Rwanda - Second Mutara Agricultural and Livestock Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 2221-RW RWANDA MUTARA AGRICULTURAL AND LIVESTOCK DEVELOPMENT PROJECT - PHASE II STAFF APPRAISAL REPORT May 25, 1979 Eastern Africa Region Central Agricultural Division This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without Worid Bank authorization.  CURRENCY EQUIVALENTS Current Unit = Franc Rwandais US$1.00 = RF 91.91 RF 100 = US$1.09 WEIGHTS AND MEASURES Metric System Metric British/US Equivalents l meter (m) - 3.3 feet ilhectare (ha) 2.47 acres 1 are 100 m = 0.02 acres 1 kilometer (km) 2 0.62 miles 1 square kilometer (km ) 0.39 square mile (sq. mi.) 1 kilogram (kg) 2.2 pounds (lb) 1 liter (1) 0.26 US gallon (gal) 0.22 British gallon (lmp gal) 1 metrie ton (m ton) 2,204 pounds (lb) GLOSSARY AND ABBREVIATIONS AIDR Association Internationale de Developpement Rural (International Association for Rural Development) BGM - Bugesera/Gisaka/Migongo Project ISAR - Institut des Sciences Agronom:iques du Rwanda (Agricultural Research Institute of Rwanda) OBM - Office du Bugesera et Mayaga (Agency for Bugesera and Mayaga) OCIR - Office des Cultures Industrielles du Rwanda (Agency for Industrial Crops of Rwanda) OPROVIA - Office National pour le Developpement et la Commercialisation des Produits Vivriers et des 'Productions Animales (National Agency for Development and Marketing of Foodcrops and Animal Products) OVAPAM - Office pour la Valorisation Pastorale et Agricole du Mutara (Agency for Agricultural and Livestock Development of Mutara) SSS - Service des Semences Selectionnees (Selected Seed Service) UN/HCR - United Nations - High Commission for Refugees CIDA - Canadian International Development Agency  FOR OFFICIAL USE ONLY RWANDA APPRAISAL OF THE MUTARA AGRICULTUFAL AND LIVESTOCK DEVELOPMENT PROJECT PHASE II TABLE OF CONTENTS Page No. I. BACKGROUND ..................................................... 1 A. Project Background . ........ ........................... 1 B. The Agricultural Sector .................................... 2 C. Crop and Livestock Marketing and Processing .. .............. 5 D. Institutions Responsible for Agricultural and Livestock Development ....... ............................ 6 E. Agricultural Credit ... ........... ............. a .. ....... 8 F. Government Strategy and Plans for agricultural and Rural Development ....... ............................ 8 II. THE PROJECT AREA ........................... 9 A. Location and Physical Resources ..... ............. 9 B. Government Services ...... .............. .................... 10 C. Sociological Aspects ...................................... il D. Agricultural Development Project - Phase I (Credit 439-RW) . 13 E. Marshland Development Project- ............... ............... 18 III. THE PROJECT ................................................. 19 A. Project Concept ............................................. 19 B. General Description ......................................... 20 C. Detailed Features ........................................... 21 D. Project Costs ............................................... 27 E. Financing ................................................. 28 F. Procurement ................................................ 29 G. Disbursement ............................................... 30 H. Accounts and Audit ......................................... 31 I. Environmental Impact ....................................... 31 IV. ORGANIZATION AND MANAGEMENT .................................... 32 A. Project Organization and Management ...... .................. 32 B. Project Implementation Schedule ....... ..................... 38 C. Monitoring and Evaluation .......................... ....... 38 This report is based on the findings of an IDA appraisal mission to Rwanda in April-May 1978, composed of D. Lallement, H. Marticou, D. Coyaud (IDA), R. Lemarchand and J. Le Hasif (consultants). H. Sanger (RMEA) and R. Randriamandranto (IDA Washington) also participated in the appraisal. D. Lallement held post-appraisal discussions in Kigali in December, 1978 with participation of C. Blanchi and H. Sanger (RMEA). This documnent has a restricted distribution and rmay be used by recipients only in the performance of their official dulies. Its contents may not otherwise be disclosed without Worid Bank authorization.  TABLE OF CONTENTS (Continued) Page No. V. AGRICULTURAL PRODUCTION ....................................... 39 A. Technical Features ........................................ 39 B. Producer Benefits ......................................... 42 C. Production Estimates ...................................... 43 VI. GOVERNMENT BENEFITS ..................... ...................... 43 VII. MARKETS AND PRICES ...................... ...................... 44 VIII. ECONOMIC BENEFITS AND JUSTIFICATION ........... .. .............. 44 IX. SUMMARY OF AGREEMENTS TO BE REACHED ON CREDIT CONDITIONS ...... 48 Schedule 1 - Proposed Staffing for the Project ........ ........ 52 SUPPORTING TABLES AND CHARTS Table 1 - Total Costs - Phase I Project ........................ 53 Table 2 - Summary of Annual Project Costs - Phase II Project ... 54 Table 3 - Estimated Schedule of Disbursements .... ............. 55 Table 4 - Economic Analysis ................................... 56 Table 5 - Impact on Government Budget ......................... 57 Chart 1 - Project Organizational Structure .................... 58 Chart 2 - Project Implementation Schedule ..................... 59 Material available in Implementation Volume ................... 60 MAPS 1. Rwanda Country Map - Location of Project Area IBRD 13865 2. Project Area - IBRD 13866  RWANDA MUTARA AGRICULTURAL AND LIVESTOCK DEVELOPMENT PROJECT - PHASE II I. BACKGROUND A. Project Background 1.01 IDA has been involved in the Mutara Region since 1974 when the first Mutara Agricultural Development Proje,-t was initiated. Due to several imple- mentation difficulties including cost overruns, IDA agreed with Government to reduce the scope of the project. Following an in-depth Review mission which visited Rwanda in late 1975, it was agreed in principle that the project would in future be considered a Phase I project to be followed by a Phase II project. Government requested that ID)A consider financing for the Phase II Project in March 1978. The Project proposal was prepared for Government by IFAGRARIA, the consulting firm which provided technical assistance for the implementation of the Phase I project, with assistance from RMEA. The pre- paration report was reviewed by a tripartite Government Committee, including representatives from the Ministries oai Finance, Planning, and Agriculture. 1.02 This report is based on the findings of an appraisal mission which visited Rwanda in May 1978. The mission was composed of D. Lallement, H. Marticou, D. Coyaud (IDA-Washingtorn), and J. LeHasif, R. Lemarchand (con- sultants). H. Sanger (RMEA) and R. RaLndriamandranto (IDA-Washington) also par- ticipated in the appraisal. D. Lallement held post-appraisal discussions with Government in Kigali in December 1978, with the participation of C. Blanchi and H. Sanger (RMEA). 1.03 The principal modifications to the preparation report agreed upon during the appraisal were: (a) the reduction of the Project period from five to four years to allow for the experimental nature of the Project; (b) the extension of the Project area by three new administrative sectors, in order to provide extension services to all farmers of a same Commune; (c) the attribution of the implementation responsibility to a Project Unit within the Ministry of Agriculture as the Office pour la Valorisation Pastorale et Agricole du Mutara (OVAPAM), a parastatal company established for the imple- mentation of the Phase I project, could not be self-financing as required by the law on parastatal companies; (d) a substantial reduction in the scope of the Project to put more emphasis on the development of extension services rather than additional infrastructures; and (e) a substantial reduction in Project cost. During the post-appraisal discussions, the Government informed IDA of their decision to maintain OVAPAM even though it could not be self- financing; they proposed to amend the Arrete Presidentiel establishing OVAPAM accordingly. 1.04 To date, the Bank Group has financed three agricultural development projects: the Mutara Agricultural Development Project - Phase I (Credit  439-RW) which received an IDA Credit of US$3.8 million; The Cinchona Project (Credit 656-RW) which received an IDA Credit of US$1.8 million; and the Bugesera/Gisaka/Migongo Project (Credit 668-RW) for an IDA Credit of US$14.0 million. The Mutara-Phase I Project is reviewed in Chapter II(D). Implementation of the two other projects is proceeding satisfactorily. B. The Agricultural Sector 1.05 Rwanda is one of the smallest countries in Africa (26,340 km ) but one of the most densely populated, and one of the poorest. It is completely landlocked, being bordered by Uganda in the North, Tanzania in the East, Burundi in the South, and Zaire in the West. Its topography is characterized by rolling hills with intensively cultivated slopes, and marshy valley floors. Its equatorial climate is moderated by altitude; the temperature only varies between 15 and 250 C. Rwanda today faces a critical problem of land degrada- tion despite the natural fertility of most of its soils. Excessive clearance of forest has led to large scale erosion and overutilization is rapidly exhausting the land. This trend is accelerated daily by the growing population's demands. Unsuitable marginal lands are cropped and erosion increases while soil fertil- ity declines. The shortage of land serves to aggravate this situation to the point where the rate of degradation approaches an exponential rate. Other natural resources are limited. Rwanda's population was estimated at 4.2 million in 1976, and is growing at2an estimated rate of 2.8% annualiy. There were on average 160 inhabitants/km in 1976, and 350 inhabitants/km of arable land, pastu5e land included. Population densities vary wideiy, between 372 inhabitants/km in the most fertile areas to 109 inhabitants/km in the savannah areas which are more hostile to settlement because of tse-tse in- festation. Rwanda's GNP at 1976 market prices was estimated at US$480 million in 1976, or US$110 per capita. 1976 estimates show the absolute poverty level for the rural population to be US$70 per capita per annum and US$130 per capita per annum for the urban population. About 60% of the rural population are below these levels. 1.06 Agriculture is Rwanda's most important single resource. In 1976 it contributed about 80% of GNP, accounted for more than 99% of its expo5t earnings, and supported over 95% of its people. About 70% (17,760 km ) of Rwanda's land area is theoretically suitable for agriculture, the remainder being covered by lakes, ri ers and forests. Discounting degraded or rocky areas, only 54% (13,809 km ) of the total area is available for agriculture, of2which 58% (8,000 km ) is suitable for cultivation; the remaining 42% (5,800 km ) is natural pasture. In 1976, the area under cultivation was given over 95% to subsistence crops and 5% to commercial crops; it includes about 1,000 km of marginal lands. Total foodcrop production falls short of the country's basic requirements. To make up its chronic deficit, Rwanda spends about 30% of its agricultural exports revenues on foodcrop imports.  -3- Crops 1.07 Foodcrops. Bananas, sorghun, beans, peas, sweet potatoes, maize, groundnuts and manioc are the principal foodcrops. Banana production is in- creasing because beer is an important source of cash income, and plays an important role in social activities. Diversification efforts are now underway to promote some planting of alternative crops such as wheat, rice, sugar ca-ne, and soya. Wheat production is increasing. Rice production reached about 2,600 tons in 1976, thanks to several marshland reclamation projects. There is little reliable data on food crop ields, but what is available indicates considerable scope for improvement. Generally, yields achieved at research stations are double or triple the national average, and yields at the paysannats (para 1.10) have also been most encouraging. 1.08 Commercial Crops. Coffee, l:ea, pyrethrum, and cinchona are Rwanda's main commercial crops. Coffee is by far Rwanda's most important export crop, accounting for well over 50% of all exports and generating about 20% of all Government revenues. In 1976, coffee production reached 27,000 tons. Almost all coffee is of the Arabica type, and is exported. Tea production has ex- panded rapidly in Rwanda and reached 4,976 tons in 1976. Pyrethrum produc- tion declined in 1976 with only 3,032 ha under cultivation compared to 4,210 ha in 1975; farmers were apparently diLscouraged by delays in payments for the crops. The production of dried flowers reached 1,753 tons in the latter year. Farming and Production Systems 1.09 Most of the land in Rwanda is traditionally farmed on a family basis. Before the demographic explosion, production was guaranteed by clear- ing new plots of land each year. These plots would be cultivated for a few years under multiple intercropping then abandoned when fertility dropped below acceptable levels. Multiple intercropping presented several advantages: the farmer was protected against climatic hazards as all plant behaviors are diver- sified; the soil was protected against too much sunshine which dries it up and raises its temperature, which in turn accelerates humus decomposition. Moreover the soil mineral resources were optimally utilized by the various crops under intercropping. Natural feallow permitted a satisfactory regenera- tion of the soils. Certain forms of vegetation use the deepest soil strata unused by cultivation and bring mineraLls back to the surface thereby regener- ating the soils. Livestock was a totaLlly separate activity from cultivation. This system is still largely practiced although it has been disrupted by recent demographic growth, the introduction of commercial crops, and the fact that crop cultivation is competing for land with livestock. 1.10 Since 1953, the Government has organized a number of settlement schemes called Paysannats, in the less densely populated areas. The objec- tives of such schemes are to remedy the general shortage of land, and to increase productivity by intensifying the delivery of technical and social services (access roads, schools, dispensaries). Farmers enter into a con- tract with the State represented by the Commune. Farmers have the usufruct of a 2 ha cultivation plot, receive more intensive extension services, but must conform to set agricultural practices. In 1976, an estimated 62,000 families had been settled in this manner. The Paysannat schemes appear to  have achieved increases in yields, production levels, and gross incomes for participating farmers. They have also been active vehicles for the promotion of new cash crops. Livestock 1.11 Animal husbandry is an important activity in Rwanda. According to the 1976 census Rwanda had 630,000 cattle, 680,000 goats and 250,000 sheep. Livestock development is beginning to play a new role in Rwanda, moving from extensive to intensive patterns. 1.12 Cattle. The herd has been declining since 1971 (1.5% p.a.), and more rapidly so since 1975 (7.2% p.a.). This is probably due to a combination of three factors: (a) substantial losses of herds to neighboring countries, largely due to the reduction in available pastures; (b) the high mortality of young animals resulting from the poor health situation; and (c) overexploita- tion of herds, in consequence of the rising demand for meat. Cattle raising in Rwanda is still organized along very traditional lines without distinction between meat and milk herds. Most cattle holdings belong to farmers, and comprise more than a dozen head. A few large herds (200-300 head) are left in the pastoral zones, mostly in the hands of Hima and Tutsi herders. There are no commercial ranches in Rwanda. Some herds on the ISAR (Karama and Rubona) and AIDR (Rusumo) research stations produce top-quality animals. Two small fattening ranches, one in Bugesera Est and one in the Gisaka/Migongo region, will be developed as part of the Bugesera/Gisaka/Migongo project, while a military ranch is also to be developed in the north of Mutara. 1.13 Rwandese cattle are in general the result of the uncontrolled crossing of two distinct strains: a small, short-horned Zebu and the red Tutsi (Ankole) animals which are large, with long straight horns. Parasitism and underfeeding make for poor breeding performance and low yields of meat and milk, but the animals are hardy and respond well to remedial action. The health situation is generally poor, and worst where there are concentra- tions of tse-tse fly, mostly in the east of the country. East Coast Fever, Bovine Trypanosomiasis and varieties of Helminthiasis do the most damage. Rwanda has a fairly extensive network of dipping tanks and sprays but they are greatly hampered by lack of resources. At the moment dipping is done free of charge. 1.14 Goats and sheep are the other two major types of livestock. The flocks are small, and the animals are generally kept tethered. Pigs are not numerous, about 70,000, almost all concentrated in the south. The Rwandese pig is hardy, undemanding and well adapted to its environment. Poultry. Chicken and eggs are still not popular and poultry-raising is limited but there were 800,000 chickens in 1976 (increasing by 20% per year) which shows some progress.  - 5 - 1.15 Meat Production. Estimates for meat production appear to be little more than guesswork. Livestock Service estimates of 1976 production were: 10,000 tons of beef (carcass) and 6,000 tons of goat meat and mutton collec- tively. However, the same Service issued figures of 38,160 cattle, 25,500 goats and 2,241 sheep for regulated slaughtering in the same year. Estimated exports of 40-50,000 head of cattle and 25-30,000 head of sheep and goats - collectively should also be treated with scepticism, and may include illegal exports. C. Crop and Livestock Marketing and Processing Crops 1.16 Most of the subsistence crops are marketed in traditional markets held periodically, where marketing is almost exclusively in the hands of private traders. Marketable surpluses are sold by farmers on a strict cash basis, generally immediately after the harvest; post harvest prices are usually depressed, rising steeply just before the next harvest. Prices also vary regionally. Commercial crops are marketed through private traders, and public companies such as the National Coffee Board. Government is attempting to develop better marketing channels to lessen seasonal price fluctuations and develop the country's storage capacity both in surplus and deficit regions (para 1.22). Livestock and Meat 1.17 Live cattle are marketed on small rural markets, and on large markets where the herds are made up for export (mainly to Zaire) or to supply the major centers of consumption. Almost all live cattle transactions take place on a cash basis. A "presentation" taxi of RF 30 per head on average and a "transaction" tax of RF 150 to 250 per head are levied by the Communes. Sales of breeding stock are very limited. Sheep and goats are bought and sold at general weekly or monthly markets. Prices vary with markets and are subject to seasonal fluctuations. Prices have risen very sharply in the last four or five years, by 50 to 100% or more according to the type of animal. 1.18 Meat is marketed on both the traditional markets where the greatest volume is in bone-in beef, and goat meat and on the high-grade meat market in Kigali with a turnover of approximately 40-50 animals per month. Official prices are, in theory, established by the Ministry of National Economy but there is no effective enforcement systeem, and butchers can easily cheat on weights. Slaughtering is mostly carried out in the rural areas at numerous stations either totally lacking in equipment or with the most rudimentary facilities; few are subject to health control. There are three abattoirs (Kigali, Butare and Kibungo). The Kigali abattoir came into service about one year ago, but is far from being operational.  - 6 - D. Institutions Responsible for Agricultural and Livestock Development 1.19 Several institutions have responsibility for agricultural and livestock production and development. The major institutions and their respective responsibilities are as follows: 1.20 The Ministry of Agriculture and Livestock is responsible for policy direction and coordination in the agriculture and livestock sector. The Ministry has a Secretariat and several operating departments. 1/ The Secretariat includes the Studies and Projects Division, and three offices (archives, budget management, and agrarian reform). The Agronomy Department is divided into the Extension Department essentially concerned with all exten- sion services and with plant protection, and the Forestry and Water Resources Department. The Rural Engineering (Genie Rural) and Soil Conservation Depart- ment develops marshland reclamation projects, and provides various other ser- vices, mainly topographic. The Livestock Department manages the country's Veterinary Services and is responsible for promoting animal production. The Land Department supervises land registration and allocation. The Ministry has recently taken responsibility for some projects, like the Office du Bugesera et Mayaga (OBM), formerly established as parastatals and which were left with inadequate management and institutional structures upon cessation of foreign assistance. In addition, the Ministry is the umbrella organiza- tion for several parastatal companies such as OCIR, (para 1.21), and OPROVIA, (para 1.22). The Ministry has over 2,000 staff members, most of whom are field extension workers (over 1,500) or veterinary field staff (about 400), operating in all the Prefectures and Communes. In general, field extension services are weak, as the staff is inadequately trained, poorly supervised and the logistical support insufficient. The Ministry's operating and development budgets amounted to RF 287 million and RF 1,291 million respectively in 1977, representing 4.2 and 6.6% of the Government s operatiag and development budgets. 1.21 The Office des Cultures Industrielles du Rwanda (OCIR) is essen- tially responsible for quality control, conditioning and marketing control and promotion, of such exports as coffee, tea, pyrethrum, and cinchona. It is fully responsible for tea production, including the management of tea plantations, of production promotion among small farmers, and of processing plants. OCIR does not market coffee but controls the one authorized exporter, and producer prices. It operates a stabilization fund for coffee prices to lessen the effects of international price fluctuations on the producer. 1.22 The Office National pour le Developpement et la Commercialisation des Produits Vivriers et des Productions Animales (OPROVIA) was established in 1975 as a parastatal agency broadly responsible for the production, processing 1/ Department - Direction Generale.  and marketing of foodcrops and livestock. To date, it has mostly been active in handling and selling food products imported with bilateral or international grants. In the medium-term it should take over several on-going projects such as: (i) GRENARWA, Greniers Agricoles du Rwanda, a USAID financed marketing project, which is currently building silos in several prefectures to develop the country's storage capacity of subsistence crops and lessen seasonal price fluctuations; (ii) a Belgian aid projsect for the extension and rehabilitation of the Kigali slaughterhouse plant; and (iii) fattening ranches in the Bugesera, which are to be developed under the I])A/BADEA/FAC financed project. OPROVIA is poorly staffed, and unless it soon receives proper technical assistance to develop its managerial and marketing capability, it will continue to stagnate. 1.23 The Institut des Sciences Agironomiques du Rwanda (ISAR) is responsi- ble for agricultural research. It is largely financed by Belgian aid, and includes a large Belgian technical assistance staff. It operates several research stations, and has carried out extensive research orn most crops grown in several ecological zones in Rwanda. on tse-tse eradication, pasture improve- ment, and livestock development. 1.24 Agricultural Training is mostly carried out by three educational establishments: The Faculte d'Agronomie of Butare University is being devel- oped with Canadian aid to train agricultural engineers; the first class grad- uated in 1977. The School of Agriculture in Butare, largely supported with French technical assistance, operates a four-year post secondary course and trains agricultural technicians and veterinarians. The school at Kibuye operates a two-year course, and trains lower level assistant agronomists. There is no program to train lower level assistant veterinarians but the training centre which is to be built under the Bugesera/Gisaka/Migongo proj- ect would start such a program. Overseas training programs remain the main source of agricultural engineers and veterinarians. 1.25 Local Governument Institutionus have played a rather limited role in agricultural and rural development in the past. However, since 1976, the Government has established guidelines for them to take a greater share of development responsibilities. The Commune is the smallest administrative unit and groups a number of Collines (hills). It is divided into two or three sectors and is headed by a Bourgmestre, appointed by the Minister of the Interior. The Communal Council (Conseil Communal) which includes represen- tatives of the population, is the main executive body; the Development Council and the Technical Committee are bodies; in charge of executing and proposing development projects. The Communal Action Fund is one of the sources of financing for communal development projects; it is funded by a 10% contribu- tion of the Communes' taxation income. The Umuganda whereby each adult contributes one day of labor a week, is the main instrument for communal development projects. The Prefecture is the second administrative unit, (there are 10 Prefectures in Rwanda), and the largest ones are subdivided into sub-prefectures. The Prefecture is headed by a Prefect, also appointed by the Minister of the Interior. To date, little coordination has taken place between the local and national institutions in programming and implementing development projects.  - 8 - E. Agricultural Credit 1.26 Institutional agricultural credit is not well developed in Rwanda. The Development Bank of Rwanda, established in 1968, extends credit primarily for smallholder tea, sugar, and other industries in the private sector. Other commercial bank credit has been extended mainly to finance the coffee industry. Interest rates by all banks vary between 7 and 10%. Informal credit is still quite important in the rural areas; most commonly, livestock is lent to a farmer so that he can get milk or start his own herd. These loans are usually transacted within a complex structure of social links and obligations. Gov- ernment is now actively considering the promotion of agricultural credit. A comprehensive agricultural credit study is being carried out under the IDA financed Bugesera/Gisaka/Migongo project. The recommendations of the study would be used to promote a nationwide systematic agricultural credit program. F. Government Strategy and Plans for Agricultural and Rural Development 1.27 The Government of Rwanda has defined the broad objectives for agricultural and rural development in the Second Five Year Plan. Four essen- tial goals are identified for the five year period 1977-1981: (i) meeting the population's nutritional requirements; (ii) promoting a better utiliza- tion of human resources; (iii) improving individual and collective living conditions; and (iv) improving Rwanda's standing vis-a-vis her foreign part- ners. The Plan emphasizes the priority to be given to the agricultural sec- tor, in particular to integrated rural development as one of the means of agricultural growth. The Plan recognizes, however, that diversification is necessary both within the sector to lessen reliance on a few crops, and with other sectors to lessen reliance on agriculture for economic growth. 1.28 The specific objectives for agriculture for 1977-1978 are as follows. The agricultural sector as a whole would grow at 5.0% p.a. as compared to 7.1% and 7.2% respectively for the secondary and tertiary sectors. Subsistence crops would increase at 3.8% p.a. as compared to 12.7% for export/industrial crops, 11.9% for animal products, and 37% for fish. Value added for new traditional crops such as oilseeds (groundnuts, soybeans), potatoes, wheat, rice and vegetables would grow at higher average rates than other subsistence crops. Likewise for livestock products: beef and mutton would only grow at 3.6% p.a. as compared to 22.6% p.a. for pork, poultry, eggs, milk and fish. Among export crops, value added for relatively new crops such as cinchona, cotton and sisal, medicinal or essence plants, tobacco, etc., would grow at higher rates than traditional crops such as coffee, tea and pyrethrum. Rural income would increase by 4.2% p.a. For the 1977-81 period, the share of agri- culture in the national development budget would average 8.5%, of which 89% would be financed from external sources, both public (65%) and private (35%).  - 9 - 1.29 The Plan recognizes that these targets will be hard to achieve. The only chances of success would come from the combined efforts of the population. The Communes should become the major catalyst for development, as they provide a better framework for the people to work on improving their environment. The Umuganda (para 1.25), should be actives in developing social services in local communities. The development of cooperatives would also be emphasized, mainly to encourage the production and markelting of agricultural inputs and produce. The Plan also emphasizes the need to develop better technical services for agriculture. The Communes would assume part of the training responsibilities by developing centers where training would be given to selected farmers who would then become promoters of techno:Logical improvements in their colline. 1.30 Because of the severe constraints on land, livestock and crop development would focus on intensive production. Erosion control should receive priority in order to preserve the country's production potential. Because of the scarcity of foreign exchange, organic fertilizer and improved seed varieties are two major inputs that could be used to maintain soil fer- tility and increase productivity. ThiLs would be achieved by integrating livestock to agriculture, and by carrying out further research on higher yielding varieties. Livestock productivity itself would be increased by preserving existing pastures, expanding extension and veterinary services, developing fodder production, and upgrading the genetic value of the herd. II. THE PROJECT AREA A. Location and Physical Resources 2.01 Location. The Mutara region, located in the northeast of Rwanda, is bordered by Uganda on the West, and Tanzania on the East (Map 13865). The altitude averages 1400 to 1500 meters, and the rainfall varies between 700 and 1200 mm, making the region one of the driest in Rwanda. The temperature ranges from 12 to 33 C . The region is not an administrative unit and is not stricly defined (para 2.07). Roughly it covers the three Communes of Muvumba, Ngarama, and Gitaza in the Byumba Prefecture. Before Independence, the Eastern half of the Mutara was designated as a National Park covering 185,000 ha (55% of the Mutara); in addition, some 40,000 ha in the south have been earmarked as a hunting reserve. As a result, only 70,000-80,000 ha are left for development. Since 1974, this remaining area has been the object of two major Government- sponsored development projects: the IDA-financed Agricultural Development Project, covering approximately 51,000 ha, and the CIDA-financed Marshland Development Project, covering approximately 9,000 ha. Recently 5,000 ha in the North have been set aside to be developed as a military ranch (Map 13866).  - 10 - 2.02 The proposed Agricultural and Livestock Development Project-Phase II would cover the same area as the Phase I project, increased in the south by three administrative sectors, Ngarama and Kigasha in the Ngarama Commune, and Karama in the Muvumba Commune. The Project area would cover about 60,000 ha total, including the whole Commune of Muvumba, and most of the Commune of Ngarama, except for the mountainous zone covering the sectors of Gatsibo, Gatinda, and Manishya; no topographical survey is available at present. The Project area is bordered by Uganda on the West, the National Park on the East, and the Hunting Reserve in the South east. The Kigali-Kagitumba road, a major access road to Uganda, is the actual border between the Project and the Park. The marshlands of the Kagitumba and Muvumba rivers on the perimeter are excluded from the project as they are part of the CIDA-financed Marshland Development Project (see Map 13866). 2.03 The Project area can be divided into two main zones: (a) the roll- ing savannahs north of the Muvumba river, and the steep cultivated hills to the south. Because of its pastoral character, the northern area was selected by the Phase I project to develop group ranches while the southern area, which is more humid-and has better water resources, was selected to develop a paysannat (paras 1.10 and 2.17). There are no urban centers in the Project area. The closest town is Byumba, the center of the Prefecture south of the Project area, but a number of small markets and trading centers are appearing. Access to the Project area is via the Kigali-Kagitumba road, passable through- out the year, and the Kigali-Byumba-Ngarama road, hardly passable during the rainy season. B. Government Services Technical Services 2.04 The Communes, Prefectures and Sub-Prefectures are supposed to pro- vide some technical services. They consist at present of road maintenance (mostly done by hand), and agricultural and veterinary services (reaffor- estation, distribution of improved seeds, development of the Prefectural or Communal demonstration plots, the dipping and vaccination programs). In fact, with the Phase I project, OVAPAM (para 2.17) has provided most agricul- tural services in the Muvumba and Ngarama Communes while the communal agro- nomists and veterinarians focussed on the communal sectors outside the project area. Social Services 2.05 They are limited to: (a) two health centers, one at Rukumo and one at Nyagahita, both in the Paysannat; neither functions properly for lack of medicine, although each has a nurse/social worker; and (b) a few schools, in various degrees of maintenance: two in the Paysannat, of which only one has a complete cycle (Rukumo), and two in the Group Ranches, at Nyagatare and Rwempasha. Buildings and equipment are rudimentary.  -. il - 2.06 Some Development Centers are now being built on the initiative of the Byumba Prefect to provide the population with training sessions on such subjects as health care, nutrition, and agricultural techniques. The objec- tive is to have one per Commune. All the professional workers from the Commune would contribute to the courses and seminars. C. Sociological Aspects 2.07 Historical Background. Once part of the old Ndorwa-Ankole Kingdom, the Mutara became incorporated into the expanding nuclear kingdom of Rwanda in the late 18th century. It remained an army fief held by Tutsi warriors until the inception of colonial rule at the turn of the century. Even during the period of German and Belgian colonial rule, it was always regarded as a marginal tse-tse infested area, mostly used by Tutsi and Hima pastoralists to graze their cattle. Following Independence and the Civil War (1959-1962), the Mutara became a refuge for Tutsi elements; later Hutu cattle herders and farmers started settling in the south, because of population pressure and land scarcity in other parts of the country. 2.08 Population. No reliable demographic data exist for the Project area. From the communal archives, sample surveys and other data collected by the Phase I project, the population of the Project area in 1977 can be estimated at 58,000 people, roughly distributed as follows: Group ranches 10%, Paysannat 65%, and non-Paysannat sectors 25%. Trends are also extremely difficult to assess; several migratory movements have taken place in recent years. With the Phase I project, pastoral families first moved out of the Mutara (para 2.18), but some are now coming back. Farming-families immi- grated much faster than Government had anticipated. 2.09 The Mutara is ethnically more diverse than other parts of Rwanda. Traditionally inhabited by Hima pastoralists, it was later occupied by Tutsi, Hutus and Twas. In addition, since 1972 some 900 Barundi families have been been installed in the project area (Muvumba Commune). 2.10 According to a survey done in the Paysannat, the average family size is 4.6 persons, much below the national average of 5.6 - 6 persons, reflecting the fact that most immigrant families are young. However, this sample survey only included children of the family head and one wife. Polygamy is reputedly quite frequent but details are limited. The age structure is also quite char- acteristic of a newly installed population. Children (1-18) represent 55% of the total population, and the age composition is as follows: 76% from 1 to 12, 14% from 12 to 15, and 10% from 15 to 18 years of age. 2.11 The number of households in the area may be much higher than reported. Many Paysannat beneficiaries admit to using hired labor. Most typically this is provided by a family member who comes to live on the plot, at first alone, but subsequently with all his own family. Local authorities deny this practice, but project staff confirmed that it does, in fact, prevail.  - 12 - 2.12 Land Tenure. The allocation of land set aside for development under the Phase I project is virtually completed. The Paysannat and Group Ranch systems established give relative security of tenure. Beneficiaries only have usufruct rights on the land, not ownership. Hereditary rights exist and are well-defined in the case of the Paysannat where plots are inherited in whole by a single child, usually the eldest son. In the Group Ranches, hereditary rights are not yet defined clearly, as the experience is new to the country. Provided the whole herd remains constant per ranch, the subdivision of herds between children will be accepted, and has been de facto accepted during installation. Theoretically, only one family can settle on a plot in the Paysannat. In practice, Government will have to reconsider this in view of the extended family system and pressure on land. In older Paysannats, the authorities allow one more family per plot, provided all Rugos (houses and farm buildings) are grouped in order to leave maximum land for cultivation. The land tenure system for the Marshland Development Project has not yet been fully devised. 2.13 Land distribution within the Paysannat and Group Ranches is fairly equitable but inequalities do exist between the Paysannat, Group Ranches, and farms not integrated into the Phase I project. In the Paysannat, all the cultivation plots are 2 ha, except those of the Barundi refugees which are only 1.5 ha. In comparison, farmers who have not yet benefitted from such projects in Rwanda have an average of 1.3 ha of land. Not all plots have the same production potential, and obviously not all can be at the same distance from existing services. In the Group Ranches all families have 2 ha of grazing land per animal unit and a 1.5 ha cultivation plot per family. However, some families have more cattle than others; cattle ownership ranges from a minimum of 5 head to 100 head or more. Some ranches have a much better pasture endowment than others. 2.14 Settlement. Traditionally, farmers and pastoralists of the Mutara lived in extremely scattered settlements. This has radically changed with the Phase I project as farmers have been regrouped along a road in the Paysannat, and even more in the Group Ranches where they have been regrouped into villages. These changes appear to have been reasonably well accepted by participants, although a further regroupment of farmers in the Paysannat, proposed during 1975, had to be abandoned because of farmers' reluctance to live in villages. This may change, however, when some social services, namely schools and health centers, become available. Spontaneous regroup- ment is already noticeable around market places. 2.15 Production Systems. One must make a clear distinction between farming and pastoral activities. (i) Farming System. Cultivation of the plot is mostly done with family labor. Due to the current age structure of the population, labor is relatively scarce and may be one of the constraints on faster and more intensive resource development of the plot. Men are usually responsible for building the house, clearing land, and maintaining the banana plantation. W