural Development Report N° 1). AUN: Aas: Norway.

Khaldi N. 1984. Evolving food gaps in the Middle East/North Africa: prospects and policy implications (Research Report N° 47). International Food Policy Research Institute: Washington DC, USA.

Kizirian HM. 1970. A linear programming model for optimum agricultural crop enterprise selection in the Central Beqa'a Valley region of Lebanon (MS Dissertation). Department of Agricultural Economics, American University of Beirut: Beirut, Lebanon.

Kosilla V. 1988. The availability of crop residues in developing countries in relation to livestock populations. In: Reed JD, Capper BS and Neate PJH (eds) Plant breeding and the nutritive value of crop residues. International Livestock Centre for Africa: Addis Ababa, Ethiopia. 29-39.

Maerz U. 1990. Farm classification and impact analysis of mixed farming systems in Northern Syria. In: Doppler W (ed) Farming systems and resource economics in the tropics, Volume 7. Wissenschaftsverlag Vauk: Kiel. Germany.

Mahmood K and Rodríguez A. 1993. Marketing and processing of small ruminants in highland Balochistan, Pakistan. Small Rumin. Res. 10: 93-102.

Mitchell DO and Ingco MD. 1993. The world food outlook. November 1993. International Economics Department, World Bank: Washington DC, USA.

Mitri J. 1994. Medic pasture establishment using pods sown into barley in the year prior to the pasture phase (M. Sc. Dissertation). Lebanese University: Beirut, Lebanon.

Nasser S. 1994. Agricultural systems analysis in dryland areas: Al-Hassakeh Province (M. Sc. Dissertation). Department of Agricultural Economics, Agriculture Faculty, University of Damascus: Damascus, Syria.

Noor M. 1987. Rangeland management in Pakistan. International Center for Integrated Mountain Development: Kathmandu, Nepal.

Nordblom TL. 1987. Farming practices in southern Idleb Province, Syria: survey results in 1985 (ICARDA-107). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Nordblom TL. 1988. The importance of crop residues as feed resources in West Asia anfd North Africa. In: Reed JD, Capper BS and Neate PJH (eds) Plant breeding and the nutritive value of crop residues. International Livestock Centre for Africa: Addis Ababa, Ethiopia. 41-63.

Nordblom TL. 1992. Land tenure, social systems and institutions for dryland fanning. Paper presented at the seminar on natural resources and environmental management in the dry areas, 16-27 February 1992. International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Nordblom TL. 1993. Ley farming in the Mediterranean from an economic view. In: Christiansen S, Materon L, Falcinelli M and Cocks PS (eds) Introducing ley farming into the Mediterranean basin (ICARDA-043). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Nordblom TL and Shomo F. 1993. Livestock and feed trends in West Asia and North Africa: past, present and future. In: The situation of agriculture in Mediterranean countries, N° 5, Volume 1, Agricultural research and food security. Centre International des Hautes Etudes Mediterrannéennes: Paris, France. 15-30.

Nordblom TL and Shomo F. 1994. Disaggregated food and feed trends to 2010 in the West Asia/North Africa region. Paper prepared for the roundtable meeting on population and food in the early 21st Century: meeting future food needs of an increasing world population, 14-16 February 1994. International Food Policy Research Institute: Washington DC, USA.

Nordblom TL and Thomson E 1987. A whole-farm model based on experimental flocks and crop rotations in northwest Syria (ICARDA-102). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Nordblom TL, Christiansen S. Nersoyan N and Bahhady F. 1992. A whole-farm model for economic analysis of medic pasture and other dryland crops in two-year rotations with wheat in northwest Syria (ICARDA-066). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria.

Nordblom TL, Dendal K, Sawmy Edo H, Nasser S, Shomo F, Bahhady F and Christiansen S. 1993a. Survey of crop/livestock systems in Hassakeh Province in northeast Syria (Pasture, Forage and Livestock Program Annual Report, 1992. ICARDA-070). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria. 161-170.

Nordblom TL, Goodchild AV and Wachholtz R. 1993b. Rapid characterization of feeding and grazing calendars: techniques for single-visit interviews, summary and graphic presentation (Pasture, Forage and Livestock Program Annual Report, 1992. ICARDA-070). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria. 1-12.

Nordblom TL, Pannell DJ, Christiansen S. Nersoyan N and Bahhady F. 1994. From weed to wealth? Prospects for medic pastures in the Mediterranean farming system of north-west Syria. Agric. Econ. 11: 29-42.

Nordblom TL, Pannell DJ, Nersoyan N, Bahhady F, Osman AE and Christiansen S. 1995. Where does medic pasture fit in to the risky dryland farming and rangeland environment of northwest Syria? An economic analysis. International Center for Agricultural Research in the Dry Areas: Aleppo, Syria (mimeo).

Oram P and Belaid A. 1990. Legumes in farming systems, a joint ICARDA/IFPRI report (ICARDA-160). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria (mimeo).

Osman AE. 1992. Response of Mediterranean grassland to phosphate fertilization: results of 1989/1990 and 1990/1991 seasons (Pasture, Forage and Livestock Program Annual Report for 1990/1991. ICARDA-039). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria. 126-135.

Osman AE and Bahhady F. 1993. Effects of residual phosphate on marginal lands (Pasture, Forage and Livestock Program Annual Report for 1992. ICARDA-070). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria. 171-182.

Osman AE, Ibrahim MH and Jones MA (eds) 1990. The role of legumes in the farming systems of the Mediterranean areas. Kluwer Academic Publishers: Dordrecht, The Netherlands.

Osman AE, Cocks PS, Russi L and Pagnotta MA. 1991. Response of Mediterranean grassland to phosphate and stocking rates: biomass production and botanical composition. J. Agric. Sci., Cambridge 116: 37-46.

Qureshi AR. 1975. Optimal crop and dairy enterprise selection for small farmers in Sialkot District, Punjab, Pakistan (MS Dissertation). Department of Agricultural Economics, American University of Beirut: Beirut, Lebanon.

Qureshi AW. 1991. Biological limits to animal production from pastoralism in the Near East. Paper presented at the FAO/CARDNE workshop on pastoral communities in the Near East: traditional systems in evolution, 1-5 December 1991, Amman, Jordan.

Rabo A. 1986. Change on the Euphrates: villagers, townsmen and employees in Northeast Syria (Studies in Social Anthropology N° 15). Department of Social Anthropology, University of Stockholm: Stockholm, Sweden.

Rodríguez A, Ali I, Afzal M and Shah NA. 1993. Marketing of goat and sheep skins in highland Balochistan, Pakistan. Small Rumin. Res. 12: 259-270.

Rosegrant MW and Agcaoili M. 1994. Global and regional food demand, supply and trade prospects to 2010. Paper prepared for the roundtable meeting on population and food in the early 21st century: meeting future food needs of an increasing world population, 14-16 February 1994. International Food Policy Research Institute: Washington DC, USA.

Sarma JS. 1986. Cereal feed use in the Third World: past trends and projections to 2000 (Research Report N° 57). International Food Policy Research Institute: Washington DC, USA.

Satana S. 1974. Farm planning under risk in Mustafakemalpasa County, Bursa Province, Turkey (MS Dissertation). Department of Agricultural Economics, American University of Beirut: Beirut, Lebanon.

Shomo F. 1992. Application of computing methods to food security problems in North Kordofan Province: case study of Um Ruwaba District (M. Sc. Dissertation). Department of Rural Economy, Faculty of Agriculture, University of Khartoum: Khartoum, Sudan.

Shomo F, Nordblom TL, Pannell D, Redman E, Nasser S, Dendal K, Melki G, Bahhady F, Sawmy Edo H and Christiansen S. 1995. A whole-farm risk model for economic analysis of medic pasture and other dryland crops in two-year rotations with wheat in northeast Syria. International Center for Agricultural Research in the Dry Areas: Aleppo, Syria (mimeo).

Slayman F, Daghir N and Saoud N. 1986. Encyclopedia of animal resources in the Arab countries. Volume 14, Lebanon. Arab Center for the Studies of Arid Zones and Dry Lands: Damascus, Syria [Arabic].

Spiller GA. 1991. The Mediterranean diets in health and disease. Van Nostrand Reinhold: New York, USA.

Sting J. 1987. The economy of the integration of livestock and trees in a dry area of Syria (M. Sc. Dissertation). Institute of Agricultural Economics and Social Sciences in the Tropics and Sub-Tropics, University of Hohenheim: Stuttgart, Germany.

TAC/CGIAR. 1992. Review of CGIAR priorities and strategies. Part 1. TAC Secretariat, Food and Agriculture Organization: Rome, Italy (mimeo).

Tutwiler RN and Bailey E. 1991. Agricultural growth and sustainability: conditions for their compatibility in the semi-arid tropics (Middle East). Paper prepared for DSE/IFPRI conference on "Agricultural sustainability, growth and poverty alleviation: issues and policies", 23-27 September 1991, Feldafing, Germany.

Umali DL, Feder G and de Haan C. 1994. Animal health services: finding the balance between public and private delivery. World Bank Res. Obs. 9: 71-96.

UN. 1993. World population prospects: the 1992 revision. Department for Economic and Social Information and Policy Analysis, United Nations: New York, USA.

USDA. 1990. World agriculture trends and indicators, 1970-89 (Statistical Bulletin N° 815). Economic Research Service, United States Department of Agriculture: Washington DC, USA.

USDA. 1993a. Africa and the Middle East. International agriculture and trade reports (Situation and Outlook Series RS-93-3). Economic Research Service, United States Department of Agriculture: Washington DC, USA.

USDA. 1993b. World agriculture trends and indicators, 1970 91 (Statistical Bulletin N° 861). Economic Research Service, United States Department of Agriculture: Washington DC, USA.

USDA. 1994. Production, supply and distribution: database and software on diskette (PS&D VIEW '92 Users Manual and Database). Economic Research Service, United States Department of Agriculture: Washington DC, USA.

Wachholtz R. Nordblom TL and Arab G. 1993. Characterization of year-round sheep feeding and grazing calendars of bedouin flocks in the NW Syrian steppe (Pasture, Forage and Livestock Program Annual Report, 1992. ICARDA-070). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria. 215-236.

Wardeh MF, Abou-Raya AK, Abou-Akkada AR and Hassan NI. 1982. Evaluation of present status and potential development of animal feed resources in Arab countries. Volume 20. The Sultanate of Oman. Arab Center for the Studies of Arid Zones and Dry Lands: Damascus, Syria [Arabic].

Warriner D. 1962. Land reform and development in the Middle East: a study of Egypt, Syria and Iraq (Second Edition). Oxford University Press: London, UK.

White CL. 1994. A study of the vitamin and mineral status of sheep in West Asia (Syria, Jordan and Turkey). International Center for Agricultural Research in the Dry Areas: Aleppo, Syria (mimeo).

World Bank. 1993. World tables 1993. Johns Hopkins University Press: Baltimore, USA.

Appendix Table A Trends in livestock numbers (millions) in WANA, 1968-1972 to l988-1992

Sources: Livestock numbers from FAO, 1993a; livestock factors from Jahnke, 1982 - except buffalo from Kosilla, 1988

Appendix Table B Feeds and pasture offtakes ('000 tonnes) for livestock in selected WANA countries in 1968-1972 and 1988-1992

Notes:

^a) Pasture offtake is for single years, 1970 and 1990 and is estimated for Algeria at 300 kg/ha; Egypt, 90; Jordan, 79; Libya, 100; Iraq, 110; Morocco, 500; Saudi Arabia, 20; Sudan, 500; Syria, 200; and Tunisia, 600 (ACSAD, 1985); Pakistan, (660 Noor, 1987); Lebanon, 470 plus 450000 tonnes forest grazing (Slayman et al, 1986); Afghanistan, Ethiopia, Iran and Turkey assumed at 500; Yemen, 500 (Hassan et al, 1982); Oman, total estimate 218000 tonnes (Wardeh et al, 1982)

^b) Million tonnes

^c) Crop yield data from FAO, 1983a multiplied by these factors based on Kosilla, 1988 and Nordblom, 1988

Sources: FAO, 1993a except berseem in Egypt (Abu Akkada, 1984), Lebanon (Slayman, 1986), Oman (Wardeh et al, 1982) and Pakistan (Noor, 1987) and teff in Ethiopia (Kategile et al, 1981)