Previous Page Table of Contents Next Page

1.4 Rangeland development

1.4.1 Overview of livestock development projects
1.4.2 History of lowlands development and the TLDP
1.4.3 The SERP and the Pilot Project
1.4.4 Has national range development been successful?
1.4.5 Development of the southern rangelands as coordinated by SORDU
1.4.6 Collaboration among research and development institutions in the southern rangelands
1.4.7 Interaction between research and development and project impact

1.4.1 Overview of livestock development projects

Ethiopia has long collaborated with the World Bank, African Development Bank (ADB), African Development Fund (ADF), International Development Association (IDA) and other lending institutions in economic development programmes. This has included assistance with a series of livestock development projects that continues today. Lenders have commonly provided over two-thirds of the operating funds for any given project, with the remainder contributed by the Ethiopian Government. In most instances projects have been intended to improve economic linkages between highland and lowland systems. Historical background on poorly documented projects was solicited from Solomon Desta (TLDP (Third Livestock Development Project) economist, personal communication).

The First Livestock Development Project (1958-63) was narrowly focused and created the Dairy Development Agency (DDA) in the highlands. The Second Livestock Development Project (SLDP) was initiated by the Livestock and Meat Board (LMB) and budgeted at 14.7 million Ethiopian Birr (EB). The SLDP ran from 1973-81. The SLDP was only loosely affiliated with the Ministry of Agriculture (MoA). It was directed by the LMB because the project emphasised development of a marketing and infrastructure network to promote sales and processing of livestock. This was supposed to initiate commercial links between the lowlands and highlands. Only half of the original budget was eventually used because of administrative problems and Ethiopia's conflict with Somalia, which interrupted projects. The SLDP did succeed, however, in building a number of primary and terminal markets and slaughterhouses and 600 km of roads.

After the SLDP was initiated the LMB funded studies of several pastoral areas that were thought to offer potential for supplying animals for the newly created infrastructure. The consultancy firm AGROTEC/CRG/SEDES Associates (see AGROTEC/CRG/SEDES Associates, 1974a-1) was chosen to study the southern Borana rangelands because this was considered the most important region. Other consulting firms and experts supplied by the United States Agency for International Development (USAID) conducted surveys in two other rangelands to the east (LMB, 1974a) and north-east (LMB, 1974b). These studies included surveys of population demography, vegetation, water resources, pastoral socio-economics and animal husbandry. The final reports were used to generate proposals to finance a range project called the Third Livestock Development Project (TLDP), headquartered in Addis Ababa. Budgeted at EB 88 million, the TLDP was initiated in 1975 with the primary objective of developing infrastructure and natural resources to support livestock production and marketing. The three target regions totalled 203000 km2. Details of the TLDP are reported in Section 1.4.2: History of lowlands development and the TLDP.

The TLDP has traditionally operated as a semi-autonomous entity outside of the MoA. The general manager of TLDP has reported directly to the Vice Minister for Animal and Fisheries Resources Development Main Department (AFRDMD), who in turn has been charged with overseeing all aspects of livestock development as one of four vice ministers in the MoA. The TLDP received a couple of extensions to enable full use of the original funds, allowing it to operate through 1987. The TLDP continues to function at the time of writing this, however, with the Ethiopian Government funding much of the core administrative activity. Additional funds have also come into TLDP from the Fourth Livestock Development Project (FLDP), operational since 1988. The FLDP is very diverse and has focused on forage development, livestock epidemiology and livestock marketing in mixed farming systems of the highlands (FLDP, 1987). A small portion of FLDP funds, however, were allocated to the Pilot Project, which operates with TLDP staff. The Pilot Project has been based in the southern rangelands since 1988 and has focused on institution building and development of extension and monitoring capabilities for better outreach to the Borana pastoral community (Hogg, 1990a).

It was originally intended that the TLDP would gradually be phased out by the mid-1980s, but as of 1992 the TLDP remains as the only corps of national range professionals in Ethiopia. It has subsequently become the management entity for the South-east Rangelands Project (SERP), initiated in fiscal 1990-91 with funding from ADF. SERP will operate in what have been the Eastern Hararghe Administrative Region and Ogaden Autonomous Region. It is intended to be a hybrid of previous range development projects, combining the infrastructural development emphasis of TLDP with the outreach approaches of the Pilot Project (ADF, 1989).

When the TLDP is phased out there will be no permanent organisation to represent rangeland interests within the MoA. It is possible that either a new range department would be created within the MoA, or that range development would fall under another semi-autonomous authority (Solomon Desta, TLDP economist, personal communication).

The problems of merging rangeland development interests within the farming-oriented MoA lies in important distinctions between lowland and highland projects in terms of staff skills, staff management and implementation of development activities (Tafesse Mesfin, TLDP General Manager, personal communication).

A number of other rural development projects are currently operating in Ethiopia. These include smallholder dairying in the highlands and highland reclamation. A concise review of these and other projects is provided in FLDP (nd: pp 20-21).

1.4.2 History of lowlands development and the TLDP JIRDU NERDU SORDU Infrastructural improvements

Interactions among highlanders and lowlanders in Ethiopia historically have been characterised by a mix of trade and warfare (Luther, 1961; Kaplan et al, 1971; Wilding, 1985a). The establishment of contemporary trade routes between the highlands and lowlands is commonly attributed to Emperor Menelik. Following his victory over Italian forces at Adowa in 1896, he sent his armies to consolidate a grip over the lowlands by 1908. Modern roads followed such military routes in many cases (Ethiopian Road Transport Authority, unpublished data). Gravel roads were constructed by Italian companies during 1943-53 for five arteries from Addis Ababa to the lowlands. During 1960-70 some of these roads were rehabilitated and asphalted by the Ethiopian Transport Construction Authority. These included roads from Addis Ababa to Negele, Moyale, Jijiga and Assab.

One of the first attempts at infrastructural development for livestock production in the lowlands was initiated in 1965 by the Ethiopian Government and USAID. Tilaye Bekele (1987: p 16) mentions, however, that some stock ponds were built in the southern rangelands by the Ethiopian Government in the 1950s. The joint Ethiopian-USAID project was referred to as the Pilot Rangeland Development Project (PRDP) and the Ethiopian side of the project was conducted through what was then the Range Development Unit in the Livestock Department of the MoA. The intervention concept focused on development of large ponds to improve access of livestock to some 1600 km2 of Themeda and Acacia spp savannah within 50 km of the town of Yabelo on the Borana Plateau, about 570 km south of Addis Ababa. Traditionally Borana pastoralists and their cattle had relied on ephemeral, rain-fed ponds in wet seasons and deep wells in dry seasons (see Section Water resources). Pond development in the PRDP was intended to relieve pressure on wet-season grazing and improve efficiency of range use overall. About 20 large ponds were constructed using heavy machinery that removed some 200000 m3 of soil. Some of these ponds became perennial rather than ephemeral, however, and resulted in a large exodus of people and stock from the central Borana Plateau that had become degraded over several hundred years of use (Billé and Assefa Eshete, 1983b). Over the next 25 years, pastoralists settled and became permanent residents in several areas that had been opened up. The implications of this for the local ecology and pastoral development are reviewed, respectively, in Section 3.3.2: Long-term vegetation change end Section 7.1.3: Review of dynamics and past interventions.

The preliminary results of the PRDP were considered encouraging and led the MoA to formulate a more comprehensive strategy on pastoral development. This, in conjunction with activities of the LMB reported in Section 1.4.1 (Overview of livestock development projects), led to the selection of the Southern Rangelands Development Unit (SORDU), North-east Rangeland Development Unit (NERDU) and the Jijiga Rangeland Development Unit (JIRDU) as the basis of the proposal for the TLDP in 1974. These target areas were considered superior because of their proximity to highland markets, their generally higher stocking potential and because they possessed the highest quality animal breeds in the largest numbers. They also offered good proximity to export markets and meat packing plants. The NERDU area was close to the port of Assab; the JIRDU area had rail access to Djibouti and the SORDU area was bisected by a tarmac road conceived as part of a transcontinental highway system. NERDU was close to the Kombolcha meat packing plant near Dessie; JIRDU was near a plant in Dire Dawa and SORDU was about 200 km south of the Melge-Wondo plant near Shashamene.

Despite the excellent grazing potential of the lowlands to the west and south-west, these could not be considered for the TLDP because of remoteness and prevalence of trypanosomiasis (UNDP/RRC, 1984). The three TLDP sub-projects thus incorporated 27% of the lowlands in total, home to nearly one million pastoralists herding some three million TLUs in 1974. The overall purpose of each sub-project was to develop infrastructure (roads, market facilities, veterinary clinics) and natural resources (water and forage) to stimulate animal production and offtake and to increase incomes and welfare of pastoral producers (UNDP/RRC, 1984). The sub-projects are described below. JIRDU

Headquartered in Jijiga, this sub-project has been responsible for about 33000 km2 of semi-arid (60%) and arid (40%) land in the eastern half of Ethiopia (Figure 1.2). In 1974 the human population was estimated at about 500000, with the majority being semi-nomadic Somali-speaking pastoralists. The livestock population was estimated at 600000 cattle (57% of TLU), 1.3 million small ruminants (12%) and 200000 camels (31 %) for a total of over one million TLU (LMB, 1974a). This represented an average of 32 TLU/km2 in wet seasons and a ratio of TLU to humans of 2.1:1. Livestock numbers change dramatically depending on season, however. During the rainy season the population may be almost twice that in the dry season. Rainfall and forage production tend to decrease to the south and south-east but local forage conditions are greatly influenced by landscape. Of particular importance are the large valleys that extend west into the highlands near Harar. These collect soil moisture and offer higher forage production than the rest of the JIRDU area. These valleys have been traditionally used as dry-season grazing reserves for livestock which spend the rest of the year on the dry tablelands. The cattle population is dominated by a short-horned Bos indicus breed regarded as a good dual-purpose animal well adapted to difficult conditions. It also has a commendable export value to the Middle East (Girma Bisrat, PADEP Coordinator, personal communication). The cattle are concentrated more to the north in the large valleys, while the smallstock and camels are more abundant to the south and south-east. Except for areas traditionally prioritised for cattle, access to sub-surface water using traditional means is very difficult. Market access to Jijiga and Harar is fair, but it is thought that the vast majority of animal offtake is illegally sold to Somalia (Girma Bisrat, PADEP Coordinator, personal communication). NERDU

Headquartered in Weldia, this sub-project has been responsible for about 75000 km2 of arid (85%) and semi-arid (15%) land in north-central Ethiopia (Figure 1.2). In 1974 the human population was estimated at 225000, the majority of whom were nomadic Afar pastoralists. The livestock population was estimated at 734000 cattle (62% of TLU), 1.2 million small ruminants (10%) and 206000 camels (28%) for a total of over 1.18 million TLU (LMB, 1974b). This was equivalent to 16 TLU/km2 and a ratio of TLU to humans of 5.3:1. Severe drought in 1973-74 probably had reduced livestock numbers substantially compared to previous years (LMB, 1974b). The less-predictable nature of rainfall and forage production mitigate against reliable animal production and offtake in NERDU, despite good access to large markets in the region (UNDP/RRC, 1984). Herbaceous forage production and dominance of cattle typically increase with greater proximity to the highland escarpment. Sites in the Teru Depression and basins of the Awash and Mille rivers have traditionally been dry-season retreats for livestock. The main development objectives for NERDU were similar to those for the other sub-projects except for a great emphasis on rehabilitation of drought-stricken pastoralists. This rehabilitation was intended to include irrigation schemes as an alternative life-style for those who had lost access to dry-season grazing because of irrigated cultivation of cash crops along the Awash river (LMB, 1974b).

Figure 1.2. Lowland typology development regions in Ethiopia. - Source: TLDP (unpublished data) and EMA (1988). SORDU

Headquartered in Yabelo, this sub-project has been responsible for about 95000 km2 of semi-arid (70%) and arid (30%) land in southern Ethiopia (Figure 1.2). In 1974 the human population was estimated at 500000, dominated by the Boran (to the west) and Somali (to the east) whose life-styles vary from semi-nomadic to semi-settled. The livestock population was estimated at 1.3 million cattle (74% of TLUs), three million small ruminants (17%) and 94000 camels (9%) for a total of over 1.75 million TLU (AGROTEC/CRG/SEDES Associates, 1974f, 9). This equated to 11 TLU/km2 and a ratio of TLUs to humans of 3.5:1. SORDU was considered to have the highest ecological potential for livestock production of the three sub-project areas because of higher rainfall and lower temperatures (Billé 1983). The more productive environment and reliance on wells for dry-season water also influenced the Borana people to be more sedentary and socially organised, which was expected to improve prospects for animal offtake. In addition, the Boran breed of cattle was considered of high value for domestic use and export (Alberro, 1986; Girma Bisrat, PADEP Coordinator, personal communication).

At its height in the early 1980s, the TLDP supported a permanent staff of over 1000 and a temporary staff of about 4000 (Girma Bisrat, PADEP Coordinator, personal communication). SORDU had the largest staff due to concentration of activities in the south and the absence of civil unrest there. Thus, SORDU used 44% of the TLDP budget (Girma Bisrat, PADEP Coordinator, personal communication). Until the change of government in June 1991, the region around NERDU had been a focal point of armed conflict. Administrative and natural resources at JIRDU have been strained in the last few years because of 250000 refugees who have fled Somalia (A. Moussa, UNHCR Senior Programme Officer, personal communication). Infrastructural improvements

Although JIRDU, NERDU and SORDU shared common development goals, local variation in resource constraints required that different strategies be emphasised. Table 1.1 illustrates some of the achievements of the three sub-projects from 1976 to 1986 and a few details are presented below.


The SORDU area in particular was assessed to have problems regarding access to surface water by livestock. Although the NERDU area is more arid, animals there can drink from perennial rivers originating in the highlands. This helps explain, for example, why 78% of the 122 ponds constructed were in the SORDU area, followed by JIRDU (15%) and NERDU (7%). In the JIRDU area the presence of accessible sub-surface water in the foothills near Harar led to promotion of shallow wells in addition to ponds. The pond programme in JIRDU was constrained by soils that had high rates of seepage (Menwyelet Atsedu, Colorado State University, personal communication). Until 1986, additional efforts to develop water in the JIRDU area focused on promotion of cisterns (birka) to collect run-off water. Other water development in the SORDU area has involved maintenance and re-excavation of the traditional deep wells on the western half of the Borana Plateau (see Section Water resources and Section Water development activities).


The NERDU area had the greatest constraints in forage supply, and it was intended to take advantage of the river system and landscape to promote water spreading (irrigation) to increase forage production (Girma Bisrat, PADEP Coordinator, personal communication). JIRDU also faced significant forage constraints. Most of the key forage supplies occurred in the large valleys, but this was under threat from encroachment by farmers, especially in the Fafen Valley north of Jijiga. In contrast, SORDU was not perceived to have a major problem with forage supply.

Livestock disease control

Out of 23 million vaccinations over 10 years, only 8% were given by NERDU (Table 1.1), primarily because NERDU had fewer problems with livestock diseases than the other areas; and the lowest density of easily accessible cattle (Girma Bisrat, PADEP Coordinator, personal communication). Cattle was the species targeted most often in vaccination campaigns.


More attention was paid to road construction in SORDU than in the other areas because it had fewer roads. Of 3952 km constructed, 75% were in SORDU area followed by JIRDU (15%) and NERDU (10%).

Cattle marketing networks

In recognition of the more productive forage base in the southern rangelands, three ranches were established under the administration of SORDU. It was intended that cattle would be purchased in a lean condition from the pastoralists in the warm dry season (December through March), fattened during the ensuing long rains (April to May) and sold at a profit in June or July to highlands organisations. Part of the profit was to be shared with the Pastoralists in a cooperative venture (GRM, nd). In addition, a smallholder fattening programme was intended to supply animals directly to highland cooperatives for finishing in a stratified marketing structure. The JIRDU area has been the largest supplier of cattle (57%) for the smallholder fattening programme, with SORDU supplying the remainder of the 8662 head distributed (Table 1.1).

Table 1.1. Infrastructure development and service statistics for the JIRDU, NERDU and SORDU sub-project sites of the Third Livestock Development Project from 1976-1986.1

Development component





Water development

- ponds (number)




- cisterns (number)




- shallow wells (number)




- deep wells (number)




Forage development

- water spreading (ha)




- drought fodder reserves2 (ha)




Veterinary service3

- vaccination total (millions)




Road development4

- trade roads (km)




-access track (km)




Holding ranches

- established (number)




- marketed cattle (number)




Smallholder fattening programme

- purchased cattle (number)




- distributed cattle (number)





- veterinary scouts (number)




- range wardens (number)




- dip/crush attendants (number)




Trials and studies

Meteorology stations (number)




1 Where JIRDU = Jijiga Rangelands Development Unit; NERDU = North-east Rangelands Development Unit; and SORDU = Southern Rangelands Development Unit.

2 Drought-resistant genera such as Atriplex spineless Opuntia and various Acacias have been examined in joint trials with FAO and the Ethiopian Ministry of Agriculture (MoA). These have not been implemented through extension, however. Other forage screening trials involving Leuceana spp have been conducted in joint trials with the MoA at SORDU and JIRDU. Leuceana is envisioned as an intervention for years of average rainfall on water-collecting landscapes. See Section Forage improvements.

3 Vaccination totals are largely dominated by cattle. See text.

4 Where trade roads are 6 m wide and covered with transported materials when local materials were unsuitable and includes drainage structures. Access track is 4 m wide with passing bays every 300 m; these have no drainage structures and only local surfacing materials were employed.

5 Veterinary scouts are males selected from the pastoral community and trained up to 45 days in topics such as identification of major disease symptoms. They are also responsible for organizing the local people to participate in vaccination campaigns. Range wardens were men selected from the community and trained up to 21 days, mainly to guard newly constructed ponds.

Source: Girma Bisrat et al (TLDP, unpublished data).

Human resources

A total of 484 individuals were trained to provide field support for veterinary service (66%) and range management (30%). SORDU has led these efforts with 49% of trainees, followed by JIRDU (39%) and NERDU (12%). Staff established an average of nine weather stations in each region as part of the national meteorological network with assistance from ILCA (Table 1.1). The National Metereological Agency trained field staff to collect weather data.

Only SORDU has achieved or exceeded targets for most aspects of its development programme. Civil disturbances in the JIRDU and NERDU areas have undermined development efforts. Some of the project components listed in Table 1.1 have been sustainable but others have not. This is discussed further in Section 1.4.5: Development of the sot/them rangelands as coordinated by SORDU and Chapter 8: Synthesis and conclusions.

1.4.3 The SERP and the Pilot Project

The SERP (introduced in Section 1.4.1: Overview of livestock development projects) will almost double the lowlands area under development in Ethiopia (Figure 1.2). In the new organizational format JIRDU was brought under the auspices of the SERP in 1991. SORDU and NERDU will continue much as before, except that SORDU should include the Pilot Project at least through 1992 (Solomon Desta, TLDP economist, personal communication). The TLDP staff may administer all of these components until 1996-97 when the first phase of the SERP expires.

Although a small component of FLDP in terms of funding, the Pilot Project at SORDU is an important addition in terms of philosophy and strategy of pastoral development. Material reviewed here is only a brief synopsis, largely drawn from Hogg (1990a; 1990b; 1990c).

The Pilot Project was initiated in 1988 to establish viable and sustainable service cooperatives (SCs) among Borana pastoralists. It was intended to continue until 1992 but extension to 1995 may be needed to give enough time for adequate impact and evaluation (Hogg, 1990c). The Pilot Project is attempting to work in about one-third of the SORDU area (or 34000 km2), with a human population of 150000. The target area is largely around the town of Yabelo and south to the village of Dubluk, with a large portion to the far west near the Kenya border (see Figure 2.10). The Pilot Project involves a fundamental change for SORDU. The administration at SORDU used to plan and conduct infrastructural improvements and veterinary campaigns largely in isolation from the pastoralists. In the Pilot Project, however, SORDU is to become more of an enabling institution that facilitates implementation of community projects that the pastoralists have prioritised for themselves. The Boran are also expected to pay for much of the cost of the new activities on a contractual basis. This is intended to make the interaction more sustainable by reducing dependency on external funding and should provide better indicators of whether projects are really desired by the community (Hogg, 1990c). It is important to note that formation of SCs in the lowlands has traditionally been the domain of the MoA, while SORDU was in charge of the development of pastoral and range resources (see below). Defining precisely the duties of these agencies has emerged as a more critical problem as development issues become more complex and interconnected (see Chapter 8: Synthesis and conclusions).

The SC concept that has evolved in the Pilot Project is a modified version of SCs that were implemented throughout rural Ethiopia during 1975-90 (Hogg, 1989; Hogg, 1990c). The original SCs were based on socialist agrarian strategies. The rural organizational structure during 1975-90 consisted of a system of Peasant/Pastoral Associations (PAs) which served administrative functions. Some of these functions were unpopular and included tax collection, filling government quotas for crops and livestock at below market prices, raising funds for building schools or clinics and recruiting young men for the army. Service cooperatives also served to inculcate political or administrative indoctrination. Membership in PAs was mandatory. Despite the pervasive influence of PAs in daily life, they did not exert much influence over traditional legal and social mechanisms for problem solving among the Boran (Hogg, 1990c; Takele Tilahun, TLDP/ILCA post-graduate researcher, unpublished data). However, the PAs may have had some positive aspects with regard to augmenting traditional influence over resource allocation in the face of rising population pressure (see Section: Site reclamation).

The PAs formed the foundation upon which cooperatives were created. Unlike PAs, cooperatives were intended to provide economic services and membership was more optional. Cooperatives were to be formed from one or more PAs, depending on population density and resources, and were to use contributed labour and capital to create regional and national networks for the production and distribution of basic commodities. There were two main kinds of cooperatives: (1) SCs that focused more on the organization needed to procure basic goods for the inhabitants of often remote locations; and (2) producer cooperatives (PCs) that focused more on organising the production of a single commodity or a group of related commodities.

The cooperatives were designed and implemented using a top-down approach. The core activity of the SCs was to obtain basic goods from government warehouse networks at subsidised prices. Theoretically, only cooperatives could gain access to these subsidised goods. The members of the SC would also constitute a legally recognised entity that could apply to the national banking system for low-interest group loans for local development projects. One problem in the lowlands was that pastoralists were not considered able to form the permanent residential groups upon which SCs were to be based. Thus pastoral SCs would not be legally recognised and could not apply for loans. In sum, SCs and PCs were considered to be somewhat complementary and were supposed to be important conduits that avoided problems of corrupt middlemen and deficiencies in free markets. Such deficiencies theoretically included inability to efficiently network producers and consumers in remote areas.

The PAs, PCs and SCs throughout the country largely began to collapse with the demise of central authority in 1990. Government bureaucracy, poor local management and shortages of desired goods effectively suffocated cooperatives in the highlands and it has been asserted that cooperatives never really emerged to a significant degree in the lowlands (Hogg, 1990c). Good examples of how unresponsive the cooperative system really was in meeting modest demands for grain and hand tools in the southern rangelands are reported in Hodgson (1990: pp 83-117).

The Boran were organised into PAs in the 1975-76 Zemecha or Students' Campaign. Membership in a PA was based on residence in a particular madda, which is a watering and grazing entity in the traditional Borana system (see Section Water resources). One problem that emerged was that herdowners were expected to pay taxes to the original PA in which they were registered even though they may have moved elsewhere in their search for better grazing or water (Hogg,1990c). The first SC in the southern lowlands was formed from several PAs and was established by the local branch of the MoA near Teltele in 1977-78. By 1987 there were 112 PAs and 17 SCs in the southern rangelands, with average memberships of 336 and 534 household heads, respectively (Hogg,1990c). The average number of PAs per SC was 3.6. While PAs provided blanket regional coverage, the few SCs were concentrated near urban areas in wetter upper semi-arid and subhumid regions, where membership was skewed towards farmers and agropastoralists. This was largely because the local MoA had insufficient resources to extend very far from their branch offices and because of the natural bias of the MoA to work in farming systems (Hogg,1990c).

One objective of the Pilot Project is to retain and modify the SC concept as a development tool for the southern rangelands. Despite problems in the past, the SC is still seen as a viable marketing aid in remote areas where commercial traders have not established. Fortunately, previous experience with cooperatives in the lowlands has been minimal but still some of the negative experiences with the PAs will probably be a constraint in getting the people to overcome their suspicions of new development organizations (Hodgson, 1990; Hogg, 1990c). The reason for continuing to use SCs for pastoral development is based on the proposition that improved market access is essential in stimulating monetization and more commercial use of livestock, which should facilitate economic diversification and relieve some pressure on the grazing environment (see Section 3.4.2: Environmental change).

The proposed SCs are quite unlike the old SCs. In an effort to make SCs more meaningful to the people, the Pilot Project has focused on implementing them with attention to the following details (Hogg,1990c): (1) they are being structured to reflect traditional organization and leadership; (2) they are being organised to help provide members with goods they really want and not those that used to be forced upon them; and (3) they are being designed to operate in a free-market setting, planned and implemented from the bottom up and are thus less encumbered by bureaucracy. The management and placement of SCs is being carefully rationalised in a pastoral (not farming) context, including a review of those that have existed for some time. Hogg (1990c) emphasised that there is no standard menu for success, as each SC must respond to the local situation in terms of management constraints, the demand for commodities and the costs of meeting demand. This process of institution building is arduous and will take many years. Constraints of an uneducated and skeptical pool of pastoralists are dominant. This strategy should also not be viewed as a "quick fix"; only four new SCs were established in two years (Hogg, 1990c).

For more details of the Pilot Project the reader is referred to Hogg (1990a; 1990c) and FLDP (1987). Various sections in Chapter 7 (Development intervention concepts) review examples of pastoral organization as they pertain to specific development concepts in animal marketing, grain storage and water and land management.

1.4.4 Has national range development been successful?

Considering the atmosphere of regional insecurity, political change and administrative problems that has prevailed for the TLDP during the last 18 years, it is remarkable that TLDP has had as much success as it has. It is important to recognise, however, that the enumeration of total vaccinations or lengths of roads put in does not reveal whether the overall strategy has been effective, i.e. whether animal offtake has increased or the welfare of pastoralists has improved. Unfortunately, these questions cannot be rigorously addressed because hard data are lacking and many factors beyond TLDP performance influence outcomes. Monitoring of project impacts should improve evaluation but this has only recently been added to the TLDP under the Pilot Project (FLDP, 1987).

The impact of TLDP interventions in the SORDU area was assessed by ILCA using detailed interviews of pastoral leaders in 1990. Although this information is based on subjective perceptions, it is argued (see Section 7.1.3: Review of dynamics and past interventions) that improvements of infrastructure and veterinary service in the southern rangelands since the 1960s have had fundamental effects on Borana society. These effects are important and widespread, but also often subtle and hard to detect in a superficial manner. Many of the effects may be positive, but more in the sense of providing a cushion for population growth, delaying the onset of poverty for a portion of a rapidly growing population and acting as a catalyst for future social and economic change.

In addition, whether project objectives are viewed as having been met is complicated by assumptions used in the project preparation stage. For example, suppose that animal offtake is increasing (see Section 7.2: A theory of local system dynamics). This may indeed raise cash incomes, but it does not mean that the people are better off if essential goods to be purchased are too expensive or in short supply. A similar paradox arises in connection with the types of animals sold. Higher offtake rates of cattle less than two years old may be interpreted to mean that the system is becoming more "modernised" in terms of incorporating Western production ideals. Alternatively, it can suggest that older male cattle are in shorter supply and the human population is becoming poorer and more vulnerable to droughts. The Boran traditionally preferred to sell older and larger male animals because the producer can buy goods they want plus a young replacement animal (Coppock, 1992a; see Section Marketing attitudes). Older males have the lowest risk of death during drought of any class of cattle, which makes them an important reserve of wealth during times of stress (see Section 6.3.11: Livestock dispersal and herd composition). Such inferences seem radical compared with developed production systems, but this reflects fundamental differences between developed versus traditional African livestock operations (Behnke, 1984; Coppock et al, 1985; Coppock, 1992a). Both are logical when viewed within their own economic context.

A common perception among TLDP senior staff has been that the roads, markets and veterinary campaigns have not stimulated animal offtake per se and that veterinary campaigns in particular have contributed to a greater inventory of animals now in a better position to over-utilise the land (Sileshi Zewdie, SORDU veterinarian, and Solomon Dessalegn, former NERDU Manager, personal communication). These opinions are in agreement with mainstream views generated from rangeland development projects elsewhere in sub-Saharan Africa (see Introduction of Ellis and Swift, 1988; de Haan, 1990).

It is concluded in Chapter 8: Synthesis and conclusions that while the TLDP has made some important contributions in SORDU that should catalyse future opportunities for pastoral development, this development is highly dependent on other social and economic processes that operate at national scales beyond the domain of the TLDP. Economic development and improved human welfare for the Boran are defined in Section 7.1.2: Development philosophy for the Boran and can be condensed as trends toward: (1) a sustainable level of per capita milk production and per capita asset accumulation largely in the form of livestock; (2) improved food security during times of environmental perturbation; and (3) fewer risks of producers being squeezed out of the pastoral system. All of these trends should occur within a framework that also conserves valued aspects of the traditional culture. Given this scenario, it is hypothesised based on several indicators, that economic development is not occurring in the southern rangelands (see Section 7.2: A theory of local system dynamics). The Boran appear to be gradually altering some of their attitudes towards cultural and economic changes, but this is not being effectively tapped in a comprehensive development process. An apparently high rate of human population growth, in conjunction with environmental limits on cattle production, suggest that food security, per capita production and other aspects of human welfare will decline. The system is thus overpopulated and education and urban job opportunities are needed to help release pressure and allow remaining residents to have a chance of improving their living standards. Also needed are other investment opportunities that give returns comparable with those from livestock production. Policies are needed to help open regional markets in order to promote efficient interregional trade and reduce the likelihood of local food shortages and famine.

In sum, this is not to say that some technologies or management strategies within the domain of the TLDP are not important but that they make up only a very small part of a multifaceted, long-term development strategy. While all of the lowland areas have unique constraints to economic development and are probably at various points along a Continuum of induced change (Kidane Wolde Yohannes, TLDP range scientist, personal communication) it is likely that the development problems observed in SORDU are relevant to the Ethiopian lowlands as a whole.

1.4.5 Development of the southern rangelands as coordinated by SORDU Range management Water development strategy Livestock health Roads Ranch development Smaliholder fattening programme Training Trials and studies

The focal point of this report is SORDU and the Borana pastoral community that it serves (Plate 1.2 a, b). Besides the sub-project headquarters in Yabelo, SORDU also has sub-offices at Negele to the east in the Borana Administrative Region and Awassa to the north in the Sidama Administrative Region. The base in Awassa helps coordinate linkages in animal marketing between the highlands and lowlands. This section briefly outlines SORDU activities in relation to eight original project components as outlined by Girma Bisrat (1988). Material here also includes comments on objectives and observations on recent achievements. Range management

The objective of this component was to develop a management plan that would promote a sustainable level of resource use. It was hoped to describe various range areas in terms of vegetation composition, net primary production and stocking rates for livestock that would promote an optimal level of productivity in relation to maintenance of a "desirable" vegetation cover. It was also intended to prohibit "undesirable" land-use practices such as indiscriminate burning of range vegetation, wood cutting and cereal cultivation.

There have been two major constraints to implementing of this activity, namely lack of land-use plans for the lowlands and shortages of trained manpower. Land-use plans are primarily needed to identify sustainable farming areas near and within the rangelands. This would permit evolution of pockets of agropastoralism within the Borana system and provide a legal basis for rationalising user conflicts between farmers and pastoralists in contested areas on the periphery of the southern rangelands (see Chapter 8: Synthesis and conclusions). SORDU has not been able to deploy enough trained manpower in range management over the past 18 years. Hacker (1988a; 1988b) devised monitoring methods for range trends. There is concern, however, that even if levels of optimal use and cattle density could be identified, there is insufficient means to enforce land-use regulations, especially under increasing population pressure and the short-term need of the people for maximum cattle production (see Chapter 8: Synthesis and conclusions). Whether the local administration or SORDU can control even the recent spread of maize cultivation is debatable (Kidane Wolde Yohannes, TLDP range scientist, personal communication). Water development strategy

The traditional pastoral system was based on wet-season grazing with cattle watering at ephemeral ponds and dry-season grazing close to deep wells (see Section Water resources). The longer animals could stay on wet-season range the better, because this would help conserve dry-season forage and delay use of the wells. This in turn would conserve ground water and postpone a high commitment of labour (see Section 4.3.2: The encampment and the role of cooperative labour).

Studies showed that some vegetation far from wells was not evenly utilised and that new ephemeral ponds could improve access to this forage in wet seasons. Thus, during most of the first decade of SORDU, great emphasis was put on building ephemeral ponds that could theoretically double the length of grazing in "wet-season areas" from two to five months in average rainfall years (Girma Bisrat, PADEP Coordinator, personal communication).

About 95 ponds were constructed in under-utilised sites throughout the region using heavy machinery, paid for in full by SORDU. The ponds ranged in size from 10000 to 60000 m3. While some ponds functioned as intended and led to a more balanced and diverse pattern of resource use (Hodgson, 1990: p 52; see Section Grazing management), others suffered from high infiltration rates or high rates of siltation (Tilaye Bekele, 1987). Consultants prescribed a variety of improved management and siting methods but many of these were difficult to implement. Despite problems, the water development programme remains as one of the most popular activities with the Boran.

Today SORDU focuses more on trying to help the people maintain existing ponds and wells, since opportunities to expand the area under grazing are now very limited. For several years development agents tried to promote animal-drawn scoops (Abiye Astatke et al, 1986) so the people could desilt ponds themselves and conserve fuel and spare parts for heavy machinery. However, the Boran much prefer to collect money and pay for use of heavy machinery rather than use their draft animals to pull scraps (see Section Water development activities). Selling stock to pay for maintenance of water points is emerging as a major form of monetization and participation of beneficiaries in the development of the southern rangelands.

Plate 1.2 a. Borana men and a Borana woman of southern Ethiopia. - Photograph: JEPSS

Plate 1.2 b. Borana men and a Borana woman of southern Ethiopia. - Photograph: JEPSS Livestock health

The livestock health programme started strongly but has had recent problems sustaining activities. One central and eight satellite clinics were established in towns during the late 1970s to provide health treatments for animals that could be walked in. Cattle were periodically reached in the field through large-scale vaccination campaigns against rinderpest and anthrax carried out up to twice a year. Cattle dips for tick control were established in several locations and field clinics were conducted for small ruminants and camels (Hill, 1982). However, activities have declined markedly since the early 1980s, mostly due to factors beyond the control of TLDP.

By 1990 there were chronic shortages of imported drugs, reportedly because of procurement problems within the MoA. When TLDP was financially independent (until 1987), it handled its own procurement and drugs were imported more reliably. Now with other elements in the MoA controlling procurement for TLDP, it is felt that the animal health service in the lowlands has been compromised. This has been due to bureaucratic problems and attempts to save money by ordering inferior drugs (Sileshi Zewdie, SORDU veterinarian, personal communication). The greatest demand for veterinary drugs in the southern rangelands is reportedly for acaricides and antihelminthics, but these must be imported from Europe (Sileshi Zewdie, SORDU veterinarian, personal communication). Vaccines for rinderpest, blackleg and anthrax are produced at Debre Zeit in the Ethiopian highlands but supplies have been reduced in recent years. Contagious caprine pleuropneumonia (CCP) is a major constraint to goat production in the southern rangelands (Section Sheep and goats) and it is planned to produce this vaccine in Debre Zeit in the near future (Sileshi Zewdie, SORDU veterinarian, personal communication). Locally made antibiotics have also been available for human ailments through the Ministry of Health but it has been reported that Borana pastoralists obtain these from clinics and use them for livestock (Coppock, ILCA, personal observation). Acaricides and vaccines have also been reportedly smuggled from Kenya.

Today, there are severe shortages of vaccines, manpower and transport. SORDU staff are unable to engage in preventative measures, responding only to disease outbreaks and this is when the shortages are felt most acutely (Sileshi Zewdie, SORDU veterinarian, personal communication). It is now envisioned to have the Boran pay for all health services, including vaccination campaigns. This would reduce the frequency of false alarms of disease outbreaks and improve the financial basis of the service (Sileshi Zewdie, SORDU veterinarian, personal communication). Roads

By 1987, nearly 2800 km of trade and access roads had been constructed throughout the SORDU area, with about 30% built by hand labour and the remainder using heavy machinery. Ongoing activities focus on maintenance. This has probably had major impacts on growth of small towns in the region, provided jobs and facilitated grain distribution during drought. This remains as an outstanding achievement that will help catalyse future change (see Section 7.1.3: Review of dynamics and past interventions). Ranch development

Early on in the SORDU sub-project, land was set aside in three regions for use as holding and fattening areas for cattle to be sold to highland operations. The ranches were named Sarite (17000 ha), Dembel Wachu (12000 ha) and Wollenso (25000 ha). Animals were bought from local producers and typically sold to the Ministry of State Farms which used them for domestic purposes or export. One justification for SORDU assuming control over these particular ranch sites was that they had inadequate water supplies in the dry season. It was anticipated that, using SORDU's resources, water supplies could be developed on the ranches. However, difficulties were experienced at Dembel Wachu (Billé and Assefa Eshete, 1983a) and Wollenso (Girma Bisrat, 1988) in particular. The five ponds established at Dembel Wachu had seepage problems, while attempts to drill boreholes at Wollenso were constrained by a very deep water table.

The intent was to manage the ranches jointly through a collaboration of local PA members and SORDU staff and introduce concepts of ranch and range management to the Boran. Through a profit-sharing scheme, the project was supposed to provide money from livestock sales to the local community for development projects. It was also intended to eventually turn over the ranches to the local people after a demonstration period of several years.

SORDU directed ranch management, with Sarite having its first intake of cattle in 1979-80, followed by Dembel Wachu in 1980-81 and Wollenso in 1985 (GRM, nd). It was anticipated that the ranches would handle a total throughput of 36000 head between 1981 and 1987 (6000 head/annum), but the actual number was only about 25% of this (Girma Bisrat, 1988). Low throughput and associated management and marketing problems led to a net loss of over EB 300000 by June 1988 (GRM, nd). This lack of profitability, as well as a low popularity with the Boran (GRM, nd), led to the ranch concept being abandoned. It was then decided to return the land to pastoral management. Hogg (1988) proposed that Sarite be turned over to the new local SC developed in the Pilot Project (see Section 1.4.3: The SERP and the Pilot Project). He suggested that the ranch size should be halved and an SC committee would decide whether the ranch could be used in a different way to produce animals in more of a free-market setting.

In sum, the outcome of the ranch experiment at SORDU has been the same as elsewhere in pastoral Africa, i.e. the Western ranching concept has failed to transform traditional pastoralism (de Haan, 1990). It was also pointed out by GRM (nd) that the ranch strategy in SORDU ran counter to the survival tactics of the Boran, which included a low priority for selling younger stock (Coppock, 1992a; see Section Marketing attitudes). Smaliholder fattening programme

Links among cooperatives in the highlands and lowlands were expected to form the basis of reciprocal help in the exchange of lowland livestock for highland grain. Young bulls from the rangelands were to be purchased on credit and distributed to highland farmers, who would finish them on grazing and crop residues and sell them for slaughter or use them as draught animals. The primary beneficiaries of this programme were thus highland smallholders (Menwyelet Atsedu, Colorado State University, personal communication). Cash could, however, be used by the Boran to buy grain directly from highland cooperatives, thereby avoiding expensive middlemen and saving the Boran money.

By 1988 the programme involved 21 PAs and 74 cooperatives overall and nearly 5000 cattle were sold to the southern highlands (Girma Bisrat, 1988). Manpower constraints limited the success of the programme, however. More staff were needed to follow up on collection of loan repayments. Lack of sustained effort led to many smallholders defaulting on their loans and many cases ended up in court (Menwyelet Atsedu, Colorado State University, personal communication).

Another major constraint to the programme was shortage of transport. It was intended that cattle trucks from the rangelands would return from the highlands with grain (Girma Bisrat, 1988). The TLDP purchased its own fleet of trucks in the late 1980s. Privatisation of this activity today is hampered by inefficient remnants of government trucking monopolies, which remain a major constraint to interregional trade (Tafesse Mesfin, TLDP General Manager, personal communication). The Pilot Project envisions that trade between the highlands and the lowlands needs to be stimulated by the establishment of large grain stores at SCs on the Borana Plateau (Hogg, 1990c). Training

Project staff have been educated at home and abroad and pastoralists have been trained to provide field support to animal health activities (Table 1.1). One novel approach was to take over 100 Boran on trips to other pastoral and urban areas so they could get a better feel for the diversity of the nation (Girma Bisrat, 1988). Difficulties in training have occurred, however, indicating problems in sustaining field veterinary programmes and the low return rate (about 20% to date) of senior personnel sent overseas for further education since 1987. Those who have returned have also typically become administrators rather than scientists or resource managers (Coppock, ILCA, personal observation). Trials and studies

The programme of trials and studies was, intended to help SORDU collect data to monitor and guide development projects, has always been constrained by a lack of trained manpower. This was one reason why ILCA was brought in to collaborate with SORDU on research in the early 1980s and to help train post-graduates in 1987-90 (see below). It is also why TLDP has spent considerable amounts on consultants. Recent activities of the Trials and Studies Section of SORDU have included collection of weather data, preliminary establishment of a range trend monitoring network, initial trials on bush control with prescribed fire and hand-clearing methods, roadside sowing trials with Stylosanthes spp, observation of a few field plots of fodder trees (i.e. Leuceana leucocephala cv Cunningham and cv Peruvian established at Dembel Wachu ranch in 1982) with the MoA, and establishment of drought-hardy forages (Prosopis, Opuntia and Atriplex spp established at Dembel Wachu and other sites in 1987) in collaboration with FAO. All of this work is constrained by shortages of fuel, vehicles and operating funds (Tamene Yigezu, SORDU Manager, personal communication). Details of forage work at SORDU are reviewed in Section Forage improvements.

In sum, the success achieved by various project components varies. For some it is still too early to judge impact. TLDP management has routinely evaluated Project activities and dropped even high-profile activities (such as the ranches) that have not been successful (Coppock, ILCA, personal observation). Shortages of educated manpower constrain management and research capability and problems with support services and operating funds commonly impede programme implementation.

1.4.6 Collaboration among research and development institutions in the southern rangelands

The TLDP and ILCA have conducted various joint activities since 1976, when an intensive land-use survey in the Jijiga rangelands was started but abruptly curtailed because of the Somali invasion. This was followed by a joint ecological survey of the north-east rangelands a few years later. Under the umbrella of the general memorandum of agreement between the Government of Ethiopia and ILCA that was signed on 15 May 1975, the first formal Memorandum of Understanding (MoU) between TLDP and ILCA covered the Cooperative Monitoring Project for Ethiopia (CMPE), in effect from 1980-82. This agreement was replaced by the MoU for the Joint Ethiopian Pastoral Systems Study (JEPSS) from 1982-85.

The JEPSS was intended to provide a framework for research and debate on lowland development strategies. It was to focus initially on the Afar and Borana pastoral systems. Field research for the JEPSS in NERDU and JIRDU was abandoned in the early 1980s because of security problems. Some of this early work at NERDU is found in ILCA (1980), Donaldson (1982), Cossins (1983a), Billé (1983) and Negussie Tilahun (1983a; 1983b). Reports concerning JlRDU include Cossins et al (1984a-d) and Cossins and Billé (1984). The JEPSS continued to work in SORDU until 1985.

After the JEPSS ended in 1985, CARE-Ethiopia (a non-governmental development and relief organization) became a major partner in the SORDU area through establishment of the Southern Sidamo Rangelands Project (SSRP). This project was based on a MoU among the MoA/TLDP, the Relief and Rehabilitation Commission (RRC) of the Ethiopian Government, CARE-Ethiopia and ILCA from 1985 to 1988. CARE and ILCA combined staff and resources in field work, while the MoA/TLDP and RRC participated in liaison and strategy. The objectives of the SSRP were to provide tests of research hypotheses generated in the JEPSS for production interventions in the Borana system. For those interventions that appeared promising, preliminary modes of extension were also developed. ILCA focused on research while CARE worked more on extension and development. The main activities initially dealt with improved strategies for calf management, including water and forage intervention. This was later broadened to include many more development activities (Hodgson, 1990).

The SSRP ended in late 1988. While close administrative collaboration also ended between CARE-Ethiopia and ILCA at this time, informal ties between research and extension remained strong. CARE-Ethiopia then joined in another MoU with MoA/TLDP and the RRC starting in 1989 and the main development activities were collated under what was termed the Borana Rangelands Project. In this phase CARE's extension work still had a strong influence on helping shape ILCA's rangeland research (see below). CARE-Ethiopia also participated in institution building with SORDU to help establish a new SC in Dubluk madda as part of the Pilot Project (see Section 1.4.3: The SERP and the Pilot Project).

After 1988 ILCA staff designed research to complement the grass-roots activities of CARE and helped train postgraduates in a new MoU with FLDP/TLDP and the Institute of Agricultural Research (JAR of the Ethiopian Government) from 1987-90. This MoU formed the basis for the Cooperative Rangelands Pilot Research Training Programme (CRPRTP), intended to address problems of TLDP in training staff capable of conducting research. Master's-level candidates from TLDP were integrated into the field research, with course work performed at universities abroad. This project was funded by FLDP and ended in 1991 with six students completing degrees. Much of this research is documented in this volume.

1.4.7 Interaction between research and development and project impact

In theory, the SSRP was to follow a standard formula: (1) research providing the understanding of the pastoral community using a farming systems research (FSR) approach and (2) developers following with extension (see Harwood, 1979). Research was supposed to have defined interventions for the production system and their entry points.

By 1987, however, the nature of the SSRP began to change as work evolved into a more reciprocal partnership. Extension agents began to have a critical role in reshaping the research agenda. While research had successfully identified some key issues prior to 1985, it turned out that intervention was far more complex than anticipated. This complexity was revealed by the in-depth interaction of extension agents with the community. It is fair to say in retrospect that the researchers never implemented a true FSR approach on their own. The extension agents completed the circle by forming a more effective feedback loop with the Boran. Research was then redesigned on the basis of this improved information (Coppock, 1990a). Extension agents ended up being more important than researchers in implementing the FSR process. Research then followed up by putting some development issues into a more dynamic systems context (see Section 7.2: A theory of local system dynamics). Research also has had a key rode in dissemination of project information.

In sum, the model that evolved from the original SSRP turned out to be rather unlike that envisioned for FSR in Harwood (1979). It became more like the situation in the United States where extension is supposed to be the "eyes and ears" for research by providing timely feedback and new ideas (G.A. Rasmussen, Extension Specialist, Utah State University, personal communication). Our experience documents the crucial importance of clever extension agents as well as researchers who recognise that technical expertise must be shaped by development values if impact is to be rapidly achieved (see Chapter 8: Synthesis and conclusions).

Previous Page Top of Page Next Page